text
stringlengths 87
880k
| pmid
stringlengths 1
8
| accession_id
stringlengths 9
10
| license
stringclasses 2
values | last_updated
stringlengths 19
19
| retracted
stringclasses 2
values | citation
stringlengths 22
94
| decoded_as
stringclasses 2
values | journal
stringlengths 3
48
| year
int32 1.95k
2.02k
| doi
stringlengths 3
61
| oa_subset
stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00343-4
10.1016/j.jiph.2022.12.004
Original Article
The prevalence of adverse reactions among individuals with three-dose COVID-19 vaccination
Wang Yu-Ying a1
Zhang Yu-Jie a1
Zhang Meng ab1
Zhang Xiao-Yu c
Li Hai-Bin d
Wang You-Xin a
Wang Wei e
Ji Jian-Guang f⁎
Wu Li-Juan a⁎
Zheng De-Qiang af⁎⁎
a Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing 100069, China
b Department of Epidemiology and Bio-statistics, School of Public Health, Peking University, Beijing 100191, China
c Department of Anesthesiology, Sanbo Brain Hospital, Capital Medical University, Beijing 100093, China
d Heart Center & Beijing Key Laboratory of Hypertension, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
e Centre for Precision Health, Edith Cowan University, Perth, WA 6027, Australia
f Center for Primary Health Care Research, Lund University/Region Skåne, 20 502 Malmö, Sweden
⁎ Corresponding authors.
⁎⁎ Corresponding author at: Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing 100069, China
1 These authors contributed equally to this work.
6 12 2022
6 12 2022
14 9 2022
17 11 2022
5 12 2022
© 2022 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
Considering the adverse reactions to vaccination against coronavirus disease 2019 (COVID-19), some people, particularly the elderly and those with underlying medical conditions, are hesitant to be vaccinated. This study aimed to explore the prevalence of adverse reactions and provide direct evidence of vaccine safety, mainly for the elderly and people with underlying medical conditions, to receive COVID-19 vaccination.
Methods
From 1st March to 30th April 2022, we conducted an online survey of people who had completed three doses of COVID-19 vaccination by convenience sampling. Adverse reaction rates and 95% confidence intervals were calculated. In addition, conditional logistic regression was used to compare the differences in adverse reactions among the elderly and those with underlying medical conditions with the general population.
Results
A total of 3339 individuals were included in this study, of which 2335 (69.9%) were female, with an average age of 32.1 ± 11.4 years. The prevalence of adverse reactions after the first dose of inactivated vaccine was 24.6% (23.1%–26.2%), 19.2% (17.8%–20.7%) for the second dose, and 19.1% (17.7%–20.6%) for the booster dose; among individuals using messenger RNA vaccines, the prevalence was 42.7% (32.3%–53.6%) for the first dose, 47.2% (36.5%–58.1%) for the second dose, and 46.1% (35.4%–57.0%) for the booster dose. Compared with the general population, the prevalence of adverse events did not differ in individuals with underlying medical conditions and those aged 60 and above.
Conclusions
For individuals with underlying medical conditions and those aged 60 and above, the prevalence of adverse reactions is similar to that of the general population, which provides a scientific basis regarding vaccination safety for these populations.
Abbreviations
COVID-19, coronavirus disease 2019
SARS-CoV-2, severe acute respiratory syndrome coronavirus 2
mRNA, messenger RNA
CI, confidence interval
RBD-subunit vaccine, receptor-binding domain subunit vaccine
Keywords
COVID-19 vaccine
SARS-CoV-2
Special populations
Booster dose
Adverse reactions
==== Body
pmc1 Introduction
The coronavirus disease 2019 (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and the presenting symptoms of most infected people typically include fever, cough, dyspnea, myalgia or fatigue [1]. More than 603 million people worldwide were infected with SARS-CoV-2 until September 2022 [2], [3], while the infected cases in China were more than 6 million [4]. Currently, there is no effective treatment for COVID-19, and vaccination is still one of the most efficient control programs [5], [6]. Safe vaccines are available to provide effective protection against serious disease and death [7]. Several vaccines that differ in form and effectiveness have been approved for vaccination. In general, these vaccines are based on inactivation technology, such as Sinopharm (BBIBP-CorV) and Sinovac (CoronaVac), or new messenger RNA (mRNA) technology, such as Pfizer-BioNTech (BNT162b2 [equivalent to Fosun-BioNTech]) and Moderna (Spikevax mRNA-1273). Others, such as Anhui Zhifei Longcom (ZF2001), are RBD-subunit vaccines [8], [9], [10], [11], [12]. In China, the two inactivated vaccines described above are generally accepted [13].
Previous research suggests that older adults aged 60 and above are at high risk of COVID-19 [14], [15]. Additionally, individuals with underlying medical conditions, such as hypertension and coronary heart disease, are at an increased risk of infection with COVID-19 and mortality [14], [16]. Vaccination is the most effective way to prevent the development of severe disease and death due to COVID-19 [6], [17]. Several studies suggest that the elderly and people with underlying medical conditions should be vaccinated as soon as possible, but some are still hesitant to vaccinate [18], [19]. The Strategic Advisory Group of Experts on Immunization of WHO defines vaccine hesitancy as “a delay in acceptance or refusal of vaccines despite the availability of vaccination services” [20], [21]. In China, the elderly and people with underlying medical conditions are more prone to vaccine hesitancy, and their vaccination rate is lower than that of the general population. In Shanghai, the overall vaccination rate is now over 90%, but only 62% of people aged over 60 years have received the vaccination, and only 38% have received a booster dose [19]. The main reason for vaccine hesitancy may be concerns about vaccine safety, i.e., the risk of adverse reactions [22], [23].
Thus, it is important to have concrete evidence regarding vaccine safety in different populations, which could serve as the basis for reducing vaccine hesitancy [19], [23]. V-safe, a voluntary, smartphone-based safety surveillance system, was established in 2020 to monitor mRNA vaccine safety in the USA and provide information on adverse reactions after vaccination [24], [25]. However, most vaccines used in China are inactivated, thus, the V-safe report cannot be generalized to Chinese citizens [13]. In addition, few studies have focused on the adverse reactions to inactivated vaccines in China, particularly for individuals who received three doses of the vaccine.
Therefore, we conducted an online survey of adults who had completed three doses of COVID-19 vaccination using convenience sampling. Three types of vaccines were involved in our survey, including inactivated vaccines (Sinopharm and Sinovac), mRNA vaccines (Pfizer-BioNTech, Fosun-BioNTech and Moderna), and RBD-subunit vaccines (Anhui Zhifei Longcom). This study aimed to explore (1) the overall prevalence of injection site and systemic adverse reactions after vaccination, (2) whether the prevalence of adverse reactions varies by vaccine type or dose, and (3) whether the prevalence is higher among the elderly or people with underlying medical conditions compared with the general population.
2 Methods
2.1 Participants
This is a cross-sectional study. We conducted an online questionnaire survey on a convenience sample of adults who completed three doses of the COVID-19 vaccination. All respondents were recruited online. This study conformed to the requirements of medical ethics which was approved by the ethics committee of the study institution, and the questionnaire did not involve personal information, such as name, which conforms to the requirements of relevant laws and regulations. Participants provided online informed consent. Each IP address can only answer once to avoid duplicate answers. In total, 3467 respondents filled out and submitted the online questionnaire. Among these, 128 invalid questionnaires were excluded. Finally, 3339 participants were included in the analysis, with an effective response rate of 96.31%.
2.2 Measures
Data were collected using a self-made questionnaire comprising four parts [25] (Additional file 1). The first part included sociodemographic information, including sex, age, height, weight, education, underlying medical conditions, BMI (calculated from self-reported weight and height, BMI [in kg/m2] = weight [in kg]/ height2 [in m2]), and the frequency of dizziness and headache within 1 year before injection.
The second part was the type of vaccine received for each of the three doses: Sinopharm, Sinovac, Pfizer-BioNTech, Fosun-BioNTech, Moderna, and Anhui Zhifei Longcom.
The third part included adverse reactions after each dose of the vaccine, including injection site adverse reactions (itching, pain, swelling, induration, and redness), systemic adverse reactions (headache, dizziness, fever, fatigue, myalgia, joint pain, cough, nausea and vomiting, diarrhea, palpitations, insomnia, drowsiness, anaphylaxis, and others), and health impacts (needed medical care, unable to perform normal daily activities, and unable to work or attend school). Anaphylaxis includes angioneurotic edema, Henoch-Schonlein purpura, and anaphylactic shock.
The fourth part was the time of occurrence and duration of each adverse reaction. The occurrence times were as follows: <30 min, 30 min to <1 day, 1–2 days, 3–4 days, 5–7 days, and >7 days. The duration included <1 day, 1–2 days, 3–5 days, >5 days, and non-remission until the survey.
2.3 Outcome
The outcome of this study was adverse reactions, including injection site and systemic adverse reactions. We then combined the injection site and systemic adverse reactions into one new group (any adverse reactions) as another outcome.
2.4 Statistical analysis
The prevalence of adverse reactions was defined as the ratio of individuals with one or more adverse reactions to the total number of participants. Categorical variables were presented as counts (%) and continuous variables as means ± standard deviation (mean ± SD). The χ2 or Fisher’s exact test was used to compare the proportions in different subgroups.
Based on the number of responses, the occurrence time was consolidated into <1 day, 1–4 days, and >4 days, and the duration was consolidated into <1 day, 1–2 days, 3–5 days, and >5 days.
To explore whether the adverse reaction was associated with age, we selected participants aged 60 and above as the exposure group, matched with those younger than 60 years based on sex, education, and experience of any adverse reaction. To study the association between underlying medical conditions and adverse reactions, we selected individuals with medical conditions as the exposure group and matched those without medical conditions based on age (± 5 years), sex, education, and experience of adverse reactions. Additionally, we explored whether the subsequent adverse reactions (such as reactions after the second or booster dose) could be affected by the previous reactions, either in the first or second doses. The study participants were matched based on age (± 5 years), sex, education, underlying medical conditions, and dizziness or headache. Conditional logistic regression analysis was used to explore the association between these characteristics and adverse reactions.
The analyses were performed using R (version 4.1.3) and SPSS (version 25.0). Graphing was performed using GraphPad Prism (version 8.4.2). A two-tailed P-value <0.05 was considered statistically significant.
3 Results
A total of 3339 valid questionnaires were collected, including 1004 males (30.07%) and 2335 females (69.93%) ( Table 1). The mean age was 32.1 ± 11.4 years. As shown in Table 1, 1278 participants had one or more adverse reactions, and the prevalence (95% confidence interval [CI]) was 38.27% (36.62%–39.95%). The prevalence of adverse reactions was different between both sexes (P < 0.001) and was generally higher in females than in males. The highest prevalence was noted among individuals aged 40–49 years, which was 46.73% (42.38%–51.12%). We also found a significant difference among individuals with different educational levels (P < 0.001); the prevalence was higher among people with postgraduate education, which was 45.10% (42.38%–47.84%). There was no significant association between individuals with or without underlying medical conditions (P = 0.584), and the prevalence was 39.57% (34.58%–44.73%) and 38.11% (36.36%–39.89%), respectively. The comparison of the prevalence of underlying medical conditions among three different types of vaccines in age wise criteria was shown in Table S1 (Additional file 2: Table S1).Table 1 Characteristics and prevalence of adverse reactions (N = 3339).
Table 1Demographics Total (n=3339) Male (n=1004) Female (n=2335)
Total, n Adverse reactions, n (%) Pvalue Total, n Adverse reactions, n (%) Pvalue Total, n Adverse reactions, n (%) Pvalue
Age group (yrs) <0.001 0.261 <0.001
18–29 1782 632 (35.47%) 510 145 (28.43%) 1272 487 (38.29%)
30–39 693 276 (39.83%) 217 63 (29.03%) 476 213 (44.75%)
40–49 520 243 (46.73%) 156 53 (33.97%) 364 190 (52.20%)
50–59 288 108 (37.50%) 96 21 (21.88%) 192 87 (45.31%)
≥60 56 19 (33.93%) 25 5 (20.00%) 31 14 (45.16%)
BMI 0.327 0.029 0.229
<18.5 322 110 (34.16%) 53 9 (16.98%) 269 101 (37.55%)
18.5–23.9 1976 777 (39.32%) 475 127 (26.74%) 1501 650 (43.30%)
24–27.9 714 268 (37.54%) 316 93 (29.43%) 398 175 (43.97%)
≥28 327 123 (37.61%) 160 58 (36.25%) 167 65 (38.92%)
Education <0.001 <0.001 <0.001
Junior and below 91 10 (10.99%) 57 4 (7.02%) 34 6 (17.65%)
Senior 137 32 (23.36%) 53 9 (16.98%) 84 23 (27.38%)
College 1805 647 (35.84%) 494 121 (24.49%) 1311 526 (40.12%)
Postgraduate 1306 589 (45.10%) 400 153 (38.25%) 906 436 (48.12%)
Underlying medical conditionsa 0.584 0.424 0.027
CVD and CVA 152 67 (44.08%) 76 25 (32.89%) 76 42 (55.26%)
Dyslipoproteinemia 71 27 (38.03%) 33 9 (27.27%) 38 18 (47.37%)
Endocrine disease 136 45 (33.09%) 43 7 (16.28%) 92 38 (41.30%)
Autoimmune disease 22 7 (31.82%) 11 3 (27.27%) 11 4 (36.36%)
Others 101 48 (47.52%) 43 13 (30.23%) 58 35 (60.34%)
Adverse reaction: Any adverse reaction was defined as the occurrence of any injection site adverse reaction or systemic adverse reaction.
CVD and CVA: hypertension, stroke, heart disease (angina pectoris, heart failure, congenital cardiovascular disease, myocarditis, pulmonary hypertension).
Endocrine disease: diabetes mellitus, thyroid disorder (hyperthyroidism, hypothyroidism).
Autoimmune disease: HIV, rheumatoid arthritis, systemic lupus erythematosus, vasculitis, ankylosing spondylitis.
Others: chronic lung disease (chronic obstructive pulmonary disease, emphysema, chronic bronchitis, idiopathic pulmonary fibrosis, pulmonary tuberculosis, bronchial asthma), liver disease (hepatitis, liver cirrhosis), hyperuricemia, uterine myoma, epilepsy, etc.
Bold indicates statistically significant <0.05.
a The P value of the underlying medical conditions was used to indicate whether there was a statistically significant difference between groups (general population, male, female) with or without underlying medical conditions
Table 2 shows the prevalence of adverse reactions in participants who received three doses of inactivated, mRNA, or RBD-subunit vaccines. We found that vaccine types were significantly associated with any adverse reactions in each dose (all P < 0.001). The prevalence (95% CI) of any adverse reaction in different types were as follows: first doses 24.58% (23.06%–26.15%), 42.70% (32.26%–53.63%), and 19.70% (10.93%–31.32%) in the inactivated, mRNA, and RBD-subunit vaccines, respectively; second doses 19.22% (17.83%–20.66%), 47.19% (36.51%–58.06%), and 18.18% (9.76%–29.61%), respectively; booster doses 19.12% (17.73%–20.56%), 46.07% (35.44%–56.96%), and 18.18% (9.76%–29.61%), respectively. The pain was the most common injection site adverse reaction for both inactivated and mRNA vaccines, whereas itching was most frequently reported for RBD-subunit vaccines. The most common systemic adverse reactions for those who received inactivated vaccines were fatigue, myalgia, and drowsiness, whereas fatigue, myalgia, and fever were more common in those who received mRNA vaccines. Only two participants who received RBD-subunit vaccines had systemic adverse reactions (2/66). Among all the participants, the prevalence (95% CI) of any adverse reaction, any injection site reaction, and any systemic reaction that appeared in all three doses were 8.39% (7.43%–9.43%), 4.87% (4.13%–5.70%) and 2.07% (1.60%–2.64%) for inactivated vaccine, respectively; being 31.46% (22.03%–42.17%), 14.61% (8.01%–23.68%) and 11.24% (5.52%–19.69%), respectively, for mRNA vaccine; being 9.09% (3.41%–18.74%), 4.55% (0.95%–12.71%) and 1.52% (0.04%–8.16%), respectively, for RBD-subunit vaccine.Table 2 Prevalence of adverse reactions to different types of vaccines in three doses.
Table 2Adverse reaction Dose 1 Dose 2 Dose 3
INACT VAC (n=3039) mRNA VAC (n=89) RBD-SU VAC (n=66) pvalue INACT VAC (n=3039) mRNA VAC (n=89) RBD-SU VAC (n=66) pvalue INACT VAC (n=3039) mRNA VAC (n=89) RBD-SU VAC (n=66) pvalue
Any adverse reaction (%) 24.58 42.70 19.70 <0.001 19.22 47.19 18.18 <0.001 19.12 46.07 18.18 <0.001
Any injection site reaction (%) 14.41 32.58 13.64 <0.001 12.47 25.84 13.64 <0.001 13.33 20.22 13.64 0.172
Itching (%) 1.55 3.37 6.06 1.25 1.12 9.09 1.61 3.37 10.61
Pain (%) 12.60 30.34 7.58 10.66 23.60 7.58 11.62 16.85 4.55
Swelling (%) 3.78 7.87 6.06 2.44 4.49 7.58 2.50 5.62 7.58
Induration (%) 2.11 3.37 3.03 1.74 2.25 6.06 2.20 2.25 10.61
Redness (%) 0.46 1.12 1.52 0.30 1.12 7.58 0.49 2.25 10.61
Any systemic reaction (%) 11.48 17.98 3.03 0.016 7.04 31.46 3.03 <0.001 6.42 39.33 1.52 <0.001
Headache (%) 2.11 8.99 0.00 0.92 13.48 1.52 0.86 17.98 0.00
Dizziness (%) 2.67 6.74 0.00 1.45 11.24 1.52 1.22 11.24 0.00
Fever (%) 1.45 11.24 0.00 0.79 25.84 0.00 0.86 29.21 0.00
Fatigue (%) 7.01 11.24 1.52 3.49 21.35 3.03 3.09 24.72 1.52
Myalgia (%) 3.55 8.99 0.00 1.78 21.35 1.52 2.14 15.73 0.00
Joint pain (%) 0.36 1.12 0.00 0.16 5.62 1.52 0.26 3.37 0.00
Cough (%) 0.23 1.12 0.00 0.30 0.00 0.00 0.26 0.00 0.00
Nausea and Vomiting (%) 0.49 0.00 0.00 0.39 0.00 0.00 0.46 1.12 0.00
Diarrhea (%) 0.43 0.00 0.00 0.10 0.00 0.00 0.20 0.00 0.00
Palpitation (%) 0.16 2.25 0.00 0.26 1.12 1.52 0.13 1.12 0.00
Insomnia (%) 0.13 0.00 0.00 0.10 1.12 0.00 0.20 2.25 0.00
Drowsiness (%) 4.48 4.49 3.03 2.44 6.74 1.52 1.71 6.74 1.52
Anaphylaxis (%) 0.33 1.12 0.00 0.16 1.12 0.00 0.36 0.00 0.00
Mild allergy (%) 0.13 0.00 0.00 0.13 0.00 0.00 0.13 0.00 0.00
Menoxenia (%) 0.23 0.00 0.00 0.23 0.00 0.00 0.13 0.00 0.00
Hearing loss (%) 0.03 0.00 0.00 0.07 0.00 0.00 0.03 0.00 0.00
Tinnitus (%) 0.00 0.00 0.00 0.00 0.00 0.00 0.07 0.00 0.00
Any health impact (%) 1.02 4.49 0.00 0.027 1.25 14.61 1.52 <0.001 1.41 14.61 0.00 <0.001
Needed medical care (%) 0.53 2.25 0.00 0.95 5.62 0.00 0.95 4.49 0.00
Unable to perform normal daily activities (%) 0.33 1.12 0.00 0.20 5.62 1.52 0.39 6.74 0.00
Unable to work or attend school (%) 0.33 1.12 0.00 0.13 5.62 1.52 0.13 5.62 0.00
Others (%) 0.86 0.00 3.03 0.59 1.12 3.03 0.53 0.00 3.03
INACT VAC Inactivated vaccine, mRNA VAC mRNA vaccine, RBD-SU VAC Receptor-binding domain subunit vaccine.
Others: herpes zoster, chest pain, astigmatism, immunologic conjunctivitis, etc.
We then analyzed the prevalence of adverse reactions in two age groups (18–59 and above). We found that in the age group of 18 to 59, the prevalence of any adverse reactions was significantly associated with vaccine types in each dose (all P < 0.001), and it was generally higher in mRNA vaccines than in others (Additional file 3: Table S2). As in the age group of 60 and above, most participants received three dose of inactivated vaccines (55/56), we only compared the adverse reactions to inactivated vaccines in three doses and found that there was no significant difference between three vaccinations (all P > 0.05) (Additional file 4: Table S3).
The prevalence of any adverse reactions to the inactivated vaccine decreased with each injection (total P < 0.001), whereas the prevalence of any adverse reactions to the mRNA or RBD-subunit vaccine was not different among the three vaccinations (Additional file 5: Fig. S1a).
The most prevalent injection site adverse reaction for both inactivated and mRNA vaccines was pain (Additional file 6: Fig. S2a). The prevalence of fatigue decreased with each dose of inactivated vaccine (total P < 0.001), whereas it increased with each dose of mRNA vaccine (total P = 0.022) (Additional file 6: Fig. S2b).
Some common adverse reactions (pain at the injection site, fatigue, myalgia, and drowsiness) of the three doses of inactivated vaccines mainly occurred in less than 1-day post-vaccination (62.50%–83.08%) ( Fig. 1a, b, c, e), and the duration was usually <1 day (13.84%–37.84%) or 1–2 days (33.78%–55.09%) (Additional file 7: Fig. S3a, b, c, e). Meanwhile, some common adverse reactions of the three doses of mRNA vaccines (pain at the injection site, fatigue, myalgia, and fever) mainly occurred in <1 day (50.00%–100.00%) (Fig. 1a, b, c, d), and the duration was usually <2 days (52.38%–92.31%) (Additional file 7: Fig. S3a, b, c, d). Because only two participants had systemic reactions to the RBD-subunit vaccines, the occurrence time and duration were not analyzed.Fig. 1 The occurrence time of common adverse reactions of inactivated vaccine and mRNA vaccine.
Fig. 1
We also analyzed the association between the different types of booster vaccines and adverse reactions after the booster dose ( Fig. 2). For those who received inactivated vaccine for the first two doses, the prevalence of injection site adverse reactions was 13.33% (12.14%–14.59%) and 38.60% (29.63%–48.17%), respectively, for a booster dose with inactivated or mRNA vaccine; systemic adverse reactions were 6.42% (5.57%–7.35%) and 57.02% (47.41%–66.25%), respectively.Fig. 2 The association between different types of booster vaccines and adverse reactions after the booster dose.
Fig. 2
As shown in Fig. 3, age and underlying medical conditions were not significantly associated with injection site or systemic reactions for all three doses (Fig. 3). Individuals with any adverse reaction after the first dose had an increased risk of injection site (OR=8.47, 95% CI, 6.21–11.55, Fig. 3b) and systemic reactions (OR=5.88, 95% CI, 4.07–8.50, Fig. 3b) for the second dose. In addition, those who developed adverse reactions in the first two doses had a higher risk of injection site reactions (OR=7.48, 95% CI, 5.58–10.03, Fig. 3c) and systemic adverse reactions (OR=4.14, 95% CI, 2.88–5.94, Fig. 3c) after receiving the booster dose.Fig. 3 Associations of age, underlying medical conditions, and experience of reactions with adverse reactions. OR: odds ratio; CI: confidence interval.
Fig. 3
4 Discussion
In this cross-sectional study, we identified 3339 participants who had completed three doses of the COVID-19 vaccination. The most common injection site adverse reaction of both inactivated and mRNA vaccines was pain, whereas itching was the most common for RBD-subunit vaccines. Fatigue and myalgia were the most common systemic adverse reactions to inactivated and mRNA vaccines, respectively. Age and underlying medical conditions were not significantly associated with injection site or systemic reactions. People who developed adverse reactions during the first two doses had a higher risk of adverse reactions after the booster dose.
Compared with previous studies on adverse reactions in China [26], the prevalence of adverse reactions in our study was higher (15.60% vs. 24.58% in the first dose and 14.60% vs. 19.22% in the second dose). This may be due to the detailed adverse reactions and higher education levels of the participants in our study. Individuals with higher education levels may pay more attention to their physical condition and be more sensitive to adverse reactions [27]. The prevalence (95% CI) of anaphylaxis or hospitalization with the three doses was 0.39% (0.21%–0.66%), 0.24% (0.10%–0.47%), and 0.33% (0.16%–0.59%) in our study, which was similar to previous studies [25], [28], [29]. Fever, as a measurable objective indicator, is often a concern after vaccination. In our study, the prevalence of fever after mRNA vaccination was 11.24% and 25.84% in the first and second doses, respectively, which was similar to 9.5% and 29.6% in the US study [30]. However, fever was uncommon after the inactivated vaccine, with a prevalence of 2% and 1%, respectively, for the first and second dose in a clinical trial done by India and less than 1% in the booster dose in Brazil, similar to our study (1.45% in the first dose, 0.79% in the second dose and 0.86% in the booster dose) [31], [32]. The prevalence of adverse reactions in our study was higher in females than in males, and these findings align with that of previous studies which showed that females typically developed higher antibody responses and were more sensitive [33], [34].
The WHO reported that COVID-19 vaccines are safe and effective for people with underlying medical conditions such as hypertension and diabetes [17]. Previous studies also showed that the prevalence of injection site and systemic adverse reactions to inactivated vaccines was relatively low [35], [36], and few serious adverse reactions were reported [26], which were similar to the results of our study.
Our study showed that patients with adverse reactions to the first or second dose had a higher risk of adverse reactions to the booster dose. Participants who had an adverse reaction to the first two doses may have had an immunocompromised immune system [37] or a stronger antibody-mediated hypersensitivity to the COVID-19 vaccine [38]. In addition, multiple patterns of protective immunity may exist after vaccination, and antigen-specific CD4 and CD8 T-cell reactions may be associated with additional adverse reactions [39], [40]. Thus, these patients may be more prone to adverse reactions after a booster dose.
In our study, individuals who received a booster dose of mRNA vaccine had a higher risk of injection site and systemic reactions than those who received the inactivated vaccine, suggesting the necessity for adverse reaction monitoring and seeking medical care in cases of severe reactions after vaccination with a booster dose of mRNA vaccine.
4.1 Strengths
This study has several strengths. First, we were the first to collect and analyze data on adverse reactions in participants who had completed three doses of vaccination, which can provide scientific evidence for the safety of vaccination, particularly for booster doses. Second, our questionnaire included some adverse reactions with a low tendency to report (such as fatigue and drowsiness), which were rarely mentioned in previous studies, as well as detailed information on the occurrence time and duration of each adverse reaction analyzed. Moreover, we compared the difference in adverse reactions between the general population and the elderly or people with underlying medical conditions, providing scientific evidence for these populations.
4.2 Limitations
First, since we conducted an online questionnaire survey using convenience sampling and not a random sampling, the results were unlikely to be generalizable to all vaccinated populations. Second, adverse reactions were collected using a self-reported questionnaire, which probably led to a high reporting rate and bias. Third, the power for comparing the difference between inactivated and RBD-subunit vaccines was relatively lower (<80%) due to the relatively small samples of RBD-subunit vaccination and the similar prevalence of the adverse reactions, suggesting that a larger sample size for the RBD-subunit vaccine is needed to get more accurate and comparable results in the future study. However, the power of detecting the difference of adverse reactions between the inactivated and mRNA vaccination was relatively high (0.96), suggesting that the results from our study might be relatively stable and reasonable.
5 Conclusions
Among those who received three doses of the COVID-19 vaccine, the most common injection site adverse reaction was pain, whereas fatigue and drowsiness were the most common systemic adverse reactions. The prevalence of anaphylaxis and hospitalization was low. Participants aged 60 and above and those with underlying medical conditions were not at a higher risk for adverse reactions, which provides a scientific basis for strengthening vaccination safety for the elderly and individuals with underlying medical conditions.
Conflict of interest
We have no conflict of interest to declare.
Appendix A Supplementary material
Supplementary material
Supplementary material
Supplementary material
Supplementary material
Supplementary material
Supplementary material
Supplementary material
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jiph.2022.12.004.
==== Refs
References
1 Ochani R. Asad A. Yasmin F. Shaikh S. Khalid H. Batra S. COVID-19 pandemic: from origins to outcomes. A comprehensive review of viral pathogenesis, clinical manifestations, diagnostic evaluation, and management Infez Med 29 1 2021 20 36 33664170
2 Onyeaka H. Anumudu C.K. Al-Sharify Z.T. Egele-Godswill E. Mbaegbu P. COVID-19 pandemic: A review of the global lockdown and its far-reaching effects Sci Prog 104 2 2021 368504211019854
3 Ge H. Wang X. Yuan X. Xiao G. Wang C. Deng T. The epidemiology and clinical information about COVID-19 Eur J Clin Microbiol Infect Dis 39 6 2020 1011 1019 32291542
4 WHO Coronavirus (COVID-19) Dashboard. Available from: 〈https://covid19.who.int/〉.[accessed 10 September 2022]
5 Bartoli A. Gabrielli F. Alicandro T. Nascimbeni F. Andreone P. COVID-19 treatment options: a difficult journey between failed attempts and experimental drugs Intern Emerg Med 16 2 2021 281 308 33398609
6 Sultana J. Mazzaglia G. Luxi N. Cancellieri A. Capuano A. Ferrajolo C. Potential effects of vaccinations on the prevention of COVID-19: rationale, clinical evidence, risks, and public health considerations Expert Rev Vaccines 19 10 2020 919 936 32940090
7 COVID-19 advice for the public: Getting vaccinated. Available from: 〈https://www.who.int/emergencies/diseases/novel-coronavirus-2019/covid-19-vaccines/advice〉. [accessed 10 September 2022]
8 Doroftei B. Ciobica A. Ilie O.D. Maftei R. Ilea C. Mini-Review Discussing the Reliability and Efficiency of COVID-19 Vaccines Diagnostics (Basel) 4 2021 11 35054180
9 Kim E. Erdos G. Huang S. Kenniston T.W. Balmert S.C. Carey C.D. Microneedle array delivered recombinant coronavirus vaccines: Immunogenicity and rapid translational development EBioMedicine. 55 2020 102743
10 Lurie N. Saville M. Hatchett R. Halton J. Developing Covid-19 Vaccines at Pandemic Speed N Engl J Med 382 21 2020 1969 1973 32227757
11 Sreepadmanabh M. Sahu A.K. Chande A. COVID-19: Advances in diagnostic tools, treatment strategies, and vaccine development J Biosci 45 1 2020
12 SAGE), COVID-19 vaccines technical documents. Available from: 〈https://www.who.int/groups/strategic-advisory-group-of-experts-on-immunization/covid-19-materials〉. [accessed 10 September 2022]
13 Lin Z.Q. Wu J.N. Huang R.D. Xie F.Q. Li J.R. Zheng K.C. Comparison of Safety of Different Vaccine Boosters Following Two-Dose Inactivated Vaccines: A Parallel Controlled Prospective Study Vaccines (Basel) 10 4 2022
14 Chen N. Zhou M. Dong X. Qu J. Gong F. Han Y. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study Lancet. 395 10223 2020 507 513 32007143
15 Chen Y. Klein S.L. Garibaldi B.T. Li H. Wu C. Osevala N.M. Aging in COVID-19: Vulnerability, immunity and intervention Ageing Res Rev 65 2021 101205
16 Ejaz H. Alsrhani A. Zafar A. Javed H. Junaid K. Abdalla A.E. COVID-19 and comorbidities: Deleterious impact on infected patients J Infect Public Health 13 12 2020 1833 1839 32788073
17 COVID-19 advice for the public: Getting vaccinated. Available from: 〈https://www.who.int/emergencies/diseases/novel-coronavirus-2019/covid-19-vaccines/advice〉. [accessed 10 September 2022]
18 Cai J. Deng X. Yang J. Sun K. Liu H. Chen Z. Modeling transmission of SARS-CoV-2 Omicron in China Nat Med 2022
19 Zhang X. Zhang W. Chen S. Shanghai's life-saving efforts against the current omicron wave of the COVID-19 pandemic Lancet. 2022
20 MacDonald N.E. Vaccine hesitancy: Definition, scope and determinants Vaccine. 33 34 2015 4161 4164 25896383
21 Vaccine hesitancy: A growing challenge for immunization programmes. Available from: 〈https://www.who.int/news/item/18-08-2015-vaccine-hesitancy-a-growing-challenge-for-immunization-programmes〉. [accessed 10 September 2022]
22 Siu J.Y. Cao Y. Shum D.H.K. Perceptions of and hesitancy toward COVID-19 vaccination in older Chinese adults in Hong Kong: a qualitative study BMC Geriatr 22 1 2022 288 35387602
23 Thunström L. Ashworth M. Finnoff D. Newbold S.C. Hesitancy Toward a COVID-19 Vaccine Ecohealth. 18 1 2021 44 60 34086129
24 Gee J. Marquez P. Su J. Calvert G.M. Liu R. Myers T. First Month of COVID-19 Vaccine Safety Monitoring - United States, December 14, 2020-January 13, 2021 MMWR Morb Mortal Wkly Rep. 70 8 2021 283 288 33630816
25 Hause A.M. Baggs J. Gee J. Marquez P. Myers T.R. Shimabukuro T.T. Safety Monitoring of an Additional Dose of COVID-19 Vaccine - United States, August 12-September 19, 2021 MMWR Morb Mortal Wkly Rep 70 39 2021 1379 1384 34591835
26 Zhang M.X. Zhang T.T. Shi G.F. Cheng F.M. Zheng Y.M. Tung T.H. Safety of an inactivated SARS-CoV-2 vaccine among healthcare workers in China Expert Rev Vaccines 20 7 2021 891 898 33929930
27 Hahn R.A. Truman B.I. Education Improves Public Health and Promotes Health Equity Int J Health Serv 45 4 2015 657 678 25995305
28 Risma K.A. COVID-19 mRNA vaccine allergy Curr Opin Pediatr 33 6 2021 610 617 34670264
29 Xia S. Duan K. Zhang Y. Zhao D. Zhang H. Xie Z. Effect of an Inactivated Vaccine Against SARS-CoV-2 on Safety and Immunogenicity Outcomes: Interim Analysis of 2 Randomized Clinical Trials Jama. 324 10 2020 951 960 32789505
30 Rosenblum H.G. Gee J. Liu R. Marquez P.L. Zhang B. Strid P. Safety of mRNA vaccines administered during the initial 6 months of the US COVID-19 vaccination programme: an observational study of reports to the Vaccine Adverse Event Reporting System and v-safe Lancet Infect Dis 22 6 2022 802 812 35271805
31 Costa Clemens S.A. Weckx L. Clemens R. Almeida Mendes A.V. Ramos Souza A. Silveira M.B.V. Heterologous versus homologous COVID-19 booster vaccination in previous recipients of two doses of CoronaVac COVID-19 vaccine in Brazil (RHH-001): a phase 4, non-inferiority, single blind, randomised study Lancet. 399 10324 2022 521 529 35074136
32 Ella R. Vadrevu K.M. Jogdand H. Prasad S. Reddy S. Sarangi V. Safety and immunogenicity of an inactivated SARS-CoV-2 vaccine, BBV152: a double-blind, randomised, phase 1 trial Lancet Infect Dis 21 5 2021 637 646 33485468
33 Fischinger S. Boudreau C.M. Butler A.L. Streeck H. Alter G. Sex differences in vaccine-induced humoral immunity Semin Immunopathol 41 2 2019 239 249 30547182
34 Vassallo A. Shajahan S. Harris K. Hallam L. Hockham C. Womersley K. Sex and Gender in COVID-19 Vaccine Research: Substantial Evidence Gaps Remain Front Glob Womens Health 2 2021 761511
35 Al Khames Aga Q.A. Alkhaffaf W.H. Hatem T.H. Nassir K.F. Batineh Y. Dahham A.T. Safety of COVID-19 vaccines J Med Virol 93 12 2021 6588 6594 34270094
36 Zhang Y. Zeng G. Pan H. Li C. Hu Y. Chu K. Safety, tolerability, and immunogenicity of an inactivated SARS-CoV-2 vaccine in healthy adults aged 18-59 years: a randomised, double-blind, placebo-controlled, phase 1/2 clinical trial Lancet Infect Dis 21 2 2021 181 192 33217362
37 Su J.R. Ng C. Lewis P.W. Cano M.V. Adverse events after vaccination among HIV-positive persons, 1990-2016 PLoS One 13 6 2018 e0199229
38 Stone C.A. Jr. Rukasin C.R.F. Beachkofsky T.M. Phillips E.J. Immune-mediated adverse reactions to vaccines Br J Clin Pharmacol 85 12 2019 2694 2706 31472022
39 Goel R.R. Painter M.M. Apostolidis S.A. Mathew D. Meng W. Rosenfeld A.M. mRNA vaccines induce durable immune memory to SARS-CoV-2 and variants of concern Science. 374 6572 2021 abm0829 34648302
40 Rydyznski Moderbacher C. Ramirez S.I. Dan J.M. Grifoni A. Hastie K.M. Weiskopf D. Antigen-Specific Adaptive Immunity to SARS-CoV-2 in Acute COVID-19 and Associations with Age and Disease Severity Cell. 183 4 2020 996 1012 e19 33010815
| 36516647 | PMC9724502 | NO-CC CODE | 2022-12-13 23:16:25 | no | J Infect Public Health. 2023 Jan 6; 16(1):125-132 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.12.004 | oa_other |
==== Front
Int J Med Inform
Int J Med Inform
International Journal of Medical Informatics
1386-5056
1872-8243
Elsevier B.V.
S1386-5056(22)00255-6
10.1016/j.ijmedinf.2022.104941
104941
Article
Mining the vaccination willingness of China using social media data
Ding Jiaming ab
Wang Anning ab⁎
Zhang Qiang ab
a School of Management, Hefei University of Technology, Hefei 230009, China
b Key Laboratory of Process Optimization and Intelligent Decision-making, Ministry of Education, Hefei 230009, China
⁎ Corresponding author.
6 12 2022
6 12 2022
10494129 6 2022
15 10 2022
26 11 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objective
Vaccination is one of the most powerful and effective protective measures against Coronavirus disease 2019 (COVID-19). Currently, several blogs hold content on vaccination attitudes expressed on social media platforms, especially Sina Weibo, which is one of the largest social media platforms in China. Therefore, Weibo is a good data source for investigating public opinions about vaccination attitudes. In this paper, we aimed to effectively mine blogs to quantify the willingness of the public to get the COVID-19 vaccine.
Materials and Methods
First, data including 144,379 Chinese blogs from Weibo, were collected between March 24 and April 28, 2021. The data were cleaned and preprocessed to ensure the quality of the experimental data, thereby reducing it to an experimental dataset of 72,496 blogs. Second, we employed a new fusion sentiment analysis model to analyze the sentiments of each blog. Third, the public’s willingness to get the COVID-19 vaccine was quantified using the organic fusion of sentiment distribution and information dissemination effect.
Results
(1) The intensity of bloggers’ sentiment toward COVID-19 vaccines changed over time. (2) The extremum of positive and negative sentiment intensities occurred when hot topics related to vaccines appeared. (3) The study revealed that the public’s willingness to get the COVID-19 vaccine and the actual vaccination doses shares a linear relationship.
Conclusion
We proposed a method for quantifying the public’s vaccination willingness from social media data. The effectiveness of the method was demonstrated by a significant consistency between the estimates of public vaccination willingness and actual COVID-19 vaccination doses.
Keywords
Vaccination willingness
Social media
COVID-19
Sentiment analysis
Information dissemination effect
==== Body
pmcIntroduction
COVID-19 is a highly infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1], [2], [3]. Vaccination is widely regarded by health organizations as one of the most effective ways to protect people from infectious diseases [4], [5]. Presently, the COVID-19 virus is mutating and may persist for a long time [6]. To effectively reduce the prevalence and incidence of vaccine-preventable diseases, it is necessary to achieve acceptable levels of protection and sustained herd immunity through widespread vaccination [7]. While more than ten COVID-19 vaccines are currently available, understanding public vaccination willingness is critical to a deeper understanding of the global state of COVID-19 vaccination.
During the early stages of vaccination, there are compulsory vaccination measures for people working in various public places, but for the public, vaccination must follow the basic principles of ”informed, consent, and voluntary”. The development and marketing of vaccines require several years of clinical experience and strict regulatory mechanisms before they can be produced and used [8]. As COVID-19 spreads globally, targeted vaccines must be developed to prevent viral infection with limited cycles. Due to the small number of clinical trials of COVID-19, its safety and effectiveness is not always widely recognized by the public, and the public can therefore be hesitant to new vaccines [9]. A cross-sectional study in China found that 9% of the participants reportedly refused the COVID-19 vaccination, whereas 35.5% reported vaccine hesitancy [10]. In the United Kingdom, 27% of the participants reported vaccine hesitancy and 9% were resistant to getting the COVID-19 vaccine [11]. Meanwhile, vaccination willingness is also a key factor for vaccination coverage, which is typically collected by surveys [12], [13]. However, it is costly and time consuming. With vaccination willingness changing over different events and times, the survey results may quickly become outdated [14]. Therefore, it is crucial to investigate a novel, timely, and automated method for mining the public’s vaccination willingness.
Nowadays, many methods for sharing information have been incorporated into several social media platforms with high speed and penetration [15]. Sina Weibo is one of the largest social media platforms in China. There has been an increasing number of discussions on COVID-19 vaccines, offering a promising data source to mine the public’s vaccination willingness. Furthermore, accessibility, interactiveness, and spontaneity of such data give this study a great opportunity.
In this paper, we proposed a quantitative vaccination willingness (QVW) model to mine the vaccination willingness of China, which consists of a sentiment analysis model and information dissemination effect. First, we performed a fusion sentiment analysis model based on bidirectional encoder representations from transformers-bidirectional long short-term memory network-convolutional neural network (BERT-Bi-LSTM-CNN) architecture, and obtained the sentiment probability of each blog. Second, we used an information dissemination effect index to measure the dissemination effect of each blog. Third, the combination of positive sentiment probability and corresponding information dissemination effect was used to quantify the public’s vaccination willingness.
To validate the effectiveness of our proposed method, we use it to calculate the public’s vaccination willingness estimates for 36 days from March 24 to April 28, 2021, and directly collect the actual vaccination doses. The results show that the correlation coefficient of the public’s willingness with actual vaccination doses is 0.67, indicating that the proposed model provides a reliable signal of the public’s vaccination willingness.
Our proposed method is motivated by a wealth of research in the social and computational sciences, suggesting that the influence intensity in a social network is reflected through social activities among users. Furthermore, the content posted on social media can influence other people’s decisions. The core contribution of this paper is a new methodological tool for quantifying the vaccination willingness of the public. Our method is a real-time, low-cost alternative to existing methods that researchers can use it for quantifying public willingness for different vaccinations, such as HIV, HPV, and H7N9 vaccines [16], [17], [18]. The results of this study can better help governments, policymakers, and healthcare providers take effective steps to drive a successful COVID-19 campaign.
Related works
At present, online social media such as Weibo, Facebook, and Twitter spread a great deal of information about the COVID-19 vaccine, which includes the public’s attitudes and opinions on vaccines as well as vaccine knowledge and popularization [19]. The wide application of social media provides a good data source for scholars to carry out research [20]. Previous studies not only employed sentiment analysis in social media data to analyze public’s attitudes and opinions about COVID-19 vaccines, but also carried out related research on COVID-19 vaccination willingness [21], [22].
Sentiment analysis on COVID-19 vaccines. As the vaccines developed and rolled out, a large and growing body of literature has performed the sentiment analysis of vaccine-related social media data to understand people’s attitude toward vaccination [23]. Several studies have focused on performing sentiment analysis on different vaccines [24], [25], aiming to mine different public’s attitudes and opinions on two or more vaccines. There are also several studies investigating that changes in public attitudes toward vaccines in one country [26] or even multiple countries [27]. Collectively, the existing literature is mainly focused on reflecting the vaccination willingness from public attitudes toward vaccines. Methodologically, scholars have used different sentiment analysis methods, such as dictionary and rule-based tools [25], [27], machine learning -based models [24], [26]. However, these methods often require a great number of complex features and their performance are susceptible to sparse data.
Vaccination willingness for COVID-19. The study of vaccination willingness has recently attracted extensive attention in the fields of medicine, management, psychology and so on. The existing literature has conducted relevant research on vaccination willingness from two aspects. (1) Several studies investigated the vaccination willingness and its potential predictors within small groups through questionnaires [28], [29]. On this basis, several scholars have performed systematic reviews and meta-analyses to estimate the vaccination willingness of the broader population [30], [31]. (2) A few of researchers proposed different social network modeling to study vaccination willingness and its influence factors, such as an epidemiological model of social networks that considered population heterogeneity and different vaccination strategies [32], a model integrating multiple criteria belief modeling with social network analysis [33]. However, from a data perspective, the sample size of questionnaires is small and the collection is time-consuming, while the social network modeling uses simulated data and lacks of real data. Online social media provide a wealth of real-world data and low-cost analytical tools for research. Therefore, the use of social media data to quantify the public vaccination willingness represents a fundamental shift in measurement method.
Materials and methods
This section introduces data acquisition and the proposed QVW model in detail.The overall steps of the proposed model is shown in Fig.1 .Fig. 1 The overall steps of the QVW model.
Data acquisition
The experimental data are from the Weibo data source pool. Weibo has more than 500 million registered users. A web crawler using Python was developed to obtain blogs on Weibo. The search keywords for the blogs were . Therefore, all Chinese-language blogs related to the keywords posted during the period from March 24 to April 28, 2021 were obtained, totaling 144,379 blogs. Furthermore, the actual data for vaccination doses were obtained on the website of the national health commission of the people’s republic of China (http://www.nhc.gov.cn/).
Data preprocessing
The collected blogs are preprocessed by Python 3.7 to ensure the quality of the experimental data. The preprocessing steps mainly include: (1) we removed blogs with duplicate content mainly to eliminate the influence of multiple duplicate blogs generated by fake accounts. (2) used regular expressions re library to delete blogs containing ”HPV+vaccine”, ”rabies+vaccine”, ”H7N9+vaccine” and ”HIV+vaccine”. (3) deleted reposts and kept original blogs, as the latter can better express the real thoughts. (4) filtered URLs and HTML tags. (5) removed stop words from blogs using the stop word list posted by Harbin Institute of Technology (https://github.com/goto456/stopwords). (6) converted traditional characters to simplified characters.
Experimental datasets
After data preprocessing, we obtained an experimental dataset containing 72,496 blogs. 12,000 blogs were randomly selected and manually annotated by four doctoral students majoring in information systems and management, each of which annotated 3,000 blogs. Subsequently, the 3,000 blogs annotated by each annotator were split into three parts and assigned them to other annotators for re-annotation. The annotation consistency ratio was 84.2%, and 10,000 consistent blogs were retained as the dataset, containing three sentiment labels: positive (2), neutral (1), and negative (0). The annotated dataset is referred to as the PNN dataset, which was divided into training, test and validation sets in the ratio of 8:1:1. Table 1 lists the detailed statistics of the PNN dataset. Examples of ”positive”, ”neutral”, and ”negative” blogs are listed in Table 2 .Table 1 Statistics of the PNN dataset.
Label Training Test Validation Total (%)
Positive 3,824 483 493 4,800 (48.00)
Neutral 316 38 33 387 (3.87)
Negative 3,860 479 474 4,813 (48.13)
Total 8,000 1,000 1,000 10,000 (100)
Table 2 Examples of ”positive”, ”neutral”, and ”negative” blogs.
The proposed QVW model
Given blogs with the number of likes, comments, and retweets and the corresponding blogger with the number of followers, the number of posting blogs, and the registration time. Our goal is to develop an automated method to quantify a score of the public vaccination willingness. A sentiment analysis model was first used to get the sentiment probability of each blog and the corresponding sentiment classification. Subsequently, we calculated the information dissemination effect of each blog. Finally, a score to the public vaccination willingness is quantified by combining the sentiment probability of blogs and corresponding information dissemination effect. In this subsection, the specific steps are introduced in detail.
Sentiment analysis model
Sentiment analysis refers to the process of determining sentiment and classifying the polarity of text content [34]. We classified blogs into three categories, namely positive, negative, and neutral sentiment. The positive blogs indicate that bloggers are optimisim about vaccination, the neutral blogs show basic information or knowledge on vaccination, and the negative blogs reflect the adverse opinions of bloggers about vaccination. The proposed sentiment analysis model consists of three parts: (1) Bidirectional encoder representations from transformers (BERT). (2) Bidirectional long short-term memory network (Bi-LSTM). (3) Convolutional neural network (CNN). Fig.2 shows the overall architecture.Fig. 2 Architecture of our proposed sentiment analysis model for analyzing the blogs.
BERT. The BERT pretrained language model [35] was proposed by Google in 2017, which used to dynamically generate character embedding vectors in semantic space. It employs the ”masked language model” to pretrain multiple bidirectional transformer encoders. The deep bidirectional language representation of each character can be obtained from the forward and backward directional text information.
Given a blog B=(c1,c2,⋯,cn), the character embedding can be obtained using the BERT pretrained model as shown below:(1) (xc1,xc2,⋯xcn)=BERTc(c1,c2,⋯,cn)
where xci is the character embedding of ci.
Bi-LSTM. The Bi-LSTM layer uses two LSTMs [36] in different directions connected by the same output layer to extract the contextual features of the text. This provides the output layer with complete contextual information. The forward LSTM network learns future characteristics and gets the forward hidden state ht→ for character ct, and the backward LSTM network learns historical features and gets the backward hidden state ht←for character ct. The formula for calculating the final state ht is as follows:(2) ht→=LSTM→(xct,ht-1→)
(3) ht←=LSTM←(xct,ht+1←)
(4) ht=ht→,ht←
where the final state ht is the concatenation of ht→ and ht←. Therefore, the feature representation of the blog B=(c1,c2,⋯,cn) generated by the Bi-LSTM layer is shown as follows:(5) H=h1,h2,⋯,hn
CNN. The blogs were posted by different bloggers, who expressed sentimental tendencies through adjectives, adverbs and other sentimental words. The CNN layer extracts local semantic features expressed by sentimental words using the convolutional layer [37]. The specific formula is as follows:(6) O=CNN(H)
Output. The feature vectors obtained from the CNN layer are considered as the input of the full connection layer, and then input into the softmax classifier. The formula is as follows:(7) S=softmax(W*O+b)
where S represents the sentiment probability set of the blog, W denotes the weight coefficient matrix, b is the corresponding bias. Additionally, model parameters were adjusted by minimizing the cross entropy, and the specific formula is as follows:(8) Loss=-1D∑iyilogyi^
where D denotes the number of training samples, yi is the ground-truth label, yi^ is the prediction label of i-th blog.
Information dissemination effect
In social networks, different communication subjects will have diverse views on the same information, and the large-scale dissemination of views will affect individuals’ risk perception and produce corresponding behaviors. For example, blogs expressing a positive attitude towards vaccination published by influential individuals can use their efficient information dissemination ability to promote vaccination. Factors of dissemination effect include information content quality and information publishers’ influence. The dissemination effect of a blog is mainly reflected by its quality and the corresponding blogger influence. Fig.3 shows the main composition of blog information dissemination effect. Therefore, combining blog quality (BQ) and the corresponding blogger influence (BI), the information dissemination effect (IDE) of a blog is calculated as follows:(9) IDEi=β×BQi+(1-β)×BIi
where β is a hyper-parameter for balancing the contribution between of BQi and BIi. The value of β is empirically set to 0.4.Fig. 3 The main composition of blog information dissemination effect.
Blog quality. The blog quality is mainly reflected by the number of comments, retweets, and likes. The more comments, retweets, and likes that a blog obtains, the more attention receives and the greater its influence. Therefore, the calculation formula for blog quality using different metric contributions is as follows:(10) BQi=BCi+BRi3+BLi4
where BQi is the blog influence of blogger i,BCi,BRi,BLi denote the number of comments, retweets, and likes of the corresponding blog respectively.
Blogger influence. First, the increase in a blogger’s attention is a manifestation of its expanded influence. We use the PageRank (PR) algorithm [38] to calculate the blogger popularity. Second, the blogger’s active level reflects the blogger’s activity status on social platforms, and the frequency of bloggers following others, posting blogs, and increasing followers are regarded as influence factors of the blogger active level. Therefore, we define the influence of a blogger as:(11) BIi=PR(Vi)+∑j=1k(γj×xij)
where BIi is the blogger influence of blogger i,PRVi represents the i-th blogger popularity, xij denotes the frequency of bloggers following others, posting blogs, and increasing followers of i-th blogger, k is the number of influence factors, and γj denotes the weight of the j-th influence factor of blogger i. In Eq (11), the calculation formulas of PRVi and xij are as follows:(12) PR(Vi)=(1-α)+α∑Vt∈G(Vi)PR(Vt)L(Vt)
(13) xij=XijTcurrent-Tregistered
where GVi is the blogger’s fan set, LVt is the follower numbers of blogger t,α is set to 0.85 according to the traditional PageRank algorithm [38], Tcurrent denotes the time of data acquisition, Tregistered is the blogger i registered time, Xij is the number of j-th influence factor of blogger i active level.
When determining the values of γj in Eq (11), we used the analytic hierarchy process [39] to calculate the weight of blogger activity level. The relevant weight determination steps are illustrated in the supplementary material A.
Vaccination willingness
To quantify a score for vaccination willingness, the model first calculates the sentiment probability of each blog using Eq (7). Specifically, only the positive sentiment probability is selected for the subsequent stage. Next, the model quantifies the corresponding IDE using Eq (9). Finally, a score is calculated by combining the sentiment probability and corresponding IDE. We denote this score as the public’s vaccination willingness (PVW). The core hypothesis is that the higher the PVW, the more vaccination doses. The specific calculation formulas are as follows:(14) PVW+=1n1×∑i=1n1(Spos)i+μ×lgIDEi
(15) PVW-=1n1×∑i=1n2(Spos)i-μ×lgIDEi
(16) PVW=n1×PVW++n2×PVW-n1+n2
where (Spos)i denotes the positive probability of i-th blog, n1 and n2 represent the number of positive and negative blogs, μ∈0,1.
Experimental parameter settings
Experimental parameter settings can directly affect the experimental results. To complete the experiments efficiently, the proposed QVW model was developed on Ubuntu 20.10 using Python in the TensorFlow framework. The relevant parameter settings are shown in Table 3 .Table 3 Experimental parameter settings.
Parameter Value
(BERT) Character embedding size 128
Bi-LSTM layer 1
Bi-LSTM hidden size 128
CNN sliding window size 3, 4, 5
CNN sliding window number 128
CNN pooling method Max pooling
Initial learning rate 3e-4
Optimization Adam
Dropout 0.15
Batch_size 128
γ1 0.1919
γ2 0.1744
γ3 0.6337
Results
This section begins with an analysis of blogs collected from March 24 to April 28, 2021, to understand public sentiments on vaccination. Furthermore, PVW estimates are quantified using the proposed QVW model.
Sentiment analysis
Model evaluation
In order to evaluate the performance of the proposed sentiment analysis model, accuracy and F1-score are used during the test stage as described in Eqs.(17) to (20).(17) precision=TPTP+FP
(18) recall=TPTP+FN
(19) accuracy=TP+TNTP+FP+TN+FN
(20) F1-score=2×precision×recallprecision+recall
where TP,TN,FP, and FN are true positive, true negative, false positive, and false negative, respectively. To verify the effectiveness and necessity of each module on the sentiment analysis model, we designed three variant ablation study models. Details are shown below:
BERT: This model is the baseline.
BERT+Bi-LSTM: This model consists of BERT pre-trained model and Bi-LSTM layer.
BERT+Bi-LSTM+CNN: This is our proposed model.
We performed the different ablation study on the PNN dataset, and the experimental results of the ablation study are shown in Table 4 . The baseline achieves good performance by using the BERT model. The performance of BERT+Bi-LSTM model is significantly improved by adding the Bi-LSTM layer to the baseline model. The better performance is owing to the use of Bi-LSTM layer to extract bi-directional text information, which can significantly improve the performance of feature representation. Our proposed model achieves the optimal performance.Table 4 The experimental results of the ablation study.
Model Accuracy F1-Score
BERT (Baseline) 0.8542 0.8537
BERT+BiLSTM 0.8915 0.8907
BERT+BiLSTM+CNN 0.9129 0.9128
Numbers of sentiment criteria
The proposed sentiment analysis model categorized the collected blogs into three categories: positive, neutral, and negative. The proportion of blogs in each category are shown in Fig.4 . After analysis, we found 33,462 positive blogs (46.2%), 34,663 negative blogs (47.8%), and 4,371 neutral blogs (6.0%). The positive and negative blogs are in the majority and in roughly equal numbers. This indicates that mixed public attitudes toward COVID-19 vaccines and vaccination in general. The neutral blogs that merely contained information related to the COVID-19 vaccine were lower in frequency.Fig. 4 The proportion of sentiment criteria.
Timeline of sentiments
Fig. 5 shows how sentiments changed or shuffled over time. It can be inferred that most of the feelings expressed by the public falls under the positive and negative, with neutral feelings for a very small minority. The highest positive sentiment was roughly 61% on April 10, 2021 and the lowest positive sentiment was approximately 33% on April, 3 and 11, 2021. To understand how and what public discuss in the above three days, we performed topic modeling for the blogs of these three days and selected the hottest topic as shown in Table 5 . The introduction of topic modeling is illustrated in Appendix A. We further performed topic modeling for positive, neutral, and negative blogs separately (See Appendix A).Fig. 5 The percentage of positive, neutral and negative sentiment in blogs.
Table 5 The hottest topic discussed by blogs on 3rd,10th, and 11th of April.
As shown in Table 5, on April 3, the hottest topic was . Most bloggers expressed strong skepticism about this news from some organizations and conveyed negative sentiments. On April 10, the hottest topic was . Wenhong Zhang is an expert of infectious disease. Most blogs gave positive sentiments about this topic and expressed willingness or action on vaccination. On April 11, the hottest topic was . Many businesses and communities made it mandatory for all employees to be vaccinated against COVID-19. Most blogs opposed and were unsupportive of this topic, there conveying negative sentiments.
Sentiment of segmented population
To better understand the sentiment of different segmented populations toward COVID-19 vaccines or vaccination, we segmented all bloggers by three dimensions, including gender, certification, and age. As shown in Table 6 , first, the proportion of males with positive sentiments (50.60%) was higher than that of females (42.16%), and the proportion of females with negative sentiments was higher (51.44%). Second, the proportion of official bloggers with positive sentiments was higher (60.24%) than that of public bloggers (44.47%). Neutral sentiments were highest among official bloggers (13.75%), as some of the official blogs expressed knowledge about vaccines or vaccination. Third, among the three age groups, the ”40-” age group had the most proportion of bloggers with positive sentiments (50.22%), while the ”19-39” age group had the most proportion of bloggers with negative sentiments (49.39%), followed by the ”9-18” age group (47.38%).Table 6 The number and proportion of sentiment criteria in the segmented population.
Variables n (%) Positive (%) Neutral (%) Negative (%)
Gender Male 34,327 (47.35) 17,370 (50.60) 1,929 (5.62) 15,028 (43.78)
Female 38,169 (52.65) 16,092 (42.16) 2,442 (6.40) 19,635 (51.44)
Certification Official 7,763(10.71) 4,677 (60.24) 1,067 (13.75) 2,019 (26.01)
Public 64,728(89.29) 28,785 (44.47) 3,304 (5.10) 32,644 (50.43)
Age 9-18 4,267 (5.89) 1,972 (46.21) 273 (6.41) 2,022 (47,38)
19-39 17,867 (24.64) 8,097 (45.32) 945 (5.29) 8,825 (49.39)
40- 6,505 (8.97) 3,267 (50.22) 445 (6.84) 2,793 (42.94)
Missing data 43,857 (60.50) - - -
Note: Official certification includes government, enterprise, organization, and media certification. Public certification refers to the user certificate on Weibo through ID card or other credentials.
Public’s vaccination willingness
Timeline of public’s vaccination willingness
The change tendency of the PVW quantified by blogs and actual vaccination doses (AVD) from March, 24 to April, 28 is shown in Fig.6 . For PVW estimate and AVD, the colors used are red and green, respectively. The PVW estimate for April 10 was as high as about 0.70, while the PVW estimate for April 02 was as low as about 0.34. The average PVW estimate is about 0.53. It is clear that the PVW estimate and AVD have a certain correlation.Fig. 6 The change tendency of the PVW estimates quantified by blogs and the actual vaccination doses (AVD).
Validation results
To evaluate the overall accuracy of PVW, we computed the pearson correlation coefficient between the actual vaccination doses (AVD) and PVW estimates daily. Fig.7 shows scatter plots of PVW versus AVD for verifying its relevance. The correlation coefficient between PVW versus AVD for over 36 days is 0.67, which follows the positive skew. This implies that there is a strong correlation between them.Fig. 7 Scatter plots of PVW versus AVD.
To demonstrate the effectiveness of QVW model, we designed some of measurement methods across a range of IDE to determine how they affect correlation. Furthermore, we computed the pearson correlation coefficients between PVW estimates and AVD of days t,t+1, t+2, and t+3 to observe the delayed impact of blogs. The baseline only considers the positive sentiment probability. Table 7 displays correlations between PVW versus AVD for the several measurement methods of IDE. Based on the baseline, four different measurement methods of considering IDE outperform the baseline. It indicates the effectiveness of IDE. The correlations of PVW estimates and AVD quantified by our proposed method are optimal of days t,t+1 (r = 0.69, 0.86). The PVW estimates quantified by the measurement of blog quality (BQ) has the largest correlation value with AVD of days t+2 (r = 0.62). In aggregate, the correlations appear robust to these algorithmic decisions, indicating that the value of this method is not limited to one particular implementation. Note the experimental results of our method, the correlation coefficient between the PVW estimates and AVD of day t+1 is the highest (r = 0.86). The consistency decreases as the number of days of delay increases, and virtually no consistency is recorded by the three days delayed. The same is true for the other three methods, suggesting that the day’s blogs don’t have a timely impact, but are having the best impact the next day, and the impact will gradually decrease over time.Table 7 Pearson correlation coefficients (r) for PVW estimates of day t and AVD of day t+1, t+2, and t+3 across a range of information dissemination effect measurement methods.
Method t t+1 t+2 t+3
Senti (Baseline) 0.49 0.69 0.43 0.16
Senti+BQ 0.53 0.71 0.62 0.27
Senti+BP 0.58 0.72 0.47 0.18
Senti+BI (BP+BAL) 0.61 0.79 0.53 0.26
Ours (Senti+BQ+BI) 0.67 0.86 0.58 0.23
Note: Our proposed method is highlighted in bold. The largest values in each column are in italics. Senti is the positive sentiment probability. BP represents the blogger popularity calculated by PageRank algorithm. BAL refers to blogger activity level.
Discussion
Our analysis shows different perceptions of blogs on the COVID-19 vaccine or vaccination. Our results not only help understand individual perceptions, but also provide important implications for understanding the public stance on current public health knowledge about COVID-19 vaccines. Furthermore, we investigated a novel method to mine the public’s vaccination willingness. Experimental results also demonstrate the feasibility of the proposed approach. We encourage future researchers to verify the overall accuracy of PVW using different validation methods, such as continued questionnaires surveys.
Rich textual data from social media can be used to better identify public attitudes and views on COVID-19 vaccines. As mentioned in Section 3.2, we propose a QVW model to analyze the blogs of 72,496 bloggers from March 24, to April 28, 2021, to mine the public’s vaccination willingness. The vaccination doses also depend on health conditions, social economy, cultures, and other factors, beyond the scope of this study and will be considered in our future research work. Nevertheless, among the blogs obtained, a significant consistency was observed between AVD and the public’s willingness quantified by the blogs (see subSection 4.2), illustrating the effectiveness of our proposed QVW model.
The intensity of bloggers’ sentiments on COVID-19 vaccines changed with time (see Fig.5. Notably, the positive and negative sentiment intensities were roughly equal on most days. However, there were differences in the intensity of positive sentiment on some days. For example, the highest positive sentiment is about 61% on April, 10 whereas the lowest positive sentiment is about 33% on April, 03 and April, 11, with a range of nearly 30%. Therefore, the blogger sentiment intensity about the COVID-19 vaccine fluctuates.
Furthermore, one can see a meaningful correlation between PVW and AVD (see Fig.7. The correlation coefficient between PVW and AVD is 0.67. The row of plots indicates that PVW for AVD follows the positive skew. That is, while most days have low-to-moderate AVD, there are a few low or high PVW estimates. These include April, 03 and April, 10 (see Fig.6. We encourage future researchers to explore more relations between PVW and AVD, such as causality. Additionally, they may also explore whether AVD can be predicted using PVW estimates.
Implications for theory
The theoretical significance of this study is in at least two important ways. First, we explored the sentiments and perceptions toward vaccination in China using information from social media analytic during pandemics. Although a few researchers have explored social media-based sentiment analysis on COVID-19, vaccines, and vaccination in different countries [10], [40], [41], this study is the first to analyze public sentiments and perceptions on vaccination in China using social media data. Furthermore, we extracted main topics of positive, neutral, and negative blogs during test period using the LDA algorithm and analyzed the reasons for the extreme value. We encourage future researchers to explore fine-grained topics using different aspects, such as healthcare advisory, anxiety, entertainment, industrial, politics, social support, and economy.
Second, we mined the public vaccination willingness in China from social media data using our proposed QVW model. Nowadays, social media has become an integral part of society, with millions of users voicing their opinions that before now, were left unheard. Thus, this opens up a wide field of analysis that was impossible before. The findings indicate that social media data is a good source for mining public opinions. The vaccination willingness can be quantified by social media data during unprecedented times like a pandemic. The sentiment analysis and IDE index were combined innovatively to quantify the PVW estimates. A few researchers have explored vaccination willingness using questionnaires, but compared with these methods, our approaches are timely, convenient and accuracy. Similarly, we encourage future researchers to propose better methods for quantifying PVW and explore more direct connections between the PVW estimates and AVD.
Implications for practice
The findings of this study have a number of important implications for future practice. Although this study focuses on mining COVID-19 vaccination willingness of China, the proposed method may have a generalization to the following aspects. To begin, the proposed method is available to explore and analyze public attitudes and opinions on vaccination, which is in preparation for the promotion of vaccination. Furthermore, the researchers can use the proposed method to mine the public vaccination willingness in different countries using different social media platforms, such as Twitter. Second, the proposed method can timely monitor the public willingness to take new vaccines, as COVID-19 may persist in mutating and some never-before-seen viruses may emerge, then targeted new vaccines will be rolled out.
The results of this study also can help governments, policymakers, and healthcare providers better understand the dynamic interrelationships between the overall sentiment of society and AVD. This may be used to design better strategies to drive a successful COVID-19 campaign. Specifically, policymakers and governments can devise appropriate policies and decisions to control public sentiment and reduce effects of adverse vaccine events on social sentiment. Furthermore, healthcare providers can better understand the public’s attitudes and willingness toward vaccination to improve the quality of service. Influential bloggers should actively advocate for people to get vaccinated against COVID-19. The public must engage more forcefully and constructively in health social media.
Conclusion
The COVID-19 pandemic is one of the most significant public health problems globally that disrupted the lives of millions in many countries. All governments and researchers around the world are trying to reduce the adverse effects of the disease. Currently, vaccination is the most effective way to fight the COVID-19 pandemic. In this paper, we propose a QVW model to mine the public’s vaccination willingness in China using social media data. To verify the effectiveness of our proposed method, we calculated the correlation coefficients of the public’s willingness and AVD (r = 0.67), providing a reliable signal of the public’s vaccination willingness estimates.
We hope the findings will drive a successful COVID-19 campaign. Specifically, the decision-makers and policy developers can reasonably apply social media data to strengthen vaccination willingness and reduce vaccine hesitancy and opposition. Public health authorities may be able to work through Weibo and others to collaborate with other official media outlets to increase positive information and reduce negative information about vaccines. When the PVW estimates are persistently high, public health authorities may appropriately improve the vaccine supply capacity and increase the number of medical workers for vaccination. On the contrary, when the PVW estimates continue to be low, public health authorities may find out the reasons and develop effective strategies and interventions. Furthermore, understanding public attitudes and vaccination willingness could help public health authorities reinforce optimistic comments within positive blogs while refuting aggressive language spreading false information within negative blogs.
One limitation of our study is ignoring the effects of COVID-19 vaccine news and statistics on the overall sentiment. Another limitation is the lack of consideration of government policies, vaccine supply capacity, individual’s perceived risk of disease and other factors that influence vaccination willingness. In the future, we will conduct more research to consider more influencing factors of vaccination willingness and find the vaccination willingness of segmented population, such as gender, age, education level, region and so on.
Summary points
What was already know on the topic?• Vaccination is one of the most powerful and effective protective measures against COVID-19, however, a timely, inexpensive, and automated method for mining public vaccination willingness is desperately needed.
• Social media has become an integral part of society, with millions of users voicing their opinions that before now, were left unheard. It provides a good source for mining public opinions.
• The influence intensity in a social network is reflected through social activities among users. Furthermore, the content posted on social media can influence other people’s decisions.
What this study added to our knowledge?• The public vaccination willingness can be quantified using social media data, suggesting that it need not rely on the questionnaire surveys to obtain vaccination willingness.
• The proposed methodological tool to quantify the vaccination willingness is groundbreaking, which is instructive for future work improvement.
CRediT authorship contribution statement
Jiaming Ding: Conceptualization, Data Collection, Methodology, Software, Writing-Original Draft. Anning Wang: Conceptualization, Writing - Review & Editing, Investigation. Qiang Zhang: Supervision, Project administration, Funding acquisition.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Topic modeling of COVID-19 vaccine-related blogs
Topic modeling is an effective text-mining tool that can discover hidden semantic structures in texts; it is an unsupervised method that can extract thematic structures from large amounts of textual content. Its main advantage is that the processing of huge quantities of blog content can be performed effectively to generate correlations between words belonging to the same topology [42]. Many topic models are available, and researchers chose Latent Dirichlet Allocation (LDA) as the most representative model for the study [43]. The LDA algorithm requires an input of the number of optimal topics that are selected by evaluating the perplexity of different topic number models. We performed the LDA algorithm on positive, neutral, and negative blogs using the genism library of Python 3.7. This algorithm ultimately produced the number of topics for positive, neutral, and negative blogs to be five, three, and four as shown in Table A1 .Table A1 Top words and example blogs for each positive, neutral, and negative topic.
Note: The blogs are paraphrased to protect users’ privacy (The fourth column).
Positive topics. The positive topics reflect the optimistic opinions of bloggers about vaccination and new advances in vaccines, including ”positive emotion around vaccination,” ”global cooperation and support,” ”vaccine progress around the world,” ”progress on vaccination,” and ”opinion leader effect.” First, topic 1 focuses on positive public attitudes and actions toward COVID-19 vaccination related to personal experiences, received information, and personal values that have a positive impact on other people’s vaccine perceptions, which in turn promotes their vaccination behavior. Second, topics 2 and 3 deal with vaccines as a global issue related to the globality of COVID-19 vaccines. These topics involve active blogs about vaccine progress and global cooperation and support; these blogs release research progress on vaccines to the public in an attempt to relieve their concerns to a certain extent. Third, topic 4 centers on the progress of vaccination, including the report on the number of vaccinations and vaccine supply. Fourth, topic 5 covers the active blogs of opinion leaders on public vaccination. Opinion leaders can greatly speed up information dissemination, which in turn affects the public’s awareness of vaccines and vaccination behavior.
Neutral topics. The neutral topics focus on promoting public knowledge about vaccines and vaccination, including ”educating communities,” ”instruction on getting vaccines,” and ”vaccine rollout.” First, topic 6 disseminates important information and knowledge about vaccines and answers related questions through blogs, part of which was spreading information on vaccine knowledge sharing through webinars. Second, topic 7 provides guidance on access to vaccines, which includes information dissemination from health authorities at all levels to guide the public on vaccination. Third, as vaccine development advances, many vaccine-oriented blogs are rolling out, as described in topic 8.
Negative topics. The negative topics reflect the adverse opinions of bloggers about vaccination and compulsory measures, including ”negative emotion around vaccination,” ”special population restrictions,” ”compulsory measure,” and ”adverse reactions.” First, topic 9 focuses on negative public attitudes and actions toward the COVID-19 vaccine, partly due to concerns about its safety and efficacy, and partly due to lack of vaccine knowledge. Second, topics 10 and 11 include vaccines as controversial topics. These topics express confusion and dissatisfaction among organizations or institutions with mandatory vaccination policies. Related blogs spread primarily negative attitudes about vaccines to the public, hindering the progress of vaccination. Third, topic 12 expresses the adverse reactions after vaccination, such as arm soreness, drowsiness, allergies, fever, etc. These real feelings can cause worry and fear in the unvaccinated population, which can change their vaccinated behavior.
Acknowledgments
This work was supported by grants from the National Natural Science Foundation of China (Nos. 72101078 and 72171069) and the Fundamental Research Funds for the Central Universities (No.JZ2021HGTA0131).
==== Refs
References
1 Huang C. Wang Y. Clinical features of patients infected with 2019 novel coronavirus in Wuhan China, The Lancet 395 10223 2020 497 506 10.1016/S0140-6736(20)30183-5
2 Goldman R.D. Yan T.D. Caregiver willingness to vaccinate their children against COVID-19: Cross sectional survey Vaccine 38 48 2020 7668 7673 10.1016/j.vaccine.2020.09.084 33071002
3 Leigh J.P. Moss S.J. Factors affecting COVID-19 vaccine hesitancy among healthcare providers in 23 countries Vaccine 40 31 2022 4081 4089 10.1016/j.vaccine.2022.04.097 35654620
4 Yaqub O. Castle-Clarke S. Sevdalis N. Chataway J. Attitudes to vaccination: a critical review Social science & medicine 112 2014 1 11 10.1016/j.socscimed.2014.04.018 24788111
5 Biswas N. Mustapha T. Khubchandani J. Price J.H. The nature and extent of COVID-19 vaccination hesitancy in healthcare workers Journal of community health 46 6 2021 1244 1251 10.1007/s10900-021-00984-3 33877534
6 Callaway E. The coronavirus is mutating — does it matter? Nature 585 7824 2020 174 177 10.1038/d41586-020-02544-6 32901123
7 Yoda T. Katsuyama H. Willingness to receive COVID-19 vaccination in Japan Vaccines 9 1 2021 48 10.3390/vaccines9010048 33466675
8 Mellet J. Pepper M.S. A COVID-19 vaccine: big strides come with big challenges Vaccines 9 1 2021 39 10.3390/vaccines9010039 33440895
9 Trent M. Seale H. Chughtai A.A. Salmon D. MacIntyre C.R. Trust in government, intention to vaccinate and COVID-19 vaccine hesitancy: a comparative survey of five large cities in the United States United Kingdom, and Australia, Vaccine 40 17 2021 2498 2505 10.1016/j.vaccine.2021.06.048 34218963
10 Wang C. Han B. Vaccination willingness, vaccine hesitancy, and estimated coverage at the first round of COVID-19 vaccination in China: A national cross-sectional study Vaccine 39 21 2021 2833 2842 10.1016/j.vaccine.2021.04.020 33896661
11 Sherman S.M. Smith L.E. COVID-19 vaccination intention in the UK: Results from the COVID-19 Vaccination Acceptability Study (CoVAccS), a nationally representative cross-sectional survey Human Vaccines & Immunotherapeutics 17 6 2021 1612 1621 10.1101/2020.08.13.20174045 33242386
12 Zhu X.-M. Yan W. Patterns and influencing factors of COVID-19 vaccination willingness among college students in China Vaccine 40 22 2022 3046 3054 10.1016/j.vaccine.2022.04.013 35450782
13 Kessels R. Luyten J. Tubeuf S. Willingness to get vaccinated against Covid-19 and attitudes toward vaccination in general Vaccine 39 33 2021 4716 4722 10.1016/j.vaccine.2021.05.069 34119349
14 Culotta A. Cutler J. Mining Brand Perceptions from Twitter Social Networks Marketing science 35 3 2016 343 362 10.1287/mksc.2015.0968
15 Martí P. Serrano-Estrada L. Nolasco-Cirugeda A. Social media data: Challenges, opportunities and limitations in urban studies Computers, Environment and Urban Systems 74 2019 161 174 10.1016/j.compenvurbsys.2018.11.001
16 Connochie D. Tingler R.C. Bauermeister J.A. Young men who have sex with men’s awareness, acceptability, and willingness to participate in HIV vaccine trials: Results from a nationwide online pilot study Vaccine 37 43 2019 6494 6499 10.1016/j.vaccine.2019.08.076 31522806
17 Mascaro V. Pileggi C. Currà A. Bianco A. Pavia M. HPV vaccination coverage and willingness to be vaccinated among 18–30 year-old students in Italy Vaccine 37 25 2019 3310 3316 10.1016/j.vaccine.2019.04.081 31064676
18 Wu S. Su J. Willingness to accept a future influenza A(H7N9) vaccine in Beijing China, Vaccine 36 4 2018 491 497 10.1016/j.vaccine.2017.12.008 29246476
19 Muric G. Wu Y. Ferrara E. COVID-19 vaccine hesitancy on social media: building a public twitter data set of antivaccine content, vaccine misinformation, and conspiracies JMIR public health and surveillance 7 11 2021 e30642 10.2196/30642 34653016
20 Salathé M. Khandelwal S. Assessing vaccination sentiments with online social media: implications for infectious disease dynamics and control PLoS computational biology 7 10 2011 e1002199 10.1371/journal.pcbi.1002199 22022249
21 Yousefinaghani S. Dara R. Mubareka S. Papadopoulos A. Sharif S. An analysis of COVID-19 vaccine sentiments and opinions on Twitter International Journal of Infectious Diseases 108 2021 256 262 10.1016/j.ijid.2021.05.059 34052407
22 Lim L.J. Lim A.J. Fong K.K. Lee C.G. Sentiments regarding COVID-19 vaccination among graduate students in Singapore Vaccines 9 10 2021 1141 10.3390/vaccines9101141 34696249
23 Monselise M. Chang C.-H. Ferreira G. Yang R. Yang C.C. Topics and sentiments of public concerns regarding COVID-19 vaccines: social media trend analysis Journal of Medical Internet Research 23 10 2021 e30765 10.2196/30765 34581682
24 Nurdeni D.A. Budi I. Santoso A.B. Sentiment analysis on Covid19 vaccines in Indonesia: from the perspective of Sinovac and Pfizer in: 2021 3rd East Indonesia Conference on Computer and Information Technology (EIConCIT) 2021 IEEE 122 127 10.1109/EIConCIT50028.2021.9431852
25 Marcec R. Likic R. Using twitter for sentiment analysis towards AstraZeneca/Oxford, Pfizer/BioNTech and Moderna COVID-19 vaccines Postgraduate Medical Journal 98 1161 2022 544 550 10.1136/postgradmedj-2021-140685 34373343
26 Zhang Z. Feng G. Xu J. Zhang Y. Li J. Huang J. Akinwunmi B. Zhang C.J. Ming W.-K. The impact of public health events on COVID-19 vaccine hesitancy on Chinese social media: national infoveillance study JMIR Public Health and Surveillance 7 11 2021 e32936 10.2196/32936 34591782
27 Yin H. Song X. Yang S. Li J. Sentiment analysis and topic modeling for COVID-19 vaccine discussions World Wide Web 25 3 2022 1067 1083 10.1007/s11280-022-01029-y 35250362
28 Brailovskaia J. Schneider S. Margraf J. To vaccinate or not to vaccinate!? Predictors of willingness to receive Covid-19 vaccination in Europe, the US, and China PloS one 16 12 2021 e0260230 10.1371/journal.pone.0260230 34851986
29 Chen M. Li Y. Chen J. Wen Z. Feng F. Zou H. Fu C. Chen L. Shu Y. Sun C. An online survey of the attitude and willingness of Chinese adults to receive COVID-19 vaccination Human Vaccines & Immunotherapeutics 17 7 2021 2279 2288 10.1080/21645515.2020.1853449 33522405
30 K.R. Nehal, L.M. Steendam, M. Campos Ponce, M. van der Hoeven, G.S.A. Smit, Worldwide vaccination willingness for COVID-19: a systematic review and meta-analysis, Vaccines 9 (10) (2021) 1071, doi:10.3390/vaccines9101071.
31 Shao W. Chen X. Zheng C. Wang G. Zhang B. Zhang W. Pneumococcal vaccination coverage and willingness in mainland China Tropical Medicine & International Health 27 10 2022 864 872 10.1111/tmi.13809 35942809
32 Markovič R. Šterk M. Marhl M. Perc M. Gosak M. Socio-demographic and health factors drive the epidemic progression and should guide vaccination strategies for best COVID-19 containment Results in Physics 26 2021 104433 10.1016/j.rinp.2021.104433 34123716
33 Ni L. Chen Y.-W. de Brujin O. Towards understanding socially influenced vaccination decision making: An integrated model of multiple criteria belief modelling and social network analysis European Journal of Operational Research 293 1 2021 276 289 10.1016/j.ejor.2020.12.011
34 Medhat W. Hassan A. Korashy H. Sentiment analysis algorithms and applications: A survey Ain Shams engineering journal 5 4 2014 1093 1113 10.1016/j.asej.2014.04.011
35 J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional transformers for language understanding, arXiv preprint arXiv:1810.04805Arxiv:1810.04805.
36 Hochreiter S. Schmidhuber J. Long short-term memory Neural computation 9 8 1997 1735 1780 10.1162/neco.1997.9.8.173 9377276
37 S. Albawi, T.A. Mohammed, S. Al-Zawi, Understanding of a convolutional neural network, in: 2017 international conference on engineering and technology (ICET), IEEE, 2017, pp. 1–6, doi:10.1109/ICEngTechnol.2017.8308186.
38 Page L. Brin S. Motwani R. Winograd T. The PageRank citation ranking: bringing order to the Web 1999 Stanford InfoLab Tech. rep.
39 Vaidya O.S. Kumar S. Analytic hierarchy process: An overview of applications European Journal of operational research 169 1 2006 1 29 10.1016/j.ejor.2004.04.028
40 Manguri K.H. Ramadhan R.N. Amin P.R.M. Twitter sentiment analysis on worldwide COVID-19 outbreaks Kurdistan Journal of Applied Research 5 3 2020 54 65 10.24017/covid.8
41 Praveen S. Ittamalla R. Deepak G. Analyzing the attitude of Indian citizens towards COVID-19 vaccine – A text analytics study Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 2 2021 595 599 10.1016/j.dsx.2021.02.031
42 Vayansky I. Kumar S.A. A review of topic modeling methods Information Systems 94 2020 101582 10.1016/j.is.2020.101582
43 Jelodar H. Wang Y. Yuan C. Feng X. Jiang X. Li Y. Zhao L. Latent Dirichlet allocation (LDA) and topic modeling: models, applications, a survey Multimedia Tools and Applications 78 11 2019 15169 15211 10.1007/s11042-018-6894-4
| 36502742 | PMC9724503 | NO-CC CODE | 2022-12-08 23:16:01 | no | Int J Med Inform. 2023 Feb 6; 170:104941 | utf-8 | Int J Med Inform | 2,022 | 10.1016/j.ijmedinf.2022.104941 | oa_other |
==== Front
Thermal Science and Engineering Progress
2451-9049
2451-9049
Elsevier Ltd.
S2451-9049(22)00403-6
10.1016/j.tsep.2022.101597
101597
Article
Comparison of adaptive thermal comfort with face masks in library building in Guangzhou, China
Tang Tianwei a
Zhou Xiaoqing a
Dai Kunquan a
Fang Zhaosong a⁎
Zheng Zhimin ab⁎
a School of Civil Engineering, Guangzhou University, Guangzhou 510006, China
b Academy of Building Energy Efficiency of Guangzhou University, Guangzhou University, Guangzhou 510006, China
⁎ Corresponding authors.
6 12 2022
6 12 2022
10159722 8 2022
14 10 2022
1 12 2022
© 2022 Elsevier Ltd. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
During the COVID-19 pandemic, wearing masks in public spaces has become a protective strategy. Field tests and questionnaire surveys were carried out at a university library in Guangzhou, China, during June 2021 and January 2022. The indoor environmental parameters were observed, thermal sensation votes of students on various environmental parameters were collected, symptoms of students wearing masks were quantified, and the appropriate amount of time to wear masks was established. To identify acceptable and comfortable temperature ranges, the relationship between thermal sensation and thermal index was investigated. During summer and winter, people wearing masks are symptomatic for a certain duration. The most frequently voted symptom was facial heat (62.7% and 54.6% during summer and winter, respectively), followed by dyspnea. During summer, more than 80% of the participants subjects were uncomfortable and showed some symptoms after wearing masks for more than 2 h (3 h during winter). In the summer air conditioning environment in Guangzhou, the neutral Top was 26.4 °C, and the comfortable Top range was 25.1–27.7 °C. Under the natural ventilation environment in winter, the neutral Top was 20.5 °C, and the comfortable Top range was 18.5–22.5 °C. This study may provide guidance for indoor office work and learning to wear masks in Guangzhou.
Keywords
University library
Masks
Human body
Thermal sensation
Operative temperature
==== Body
pmcIntroduction
The World Health Organization (WHO) designated the COVID-19 outbreak a pandemic on March 11, 2020. [1]. In the United States, SARS-CoV-2 had infected more than 8 million kids as of January 2022 [2]. The COVID-19 burden among children and adolescents has increased due to new SARS-CoV-2 variant strains [3], [4], [5]. Other personal protective techniques, like as mask wearing, quarantine, vaccination, and hand cleanliness, played a significant part in epidemic mitigation in addition to public containment and closure laws [6], [7], [8], [9], [10], [11], [12], [13]. The most efficient and affordable way to stop human-to-human virus transmission and control the COVID-19 outbreak is to wear masks in public [14]. It has become customary to wear masks while traveling, at work, and while in school [15]. Some scholars also emphasize the discomfort caused by masks [16], [17], [18], [19]. Therefore, it is necessary to thoroughly investigate the effect of masks on human thermal comfort.
Literature review
Despite their recognized benefits in protecting and insulating against toxins and viruses, masks can have side effects due to the microclimate of hot and thick humid air they cause [20], [21], [22]; additionally, wearing a mask causes significant discomfort and breathing difficulties in most people [23]. Furthermore, the temperature of the air within the mask has a significant impact on human thermal sensation [24]. Davey et al. [25] described heat-related illness symptoms in healthcare workers (e.g., 40.2% dizziness, 63.4% fatigue, 79.0% headache, and 54.5% profuse sweating) and heat stress, which impairs cognitive and physical performance. Tang et al. [19] described symptoms among students in summer air conditioning conditions; 62.7% and 25.4% of the subjects voted for facial heat and dyspnea as the most commonly observed symptoms, respectively. Some subjects who wore masks for a long time experienced rapid heartbeats (9.1%) and nausea (4.1%). Peres et al. [26] established that masks were associated with discomfort (26.8%), and affected task performance (18.9%) and communication (40.9%). Another research examined at how N95 masks and medical masks create significantly different microclimates, which have a significant impact on heart rate, thermal stress, and subjective perception of discomfort. [27]. Therefore, while people should pay attention to their own protection, they should strengthen their personal health.
Several previous studies have already been conducted to improve the indoor environment and human thermal comfort [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38]. Since these investigations were carried out before the pandemic, the impact of masks was not examined. But ever since the pandemic was announced, it is now necessary to wear a mask in public to avoid contracting COVID-19 [6], [39], [40], [41], [42]. During summer and winter, the comfort range of the environment varies [30], [34], [37]. Consequently, further attention and research on mask-wearing adaptability are needed.
Research objective
A library survey was conducted during summer and winter. The study’ objectives were as follows:
(1) The physical discomfort caused by wearing masks during summer and winter was analyzed.
(2) The correlation between the duration of wearing masks and human physiological symptoms was analyzed, and suggestions were made regarding the duration of wearing masks.
(3) The thermal index and comfort range when there are two situations simultaneously were calculated (with masks or without masks).
Methods
Research environment
Guangzhou is located in South China. This research was conducted at the library of Guangzhou University (Fig. 1 ). The study was done in June 2021 and January 2022. It was conducted for 10 days during the summer and 8 days during the winter (Table 1 ). The weather information for the days is shown in Fig 2 , based on data from the Guangzhou Weather Station. During the investigation, the relative humidity ranged from 78% to 95% and 30% and 56% in June 2021 and January 2022, respectively. The average outdoor air temperature in Guangzhou reached 30 °C and 12 °C in June 2021 and January 2022, respectively.Fig. 1 Location of the testing site.
Table 1 Data of outdoor environment
Area Date Test days Outdoor temperature range (°C) Outdoor relative humidity range (%)
Guangzhou June 2021 10 24.5 – 36.5 78 – 95
Guangzhou January 2022 8 3.0 – 24.5 30 – 56
Fig. 2 The outdoor environment data in Guangzhou
Subjects and questionnaire
These experiments were carried out in accordance with the Helsinki Declaration's ethical standards, and informed consent was obtained from all the participants. We asked them about their health status and screened them. In addition, all subjects were requested to stay in the library for more than 30 min to ensure that the subjects were in thermal equilibrium [43], [44], [45], [46], [47]. This survey's participants were health students. Except for any discomfort with the mask, they had to indicate their health status before answering the questionnaire.
In the first section, the questionnaire used investigated subjective votes on the thermal sensation, air movement sensation, and humidity sensation. The subjective vote scales were based on the thermal environment comfort levels described in the ASHRAE Handbook [48] and ISO 7726 [49]. As shown in Fig. 3 , the thermal sensation scale was as follows: −3, cold; −2, cool; −1, slightly cool; 0, neutral; 1, slightly warm; 2, warm; and 3, hot. The second section of the questionnaire investigated whether students wore masks and any symptoms of discomfort, as well as the duration of mask use. Meanwhile, we suggested that subjects consider mask-wearing as the only factor causing discomfort when filling in the questionnaire. The questionnaire content is presented in the Appendix A.Fig. 3 Subjective feeling voting scale (a) and students wore masks and any symptoms of discomfort (b).
As shown in Table 2 , a total of 2,643 valid questionnaires were collected for this study, 1,602 in the summer and 1,041 in the winter. The average ages of the subjects were 20.7 years during the summer and 20.2 years during the winter. The clothing insulation values were calculated based on ASHRAE standard 55 [50]. the average total clothing insulation (I cl) was 0.39 clo in summer and 0.88 clo in winter.Table 2 Subject anthropometric data (SD: standard deviation)
Sex Number Age (years) (SD) Height (m) (SD) Weight (kg) (SD) Body surface area (m2) (SD) Icl (clo) (SD)
Male Summer: 550 20.9 (1.51) 1.73 (0.06) 62.7 (7.95) 1.74 (0.12) 0.35 (0.11)
Winter: 399 20.4 (1.93) 1.73 (0.06) 62.4 (8.69) 1.73 (0.13) 0.83 (0.21)
Female Summer: 1,052 20.6 (1.57) 1.61 (0.05) 50.2 (6.02) 1.50 (0.10) 0.41 (0.12)
Winter: 642 20.0 (2.43) 1.61 (0.05) 49.9 (6.01) 1.49 (0.10) 0.91 (0.19)
Total Summer: 1,602 20.7 (1.56) 1.65 (0.08) 54.5 (9.00) 1.58 (0.15) 0.39 (0.12)
Winter: 1,041 20.2 (2.25) 1.66 (0.08) 54.7 (9.43) 1.59 (0.16) 0.88 (0.22)
Body surface area (A): A = 0.202 w0.424 h0.725[38].
Measured parameters and instruments
According to the ISO 7726 guidelines [49], test instruments were placed near the seats at a height of 1.1 m [51], [52], [53], [54], [55], [56], [57], [58], [59], [60]. A matt black standard globe thermometer with a diameter (D) of 0.15 m (globe emissivity, εg = 0.95) was used to measure the globe bulb temperature. The Mean radiant temperature (Tmrt) was calculated using Eq. (1) [49]:(1) Tmrt=(Tg+273)4+(1.1×108×Va0.6)(εg×D0.4)×(Tg-Ta)1/4-273
where Ta is the air temperature, Tg is the globe temperature, and Va is the air velocity.
Data processing
The Standard Effective Temperature (SET*) was calculated using the CBE Thermal Comfort Tool (http://comfort.cbe.berkeley.edu/). In this study, using SET* can help to reduce the impact of the inability to control clothing [61], [62].
The operative temperature (Top) was calculated according to the conditions presented in ASHRAE Standard 55 normative [50]. The T op was calculated using Eq. (2) [50] as follows:(2) Top=ATa+1-ATmrt,
where A is determined according to Table 3 as a function of relative air speed [50].Table 3 Relationship between A and Va.
Va <0.2 m/s 0.2–0.6 m/s 0.6–1.0 m/s
A 0.5 0.6 0.7
Linear regression was used to analyze the relationship between the environmental parameters and responses to the subjective questionnaire to determine the neutral temperature (when MTSV = 0) and comfort temperature range (when MTSV = ± 0.5). All statistical analyses were performed using IBM SPSS Statistics 25 (IBM Inc., Armonk, NY, USA), and Origin 2021 (Origin Lab Corporation, Northampton, MA, USA), including the fitting of linear regression equations and the calculation of linear regression correlation index R2 and the independent sample t-tests.
Results
Thermal parameters
The measured indoor thermal parameters (Ta, RH, Va, and Tmrt) are summarized in Table 4 . The mean Ta, RH, and Va were 27.7 °C, 79.2%, and 0.17 m/s during the summer, respectively, and 22.5 °C, 66.3%, and 0.04 m/s during the winter, respectively. Based on the calculations, the mean values of Top and SET* were 27.6 °C and 20.2 °C during the summer and 26.6 °C and 21.1 °C during the winter, respectively. Because the average Va was less than 0.2 m/s, there was almost no blowing sense for the human body, which satisfies the advised value of Va in the ASHRAE Standard 55 [50].Table 4 Indoor thermal parameters (SD: standard deviation)
Seasons Variables (units) Mean Minimum Maximum SD
Summer Ta (°C) 27.7 25.9 31.1 1.15
Tg (°C) 27.5 25.6 31.3 1.05
RH (%) 79.2 74.1 85.2 1.98
Va (m/s) 0.17 0.01 0.80 0.16
Tmrt (°C) 27.6 25.8 31.4 1.04
Top (°C) 27.6 25.9 31.2 1.10
SET* (°C) 26.6 21.8 33.9 1.78
Winter Ta (°C) 20.5 16.3 22.3 0.94
Tg (°C) 20.1 16.4 21.5 0.54
RH (%) 66.3 58.0 71.5 3.85
Va (m/s) 0.04 0.01 0.44 0.08
Tmrt (°C) 20.2 15.7 22.6 0.97
Top (°C) 20.2 16.3 22.4 0.93
SET* (°C) 21.1 16.4 25.3 1.24
Effect of wearing masks on human comfort
Among the 2,643 questionnaires collected, subjects of 1,683 questionnaires who wore masks were included (1,112 in summer and 571 in winter). Of which, wearing masks made 1,239 (73.6%) of the participants uncomfortable. The most frequently voted symptom was facial heat (62.7% in summer, and 54.6% in winter) as shown in Fig. 4 , followed by dyspnea (25.4% in summer, and 35.2% in winter). All other symptoms were observed in <10% of patients. When compared to summer, the percentage of subjects experiencing dyspnea increased while the percentage of subjects experiencing facial heat decreased during the winter However, wearing masks still caused more than 50% of the subjects to feel facial heat. Wearing masks affects the comfort of the human face. In addition, compared to summer, the percentage of subjects wearing masks with dyspnea in winter increased. One of the possible reasons was that the wind speed in the indoor environment in winter is lower than that in summer, and the gas exhaled by the human body may stay in the mask for a long time.Fig. 4 The distribution of symptoms among participants wearing masks
During the test, the subjects evaluated the environmental air quality. A total of 85.7% and 91.6% of the participants believed that the air quality in the library was acceptable during summer and winter, respectively (as shown in Fig. 5 ). For participants wearing masks, discomfort and symptoms caused by air quality problems were avoided, which was confused with the impact of masks. In addition, 35.5% of the subjects (21.2% wearing masks) felt stuffy indoors in summer, indicating that the airflow speed in the environment was occasionally low during the test.Fig. 5 Evaluation of the indoor air quality
Acceptable duration of wearing masks
More than 75% (78%) of participants were anticipated to wear masks for 2.0 h or less throughout the summer (winter), based on the voting statistical distribution of respondents' acceptable time of wearing masks. As shown in Fig. 6 (Left–Y), only a few subjects could tolerate wearing masks for more than 3 hours. The acceptable duration of wearing masks increased in the winter when compared to the summer. Within a certain period, the mask can keep the face warm and reduce the heat loss caused by the ambient cold air. The longer the mask was worn, the wetter the face became, reducing the comfort of the human body. After exceeding their “acceptable duration,” the majority of participants experienced increased physical discomfort, impacting their work and learning efficiency. As shown in Fig. 6 (Right–Y), the percentage of no symptoms and the duration of wearing the mask have a good linear relationship. More than 80% of the subjects were uncomfortable and showed some symptoms after wearing masks for more than 2 h in summer and 3 h in winter.Fig. 6 Distribution of percent “acceptable duration” (Left -Y) and the relationship between the “percentage of no symptom” and duration (Right -Y) for wearing masks.
Distribution of TSV
As shown in Fig. 7 (a–b), the thermal sensation vote (TSV) results for the whole body and the face indicated that wearing a mask had some effect on the human. The percentage of subjects wearing masks who reported a TSV greater than zero was approximately 6.7% (7.2%) greater than the proportion of subjects without masks during the summer (winter). For the face, the percentage of TSV greater than 0 also increased (13.4% in summer and 9.4% in summer). During the winter, wearing masks can improve the comfort of the human face to some extent, with the percentage of TSV less than 0 reduced by 2.4% and 1.4% during summer and winter (without masks vs. with masks), respectively. There was no obvious effects observed during the summer. The percentage of subjects wearing masks decreased at TSV = 0 (vs. without masks). There was increased impact on face comfort during summer (10.9% less vs. winter 7.9%). During the summer, wearing a mask had a certain effect on thermal sensation in the head and chest, with thermal sensation shifting to greater than 0 (Fig. 7c–d), whereas wearing a mask had no obvious effect during winter. In addition, wearing masks did not have a significant effect on the human back and limbs (Fig. 7e–f).Fig. 7 Percentage distribution of TSV in the library: (a) whole; (b) face; (c) head; (d) chest; (e) back; (f) limbs.
Comparison of the whole thermal sensation and local thermal sensation
The results of the regression analyses of local and thermal sensations are shown in Table 5 . TSV on the head and face had a significant impact on the whole TSV. There is a significant relationship between the whole and local thermal sensations because the entire thermal sensation can be seen as the integration of each local thermal sensation signal via brain control [38]. These models are shown in Eq. (3) to Eq. (8). As can be seen from Eq. (5) and Eq. (8), wearing masks had an effect on the thermal comfort of the face, head, and chest in summer and on the thermal comfort of the face in winter. During colder seasons, masks may increase comfort to some extent and protect the face from the cold air.Table 5 Analysis of whole thermal sensations and local thermal sensation.
Season TSV (Whole) Face Head Back Chest Limbs Constant
Summer With masks Coef. 0.351 0.436 0.224 0.164 0.022 –0.048
P 0.001 0.001 0.001 0.001 0.01
Without masks Coef. 0.301 0.384 0.229 0.109 0.028 –0.086
P 0.001 0.001 0.001 0.001 0.01
ΔTSV Coef. 0.05 0.042 –0.005 0.055 –0.006 0.038
Winter With masks Coef. 0.266 0.342 0.188 0.163 0.132 –0.053
P 0.001 0.001 0.001 0.001 0.01
Without masks Coef. 0.205 0.339 0.184 0.166 0.138 –0.084
P 0.001 0.001 0.001 0.001 0.01
ΔTSV Coef. 0.061 0.003 0.004 –0.003 –0.006 0.031
Summer:
With masks:(3) TSVwhole=0.351TSVface+0.436TSVhead+0.224TSVback+0.164TSVchest+0.022TSVlibms--0.048.
Without masks:(4) TSVwhole=0.301TSVface+0.384TSVhead+0.229TSVback+0.109TSVchest+0.028TSVlibms--0.086
(3) - (4):
(5) ΔTSVwhole=0.05TSVface+0.022TSVhead+0.055TSVchest+0.038
Winter:
With masks:(6) TSVwhole=0.266TSVface+0.342TSVhead+0.188TSVback+0.163TSVchest+0.132TSVlibms--0.054
Without masks:(7) TSVwhole=0.205TSVface+0.339TSVhead+0.184TSVback+0.166TSVchest+0.138TSVlibms--0.084
(6) - (7):
(8) ΔTSVwhole=0.061TSVface+0.031
Effect of wearing mask on thermal preference
As shown in Table 4, the mean Va in the library was relatively low during the summer and winter. However, the subjects expected more Va during the summer. The humidity level was deemed acceptable by the majority of participants. The subjects' ability to adapt to high relative humidity in South China was the primary reason [63], [64], [65]. The thermal preference of the participants for environmental factors is shown in Fig. 8 for wearing masks and without masks. In the summer, more than half of the subjects preferred to lower the operative temperature to improve thermal comfort, especially those wearing masks. Thus, the effects of masks on thermal comfort are significant. During winter, approximately 40% of the subjects expected the Top to rise. Therefore, to improve the thermal comfort, the Top must be reduced in summer and increased in winter.Fig. 8 Percentage thermal preference distribution for environmental parameters (−1, lower; 0, no change; and +1, higher). (a) summer; (b) winter; (A) without masks; (B) wearing masks.
Correlation analysis of MTSV and Top/SET*
Within 1 °C intervals from T op/SET*, the mean thermal sensation vote (MTSV) was computed. The neutral T op/SET* was determined by using a regression equation. As shown in Table 6 and Fig. 9 , during the summer, for subjects without masks, when MTSV= 0, T op = 26.5 °C and SET* = 25.3 °C. For subjects with masks, when MTSV = 0, T op = 26.2 °C and SET* = 25.0 °C. For subjects with masks, the neutral T op/SET* of the environment was 0.3 °C lower than it was for subjects without mask. With little T op/SET* difference between the two conditions, the mean value of 26.4/25.3 °C was used as the neutral T op/SET*. Similarly, the MTSV distributions of subjects with and without masks during winter were calculated. Neutral T op (SET*) was 20.5 °C (23.4 °C). The blue box shows the proportion of respondents with no symptoms at each temperature gradient. During the summer (winter), when T op/SET* was 26.4/24.2 °C (22.3/20.0 °C), the proportion of respondents with no symptoms reached 50%. Therefore, subjects wearing masks preferred a lower temperature environment. This temperature difference is extremely small; therefore, it can be overlooked. However, wearing masks for a long period can cause symptoms in humans. However, it has little impact on the neutral temperature of the environment and whole-body heat balance, which is consistent with some previous studies [24], [66]. Consequently, more consideration needs to be given to the duration of wearing masks and health problems.Table 6 Regression equation between Top and MTSV
Condition Equation Neutral Top (°C) Comfort temperature range (°C)
Summer Without masks MTSV = 0.359Top - 9.51 26.5 25.1–27.8
With masks MTSV = 0.398Top - 10.49 26.2 25.0–27.7
Total MTSV = 0.378Top - 10.00 26.4 25.1–27.7
Winter Without masks MTSV = 0.242Top - 5.00 20.5 18.5–22.5
With masks MTSV = 0.245Top - 5.02 20.5 18.5–22.5
Total MTSV = 0.244Top - 5.01 20.5 18.5–22.5
Fig. 9 Relationship between Top/SET* and MTSV (Left -Y), Relationship between Top/SET* and the percentage of no symptomatic subjects (Right -Y): (a) Top; and (b) SET*.
Discussion
The effect of wearing masks on the human health
Adverse effects of mask use have been reported in both healthcare workers and the general public [67], including headaches [68], increased thermal discomfort [67], potential thermal physiological responses [69], and decreased quality of work [68], [70]. Masks can raise the temperature of the skin on the face as well as the heat or moisture of the inhaled air [71]. The inner layer of a long-wearing mask becomes wet due to condensation of water vapor generated by breathing and sweat evaporation [18], [24]. Such elements have been proposed to be responsible for the increased respiratory discomfort when masks are used [72], [73]. The masks were tightly attached around the participants’ face and may have collapsed, potentially increasing dyspnea [71]. Long-term mask use has been linked to an increase in subjective visual complaints [74]. According to an online survey [75], 18.3% of respondents reported having dry eye issues, particularly women and people who wore glasses or contacts. Wearing masks for an extended period of time causes secondary complications, particularly in vulnerable populations, because airflow is restricted, resulting in a high concentration of CO2 in the body [76], [77]. The amount of CO2 in the air affects the blood's pH, which can lead to numerous health risks like vertigo, dyspnea, headache and hypoxia when the level is elevated [78]. Therefore, the effect of masks on breathing cannot be overlooked. Masks should also be removed in time to eliminate adverse effects during learning. Masks typically influence the face and head of the human body, increasing the need for overall environment comfort, especially when wearing the mask for an extended period of time [19]. Long-term usage of a moist mask might also result in facial irritation and discomfort [19], [25], [26]. Therefore, the masks must be replaced regularly. This is one of the reasons for wearing a mask for no more than 2 h.
Discussion of the whole thermal sensation and local thermal sensation
There have been a number of studies that have conducted a series of experiments on the effect of local heat sensation on whole heat sensation [79], [80], [81], [82], [83]. Zhang [79] found in the study that as the gap between the local thermal sensation of the site and the thermal sensation of the whole body increased, the weight of the influence of the site on the sensation of the whole body increased linearly. Zhang et al. [80], [81], [82] developed several models to predict local and overall thermal sensation and thermal comfort in the human body. For body parts with the same skin temperature, local sensation is much warmer during the cold tests when the whole body is cold, and much colder during the warm tests when the whole body is warm [80]. Some body parts strongly influence overall thermal sensation [81]. The effect of thermal sensation on the overall thermal sensation of different parts is usually expressed in terms of the weight of the influence of thermal sensation in different parts on the sensation of heat in the whole body, and this coefficient is mostly obtained by linear regression [84]. Therefore, multiple linear regression was used to calculate the local coefficients. By comparing the two types of TSV models (without or with masks), the main effects of wearing masks in summer and winter were calculated, and Eq. (5) and Eq. (8) were obtained. Wearing masks had a significant effect on the head, face, and chest, similar to local thermal sensation vote (Section 3.4), and consistent with the mechanistic analysis in Section 4.1.
Comparison of MTSV models
The ASHRAE Standard 55 [50] provides a graphical comfort zone method for indoor environments. In the comfort zone, the activity level was maintained between 1.0 and 1.3 m, and clothing insulation was set at 0.5 clo in the summer and 1.0 clo in the winter. As shown in Fig. 10 , all data in this study were completely outside the comfort zone of ASHRAE Standard 55 in summer. More than half of the data were within the comfort zone in winter.Fig. 10 Data plotted onto the PMV−PPD index-based chart.
Considering the adaptability to humidity in the Guangzhou area, combined with the comfort top interval in Table 5, the comfort areas during summer and winter are shown in Fig. 10. It is evident that most of the data during winter are in the comfort band. During summer, the air temperature on the fifth floor of the library was warmer than it was on the lower floors due to some spots receiving direct sunlight. By analyzing the regression equation, the upper limit of acceptable T op in summer was 29.0 °C. Therefore, when considering the impact of radiation on indoor air temperature, increasing the wind speed or shading is necessary. Owing to the high temperature of the library and the small number of students, this greatly wastes space resources and even leads to further crowding on the lower floors.
Limitations
Many factors affect human health and comfort, such as the duration of wearing masks, environmental parameters, and air quality. However, because of the limitations of field investigations, these factors cannot be controlled. For example, during the test period of this study, the humidity range was small and the impact of humidity on the population wearing masks could not be considered. Therefore, the influence of each factor can be considered only while analyzing the human body. In addition, the mask material has a significant impact on breathing [17], [27], [85], which requires further research. Some of the current conclusions are obtained through field investigation. In winter, wearing a mask for a certain period of time will improve thermal comfort. The mask reduced the heat exchange between the face and around environment. However, from the field survey, some subjects still felt that their face had a fever, which was partly related to the temperature and the duration of wearing a mask. In winter, the indoor air temperature is affected by the outdoor environment significantly. The relevant tests will be conducted in a chamber. Therefore, future simulation experiments can be conducted in an experimental chamber, combining multiple factors for analyzing the impact of wearing masks on human health, and verifying and analyzing the results of this study.
Conclusions
Field tests and questionnaire surveys were carried out in a campus library in Guangzhou, China, in June 2021 and January 2022. By analyzing the relationship between the duration of wearing masks and symptoms and the relationship between the relevant thermal index and TSV, the following results were obtained:(1). People wearing masks show symptoms for a certain duration. Wearing masks for more than 2 hours, and the percentage of symptoms is more than 50%. The most frequently voted symptom was facial heat (62.7% in summer, and 54.6% in winter), followed by dyspnea (25.4% in summer, and 35.2% in winter). Timely replacement of masks and wearing a mask for a maximum of 2 h at once are recommended.
(2). The mask significantly influences the facial thermal comfort, but it has no obvious influence on the whole body. Wearing a mask for a long time also has varying degrees of impact on the comfort of the head and chest. In summer: ΔTSVwhole = 0.05TSVface + 0.022TSVhead + 0.055TSVchest + 0.038 ; In winter: ΔTSVwhole=0.061TSVface+ 0.031.
(3). In the summer air conditioning environment in Guangzhou, the neutral T op was 26.4 °C, and the comfortable T op range was 25.1–27.7 °C. Under the natural ventilation environment during winter, the neutral T op was 20.5 °C, and the comfortable T op range was 18.5–22.5 °C.
CRediT authorship contribution statement
Tianwei Tang: Data curation, Writing – original draft. Xiaoqing Zhou: Supervision. Kunquan Dai: Data curation. Zhaosong Fang: Conceptualization, Writing – review & editing. Zhimin Zheng: Methodology, Data curation.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A
,
,
Appendix B Instruments used to measure the environmental parametersEquipment Thermal comfort instrument Universal wind speed recorder
Model SSDZY-1 WFWZY-1
Parameter Ta (°C) RH (%) Tg (°C) Va (m/s)
Measuring range -20–80 °C 0.01–99.9% -20–80 °C 0.05–5.00 m/s
Accuracy ± 0.3 °C ± 2% ± 0.3 °C 5% ± 0.05 m/s
Sampling rate 30 s 30 s 30 s 30 s
Data availability
The authors do not have permission to share data.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Project No. 51978180), GuangDong Basic and Applied Basic Research Foundation (2021A1515011671). The authors express gratitude to all the subjects who participated in the survey.
==== Refs
References
1 WHO, WHO director-General’s opening remarks at the media briefing on COVID-19, March 11, 2020. https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---11-march-2020.
2 American Academy of Pediatrics, Children’s Hospital Association. Children and COVID-19: State Data Report. 2022 Jan 6. https://www.aap.org/en/pages/2019-novel-coronavirus-covid-19-infections/children-and-covid-19-state-level-data-report/ (accessed 2022 January 13).
3 M.J. Delahoy, D. Ujamaa, M. Whitaker, et al., Hospitalizations associated with COVID-19 among children and adolescents—COVID-NET, MMWR Morb. Mortal. Wkly Rep 14 states, March 1, 2020–August 14, 2021. 70 (2021) 1255-1260. https://www.cdc.gov/mmwr/volumes/70/wr/mm7036e2.htm.
4 Food and Drug Administration, FDA authorizes Pfizer-BioNTech COVID-19 vaccine for emergency use in children 5 through 11 years of age. October 29, 2021. https://www.fda.gov/news-events/press-announcements/fda-authorizes-pfizer-biontech-covid-19-vaccine-emergency-use-children-5-through-11-years-age. (accessed December 31, 2021).
5 Food and Drug Administration, Coronavirus (COVID-19) update: FDA authorizes Pfizer-BioNTech COVID-19 vaccine for emergency use in adolescents in another important action in fight against pandemic. May 10, 2021. https://www.fda.gov/news-events/press-announcements/coronavirus-covid-19-update-fda-authorizes-pfizer-biontech-covid-19-vaccine-emergency-use. (accessed July 16, 2021).
6 Chu D.K. Akl E.A. Duda S. Solo K. Yaacoub S. Schünemann H.J. COVID-19 Systematic Urgent Review Group Effort (SURGE) study authors, Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis Lancet 395 2020 1973 1987 10.1016/S0140-6736(20)31142-9 32497510
7 Nussbaumer-Streit B. Mayr V. Dobrescu A. Quarantine alone or in combination with other public health measures to control COVID-19: a rapid review Cochrane Database Syst Rev 9 2020 article CD013574
8 Hellewell J. Abbott S. Gimma A. Feasibility of controlling COVID-19 outbreaks by isolation of cases and contacts Lancet Glob Health 8 2020 e488 e496 32119825
9 L. Ferretti, C. Wymant, M. Kendall, et al., Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing, Science 368 (2020) article eabb6936. https://doi.org/10.1126/science.abb6936.
10 Sun J. Zheng Y. Liang W. Quantifying the effect of public activity intervention policies on COVID-19 pandemic containment using epidemiologic data from 145 countries Value Health 25 2022 699 708 10.1016/j.jval.2021.10.007 35500944
11 Bourouiba L. Turbulent gas clouds and respiratory pathogen emissions: potential implications for reducing transmission of COVID-19 JAMA 323 2020 1837 1838 10.1001/jama.2020.4756 32215590
12 Moore J.P. Offit P.A. SARS-CoV-2 vaccines and the growing threat of viral variants JAMA 325 2021 821 822 10.1001/jama.2021.1114 33507218
13 S.E. Oliver, J.W. Gargano, M. Marin, et al., The Advisory Committee on Immunization Practices’ interim recommendation for use of Pfizer-BioNTech COVID-19 vaccine - United States, December 2020, MMWR Morb. Mortal. Wkly Rep. 69 (2020) 1922-1924. https://doi.org/10.15585/mmwr.mm6950e2.
14 Zhang R. Li Y. Zhang A.L. Wang Y. Molina M.J. Identifying airborne transmission as the dominant route for the spread of COVID-19 Proc Natl Acad Sci U S A 117 2020 14857 14863 10.1073/pnas.2009637117 32527856
15 Cheng V.C.C. Wong S.C. Chuang V.W.M. The role of community-wide wearing of face mask for control of coronavirus disease, (COVID-19) epidemic due to SARS-CoV-2 J Infect 81 2020 2019 107 114 10.1016/j.jinf.2020.04.024
16 Liao M.R. Liu H.Y. Wang X. A technical review of face mask wearing in preventing respiratory COVID-19 transmission Curr Opin Colloid Interface Sci 52 2021 101417 10.1016/j.cocis.2021.101417
17 Suen W.S. Huang G. Kang Z. Gu Y. Fan J. Shou D. Development of wearable air-conditioned mask for personal thermal management Build Environ 205 2021 108236 10.1016/j.buildenv.2021.108236
18 White M.K. Hodous T.K. Vercruyssen M. Effects of thermal environment and chemical protective clothing on work tolerance, physiological responses, and subjective ratings Ergonomics 34 1991 445 457 10.1080/00140139108967328 1860463
19 Tang T. Zhu Y. Zhou X. Investigation of the effects of face masks on thermal comfort in Guangzhou China, Build Environ 214 2022 108932 10.1016/j.buildenv.2022.108932
20 F. Cascini, I. Hoxhaj, D. Zaçe, M. Ferranti, M.L. Di Pietro, S. Boccia, W. Ricciardi, How health systems approached respiratory viral pandemics over time: a systematic review, BMJ Glob Health 5 (2020) article e003677. https://doi.org/10.1136/bmjgh-2020-003677.
21 European Centre for Disease Prevention and Control, Coronavirus Disease, European Center for Disease Prevention and Control, Stockholm, 2019 (COVID-19) in the EU/EEA and the UK – Ninth Update, 23 April 2020.
22 S. Bandyopadhyay, R.E. Baticulon, M. Kadhum, et al., Infection and mortality of healthcare workers worldwide from COVID-19: a systematic review, BMJ Glob Health 5 (2020) article e003097. https://doi.org/10.1136/bmjgh-2020-003097.
23 Lai A.C.K. Poon C.K.M. Cheung A.C.T. Effectiveness of facemasks to reduce exposure hazards for airborne infections among general populations J R Soc Interface 9 2012 938 948 10.1098/rsif.2011.0537 21937487
24 Nielsen R. Berglund L.G. Gwosdow A.R. Dubois A.B. Thermal sensation of the body as influenced by the thermal microclimate in a face mask Ergonomics 30 1987 1689 1703 10.1080/00140138708966058 3443092
25 Davey S.L. Lee B.J. Robbins T. Randeva H. Thake C.D. Heat stress and PPE during COVID-19: impact on healthcare workers’ performance, safety and well-being in NHS settings J Hosp Infect 108 2021 185 188 10.1016/j.jhin.2020.11.027 33301841
26 Peres D. Monteiro J. Boléo-Tomé J. Medical masks’ and respirators’ pattern of use, adverse effects and errors among Portuguese health care professionals during the COVID-19 pandemic: a cross-sectional study Am J Infect Control 50 2022 618 623 10.1016/j.ajic.2021.10.002 34653529
27 Li Y. Tokura H. Guo Y.P. Wong A.S. Wong T. Chung J. Newton E. Effects of wearing N95 and surgical facemasks on heart rate, thermal stress and subjective sensations Int Arch Occup Environ Health 78 2005 501 509 10.1007/s00420-004-0584-4 15918037
28 Gong P. Y.i. Cai, Z. Zhou, C. Zhang, B. Chen, S. Sharples, Investigating spatial impact on indoor personal thermal comfort J Build Eng 45 2022 103536 10.1016/j.jobe.2021.103536
29 Zhou C. Fang Z. Xu X. Zhang X. Ding Y. Jiang X. Ji Y. Using long short-term memory networks to predict energy consumption of air-conditioning systems Sustain Cities Soc 55 2020 102000 10.1016/j.scs.2019.102000
30 Jiang J. Wang D. Liu Y. Di Y. Liu J. A holistic approach to the evaluation of the indoor temperature based on thermal comfort and learning performance Build Environ 196 2021 107803 10.1016/j.buildenv.2021.107803
31 Cheng F. Wu Y. Gao S. Liao C. Cheng Y. Experimental study of thermal comfort in a field environment chamber with stratum ventilation system in winter Build Environ 207 2022 108445 10.1016/j.buildenv.2021.108445
32 Zhang S. Ai Z. Lin Z. Occupancy-aided ventilation for both airborne infection risk control and work productivity Build Environ 188 2021 107506 10.1016/j.buildenv.2020.107506
33 Guevara G. Soriano G. Mino-Rodriguez I.M. Thermal comfort in university classrooms: an experimental study in the tropics Build Environ 187 2021 107430 10.1016/j.buildenv.2020.107430
34 Wang H. Hu S. Liu G. Li A. Experimental study of human thermal sensation under hypobaric conditions in winter clothes Energy Build 42 2010 2044 2048 10.1016/j.enbuild.2010.06.013
35 Fang Z. Zhang S. Cheng Y. Fong A.M.L. Oladokun M.O. Lin Z. Wu H. Field study on adaptive thermal comfort in typical air conditioned classrooms Build Environ 133 2018 73 82 10.1016/J.BUILDENV.2018.02.005
36 Dhaka S. Mathur J. Quantification of thermal adaptation in air-conditioned buildings of composite climate, India Build Environ 112 2017 296 307 10.1016/j.buildenv.2016.11.035
37 Singh M.K. Kumar S. Ooka R. Rijal H.B. Gupta G. Kumar A. Status of thermal comfort in naturally ventilated classrooms during the summer season in the composite climate of India Build Environ 128 2018 287 304 10.1016/j.buildenv.2017.11.031
38 Fang Z. Liu H. Li B. Tan M. Olaide O.M. Experimental investigation on thermal comfort model between local thermal sensation and overall thermal sensation Energy Build 158 2018 1286 1295
39 Lotfi M. Hamblin M.R. Rezaei N. COVID-19: transmission, prevention, and potential therapeutic opportunities Clin Chim Acta 508 2020 254 266 10.1016/j.cca.2020.05.044 32474009
40 Wang X. Pan Z. Cheng Z. Association between 2019-nCoV transmission and N95 respirator use J Hosp Infect 105 2020 104 105 10.1016/j.jhin.2020.02.021 32142885
41 Rader B. White L.F. Burns M.R. Mask-wearing and control of SARS-CoV-2 transmission in the USA: a cross-sectional study, Lancet Digit Health 3 2021 e148 e157 10.1016/S2589-7500(20)30293-4
42 Ma Q.X. Shan H. Zhang H.L. Li G.M. Yang R.M. Chen J.M. Potential utilities of mask-wearing and instant hand hygiene for fighting SARS-CoV-2 J Med Virol 92 2020 1567 1571 10.1002/jmv.25805 32232986
43 Liu H. Liao J. Yang D. Du X. Hu P. Yang Y. Li B. The response of human thermal perception and skin temperature to step-change transient thermal environments Build Environ 73 2014 232 238 10.1016/j.buildenv.2013.12.007
44 Huizenga C. Hui Z. Arens E. A model of human physiology and comfort for assessing complex thermal environments Build Environ 36 2001 691 699 10.1016/S0360-1323(00)00061-5
45 Zhao P. Zhu N. Chong D. Hou Y. Developing a new heat strain evaluation index to classify and predict human thermal risk in hot and humid environments Sustain Cities Soc 76 2022 103440 10.1016/j.scs.2021.103440
46 Jin L. Zhang Y. Zhang Z. Human responses to high humidity in elevated temperatures for people in hot-humid climates Build Environ 114 2017 257 266 10.1016/j.buildenv.2016.12.028
47 Chen Y. Tao M. Liu W. High temperature impairs cognitive performance during a moderate intensity activity Build Environ 186 2020 107372 10.1016/j.buildenv.2020.107372
48 Engineers R.A. ASHRAE Handbook: Fundamentals 2017
49 Standards S. ISO-7726–ISO-2003, Ergonomics of the Thermal Environment-Instruments for Measuring Physical Quantities.
50 ANSI/ASHRAE, Standard 55-2017: Thermal environmental conditions for human occupancy American Society of Heating Refrigerating and Air Conditioning Engineers, ASHRAE, Atlanta, Georgia, 2017.
51 Kong D. Liu H. Wu Y. Li B. Wei S. Yuan M. Effects of indoor humidity on building occupants’ thermal comfort and evidence in terms of climate adaptation Build Environ 155 2019 298 307 10.1016/j.buildenv.2019.02.039
52 Ricciardi P. Buratti C. Thermal comfort in open plan offices in northern Italy: an adaptive approach Build Environ 56 2012 314 320 10.1016/j.buildenv.2012.03.019
53 Indraganti M. Ooka R. Rijal H.B. Thermal comfort in offices in summer: findings from a field study under the “setsuden” conditions in Tokyo, Japan Build Environ 61 2013 114 132 10.1016/j.buildenv.2012.12.008
54 Wang X. Yang L. Gao S. Zhao S. Zhai Y. Thermal comfort in naturally ventilated university classrooms: a seasonal field study in Xi’an China, Energy Build 247 2021 111126 10.1016/j.enbuild.2021.111126
55 Aghniaey S. Lawrence T.M. Sharpton T.N. Douglass S.P. Oliver T. Sutter M. Thermal comfort evaluation in campus classrooms during room temperature adjustment corresponding to demand response Build Environ 148 2019 488 497 10.1016/j.buildenv.2018.11.013
56 Gerrett N. Ouzzahra Y. Coleby S. Hobbs S. Redortier B. Voelcker T. Havenith G. Thermal sensitivity to warmth during rest and exercise: a sex comparison Eur J Appl Physiol 114 2014 1451 1462 10.1007/s00421-014-2875-0 24711078
57 Branch W.T. Skin in the game Acad Med 91 2016 300 10.1097/ACM.0000000000001068 26907905
58 Nakamura M. Yoda T. Crawshaw L.I. Regional differences in temperature sensation and thermal comfort in humans J Appl Physiol 105 2008 1985 1897 1906 10.1152/japplphysiol.90466.2008
59 Crawshaw L.I. Nadel E.R. Stolwijk J.A.J. Stamford B.A. Effect of local cooling on sweating rate and cold sensation Pflugers Arch 354 1975 19 27 10.1007/BF00584500 1169755
60 Luo W. Kramer R. de Kort Y. van Marken Lichtenbelt W. Effectiveness of personal comfort systems on whole-body thermal comfort – a systematic review on which body segments to target Energy Build 256 2022 111766 10.1016/j.enbuild.2021.111766
61 Li B. Du C. Tan M. Liu H. Essah E. Yao R. A modified method of evaluating the impact of air humidity on human acceptable air temperatures in hot-humid environments Energy Build 158 2018 393 405 10.1016/j.enbuild.2017.09.062
62 Zhang Z. Lin B. Geng Y. Zhou H. Wu X. Zhang C. The effect of temperature and group perception feedbacks on thermal comfort Energy Build 254 2022 111603
63 Fang Z. Tang T. Zheng Z. Zhou X. Liu W. Zhang Y. Thermal responses of workers during summer: an outdoor investigation of construction sites in South China Sustain Cities Soc 66 2021 102705 10.1016/j.scs.2020.102705
64 Luo M. Cao B. Damiens J. Lin B. Zhu Y. Evaluating thermal comfort in mixed-mode buildings: a field study in a subtropical climate Build Environ 88 2015 46 54 10.1016/j.buildenv.2014.06.019
65 Fu C. Zheng Z. Mak C.M. Fang Z. Oladokun M.O. Zhang Y. Tang T. Thermal comfort study in prefab construction site office in subtropical China Energy Build 217 2020 1099958 10.1016/j.enbuild.2020.109958
66 Morris N.B. Piil J.F. Christiansen L. Andreas D. Prolonged facemask use in the heat worsens dyspnea without compromising motor-cognitive performance Temperature 8 2021 160 165 10.1080/23328940.2020.1826840
67 Ong J.J.Y. Bharatendu C. Goh Y. Headaches associated with personal protective equipment – a cross-sectional study among frontline healthcare workers during COVID-19 Headache 60 2020 864 877 10.1111/head.13811 32232837
68 Shumake-Guillemot J. Amir S. Anwar N. Protecting health from hot weather during the COVID-19 pandemic, Global Heat Health Information Network 2020 http://ghhin.org/heat-and-covid-19
69 Daanen H.A.M. S. Bose-O’Reilly, A.D. Flouris, COVID-19 and thermoregulation-related problems: practical recommendations Temperature 8 2021 1 11 10.1080/23328940.2020.1790971
70 Elisheva R. Adverse effects of prolonged mask use among healthcare professionals during COVID-19 J Infect Dis Epidemiol 6 2020 1 5
71 Reychler G. Straeten C.V. Schalkwijk A. Poncin W. Effects of surgical and cloth facemasks during a submaximal exercise test in healthy adults Respir Med 186 2021 106530 10.1016/j.rmed.2021.106530
72 Roberge R.J. Kim J.H. Benson S.M. Absence of consequential changes in physiological, thermal and subjective responses from wearing a surgical mask Respir Physiol Neurobiol 181 2012 29 35 10.1016/j.resp.2012.01.010 22326638
73 Kim J.H. Wu T. Powell J.B. Roberge R.J. Physiologic and fit factor profiles of N95 and P100 filtering facepiece respirators for use in hot, humid environments Am J Infect Control 44 2016 194 198 10.1016/j.ajic.2015.08.027 26476496
74 Lim E.C. Seet R.C. Lee K.H. Wilder-Smith E.P. Chuah B.Y. Ong B.K. Headaches and the N95 face-mask amongst healthcare providers Acta Neurol Scand 113 2006 199 202 10.1111/j.1600-0404.2005.00560.x 16441251
75 Boccardo L. Contact lens and anterior eye self-reported symptoms of mask-associated dry eye: a survey study of 3,605 people Cont Lens Anterior Eye 45 2022 101408 10.1016/j.clae.2021.01.003
76 Ahmad M.D.F. Wahab S. Ahmad F.A. A novel perspective approach to explore pros and cons of face mask in prevention the spread of SARS-CoV-2 and other pathogens J Saudi Pharm Soc 29 2021 121 133 10.1016/j.jsps.2020.12.014
77 Diaz Milian R. Foley E. Bauer M. Martinez-Velez A. Castresana M.R. Expiratory central airway collapse in adults: anesthetic implications (Part 1) J Cardiothorac Vasc Anesth 33 2019 2546 2554 10.1053/j.jvca.2018.08.205 30279064
78 Control Center of Disease, Strategies for Optimizing the Supply of Face Masks, Control Center of Disease, 2020.
79 Zhang H. Human Thermal Sensation and Comfort in Transient and Non-uniform Thermal Environments [D] 2003 University Of California Berkeley
80 Zhang H. Arens E. Huizenga C. T. Han Thermal sensation and comfort models for non-uniform and transient environments: Part I: local sensation of individual body parts Build. Environ. 45 2010 380 388
81 Zhang H. Arens E. Huizenga C. T. Han Thermal sensation and comfort models for non-uniform and transient environments, part II: local comfort of individual body parts Build. Environ. 45 2010 389 398
82 Zhang H. Arens E. Huizenga C. T. Han Thermal sensation and comfort models for non-uniform and transient environments, part III: whole-body sensation and comfort Build. Environ. 45 2010 399 410
83 Wang H. Xu M. Bian C. Experimental comparison of local direct heating to improve thermal comfort of workers Building and Environment 177 2020 106884 ISSN 0360–1323
84 Jin Q. Lin D. Zhang H. Li X. Xu H. Thermal sensations of the whole body and head under local cooling and heating conditions during step-changes between workstation and ambient environment Building and Environment 46 11 2011 2342 2350
85 T.W. Reader, U.W.B. Jr, Face masks including a spunbonded/meltblown/spunbonded laminate, Google Patents (2007).
| 0 | PMC9724504 | NO-CC CODE | 2022-12-15 23:18:05 | no | 2023 Jan 1; 37:101597 | utf-8 | null | null | null | oa_other |
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
The Authors. Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00321-5
10.1016/j.jiph.2022.11.027
Article
Characteristics of the population with mild COVID-19 symptoms eligible for early treatment attended in a single center in Northern Italy
P Paola Magro Magro a⁎
Degli Antoni M a
Formenti B b
Viola F a
Castelli F ab
Amadasi S a
Quiros-Roldan E a
a U.O. Malattie Infettive, ASST Spedali Civili di Brescia e Università degli Studi di Brescia
b Cattedra UNESCO “Formazione e rinforzo delle risorse umane per lo sviluppo sanitario nei Paesi a risorse limitate”, Università degli Studi di Brescia, Italia
⁎ Corresponding author.
6 12 2022
6 12 2022
26 5 2022
8 11 2022
20 11 2022
© 2022 The Authors. Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
After more than two years from the first COVID-19 detected case in Brescia, Northern Italy, monoclonal antibodies and antiviral therapy aimed at early treatment of mild COVID-19 in patients at risk of progression and of hospitalization has been approved in Italy. Here we report the characteristics of the population eligible for the COVID-19 early treatments at our COVID-19 Early Therapy Unit of the Infectious Diseases Department of the ASST Spedali Civili of Brescia, with the aim to evaluate the characteristics of the foreign and native groups. Up to March the 31st, 2022, a total of 559 patients were referred to our Unit for COVID-19 early treatment, where 7.6% were foreigners, a group significantly younger than natives (p<0.05). Particular differences are noticed between the native and the foreign population, where people aged > 65 years old were significantly more frequent among italians (39.7% vs 16.3%, p<0.01), while primary or acquired immunodeficiencies were more frequent in foreigners (55.8% vs 38.9%, p=0.03). Substantial differences are noted between native and foreign populations, where 14% and 26% (p<0.05) respectively have never been vaccinated for COVID-19. Overall, 71% of the referred patients received an early treatment for mild COVID-19, with no differences between the two groups. Overall, on day 28 after treatment, 23 (4%) patients had been hospitalized due to COVID-19 related complications and four died (0,7%), no one was foreigner. In conclusion, while the treatment offered for mild COVID-19 appears to be rather uniform between the native and the foreign populations, some differences, especially in preventive vaccination COVID-19, must be taken into account.
Keywords
Migrant health
COVID19
access to care
COVID19 vaccine
early COVID19 treatment
==== Body
pmc1 Introduction
At the beginning of the pandemic, Italy was the European COVID-19 epicenter, and the province of Brescia, Northern Italy, was one of the first and most affected Western areas.
After more than two years from the first detected case, some treatments are proved effective against SARS-CoV-2 whether initiated in the early phase of the disease in people with mild/moderate symptoms who also have other health conditions increasing their likelihood of developing progressive COVID-19 related disease [1].
Monoclonal antibodies therapy aimed at early treatment of mild/moderate COVID-19 in patients at risk of progression and of hospitalization has been approved in Italy since June 2021, while antiviral drugs (ritonavir-boosted nirmatrelvir and molnupinavir) for the same purpose were released at the beginning of 2022 [1].
Based on the impossibility to offer these therapies to all SARS-CoV-2 infected non-hospitalized patients, health authorities have recommend to prioritize the treatment of patients who are at the highest risk of clinical progression according to: age, vaccination status, immune status, and the presence of risk factors, including obesity, diabetes and severe cardiological/neurological or lung disease.
Brescia has always been characterized by being one of the Italian provinces with the highest presence of foreigners in its territory, which, in 2021, reached 12.4% of the resident population [2]. Many studies have shown that the COVID-19 pandemic has disproportionately impacted the foreign population, especially forced migrants living in high-resource countries [3]. Substantial differences in the management of SARS-CoV-2 infection were also found in the foreign population in Italy, who had lower vaccination rates and less access to diagnostic tests than the native population [4].
Here we report the characteristics of the population eligible for the COVID-19 early treatments at our COVID-19 Early Therapy Unit of the Infectious Diseases Department of the ASST Spedali Civili of Brescia. Aim of this study is to evaluate the characteristics of the foreign and native population eligible for the COVID-19 early treatment.
2 Methods
We performed an observational, retrospective, monocentric study. Patients were referred to our center after a diagnosis of COVID-19 performed by the General Practitioner or by another Specialist through a nasal swab for the research of SARS-CoV-2.
We included in our study all patients with mild COVID-19 symptoms eligible for COVID-19 early treatment according to the indications provided by Agenzia Italiana del Farmaco (AIFA) from the 22nd of aprile, 2021 up to the 31st of march 2022 [4]. Eligible patients aged 18 years or older and tested positive for SARS-CoV-2 with mild symptom onset within the prior 7 days were included. Mild COVID-19 illness is defined by mild symptoms (fever, cough, sore throat, malaise, headache, muscle pain, nausea, vomiting, diarrhea, loss of taste and smell) without dyspnea or abnormal chest imaging. In our center, patients with moderate COVID-19 - defined by the presence of clinical or radiographic evidence of lower respiratory tract infection with oxygen saturations that exceed 94% - were hospitalized, therefore they were not included in this study. [5].
They had at least one risk factor for COVID-19 progression among the following: age 65 years or older, diabetes requiring medication, obesity (body mass index >30), chronic kidney disease, chronic liver disease, cardiological disease, bronchopneumopathy (chronic obstructive pulmonary disease or severe asthma), solid organ or hematopoietic stem cell transplant, hematological disease, oncological disease or other immunodeficiencies. Patients were excluded if they were hospitalized for COVID-19 pneumonia or if they had signs or symptoms of severe COVID-19
Epidemiological, clinical and therapeutic data were retrieved from clinical charts. Results are reported as prevalences. Chi-squared test was used to compare results between the Italian and foreign population, where differences were considered significant when p<0.05 (Epi Info 7.2 software).
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Hospital (reference code NP-5363 approved on 26/04/2022)
3 Results
Up to March the 31st, 2022, a total of 559 patients were referred to our Unit for COVID-19 early treatment, where 7.6% were foreigners (n=43). The great majority of foreigners were from the WHO European Region (60.4%; n=26), all of which from Eastern European countries. Six patients were from the African Region (14%), 4 from the South-East Asian Region (9%), three patients were from the Americas (7%) and three from Eastern Mediterranean Region (7%), while only one was from the Western Pacific Region (2%).
The great majority (60.4%) were from the WHO European Region, all of which from Eastern European countries, and 14% and 9% were from the WHO African and South-East Asian Region respectively.
The proportion of female patients was 47%, with a similar balance both in Italians and foreigners (46% and 56% respectively). Median age was 63.3 in the overall population with significant differences in foreigners compared to natives (49.6 vs 65.2 yeras old, respectively, p<0.01).
Concerning the risk factor for COVID-19 progression, the referred population presented: primary or acquired immunodeficiency (40%); age over 65 years old (38%); cardio cerebral-vascular disease (35%); BMI>30 (18%);; COBP (15%); chronic kidney disease (13%); diabetes (7.5%); cancer pathology (13%); hematological diseases (4%); transplant (4%), chronic liver disease (4%) and haemoglobinopathies (1%). Particular differences are noticed between the native and the foreign population, where people aged > 65 years old were significantly more frequent among italians (39.7% vs 16.3%, p<0.01), while primary or acquired immunodeficiencies were more frequent in foreigners (55.8% vs 38.9%, p=0.03). ( Table 1).Table 1 Characteristics of the population with mild SARS-CoV-2 infection referred forCOVID-19 early treatment.
Table 1 TOTAL n (%) NATIVES n (%) FOREIGNERS n (%)
Referred population 559 516 (92.3%) 43 (7.6%)
Sex (Female) 263 (47%) 239 (46.3%) 24 (55.8%)
Age (median) 63.3 65.2 49.6
RISK FACTORS
Immunodeficiency 225 (40.2%) 201 (38.9%) 24 (55.8%)
Age over 65 years old 212 (37.9%) 205 (39.7%) 7 (16.3%)
Cardio celebral-vascular disease 196 (35%) 185 (35.8%) 11 (25.5%)
BMI>30 102 (18.2%) 102 (19.7%) 0
COPD 85 (15.2%) 81 (15.7%) 4 (9.3%)
Chronic kidney disease 73 (13%) 64 (12.4%) 9 (21%)
Diabetes 42 (7.5%) 38 (7.3%) 4 (9.3%)
Cancer pathology 71 (12.7%) 69 (13.3%) 2 (4.6%)
Hematological diseases 23 (4%) 22 (4.2%) 1 (2.3%)
Neurodevelopmental and neurodegenerative diseases 13 (2.3%) 13 (2.5%) 0
Transplant 21 (3.7%) 19 (3.6%) 2 (4.6%)
Chronic liver disease 22 (3.9%) 18 (3.4%) 4 (9.3%)
Haemoglobinopathies 6 (1%) 5 (1%) 1 (2.3%)
COVID-19 VACCINATION COVERAGE
Unvaccinated (0 doses) 85 (15%) 74 (14%) 11 (26%)
Completely vaccinated (3 doses) 282 (50.4%) 268 (52%) 14 (33%)
EARLY TREATMENT
Early treatment received 397 (71%) 367 (71%) 30 (69.7%)
Monoclonal antibodies 257 (64.7%) 237 (64.5%) 20 (66.6%)
Antiviral 140 (35.2%) 130 (35.4%) 10 (33.3%)
COVID-19 OUTCOME
Hospitalization due to COVID-19 23 (4%) 23 (4.4%) 0
Death 4 (0.7%) 4 (0.7%) 0
*COPD=Chronic obstructive pulmonary disease;
BMI=Body mass index
Half of the referred population (50.4%) had received the full three-dose COVID-19 vaccination cycle, while 27.7% had received two doses and 15% had not received any. Substantial differences are noted between native and foreign populations, where 14% and 26% (p <0.05) respectively have never been vaccinated for COVID-19. Even among those who have completed the vaccination cycle with three doses there are significant differences between the two populations (52% in Italians and 33% in foreigners, p<0.05).
Overall, 71% (N=397) of the referred patients received the early treatment for mild COVID-19, with no differences between the two groups (69% in the foreigners, 71% in native patients). In particular, 65% received monoclonal antibodies and 35% received antivirals. More specifically, for those treated with monoclonal, 50% received sotrovimab, 36,7% banlanivimab-etesevimab and 13% casirivimab; molnupinavir (40,7%), ritonavir-boosted nirmatrelvir (34,2%) and rermdesivir (25%) were the antivirals administered.
Those who did not receive any treatments, either refused (N=63; 38,8%), had exceeded the limit of days from infection (N=15; 9,2%) or did not meet criteria for the therapy (asymptomatic or hospitalized) (n=73; 45%). We did not find differences between the proportion of foreigners and natives who did not receive treatment (30% in foreigners and 29% in Italians).
Overall, on day 28 after treatment, 23 (4%) patients had been hospitalized due to COVID-19 related complications and four died (0,7%), no one was foreigner.
4 Discussion
In this study we did not find any significant differences in either treatment rates or type of early treatment between foreigners and natives patients with mild COVID-19.
However, the two populations differ for some demographics and clinical risk factors, such as younger age and higher rate of immunodeficiencies among foreigners, who were also significantly less vaccinated for COVID-19 (p<0.05).
The percentage of foreign patients who acceded to early treatment for COVID-19 in our study (8%) is slightly lower than the proportion of foreign people resident in our territory (12%) [4] (data not shown).
This difference may reflect the fact that the migrant population is often younger), and then, supposedly healthier, in comparison with the native one [4]. Another reason for this difference may be also partly attributed to the existing barriers in the access to health care services, whether due to language, document situation or cultural gap, in the foreign population [3]. Moreover, challenges in the fruition of the health care services have been exacerbated by the COVID-19 pandemic, and this may have reduced the access to health services in the non-Italian population [6].
Interestingly, either primary or secondary immunodeficiency were the first criteria for eligibility to early treatment for COVID-19 in migrants, showing that a vast part of the migrant population referred to our Unit was composed by foreign patients that were already followed in other Departments for chronic diseases and therefore already chronically linked to the National Health Care System. Despite this, foreigners showed significantly lower rates of COVID-19 vaccination (p<0.05) in comparison to Italians. These results are in accordance with our recent population-based study that included all residents in our province where 25.3% of foreign people had not received any vaccination dose, against 8.7% of the Italian population at the end of the 2021 [4]. We believe that this data should be carefully taken into account by the policy makers, as long as it proves, once again, that this population is scarcely tailored by the public health message. Our data on follow-up showing no death and no COVID-19-related hospitalization in the foreign group are probably due to the younger age of this population.
This is the first study describing characteristics of patients with mild COVID-19 symptoms, with a particular focus on geographic origin. However our study has several limitations. First of all, the retrospective nature, the monocentric design and the modest size of the study group do not consent to extend these findings to other settings. Moreover, we analyzed the foreign group without distinctions in document situation, geographical origin and duration of stay in Italy, which are some of the most important factors that determine the possibility of access to health services.
In conclusion, while the treatment offered for mild COVID-19 appears to be rather uniform between the native and the foreign populations, some differences, especially in preventive measures, must be taken into account in order to implement ad hoc tailored strategies for the two populations.
==== Refs
References
1 SIMIT - Italian Society of Infectious and Tropical Diseases, “Vademecum per la cura delle persone con infezione da SARS-CoV-2, Edizione 5.0, 2 aprile 2022” Available at: https://www.simit.org/images/regioni/lombardia/comunicazioni/Vademecum%20COVID-19/Vademecum%20COVID%20SIMIT%20Lombardia%205.0.pdf (Accessed on 16th May 2022)
2 Italian National Health Statistics (ISTAT). Demographic Statistics, elaborated by tuttitalia.it. Available at: https://www.tuttitalia.it/lombardia/provincia-di-brescia/statistiche/cittadini-stranieri-2021/#:~:text=Gli%20stranieri%20residenti%20in%20provincia,India%20(9%2C1%25). (Accessed on 16th May, 2022)
3 European Centre for Disease Prevention and Control. Reducing COVID-19 transmission and strengthening vaccine uptake among migrant populations in the EU/EEA – 3 June 2021. ECDC: Stockholm; 2021
4 Profili F. Stasi C. Silvestri C. Ferroni E. Zorzi M. Ventura M. Petrelli A. Spadea T. Rusciani R. Bartolini L. Caranci N. Cacciani L. Calandrini E. Maifredi G. Leoni O. Voller F. Gruppo di lavoro INMP Covid19 e immigrati. L’impatto della pandemia di COVID-19 nella popolazione italiana e straniera residente nelle diverse fasi: i risultati di un progetto multicentrico inter-regionale [The impact of the COVID-19 pandemic on the Italian and foreign population in the various phases: the results of an interregional multicentre project] Epidemiol Prev 46 4 2022 Jul-Aug 71 79 Italian. doi: 10.19191/EP22.4S1.058. PMID: 35862562 35862562
5 AIFA – Italian Agency for Drugs – “RACCOMANDAZIONI AIFA SUI FARMACI per la gestione domiciliare di COVID-19 Vers. 8 – Agg. 12/04/2022” Available at https://www.aifa.gov.it/documents/20142/1123276/IT_Raccomandazioni_AIFA_gestione_domiciliare_COVID-19_Vers7_12.04.2022.pdf
6 Fabiani M. Mateo-Urdiales A. Andrianou X. Bella A. Del Manso M. Bellino S. Rota M.C. Boros S. Vescio M.F. D'Ancona F.P. Siddu A. Punzo O. Filia A. Brusaferro S. Rezza G. Dente M.G. Declich S. Pezzotti P. Riccardo F. COVID-19 Working Group. Epidemiological characteristics of COVID-19 cases in non-Italian nationals notified to the Italian surveillance system Eur J Public Health 31 1 2021 Feb 1 37 44 10.1093/eurpub/ckaa249. PMID: 33416859; PMCID: PMC7851886 33416859
| 36521328 | PMC9724552 | NO-CC CODE | 2022-12-13 23:16:43 | no | J Infect Public Health. 2023 Jan 6; 16(1):104-106 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.11.027 | oa_other |
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00342-2
10.1016/j.jiph.2022.12.003
Original Article
Comorbidities prolonged viral shedding of patients infected with SARS-CoV-2 omicron variant in Shanghai: a multi-center, retrospective, observational study
Pei Lei a1
Chen Ying a1
Zheng Xiangtao a1
Gong Fangchen a1
Liu Wenbin a
Lin Jingsheng b
Zheng Ruizhi c
Yang Zhitao a⁎
Bi Yufang c
Chen Erzhen a⁎
a Department of Emergency, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
b Department of Disciplinary Development and Planning, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
c Department of Endocrine and Metabolic Diseases, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
⁎ Correspondence to: Department of Emergency, Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Ruijin No. 2 Road, Huangpu District, Shanghai, 200025, P.R. China
1 Contributed equally to this manuscript.
6 12 2022
6 12 2022
11 9 2022
15 11 2022
4 12 2022
© 2022 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
As the omicron variant of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) surges amid the coronavirus disease 2019 (COVID-19) pandemic, there is limited comorbidities data associated with viral shedding time (VST). We aimed to investigate the effect of comorbidities on VST in asymptomatic and mild patients with omicron.
Methods
A multi-center, retrospective, observational study was conducted from March 12, 2022 to May 24, 2022 in Shanghai. The analysis was adjusted for patients’ baseline demographic, using log-rank test and logistic regression model.
Results
The study enrolled 198262 subjects. The median duration of viral shedding time (VST) was 8.29 days. The number of cumulative viral shedding events was significantly lower in the chronic obstructive pulmonary disease (COPD), hyperlipidemia, diabetes, urinary system disease, and cardiocerebrovascular disease than in the no corresponding comorbidities group. Patients with comorbidities had a lower incidence of viral shedding, and the most significant independent risk factor is COPD (aOR 1.78, 95% CI: 1.53-2.08, p<0.001). Across different age ranges, the comorbidities affecting viral shedding also differ, with the greatest risk factors for viral shedding being hyperlipidemia (aOR 2.23, 95% CI: 1.50-3.31, p<0.001) and COPD (aOR 1.85, 95% CI: 1.50-2.28, p<0.001) between ages of 18-39 and 40-64, and thyroid dysfunction (aOR 2.36, 95% CI: 1.60-3.47, p<0.001) above age 64.
Conclusions
Omicron-infected patients with comorbidities might prolong the VST. The independent risk factors also differ across age ranges, suggesting that providing targeted effective prevention and control guidance and allocating appropriate resources to different populations should be a crucial strategy.
Keywords
COVID-19
SARS-CoV-2
Omicron
Comorbidity
Viral shedding time
==== Body
pmc1 Introduction
The unprecedented pandemic of coronavirus disease 2019 (COVID-19) started more than 2 years ago. So far, the alpha, beta, gamma, delta, and omicron variants of SARS-CoV-2 have emerged, and each later variant is more transmissible than the previous one [1]. After the omicron (B.1.1.529) variant of severe Acute respiratory syndrome coronavirus 2 (SARS-COV-2) was first identified in southern Africa [2], the virus has more than 30 mutations in its spike gene that made it more infectious, increasing the risk of reinfection and potentially escaping immunity [3], [4]. It was declared by the World Health Organization (WHO) to be a variant of concern on 77 November 25, 2021. Currently, the omicron variant of SARS-CoV-2 has overtaken other variants to become the predominant circulating strain, sweeping the world [5]. Previous studies indicated that omicron infection was associated with significantly shorter hospital stays and lower severity and mortality than in previous variants [6], [7], [8]. Since the outbreak of the omicron variant in late February 2022, in Shanghai, the major omicron subline is BA.2. As of May 4, 2022, 593336 cases have been identified, including 538 450 asymptomatic carriers [9]. In view of the alarming global spread and morbidity of omicron variant, the prolonged viral shedding time (VST) has attracted widespread attention, bringing greater challenges and difficulties to epidemic prevention and control. The VST is an important parameter to judge the discharge and termination of quarantine of infectious diseases and determines the transmission and duration of infectiousness.
More than half of patients with COVID-19 were reported to have at least one comorbidity on admission according to report [10]. Hypertension, diabetes, chronic obstructive pulmonary disease (COPD), and obesity were the most commonly reported. Comorbidities increase the prognosis of acute illness and the risk of exacerbating severe symptoms. However, there was insufficient evidence to prove any comorbidity had effects on VST in omicron variant infections. Most published studies on COVID-19 were retrospective and observational designs with inadequate sample sizes, making it difficult to evaluate how particular comorbidity affects viral shedding time [11]. To address this knowledge gap, therefore, we carried out a multi-center, retrospective, observational study to evaluate the effects of different comorbidities on clinical outcomes in asymptomatic carriers and mild cases infected with the omicron variant, with the hope that our study will provide updated information on the management of this variant.
2 Materials and methods
2.1 Study design and participants
This study was a multi-center, retrospective, observational study of adults (aged 18 years or older) with the omicron variant who were admitted to three Fangcang shelter hospitals from March 12, 2022 to May 24, 2022, during the omicron variants circulated in shanghai. Eligible patients included asymptomatic carriers and mild cases diagnosed according to SARS-CoV-2 diagnosis and treatment guidance(ninth edition) of the National Health Commission of China. Laboratory confirmation of SARS-CoV-2 was defined as a positive result of real-time reverse transcriptase-polymerase chain reaction assay of nasopharyngeal (NPS) or throat swab (TS). The outcome of the study was the viral shedding time among asymptomatic and mild persons testing positive for SARS-CoV-2.
2.2 Data collection and definitions
Epidemiological, demographic, comorbidities, and VST of all patients with a confirmed COVID-19 diagnosis were collected and recorded using a standardized electronic database. Viral shedding was defined as negative RT-PCR for SARS-CoV-2 on 2 consecutive NPS or TS. Persons who had received at least two doses of the COVID-19 vaccine, irrespective of vaccine type, more than 7 days before the case date were classified as vaccinated, whereas persons with an incomplete primary vaccination schedule (one or no doses) were considered unvaccinated. Defined as diagnoses in an inpatient or specialized care at any time point during the 5 years before baseline in the following disease groups: hypertension, hyperlipidemia, diabetes, COPD, thyroid dysfunction, cardiocerebrovascular disease, urinary system disease, liver diseases, and other conditions and diseases (neurological diseases, cancer or immunosuppressed states, HIV or mental health disorder).
2.3 Virological assessment and clinical management
For all included patients, the diagnosis of SARS-CoV-2 infection was confirmed on at least one respiratory specimen by the detection of SARS-CoV-2 RNA through real-time polymerase chain reaction (RT-PCR). During hospitalization, all patients underwent follow-up NPS and TS to assess viral shedding. The therapeutic management of patients was based on internal hospital protocol, national and international guidelines, and clinical judgment, according to the best evidence available at the time.
2.4 Statistical analysis
Continuous and categorical variables were expressed as medians with IQRs and as numbers (%), respectively. The cumulative probability of achieving viral shedding was estimated by Kaplan–Meier curves in people with different comorbidity. The differences between groups were compared by the log-rank statistic test. Multivariable logistic regression analysis to identify predictive factors of prolonged viral shedding. All statistical analyses were performed using R (version 4.2). A P -value <0.05 indicated conventional statistical significance.
2.5 Ethical Approvals
The ethics committee approved this study (Protocol Record SCAM2022). The ClinicalTrials.gov Identifier is NCT05375786. Patient information remained anonymous, and written consents were waived due to a major infectious disease outbreak.
3 Results
3.1 Baseline characteristics of all subjects infected with the omicron variant
In this study, 198262 subjects infected with the omicron variant of SARS-CoV-2 were enrolled from March 12, 2022, to May 24, 2022 while the variants circulated in Shanghai. Their median age was 43.00 (IQR: 32.00–54.00), 59.06% were male, and the median duration time of VST was 8.29 days (IQR: 5.33–11.27). Among them, 20,504 (10.34%) are mildly infected patients and the rest are asymptomatic carriers (177758 cases, 89.66%). We further analyzed the vaccination status of those patients: that is, one with incomplete vaccination (48999 cases, 24.71%), one with full (two-dose) vaccination (59745 patients, 30.13%), and one that received booster shots (i.e., three-dose vaccination) (89518 cases, 45.15%). Then, the subjects were divided into two subgroups: no comorbidities (164366 cases) or at least one comorbidities (33896 cases, 17.09%). In the group with one or more comorbidities, the average VST was longer than in the group without comorbidities. ( Table 1)Table 1 The characteristics of the total subjects were included.
Table 1 Overall ( n=198262) Presence of comorbidity ( n=164366) Absence of comorbidity ( n=33896) p
Sex
Male (%) 117095 (59.06) 97232 (59.16) 19863 (58.60) 0.059
Age (median [IQR]) 43.0 [32.0, 54.0] 40.0 [31.0, 52.0] 56.000 [46.0, 64.0] <0.0001
Vaccination (%) <0.0001
0/1 dose (incomplete vaccination) 48999 (24.71) 39346 (23.94) 9653 (28.48)
2 dose 59745 (30.13) 50439 (30.69) 9306 (27.45)
3 dose 89518 (45.15) 74581 (45.37) 14937 (44.07)
Viral Shedding Time (median [IQR]) 8.290 [5.330, 11.270] 8.200 [5.320, 10.580] 9.290 [6.330, 12.280] <0.0001
Diagnosis <0.0001
Mild (%) 20504 (10.34) 15023 (9.14) 5481 (16.17)
Data are N (%), mean (SD) or median (IQR). SD: Standard Deviation; IQR: Interquartile Range.
3.2 Baseline characteristics of subjects with comorbidities
Then, demographic information and distribution of comorbidities were analyzed as shown in Table 2, and baseline characteristics of patients of different age groups were compared ( Fig. 1). Among 33896 subjects with comorbidities, the median age was 56.00 (IQR: 46.00–64.00), and the median duration time of VST was 9.29 days (IQR: 6.33–12.28). The main complications of omicron infection were hypertension (65.24%), diabetes (28.39%), Cardiocerebrovascular disease (9.01%), thyroid dysfunction (5.91%), and hyperlipidemia (3.94%). Among infected patients with underlying disease, 70.99% (12.14% of the total subjects) had at least one comorbidity. (Table 2)Table 2 Baseline characteristics of subjects with comorbidities.
Table 2 Presence of comorbidity ( n=33896)
Sex = Male (%) 19863 (58.60)
Age (median [IQR]) 56.0 [46.0, 64.0]
Vaccination (%)
0/1 dose (incomplete vaccination) 9653 (28.48)
2 dose 9306 (27.45)
3 dose 14937 (44.07)
Viral shedding time (median, IQR) 9.29 [6.33, 12.28]
COPD 794 (2.34)
Hypertension 22113 (65.24)
Hyperlipidemia 1335 (3.94)
Diabetes 9622 (28.39)
Thyroid dysfunction 2003 (5.91)
Chronic liver disease 160 (0.47)
Cardiocerebrovascular disease 3053 (9.01)
Gout 133 (0.39)
Urinary system diseases⁎ 723 (2.13)
Other diseases 7219 (21.30)
Number of comorbidities
1 24063 (70.99)
2 7132 (21.04)
3 2101 (6.20)
> 3 600 (1.77)
Diagnosis
Mild (%) 5481 (16.17)
Data are N (%), mean (SD) or median (IQR).
Abbreviations: SD: Standard deviation; IQR: Interquartile range; COPD: Chronic obstructive pulmonary disease.
⁎ Urinary System Diseases: Chronic kidney disease and urinary system diseases were included.
Fig. 1 The characteristics of subjects with comorbidities stratified by age.
Fig. 1
3.3 The cumulative probability of viral shedding and associated risk factors
Over a median follow-up of 8.29 days (IQR 5.33–11.27). The shortest duration of viral shedding observed in our population was 3 days, while the longest was 30 days. After stratifying by different comorbidities, Kaplan-Meier estimates the cumulative probability of viral shedding from the upper respiratory tract (URT), as shown in Fig. 2. By log-rank test, the cumulative probability of viral shedding from URT in patients with different comorbidities during the observation was significantly higher than that in patients without comorbidities. The number of viral shedding events was significantly lower in the COPD (794 cases, p<0.0001), hypertension (22113 cases, p<0.0001), hyperlipidemia (1335 cases, p<0.0001), diabetes (9622 cases, p<0.0001), urinary system disease (723 cases, p<0.0001), cardiocerebrovascular disease (3053 cases, p<0.0001), chronic liver disease (160, p<0.005), and thyroid dysfunction (2003 cases, p<0.0001) than in no corresponding comorbidities group.Fig. 2 Kaplan-Meier curves estimating the cumulative probability of viral shedding in total population with different comorbidities.
Fig. 2
3.4 Risk factors associated with prolonged viral shedding
In this study, prolonged viral shedding time was defined as detecting SARS-CoV-2 RNA on respiratory specimens for > 8.29 days (the median duration of VST in our population). As shown in Fig. 3, when having more than three comorbidities, the risk of prolonged viral shedding increased to 2.09 times (aOR 2.09, 95% CI: 1.74-2.50, p<0.001) ( supplementary table 1). In the included subjects, Having COPD (aOR 1.78, 95% CI: 1.53-2.08, p<0.001), hypertension (aOR 1.21, 95% CI: 1.18-1.25, p<0.001), hyperlipidemia (aOR 1.32, 95% CI: 1.18-1.49, p<0.001), diabetes (aOR 1.21, 95% CI: 1.15-1.26, p<0.001) and thyroid dysfunction (aOR 1.27, 95% CI: 1.15-1.41, p<0.001) during hospitalization were significantly associated to increased odds of slower viral shedding as shown in Fig. 4 (supplementary table 2). Among different age ranges, the comorbidities affecting viral shedding also differ, with the greatest risk factors for viral shedding in 18-39 years, 40-64 years, and older than 64 years being hyperlipidemia (aOR 2.23, 95% CI: 1.50-3.31, p<0.001, Fig. 5A, supplementary table 3), COPD (aOR 1.85, 95% CI: 1.50-2.28, p<0.001, Fig. 5B, supplementary table 4), thyroid dysfunction (aOR 2.36, 95% CI: 1.60-3.47, p<0.001, supplementary table 5), respectively.Fig. 3 Predictive factors (based on number of comorbidities) of viral shedding time (> 8.29 days) by Logistic regression analysis (on 198262 patients).
Fig. 3
Fig. 4 Predictive factors (based on different comorbidities) of viral shedding time (> 8.29 days) by Logistic regression analysis (on 198262 patients).
Fig. 4
Fig. 5 Predictive factors of viral shedding time (> 8.3 days) by Logistic regression analysis. A. Age of 18-39. B. Age of 40-64. C. Above age of 64.
Fig. 5
4 Discussion
This retrospective study focused on the duration of viral shedding from the URT and the association with both comorbidities and prolonged viral shedding.
In our cohort, the median duration of viral shedding, from PCR positive onset to viral shedding, was 8.29 days. While concerning the presence of comorbidity, the median time was 9.29 days. Advanced age, presence of symptoms, and underlying comorbidities were considered independent risk factors of delayed viral shedding in our study. Although comorbidities have been identified as one of the main prognostic factors for COVID-19 severity, only a few studies have reported an association with the duration of viral shedding among people infected with the omicron variants [12]. Our study observed that patients with underlying comorbidities were more likely to have both slower viral shedding and prolonged viral detection with an increased risk for each additional comorbidity. Additionally, there were significant relations between specific comorbidities and the persistence of viral RNA, including hypertension, hyperlipidemia, diabetes, COPD, thyroid dysfunction, and so on. The comorbidities affecting viral shedding also differ across age ranges, with the greatest risk factors for viral shedding being hyperlipidemia and COPD in 18-39 years and 40-64 years, and thyroid dysfunction in the elderly group (≥ 65 years), respectively. It would be instructive for the health professionals and the community regarding the precautionary measures, comprehending the risk of comorbidities in COVID-19, and establishing management strategies to combat the pandemic situation. However, the mechanisms and pathophysiology of some comorbidities in COVID-19 patients yet need further understanding.
During this pandemic in Shanghai, only a rare proportion of critically ill or deceased patients were reported due to the omicron infection directly [9]. Patients infected with the omicron variant of SARS-CoV-2 included in this study, Similar to other studies reported from other countries [8], [13], [14], demonstrated reduced clinical severity and were mainly asymptomatic and mild. This result further mirrors the attenuated pathogenicity of the omicron variant compared to that induced by the wild-type strain or other variant.
Early studies reported by South African researchers suggested that the pathogenicity was greatly attenuated during the spread of the omicron variant [6], [7], [15]. Differing from most countries, China has high vaccination coverage but low numbers of COVID-19 infections and rare reinfections. Thus, the acquisition of immunity to SARS-CoV-2 comes primarily from effective vaccination. Reduced effectiveness of fully vaccinated individuals when fighting omicron infection compared to the Delta variant has been reported [16], [17]. In our study, it should be noted that timely vaccination (with a booster shot) did not provide a significant protective effect against viral shedding, possibly due to a series of mutations in the omicron genome spike protein involved in immune evasion [18], [19]. Reduced neutralization of the omicron variant has been reported in studies [20], [21], [22] using plasma specimens from individuals with complete (two or three doses) mRNA vaccine series, and from patients with prior SARS-CoV-2 infection, which further demonstrates that there might be a limited effect of the vaccine on clearance of the omicron variant among asymptomatic and mild patients.
Although severity is alleviated with the emergence of the BA.2 subtype of omicron variant, higher transmissibility of omicron variant infections and immune evasion from previous infection and vaccination remains a concern. The high rate of infection in the community has overwhelmed healthcare systems in Shanghai and elsewhere and has translated to high absolute numbers of hospitalizations with lower severity of infections associated with the omicron variant. Observations in Hong Kong with low immunity previously caused by infection [23] highlight the risk of severe and fatal illness due to omicron variants, although the risk of severe clinical outcomes in cases tends to be lower than with Delta variants. Underlying diseases, such as hypertension, cardiovascular disease, diabetes, and COPD, have been reported as risk factors for severe disease and also increased the mortality rate, therefore better management with special consideration must be given to these patients. Such patients need to be accurately evaluated on admission and different guidelines are designed for these patients.
As the situation is rapidly evolving, future studies with larger sample sizes will likely contribute to identifying additional comorbidities. The prevalence of chronic diseases is increasing year by year, and targeted public health interventions must be adopted to better protect people with chronic diseases from infection with SARS-CoV-2 and other respiratory viruses. Knowledge of populations at risk is critical for providing effective guidance and allocating appropriate resources.
5 Limitations and Strength
Our study has several limitations. First, the study could be prone to bias related to unmeasured confounders due to its observational nature. Such as drugs used to treat comorbidities or health care resources that could influence viral shedding. Given the recent escalation of omicron outbreaks and the increasing number of patients with no symptoms or no need for hospitalization, clinical data became less available, which further reduces the data of patients in different subgroups. Second, the estimated duration of viral shedding could have been influenced by heterogeneity in the frequency of specimen collection and the type of respiratory specimen used. Additionally, the lack of any quantitative determination of viral load such as cycle threshold prevents us from drawing conclusions about the potential infectivity of long-term shedding. Finally, there was no severe or deceased patient in our study, so we could not have access to analyze the possible risk factors associated with the severity or mortality of COVID-19 infection by the omicron variant. However, the study's main strength is its large sample size. To our knowledge, this is one of the largest cohorts in which the duration of vrial shedding has been investigated. Thus, this study provides significant information on the correlation between prolonged viral shedding and comorbidities.
6 Conclusion
Asymptomatic and mild omicron infected patients with comorbidities such as COPD, hypertension, hyperlipidemia, diabetes, and thyroid dysfunction might prolong the viral RNA shedding time. The independent risk factors also differ across age ranges, suggesting that appropriate action to protect those most at risk will therefore be integral to limiting the number of severe and fatal cases and mitigating the burden on health systems.
Funding Source
This work was supported by the Program for Outstanding Medical Academic Leader, Shanghai Shenkang Hospital Development Center of China (grant numbers SHDC2020CR1028B, SHDC22021304).
Appendix A Supplementary material
Supplementary material
Acknowledgments
The authors wish to thank the patients involved in the study as well as all medical staff who work on the frontline.
Conflicts of Interest
None.
Disclosure Statement
No competing financial interests exist.
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jiph.2022.12.003.
==== Refs
References
1 Hirabara S.M. Serdan T.D.A. Gorjao R. Masi L.N. Pithon-Curi T.C. Covas D.T. SARS-COV-2 Variants: Differences and Potential of Immune Evasion Front Cell Infect Microbiol 11 2021 781429 10.1016/j.ijid.2021.12.357
2 Viana R. Moyo S. Amoako D.G. Tegally H. Scheepers C. Althaus C.L. Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in southern Africa Nature 603 7902 2022 679 686 10.1038/s41586-022-04411-y 35042229
3 Harvey W.T. Carabelli A.M. Jackson B. Gupta R.K. Thomson E.C. Harrison E.M. SARS-CoV-2 variants, spike mutations and immune escape Nat Rev Microbiol 19 7 2021 409 424 10.1038/s41579-021-00573-0 34075212
4 Planas D. Saunders N. Maes P. Guivel-Benhassine F. Planchais C. Buchrieser J. Considerable escape of SARS-CoV-2 Omicron to antibody neutralization Nature 602 7898 2022 671 675 10.1038/s41586-021-04389-z 35016199
5 Del Rio C. Omer S.B. Malani P.N. Winter of Omicron-The Evolving COVID-19 Pandemic Jama 327 4 2022 319 320 10.1001/jama.2021.24315 34935863
6 Wolter N. Jassat W. Walaza S. Welch R. Moultrie H. Groome M. Early assessment of the clinical severity of the SARS-CoV-2 omicron variant in South Africa: a data linkage study Lancet 399 10323 2022 437 446 10.1016/S0140-6736(22)00017-4 35065011
7 Maslo C. Friedland R. Toubkin M. Laubscher A. Akaloo T. Kama B. Characteristics and Outcomes of Hospitalized Patients in South Africa During the COVID-19 Omicron Wave Compared With Previous Waves Jama 327 6 2022 583 584 10.1001/jama.2021.24868 34967859
8 Lewnard J.A. Hong V.X. Patel M.M. Kahn R. Lipsitch M. Tartof S.Y. Clinical outcomes associated with SARS-CoV-2 Omicron (B.1.1.529) variant and BA.1/BA.1.1 or BA.2 subvariant infection in southern California Nat Med 2022 10.1038/s41591-022-01887-z
9 Zhang X. Zhang W. Chen S. Shanghai's life-saving efforts against the current omicron wave of the COVID-19 pandemic Lancet 399 10340 2022 2011 2012 10.1016/S0140-6736(22)00838-8 35533708
10 Gasmi A. Peana M. Pivina L. Srinath S. Gasmi Benahmed A. Semenova Y. Interrelations between COVID-19 and other disorders Clin Immunol 224 2021 108651 10.1016/j.clim.2020.108651
11 Kahn F. Bonander C. Moghaddassi M. Rasmussen M. Malmqvist U. Inghammar M. Risk of severe COVID-19 from the Delta and Omicron variants in relation to vaccination status, sex, age and comorbidities - surveillance results from southern Sweden, July 2021 to January 2022 Euro Surveill 27 9 2022 2200121 10.2807/1560-7917.ES.2022.27.9.2200121
12 Mondi A. Lorenzini P. Castilletti C. Gagliardini R. Lalle E. Corpolongo A. Risk and predictive factors of prolonged viral RNA shedding in upper respiratory specimens in a large cohort of COVID-19 patients admitted to an Italian reference hospital Int J Infect Dis 105 2021 532 539 10.1016/j.ijid.2021.02.117 33676001
13 Abdullah F. Myers J. Basu D. Tintinger G. Ueckermann V. Mathebula M. Decreased severity of disease during the first global omicron variant covid-19 outbreak in a large hospital in tshwane, south africa Int J Infect Dis 116 2022 38 42 10.1016/j.ijid.2021.12.357 34971823
14 Modes M.E. Directo M.P. Melgar M. Johnson L.R. Yang H. Chaudhary P. Clinical Characteristics and Outcomes Among Adults Hospitalized with Laboratory-Confirmed SARS-CoV-2 Infection During Periods of B.1.617.2 (Delta) and B.1.1.529 (Omicron) Variant Predominance - One Hospital, California, July 15-September 23, 2021, and December 21, 2021-January 27, 2022 MMWR Morb Mortal Wkly Rep 71 6 2022 217 223 10.15585/mmwr.mm7106e2 35143466
15 Madhi S.A. Kwatra G. Myers J.E. Jassat W. Dhar N. Mukendi C.K. Population Immunity and Covid-19 Severity with Omicron Variant in South Africa N Engl J Med 386 14 2022 1314 1326 10.1056/NEJMoa2119658 35196424
16 Cele S. Jackson L. Khoury D.S. Khan K. Moyo-Gwete T. Tegally H. Omicron extensively but incompletely escapes Pfizer BNT162b2 neutralization Nature 602 7898 2022 654 656 10.1038/s41586-021-04387-1 35016196
17 Collie S. Champion J. Moultrie H. Bekker L.G. Gray G. Effectiveness of BNT162b2 Vaccine against Omicron Variant in South Africa N Engl J Med 386 5 2022 494 496 10.1056/NEJMc2119270 34965358
18 McCallum M. Czudnochowski N. Rosen L.E. Zepeda S.K. Bowen J.E. Walls A.C. Structural basis of SARS-CoV-2 Omicron immune evasion and receptor engagement Science 375 6583 2022 864 868 10.1126/science.abn8652 35076256
19 Kannan S.R. Spratt A.N. Sharma K. Chand H.S. Byrareddy S.N. Singh K. Omicron SARS-CoV-2 variant: Unique features and their impact on pre-existing antibodies J Autoimmun 126 2022 102779 10.1016/j.jaut.2021.102779
20 Schmidt F. Muecksch F. Weisblum Y. Da Silva J. Bednarski E. Cho A. Plasma Neutralization of the SARS-CoV-2 Omicron Variant N Engl J Med 386 6 2022 599 601 10.1056/NEJMc2119641 35030645
21 Nemet I. Kliker L. Lustig Y. Zuckerman N. Erster O. Cohen C. Third BNT162b2 Vaccination Neutralization of SARS-CoV-2 Omicron Infection N Engl J Med 386 5 2022 492 494 10.1056/NEJMc2119358 34965337
22 Pulliam J.R.C. van Schalkwyk C. Govender N. von Gottberg A. Cohen C. Groome M.J. Increased risk of SARS-CoV-2 reinfection associated with emergence of Omicron in South Africa Science 376 6593 2022 eabn4947 10.1126/science.abn4947 35289632
23 Smith D.J. Hakim A.J. Leung G.M. Xu W. Schluter W.W. Novak R.T. COVID-19 Mortality and Vaccine Coverage - Hong Kong Special Administrative Region, China, January 6, 2022-March 21, 2022 MMWR Morb Mortal Wkly Rep 71 15 2022 545 548 10.15585/mmwr.mm7115e1 35421076
| 0 | PMC9724554 | NO-CC CODE | 2022-12-07 23:20:14 | no | J Infect Public Health. 2022 Dec 6; doi: 10.1016/j.jiph.2022.12.003 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.12.003 | oa_other |
==== Front
J Hosp Infect
J Hosp Infect
The Journal of Hospital Infection
0195-6701
1532-2939
Published by Elsevier Ltd on behalf of The Healthcare Infection Society.
S0195-6701(22)00375-9
10.1016/j.jhin.2022.11.020
Article
Risk factors for hospital-acquired infections during the SARS-CoV-2 pandemic
Kwon Jennie H. 1
Nickel Katelin B. 1
Reske Kimberly A. 1
Stwalley Dustin 1
Dubberke Erik R. 1
Lyons Patrick G. 2
Michelson Andrew 2
McMullen Kathleen 3
Sahrmann John M. 1
Gandra Sumanth 1
Olsen Margaret A. 1
Burnham Jason P. 1∗
1 Washington University in St. Louis School of Medicine, Division of Infectious Diseases
2 Washington University in St. Louis School of Medicine, Division of Pulmonary and Critical Care Medicine
3 Mercy, Infection Prevention, St. Louis, MO
∗ Corresponding author: 4523 Clayton Avenue, Campus Box 8051, St. Louis, MO, 63110 Telephone number: 314-652-4100 x54419 Fax: 314-454-8687
6 12 2022
6 12 2022
27 9 2022
8 11 2022
17 11 2022
© 2022 Published by Elsevier Ltd on behalf of The Healthcare Infection Society.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objective
To evaluate risk factors for hospital-acquired infections (HAIs) in patients during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic including historical and concurrent cohorts.
Design
Retrospective cohort.
Setting
Three Missouri hospitals, data from 1/1/2017-9/30/2020.
Participants
Patients ≥18 years of age and admitted ≥48 hours.
Methods
Univariate and multivariable Cox proportional hazards models incorporating the competing risk of death were used to determine risk factors for HAIs. We performed a priori sensitivity analyses to assess the robustness of our urine, blood, and respiratory culture-based HAI definition.
Results
The cohort included 254,792 admissions, with HAIs occurring during 7,147 (2.8%) of admissions (1,661 blood, 3,407 urine, 2,626 respiratory). Patients with SARS-CoV-2 had increased risk of HAI (adjusted hazards ratio 1.65; 95% confidence interval 1.38–1.96), and this was one of the strongest risk factors for development of an HAI. Other risk factors for HAI included certain admitting services, chronic comorbidities, ICU stay during index admission, extremes of body mass index, hospital, and selected medications. Factors associated with decreased risk of HAI included year of admission (declined over the course of the study) and admitting service and medications. Risk factors for HAI were similar in sensitivity analyses that restricted to patients with diagnostic codes for pneumonia/upper respiratory infection and urinary tract infection.
Conclusions
SARS-CoV-2 was associated with significantly increased risk of HAIs.
Abbreviated title, SARS-CoV-2
COVID-19, hospital-acquired infections
==== Body
pmcIntroduction
In the early stages of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, there were no proven treatments, limited diagnostic tests, and uncertainty about coincident and secondary infections in patients with coronavirus disease 2019 (COVID-19). As a result, a large proportion of patients admitted to hospitals received antimicrobials (∼75%), in order to treat potential coincident and secondary bacterial infections.1 More recent studies have since shown that rates of coincident bacterial infections when patients present to the hospital for COVID-19 are low (∼9%),1 , 2 though antimicrobial use remains high (∼75%).1 There is now a trend towards improving antimicrobial stewardship for hospitalized patients with SARS-CoV-2.2 , 3
Recently, more studies have been published on secondary bacterial infections in hospitalized patients with COVID-19.[4], [5], [6], [7], [8], [9] However, the studies are often small,6 , 8 and/or have limited patient populations (intensive care unit [ICU] only),5 , 6 , 8 or are lacking historical or concurrent controls. There have been large studies of data reported to the National Healthcare Safety Network (NHSN) that show higher rates of central line-associated bloodstream infection (CLABSIs), catheter-associated urinary tract infection (CAUTIs), and ventilator-associated events (VAEs), but relatively stable Clostridioides difficile infection (CDI) rates during the pandemic, but these studies lack patient level data to correlate particular patient-level risk factors with the development of infections during hospitalization.4 , 7
With this gap in understanding, we sought to use pre-COVID-19 and pandemic-era data to determine patient-level risk factors for the development of hospital-acquired infections, including urine, bloodstream, and respiratory infections.
Methods
Study cohort
We assembled a retrospective cohort of all patients aged ≥18 years who were admitted between 1/1/2017 and 8/31/2020 for ≥48 hours at one of three Saint Louis, Missouri hospitals (one large tertiary care referral hospital and two community hospitals). We required that patients be discharged by 9/30/2020 to allow for complete capture of admission information. This study was approved by the Washington University School of Medicine Institutional Review Board with a waiver of informed consent.
All data including culture results were collected electronically from the BJC Healthcare Informatics database. The Informatics database has been in existence since 1994 and includes all microbiology results, other laboratory results, vital signs, medications administered in-hospital, demographics, and administrative data.10
Hospital-acquired infection (HAI) outcome definition
Hospital-acquired infections were defined as those occurring ≥48 hours after admission and were based on positive urine, respiratory, and blood cultures. The HAI date was defined as the first positive culture collection date. Infections within the repeat infection timeline were defined according to Centers for Disease Control and Prevention (CDC) standards.11
Potential risk factors for HAI
Demographics collected included sex, race, age, payer, and prior hospitalization at any of the three study hospitals in the past three months; zip-code was used to calculate zip-code level socioeconomic status (SES) measures including the U.S. Census Bureau median household income and the CDC Social Vulnerability Index (SVI).12 , 13 Additional data included comorbidities defined using ICD-10-CM (International Classification of Diseases, Tenth Revision, Clinical Modification) diagnosis codes and the Elixhauser classification,14 , 15 body mass index (BMI), procedures/surgeries performed per ICD-10-PCS codes, admitting service defined as the first non-emergency department service per admission, inpatient medications, and ICU stay. Laboratory results and vitals sign measurements in the first 24 hours of the admission were used to calculate a modified Acute Physiology and Chronic Health Evaluation (APACHE) II score, including all items except for Glascow coma score.16 We also collected COVID-19 status during the admission based on results of SARS-CoV-2 PCR testing and/or a diagnosis codes for COVID-19 (ICD-10-CM: U07.1). Medical record review was performed for admissions with a COVID-19 ICD-10-CM diagnosis code without a positive laboratory result to confirm a clinical diagnosis. ICU status, procedures, medications, and COVID-19 status were captured for the entire admission for those without an HAI, and captured through the calendar day before HAI for admissions with an HAI. Continuous variables with missing data were assigned the median value (i.e., SVI, median household income); missing BMI was retained as a category when categorizing into BMI classes.
Statistical analyses
For all potential risk factors for HAI, we calculated rates per 10,000 hospital-days, where hospital-days is the number of days to the earliest of HAI or hospital discharge. Univariate and multivariable Cox proportional hazards models incorporating the competing risk of death were used to determine risk factors for HAI.17 For multivariable models, we accounted for clustering by patient. Among variables with P<0.05 in univariate analysis, we used backward selection with a threshold of P<0.05 to construct final multivariable models.
We performed a priori sensitivity analyses to assess the robustness of our culture-based HAI definition. Under varying definitions of the HAI outcome, we fit a model using the final selected variables and assessed the change in hazard ratios. Our first sensitivity analyses required that a urine-positive culture also have a urinary tract infection diagnosis code during the admission, and a respiratory-positive culture was required to have a diagnosis of pneumonia or upper respiratory infection; all blood cultures were included (except common skin contaminants isolated in a single blood culture, unless only a single blood culture was drawn from which coagulase negative staphylococci were isolated). A second sensitivity analysis included all blood and respiratory positive HAI cultures but required that urine-positive cultures have a urinalysis (UA) test ± two days of the urine culture with ≥10 white blood cells (i.e., pyuria), to be considered a UTI. All analyses were performed in SAS v9.4 (Cary, NC).
Results
The final cohort included 254,792 adult admissions ≥48 hours across the three study hospitals from 1/1/17 through 8/31/2020. Among these admissions, patients were predominantly White (63.4%) and female (55.4%). See Table 1 for descriptive statistics of the cohort of admissions.Table 1 Characteristics of the Cohort of Hospital Admissions by HAI typea
Table 1Variable Value Total n (%) HAI- urine n (%) HAI- blood n (%) HAI- respiratory n (%) No HAI n (%)
Total 254,792 3,407 1,661 2,626 247,645
Age, mean (SD) 57.6 (19.0) 65.9 (16.2) 59.5 (15.6) 59.8 (16.0) 57.4 (19.0)
Male 113,685 (44.6) 1,164 (34.2) 963 (58.0) 1,676 (63.8) 110,177 (44.5)
Race Black 84,789 (33.3) 972 (28.5) 450 (27.1) 834 (31.8) 82,723 (33.4)
White 161,609 (63.4) 2,319 (68.1) 1,147 (69.1) 1,672 (63.7) 156,798 (63.3)
Other/missing 8,394 (3.3) 116 (3.4) 64 (3.9) 120 (4.6) 8,124 (3.3)
Hospital 1 154,737 (60.7) 2,293 (67.3) 1,345 (81.0) 1,952 (74.3) 149,594 (60.4)
2 39,126 (15.4) 377 (11.1) 177 (10.7) 371 (14.1) 38,267 (15.5)
3 60,929 (23.9) 737 (21.6) 139 (8.4) 303 (11.5) 59,784 (24.1)
Medicaid/self-pay 80,024 (31.4) 899 (26.4) 501 (30.2) 868 (33.1) 77,919 (31.5)
Year of admissionb 2017 66,136 (26.0) 908 (26.7) 386 (23.2) 763 (29.1) 64,215 (25.9)
2018 69,742 (27.4) 951 (27.9) 395 (23.8) 677 (25.8) 67,847 (27.4)
2019 74,170 (29.1) 999 (29.3) 521 (31.4) 695 (26.5) 72,134 (29.1)
2020 44,744 (17.6) 549 (16.1) 359 (21.6) 491 (18.7) 43,449 (17.5)
BMI category BMI <18.5 underweight 10,079 (4.0) 163 (4.8) 76 (4.6) 196 (7.5) 9,673 (3.9)
BMI 18.5 to <25: normal 67,973 (26.7) 877 (25.7) 451 (27.2) 761 (29.0) 66,017 (26.7)
BMI 25 to <30: overweight 69,050 (27.1) 890 (26.1) 456 (27.5) 697 (26.5) 67,151 (27.1)
BMI 30 to <35: obese class 1 47,497 (18.6) 643 (18.9) 337 (20.3) 464 (17.7) 46,155 (18.6)
BMI 35 to <40: obese class 2 25,927 (10.2) 356 (10.4) 168 (10.1) 221 (8.4) 25,238 (10.2)
BMI ≥40: obese class 3 morbid obesity 25,444 (10.0) 427 (12.5) 169 (10.2) 279 (10.6) 24,651 (10.0)
Missing BMI 8,822 (3.5) 51 (1.5) 4 (0.2) 8 (0.3) 8,760 (3.5)
ICU admissionc 48,870 (19.2) 1,415 (41.5) 576 (34.7) 2,120 (80.7) 45,171 (18.2)
Modified APACHE II score 1st quartile (lowest values) 58,980 (23.1) 564 (16.6) 262 (15.8) 227 (8.6) 57,969 (23.4)
2nd quartile 65,159 (25.6) 674 (19.8) 290 (17.5) 302 (11.5) 63,964 (25.8)
3rd quartile 71,757 (28.2) 975 (28.6) 451 (27.2) 497 (18.9) 69,950 (28.2)
4th quartile (highest values) 58,896 (23.1) 1,194 (35.0) 658 (39.6) 1,600 (60.9) 55,762 (22.5)
AIDS 1,839 (0.7) 13 (0.4) 11 (0.7) 26 (1.0) 1,791 (0.7)
Alcohol abuse 14,058 (5.5) 190 (5.6) 95 (5.7) 271 (10.3) 13,548 (5.5)
Deficiency anaemias 59,378 (23.3) 1,269 (37.2) 597 (35.9) 1,089 (41.5) 56,692 (22.9)
Rheumatoid arthritis/collagen vascular disease 11,707 (4.6) 199 (5.8) 88 (5.3) 148 (5.6) 11,301 (4.6)
Chronic blood loss anaemia 6,807 (2.7) 88 (2.6) 48 (2.9) 53 (2.0) 6,633 (2.7)
Congestive heart failure 59,145 (23.2) 1,318 (38.7) 531 (32.0) 1,113 (42.4) 56,440 (22.8)
Chronic pulmonary disease 66,044 (25.9) 1,010 (29.6) 452 (27.2) 1,168 (44.5) 63,622 (25.7)
Chronic kidney disease 57,561 (22.6) 1,234 (36.2) 567 (34.1) 887 (33.8) 55,102 (22.3)
Coagulopathy 24,603 (9.7) 794 (23.3) 626 (37.7) 1,148 (43.7) 22,318 (9.0)
Depression 49,721 (19.5) 893 (26.2) 432 (26.0) 648 (24.7) 47,897 (19.3)
Diabetes 76,724 (30.1) 1,363 (40.0) 619 (37.3) 1,033 (39.3) 73,963 (29.9)
Drug abuse 14,561 (5.7) 128 (3.8) 75 (4.5) 194 (7.4) 14,195 (5.7)
Hypertension 159,737 (62.7) 2,736 (80.3) 1,182 (71.2) 2,047 (78.0) 154,228 (62.3)
Hypothyroidism 33,583 (13.2) 713 (20.9) 265 (16.0) 417 (15.9) 32,302 (13.0)
Liver disease 18,748 (7.4) 310 (9.1) 256 (15.4) 340 (12.9) 17,933 (7.2)
Lymphoma 7,163 (2.8) 134 (3.9) 204 (12.3) 64 (2.4) 6,783 (2.7)
Metastatic cancer 18,125 (7.1) 268 (7.9) 142 (8.5) 142 (5.4) 17,599 (7.1)
Other neurological disorders 37,420 (14.7) 985 (28.9) 350 (21.1) 819 (31.2) 35,476 (14.3)
Paralysis 15,484 (6.1) 632 (18.6) 160 (9.6) 512 (19.5) 14,310 (5.8)
Peripheral vascular disease 23,215 (9.1) 551 (16.2) 282 (17.0) 563 (21.4) 21,942 (8.9)
Psychoses 17,280 (6.8) 242 (7.1) 83 (5.0) 173 (6.6) 16,821 (6.8)
Pulmonary circulation disease 5,806 (2.3) 167 (4.9) 109 (6.6) 198 (7.5) 5,381 (2.2)
Solid tumour without metastasis 15,086 (5.9) 233 (6.8) 94 (5.7) 146 (5.6) 14,640 (5.9)
Valvular disease 28,240 (11.1) 714 (21.0) 283 (17.0) 573 (21.8) 26,810 (10.8)
Abbreviations: AIDS, Acquired immune deficiency syndrome; BMI, body mass index; HAI, hospital-acquired infection; SD, standard deviation.
a HAI types during an admission were not mutually exclusive. There were 3,050 urine-only HAI admissions, 1,355 blood-only HAI admissions, 2,221 respiratory-only HAI admissions, 215 urine and respiratory HAI admissions, 164 blood and respiratory HAI admissions, 116 urine and blood HAI admissions, and 26 admissions with urine, blood, and respiratory HAIs.
b Admissions in 2020 were only through an admit date of August 31, 2020.
c ICU admission before HAI, as applicable.
Among cultures collected ≥48 hours into the index admission (i.e., HAIs), there were a total of 3,492 positive urine cultures, 1,733 blood cultures, and 2,943 respiratory cultures (Supplemental Table 1). A listing of the most common organisms isolated ≥48 hours after hospitalization can be found in Supplemental Table 1.
Before the SARS-CoV-2 pandemic, the HAI rate across the three hospitals was 2.8% (n=6,202). Among patients admitted during the pandemic, 2.8% (n=798) of admissions among patients without COVID-19 had an HAI, whereas 9.4% (n=147) admissions among patients with COVID-19 had an HAI.
Multivariable Cox models
Univariate factors associated with increased risk of HAI are shown in Supplemental Table 2. Factors associated with increased risk of HAI in multivariable analysis are shown in Table 2 . Extremes of age (18-25 years of age and >66) were associated with increased risk of HAI, as were extremes of BMI (<18.5 and ≥40). Study hospital also significantly impacted likelihood of HAI, with the lower risk for HAI at the two community hospitals, as compared to the tertiary care hospital.Table 2 Multivariable Cox Proportional Hazards Model of Significant Risk Factors for Hospital-acquired Infections
Table 2Variable Value HAI- primary analysis HR (95% CI)
Admission with COVID-19 1.65 (1.38, 1.96)
Admitting service Medicine Ref.
Neurology 1.48 (1.33, 1.65)
OB/GYN 0.60 (0.48, 0.77)
Oncology 1.28 (1.18, 1.39)
Ortho/neurosurgery 1.05 (0.94, 1.18)
Other/missing 1.06 (0.95, 1.18)
Psychiatry 0.27 (0.20, 0.36)
Surgery 1.09 (1.02, 1.16)
Age in 5 year categories 18-25 years 1.09 (0.93, 1.27)
26-30 years 0.85 (0.72, 1.01)
31-35 years 0.87 (0.74, 1.02)
36-40 years 0.91 (0.78, 1.05)
41-45 years 0.86 (0.74, 0.99)
46-50 years 0.95 (0.84, 1.07)
51-55 years 0.89 (0.80, 0.99)
56-60 years 0.97 (0.89, 1.07)
61-65 years Ref.
66-70 years 1.04 (0.95, 1.14)
71-75 years 1.11 (1.01, 1.21)
76-80 years 1.11 (1.00, 1.23)
81-85 years 1.13 (1.00, 1.26)
86+ years 0.99 (0.88, 1.13)
Body mass index (BMI) category BMI <18.5 underweight 1.20 (1.07, 1.34)
BMI 18.5 to <25: normal 0.99 (0.93, 1.06)
BMI 25 to <30: overweight Ref.
BMI 30 to <35: obese class 1 1.07 (1.00, 1.15)
BMI 35 to <40: obese class 2 1.02 (0.93, 1.11)
BMI ≥40: obese class 3 morbid obesity 1.16 (1.06, 1.26)
Missing BMI 0.68 (0.52, 0.88)
Study hospital 1 Ref.
2 0.73 (0.67, 0.79)
3 0.89 (0.83, 0.96)
Male 0.85 (0.81, 0.89)
Race White Ref.
Black 0.88 (0.83, 0.93)
Other/refused/missing 1.14 (1.01, 1.29)
Year of admission 2017 Ref.
2018 0.91 (0.86, 0.97)
2019 0.88 (0.83, 0.94)
2020 0.84 (0.78, 0.90)
Deficiency anaemias 1.12 (1.06, 1.18)
Congestive heart failure 1.15 (1.08, 1.21)
Chronic pulmonary disease 1.14 (1.08, 1.20)
Coagulopathy 1.58 (1.49, 1.67)
Depression 1.25 (1.18, 1.34)
Hypertension 1.27 (1.18, 1.35)
Hypothyroidism 1.10 (1.03, 1.18)
Other neurological disorders 1.30 (1.23, 1.38)
Paralysis 1.69 (1.58, 1.81)
Peripheral vascular disease 1.09 (1.02, 1.16)
Pulmonary circulation disease 1.26 (1.14, 1.39)
Have 1+ ICU stays for the admission 1.12 (1.06, 1.19)
Medication: ACE inhibitors 0.71 (0.66, 0.76)
Medication: H2 antagonists 1.23 (1.17, 1.29)
Medication: angiotensin II inhibitors 0.79 (0.73, 0.86)
Medication: antacids 0.76 (0.70, 0.82)
Medication: antidepressants 0.72 (0.68, 0.76)
Medication: antihypertension combinations 0.68 (0.52, 0.89)
Modified APACHE II score 1st quartile (lowest values) Ref.
2nd quartile 0.98 (0.91, 1.07)
3rd quartile 1.07 (0.99, 1.15)
4th quartile (highest values) 1.19 (1.10, 1.28)
Procedures: GI 0.74 (0.69, 0.79)
Procedures: GU 0.75 (0.64, 0.87)
Procedures: rehabilitation 0.74 (0.64, 0.86)
Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; ACE, angiotensin-converting enzyme; CI, confidence interval; COVID-19, coronavirus disease of 2019; GI, gastrointestinal; GU, genitourinary; HR, hazard ratio; ICU, intensive care unit; OB/GYN, obstetrics-gynaecology.
A full list of comorbidities associated with increased risk of HAI are shown in Table 2. Many chronic conditions, such as pulmonary and cardiovascular comorbidities increased HAI risk (Table 2). Hospital admission factors associated with increased risk of HAI included an ICU stay (HR 1.12), increasing APACHE II score, and a COVID-19 diagnosis. Among all risk factors included in the final multivariable model, a COVID-19 diagnosis was one of the strongest risk factors for HAI (HR 1.65).
Regarding sensitivity analyses, the multivariable models performed similarly without appreciable changes in hazard ratios for COVID-19 under our a priori sensitivity analysis conditions (Supplemental Table 3), with the overall analysis having the most conservative hazard ratio for COVID-19 (HR 1.65 in primary analysis, 1.85 in sensitivity analysis for diagnoses, and 1.69 for the urinalysis-based sensitivity analysis). For the first sensitivity analysis, we ran the multivariable model only among admissions with positive blood cultures, positive respiratory cultures plus a concomitant ICD-10 code for pneumonia and/or upper respiratory infection, and positive urine cultures plus an ICD-10 code for UTI. Among admissions with positive urine cultures (n=3,407), 74.6% (n=2,540) had an ICD-10 code for UTI. Among admissions with positive respiratory cultures (n=2,626), 77.8% (n=2,042) had an ICD-10 code for pneumonia and/or an upper respiratory infection.
In our second sensitivity analysis, we restricted admissions among patients with positive urine cultures further to only those with concomitant pyuria. Of the 3,407 admissions with positive urine cultures, 82.0% had pyuria ± two days of the urine HAI. The hazard ratios for COVID-19 (for the model including urine, blood, and respiratory HAIs) did not appreciably change for the model when including those only with pyuria (HR of 1.69 in sensitivity analysis).
Discussion
Our study is among the first to describe risk factors for hospital-onset infections in patients with COVID-19 that also includes the pre-COVID era and has patient-level data. Patients with COVID-19 had higher risk for an HAI (urine, blood, or respiratory), consistent with prior literature.4 , 7 , 18 As is consistent with national trends in decreased rates of HAIs over time prior to the pandemic, year of admission was associated with decreased hazard ratios for HAI as the study period progressed. A recent study showed that again in 2021, HAI rates increased, though because of data limitations, the authors were unable to correlate HAI rates with COVID-19.19
We found an association between HAIs and multiple risk factors, including comorbidities, intensive-care unit stay, and age, among others. Intensive care unit stays as a risk factor for HAIs is well described.20 Malnutrition has been demonstrated to be a risk for HAI previously.21 , 22 Comorbidities, extremes of age, admitting service, and hospital type have also been found to be associated with development of HAIs.23 , 24
We found that COVID-19 was a risk factor for the development of HAIs in our primary analysis, as well as our sensitivity analyses, with our primary analysis having the most conservative hazard ratio. This suggests that COVID-19 is a consistent risk factor for HAI development. Previous studies have shown that the correlation between laboratory parameters and ICD-10 codes may vary by site of infection.25 , 26 Our study was not designed to assess the relationship between laboratory parameters and ICD-10 codes, and this is a potential limitation, i.e. that our sensitivity analyses did not fully capture the population with true HAIs.
Our study is limited by its regional nature, with all hospitals located in Missouri, which may not be representative of patient populations and risk factors for HAIs across the country and globe. In addition, we present multivariable models based on culture data to assess risk factors for HAIs. We acknowledge that it is likely not all positive cultures are reflective of true infection, particularly for urine and respiratory cultures. To increase the likelihood that the positive culture reflected true infection we performed sensitivity analyses using ICD-10 codes for UTI and lower respiratory tract infection, and in a second analysis required pyuria for positive urine specimens. We performed these sensitivity analyses as medical record review for signs and symptoms would be impractical with the large number of admissions included in our study, and is known to lack accuracy retrospectively. These sensitivity analyses did not reveal appreciable differences in the hazard ratios for COVID-19 of the multivariable model.
Conclusions
In conclusion, COVID-19 predisposed to HAIs, along with many other risk factors typically associated with HAIs. Continued vigilance in maintaining good infection prevention practices to minimize HAIs is warranted, including hand hygiene, minimizing unnecessary device use, antimicrobial stewardship, and first and foremost, SARS-CoV-2 vaccination.
Financial disclosure
This study was funded by CDC grant 75D30120C09598 to JPB. The content is solely the responsibility of the authors and does not necessarily represent the official view of the CDC. JHK is supported by the National Institute of Allergy and Infectious Diseases, National Institutes of Health (grant 1K23AI137321 to J. H. K.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Declaration of Competing Interest
All authors declare no conflicts of interest.
Acknowledgments
The authors would like to thank Cherie Hill for her assistance in data acquisition.
==== Refs
References
1 Langford B.J. So M. Raybardhan S. Antibiotic prescribing in patients with COVID-19: rapid review and meta-analysis Clin Microbiol Infect 27 4 2021 520 531 33418017
2 Rawson T.M. Moore L.S.P. Zhu N. Bacterial and Fungal Coinfection in Individuals With Coronavirus: A Rapid Review To Support COVID-19 Antimicrobial Prescribing Clin Infect Dis 71 9 2020 2459 2468 32358954
3 Moore S.E. Wilde A.M. Bohn B.C. Song M. Schulz P. Antimicrobial stewardship in patients with confirmed coronavirus disease 2019 (COVID-19) Infect Control Hosp Epidemiol 2021 1 3
4 Weiner-Lastinger L.M. Pattabiraman V. Konnor R.Y. The impact of coronavirus disease 2019 (COVID-19) on healthcare-associated infections in 2020: A summary of data reported to the National Healthcare Safety Network Infect Control Hosp Epidemiol 2021 1 14
5 Ong C.C.H. Farhanah S. Linn K.Z. Nosocomial infections among COVID-19 patients: an analysis of intensive care unit surveillance data Antimicrob Resist Infect Control 10 1 2021 119 34384493
6 Copaja-Corzo C. Hueda-Zavaleta M. Benites-Zapata V.A. Rodriguez-Morales A.J. Antibiotic Use and Fatal Outcomes among Critically Ill Patients with COVID-19 in Tacna, Peru Antibiotics (Basel) 10 8 2021
7 Baker M.A. Sands K.E. Huang S.S. The Impact of COVID-19 on Healthcare-Associated Infections Clin Infect Dis 74 10 2021
8 Baccolini V. Migliara G. Isonne C. The impact of the COVID-19 pandemic on healthcare-associated infections in intensive care unit patients: a retrospective cohort study Antimicrob Resist Infect Control 10 1 2021 87 34088341
9 Smith L. Karaba S.M. Amoah J. Hospital-acquired infections among adult patients admitted for coronavirus disease 2019 (COVID-19) Infect Control Hosp Epidemiol 2021 1 4
10 Doherty J. Noirot L.A. Mayfield J. Implementing GermWatcher, an enterprise infection control application AMIA Annu Symp Proc 2006 2006 209 213 17238333
11 Centers for Disease Control and Prevention. Identifying Healthcare-associated Infections (HAI) for NHSN Surveillance. 2021.
12 Median household income, past 12 months (in 2019 INFLATION-ADJUSTED DOLLARS) Source: U.S. Census Bureau, 2015-2019 American Community Survey 5-Year Estimates. 2021. https://data.census.gov/cedsci/. Accessed January 31, 2021.
13 CDC/ATSDR Social Vulnerability Index. 2021. https://www.atsdr.cdc.gov/placeandhealth/svi/index.html. Accessed January 31, 2021.
14 Elixhauser A. Steiner C. Harris D.R. Coffey R.M. Comorbidity measures for use with administrative data Med Care 36 1 1998 8 27 9431328
15 HCUP Elixhauser Comorbidity Software. Healthcare Cost and Utilization Project (HCUP). us.ahrq.gov/toolssoftware/comorbidity/comorbidity.jsp. Accessed December 21, 2020.
16 Knaus W.A. Draper E.A. Wagner D.P. Zimmerman J.E. APACHE II: a severity of disease classification system Crit Care Med 13 10 1985 818 829 3928249
17 Kohl M. Plischke M. Leffondré K. Heinze G. PSHREG: a SAS macro for proportional and nonproportional subdistribution hazards regression Comput Methods Programs Biomed 118 2 2015 218 233 25572709
18 Sturm L.K. Saake K. Roberts P.B. Masoudi F.A. Fakih M.G. Impact of COVID-19 pandemic on hospital onset bloodstream infections (HOBSI) at a large health system Am J Infect Control 50 3 2022 245 249 34971717
19 Lastinger L.M. Alvarez C.R. Kofman A. Continued increases in the incidence of healthcare-associated infection (HAI) during the second year of the coronavirus disease 2019 (COVID-19) pandemic Infect Control Hosp Epidemiol 2022 1 5
20 Markwart R. Saito H. Harder T. Epidemiology and burden of sepsis acquired in hospitals and intensive care units: a systematic review and meta-analysis Intensive Care Med 46 8 2020 1536 1551 32591853
21 Rong A. Franco-Garcia E. Zhou C. Association of nutrition status and hospital-acquired infections in older adult orthopedic trauma patients JPEN J Parenter Enteral Nutr 46 1 2022 69 74 33660849
22 Coca D.J. Castelblanco S.M. Chavarro-Carvajal D.A. Venegas-Sanabria L.C. In-hospital complications in an acute care geriatric unit Biomedica 41 2 2021 293 301 34214270
23 Xie D.S. Xiong W. Xiang L.L. Point prevalence surveys of healthcare-associated infection in 13 hospitals in Hubei Province, China, 2007-2008 J Hosp Infect 76 2 2010 150 155 20692727
24 Kim B.G. Kang M. Lim J. Comprehensive risk assessment for hospital-acquired pneumonia: sociodemographic, clinical, and hospital environmental factors associated with the incidence of hospital-acquired pneumonia BMC Pulm Med 22 1 2022 21 35016645
25 Sogaard K.K. Thomsen R.W. Schonheyder H.C. Sogaard M. Positive predictive values of the International Classification of Diseases, 10th revision diagnoses of Gram-negative septicemia/sepsis and urosepsis for presence of Gram-negative bacteremia Clin Epidemiol 7 2015 195 199 25709502
26 van der Werff S.D. Thiman E. Tanushi H. The accuracy of fully automated algorithms for surveillance of healthcare-associated urinary tract infections in hospitalized patients J Hosp Infect 110 2021 139 147 33548370
| 36493966 | PMC9724556 | NO-CC CODE | 2022-12-09 23:15:14 | no | J Hosp Infect. 2022 Dec 6; doi: 10.1016/j.jhin.2022.11.020 | utf-8 | J Hosp Infect | 2,022 | 10.1016/j.jhin.2022.11.020 | oa_other |
==== Front
Parkinsonism Relat Disord
Parkinsonism Relat Disord
Parkinsonism & Related Disorders
1353-8020
1873-5126
Elsevier Ltd.
S1353-8020(22)00407-2
10.1016/j.parkreldis.2022.105238
105238
Article
Morbidity and severity of COVID-19 in patients with Parkinson's disease treated with amantadine - A multicenter, retrospective, observational study
Przytuła Filip a∗
Kasprzak Jakub a
Dulski Jarosław abc
Koziorowski Dariusz d
Kwaśniak-Butowska Magdalena ab
Sołtan Witold a
Roszmann Anna ab
Śmiłowska Katarzyna e
Schinwelski Michał f
Sławek Jarosław ab∗∗
a Neurology&Stroke Dpt, St. Adalbert Hospital, Gdańsk, Poland
b Department of Neurological-Psychiatric Nursing, Faculty of Health Sciences, Medical University of Gdańsk, Poland
c Department of Neurology, Mayo Clinic, Jacksonville, FL, USA
d Neurology Dpt, Faculty of Health Sciences, Medical University of Warsaw, Poland and Bródno Hospital, Warsaw, Poland
e Neurology Dpt, St. Barbara Hospital, Sosnowiec, Poland
f Neurocentrum Miwomed Sp. z o.o, Gdańsk, Poland
∗ Corresponding author. Neurology&Stroke Dpt, St. Adalbert Hospital, Al. Jana Pawła II 50, 80-462, Gdańsk, Poland.
∗∗ Corresponding author. Department of Neurological-Psychiatric Nursing, Faculty of Health Sciences, Medical University of Gdańsk, Poland Medical Univeristy of Gdańsk and Neurology&Stroke Dpt., St. Adalbert Hospital, Al. Jana Pawła II 50, 80-462, Gdańsk, Poland.
6 12 2022
6 12 2022
10523822 8 2022
19 11 2022
5 12 2022
© 2022 Elsevier Ltd. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
After more than 2 years of the pandemic, effective treatment for COVID-19 is still under research. In recent months, publications hypothesized amantadine's potential beneficial effect on SARS-CoV-2 infection.
Objective
To compare the groups of Parkinson's Disease (PD) patients who were administered amantadine chronically and those who did not take this medication in the context of the incidence and severity of COVID-19 infection.
Methods
An observational, retrospective, multicenter cohort study was conducted among consecutive patients with idiopathic PD. The structured questionnaires were completed during the patient's follow-up visits at the Outpatient Clinic or during hospitalization. The questionnaire included the following informations: patient's age, duration of PD, Hoehn-Yahr (H–Y) stage, comorbidities, medications, COVID-19 confirmed by reverse transcription polymerase chain reaction (RT-PCR) swab test for SARS-CoV-2 with specified symptoms and their severity (home or hospital treatment). The vaccination status was verified as well.
Results
Five hundred fifty-two (n = 552) patients participated in the study - 329 men (60%). The mean H–Y stage was 2.44 (range: 1–4) and the mean duration of PD was 9.6 years (range: 1–34). One hundred four subjects (19%) had confirmed COVID-19 infection. Subjects over 50 years of age had a significantly lower incidence of COVID-19 (17% vs 38%, p = 0.0001) with difference also in mean H–Y stage (2.27 vs 2.49; p = 0.011) and disease duration (8.4 vs 9.9 years, p = 0.007). There were no differences between patients with and without co-morbidities. In the whole analyzed group 219 (40%) subjects were treated with amantadine. Comparing COVID-19 positive and negative patients, amantadine was used by 48/104 (46%) and 171/448 (38%) respectively. 22% of patients on amantadine vs. 17% of patients without amantadine developed COVID-19. These differences were not significant. There were no differences in morbidity and severity of COVID-19 between amantadine users and non-users as well.
Conclusions
COVID-19 was less common in older (>50) with longer duration and more advanced patients. Amantadine did not affect the risk of developing COVID-19 or the severity of infection.
Keywords
Parkinson's disease
COVID-19
Amantadine
SARS-CoV-2
Morbidity
==== Body
pmc1 Introduction
The global scale of the Coronavirus disease 2019 (COVID-19) pandemic due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, the high number of deaths and complications that affect the quality of life of patients, oblige to search for effective medication. Vaccination is currently an effective method of preventing and mitigating COVID-19, but the duration of the protection and efficacy against new variants of SARS-CoV-2 are still uncertain. Several medications were tested and included into treatment protocols [1], however till now there are no highly effective therapies available. In recent months, publications hypothesized the potential beneficial therapeutic effect of amantadine in COVID-19 treatment [[2], [3], [4], [5], [6], [7], [8], [9], [10]]. Currently, three clinical trials are ongoing to assess the effect of amantadine on COVID-19 compared to placebo [[11], [12], [13]]. Furthermore several observational studies have been published suggesting that amantadine may have a beneficial effect in the treatment of SARS-CoV-2 infection [[14], [15], [16], [17]]. However, these studies have significant limitations including: small number of patients (mostly case series reports), concurrent treatment with other therapies; no information on other medications, no endpoints predefined in the study methodology for the efficacy and safety of amantadine. Wszołek and Tipton pointed out the need to analyze the population of patients with neurodegenerative diseases with regard to SARS-CoV-2 infection, especially patients taking amantadine and memantine chronically due to their potential antiviral effect [18]. Even though amantadine is considered as a relatively safe drug, it should be used with caution. Amantadine has a potency to cause side effects, sometimes life-threatening [19,20]. Amantadine is frequently self-administered in belief of antiviral effect. The time of administration differs among patients and may be crucial for its effectiveness. Therefore, we hypothesized that PD patients (older population at risk of fatal outcome) taking amantadine chronically as a part of anti-parkinsonian treatment may be protected against infection or fatal outcomes.
2 Study aim
The aim of the study was to compare the group of PD patients who were on amantadine treatment chronically and those who did not take amantadine in the context of the incidence and severity of COVID-19. The secondary goal was to assess the impact of the COVID-19 pandemic on the population of patients with PD and comorbidities.
2.1 Study design
Between November 30, 2020 and October 18, 2021 a survey was conducted among patients with PD. Consecutive patients treated in 4 centers specializing in movement disorders were included and analyzed. The treating physicians filled in the questionnaires, during the patient's follow-up visit at the Outpatient Clinic or during hospitalization. The structured questionnaire included questions on patient's age, duration of PD, Hoehn-Yahr (HY) stage, comorbidities, medications used for both PD and comorbidities, advanced PD treatments (deep brain stimulation - DBS, continuous subcutaneous apomorphine infusions- CSAI, levodopa/carbidopa intestinal gel - LCIG). Patients were asked whether they had COVID-19 confirmed by a reverse transcription polymerase chain reaction (RT-PCR) swab test for SARS-CoV-2. Patients were also questioned if they had been vaccinated against COVID-19 (partially or completed the full vaccination). The data on the severity of COVID-19 symptoms (dyspnoea, cough, fever, anosmia/ageusia, other infection symptoms), and the need of hospitalization (oxygen therapy, mechanical ventilation and hospitalization at the Intensive Care Unit) was collected as well. Patients were asked if they were taking amantadine before and during infection.
As the collected data are a part of routine history taking and there were no any interventions or additional scales, the Bioethical Committee approval was not necessary.
2.2 Statistical analysis
We calculated and presented the data as frequencies, percentages or mean. Qualitative comparisons were made using Chi-square and Yates corrected Chi-square test. Quantitative comparisons were made using Mann-Whitney U tests as all data in compared groups were not normally distributed. A p-value less than 0.05 was considered statistically significant for all comparisons. Statistical analyses were performed using Statistica 14 and SPSS 28.
3 Results
3.1 Basic demographics
Five hundred fifty two (n = 552) patients including 329 men (60%) and 223 women (40%), mean age of 64.8 years (SD ± 10.7, range: 27–89) participated in the study. The mean H–Y stage of PD was 2.44 (SD 1.15, range: 1–4). The mean duration of PD was 9.6 years (SD 6.06, range: 1–34) and the average levodopa equivalent daily dose (LEDD) was 1065 mg (SD 614, range). Three hundred nine subjects (n = 309, 56%) had at least one comorbidity.
3.2 Vaccination status
Four hundred eighteen patients (n = 418) were not vaccinated against SARS-CoV-2 at the time of completing the questionnaire. One hundred thirty-four (n = 134) were already fully vaccinated against COVID-19 before completing the survey, however, the study was performed mostly before SARS-CoV-2 vaccination availability. The first vaccination of the patient among the respondents was performed on January 23, 2021. Of the 134 subjects fully vaccinated before completing the survey, 25 developed SARS-CoV-2 infection in the time period between the onset of the pandemic and vaccination. Proportion of patients fully vaccinated prior to the study was the same in the COVID-19 positive and negative groups (both 24%) (Table 2 ).Table 1 Amantadine usage and dosages among COVID-19 positive and negative patients.
Table 1 Number of patients
COVID-19 (+): 104
Amantadine users 48 (46%)
1 × 100 mg 3
2 × 100 mg 25
3 × 100 mg 19
4 × 100 mg 1
Mean daily amantadine intake 237 mg
Amantadine non-users 56 (54%)
COVID-19 (−): 448
Amantadine users 171 (38%)
2 × 50 mg 2
3 × 50 mg 1
1 × 100 mg, 2 × 50 mg 1
1 × 100 mg 14
2 × 100 mg 58
3 × 100 mg 76
4 × 100 mg 19
Mean daily amantadine intake 257 mg
Amantadine non-users 277 (62%)
Table 2 The mean age, gender distribution, disease severity (H–Y: Hoehn-Yahr), disease duration and vaccination status at the time of completing the survey.
Table 2 COVID-19 (+) COVID-19 (−) p-value
Number of patients 104 448 –
Amantadine users 48 (46%) 171 (38%) 0.1338
Mean daily amantadine intake 237 mg 257 mg 0.1010
Mean age (years) 60.8 65.8 0.0001
Gender M: 67 (64%) M: 262 (59%) 0.2660
F: 37 (36%) F: 186 (41%)
Mean H–Y stage 2.27 2.49 0.0110
Mean PD duration (years) 8.4 9.9 0.0078
Mean LEDD 994 mg 1082 mg 0.1140
Vaccination status 25 vaccinated (24%) 109 vaccinated (24%) 0.9501
79 non-vaccinated (76%) 339 non-vaccinated (76%)
3.3 COVID-19 infection
Out of all 552 patients, 104 patients (19%) had confirmed COVID-19 infection with the RT-PCR test, 447 patients (81%) were negative, in 1 patient the result of the RT-PCR test was inconclusive twice. This subject was included in the study and listed as negative in the statistical analysis. The mean age of subjects in COVID-19 positive and negative groups differed significantly (60.8 vs 65.8 years respectively; p = 0.0001). Among the respondents, SARS-CoV-2 infection was more frequent in younger patients with shorter duration (8.4 vs 9.9 years, p = 0.007) and with less advanced stage of PD (mean H–Y stage 2.27 vs 2.49; p = 0.011) (Table 2). Subjects over 50 years of age had a significantly lower incidence of COVID-19 as compared to those under 50 (17% vs 38%, p = 0.0001).
Among all subjects, 309 patients (56%) had at least one comorbidity, of which 261 were COVID-19 negative and 48 positive. There was no statistically significant increase in the SARS-CoV-2 infection in patients with any comorbidities (Supplementary material). Forty four (n = 44) subjects were diagnosed dementia, of which only 5 had COVID-19 (11%). Of the dementia patients, 8 were taking memantine, none of which developed COVID-19.
Among patients taking other medications (levodopa, dopamine receptor agonists, MAO-B inhibitors, COMT-inhibitors, anticholinergics), no one significantly reduced the incidence of COVID-19 (Table 3 ). Ninety-five patients who were treated with DBS had the lowest COVID-19 morbidity (12%, p = 0.040), however the number of patients in this group was small and requires further investigation.Table 3 (A) Anti-parkinsonian treatment and COVID-19 (+) percentage for each treatment. (B) COVID-19 symptoms in amantadine users and non-users. DBS- Deep Brain Stimulation LCIG – Levodopa/carbidopa intenstinal gel.
Table 3A
Medication Number of patients COVID-19 (+) COVID-19 (−) p-value
Levodopa 514 96 418 0.7178
Rasagiline 223 40 183 0.2730
Amantadine 219 48 171 0.1338
Ropinirole 198 36 162 0.7672
Pramipexole 111 21 90 0.9812
Selegiline 4 1 3 0.7448
LCIG 12 2 10 0.8583
Apomorphine 5 2 3 0.8234
Rotigotine 3 0 3 0.9231
Biperiden 2 0 2 0.8234
Pridinole 2 0 2 0.8234
Entacapone 8 0 8 0.3590
DBS 95 11 84 0.0476
B
COVID-19 symptoms Amantadine users number Amantadine non-users number p-value
No symptoms 12 15 0,8360
Musculoskeletal pain 3 2 0,8597
Fatigue 8 7 0,7467
Gastrointestinal symptoms 0 3 0,2985
Dyspnoea 18 12 0,0713
Cough 24 20 0,1416
Fever (>38 °C) 22 26 0,9516
Anosmia/Ageusia 20 16 0,1617
PD symptoms deterioration 4 0 0,0907
Respiratory failure (oxygen therapy) 10 6 0,2488
Mechanical ventilation 3 0 0,1899
3.4 Amantadine
Among all patients, amantadine was taken by 219 (40%) with daily doses ranged between 100 and 400 mg (Table 1). Of COVID-19 negative subgroup amantadine was taken by 171/448 (38%) versus 48/104 (46%) in patients who were infected. The difference was not statistically significant (Table 2). Mean daily dose was similar (237 mg vs 257 mg) between COVID-19 positive and negative groups and the difference was statistically non-significant (p = 0.101).
The number of COVID-19 positive subjects was greater in patients taking amantadine (22% vs 17% without amantadine) however, this difference was not statistically significant (p = 0.133) (Fig. 1 ).Fig. 1 (A) Percentage of COVID-19 positive and negative patients, treated and not treated with amantadine. (B) COVID-19 positive patients treated and not treated with amantadine and COVID-19 negative patients with relation to age.
Fig. 1
Majority (88/104) of COVID-19 positive patients underwent mild infection only. Fifteen (n = 15) patients were hospital admitted (9 of them were taking amantadine. Oxygen therapy was required in 16 (10 were taking amantadine), while 3 patients required respiratory therapy, all were on amantadine treatment (at the age of 53, 68, 81). Severity of symptoms of SARS-CoV-2 infection was similar between patients taking and not taking amantadine (Table 3).
4 Discussion
Amantadine's antiviral effect is to inhibit the M2 proton channel of influenza A virus. The M2 proton channel is necessary to release the viral genetic material into the cytoplasm by acidifying the inside of the virus, dissociating RNA from its bound matrix proteins which allows for the release of genetic material and further replication [21,22]. Amantadine has been widely used since 1966 in the prevention and treatment of influenza A; however, due to the numerous and rapidly spreading influenza virus mutations, the effectiveness of amantadine has become low and its use is currently not recommended [23].
SARS-CoV-2 enters the host cell by the reaction of spike protein (S-protein) with the angiotensin-converting enzyme type 2 (ACE-2) receptor, which is a membrane receptor widely present in many tissues of the body mainly in the epithelium of the respiratory tract, small intestine, endothelium as well as in the heart muscle and kidneys [24]. It is now also known that the ACE-2 receptor is found in many structures of the central nervous system - both in neurons and in glial cells. The invasion of host cells by SARS-CoV-2 involves binding the viral S protein to the ACE-2 receptor using Cathepsin L (CTSL) - an endosomal cysteine protease. These reactions cause the virus fusion with the host cell membrane, releasing the viral genetic material into the cytoplasm [25]. Several hypotheses have been described regarding the possible mechanisms of action of amantadine on COVID-19 [26]. SARS-CoV-2 in its genome has sequences encoding membrane proteins with ion channel activity (Protein E, Protein 3a, ORF7b and ORF10); amantadine can inhibit the activity of protein E and ORF10, reducing viral replication and virus-dependent inflammation [7]. It was hypothesized that amantadine has the ability to bind to the E-protein of the coronavirus ion channel, inhibiting the proton channel and thus preventing the release of viral genetic material into the cytoplasm [6]. In other publications, it is hypothesized that amantadine causes a change in the lysosomal microenvironment and their dysfunction, inhibiting the key reactions needed for SARS-CoV-2 infection [4]. Amantadine use may result in an increase in the pH of endosomes and down-regulation of cathepsin-L - a protease necessary for the connection of the S protein of the coronavirus with the ACE-2 receptor and impaired viral entry and replication [3]. In-vitro studies on the Vero 6 cell line showed that amantadine might potentially inhibit SARS-CoV-2 replication by inhibiting viroporins [27]. Amantadine has a beneficial immunomodulatory effect by reaction with dopaminergic receptors on T lymphocytes, activating resting T effector lymphocytes and inhibiting regulatory T cells [28].
In recent months, studies hypothesized the potential beneficial effect of amantadine on COVID-19 severity. In one of them, a survey was conducted among people with neurological diseases (10 with multiple sclerosis -MS, 5 with PD, 7 with dementia) taking amantadine or memantine who obtained positive COVID-19 RT-PCR swab results. None of the 22 patients developed infection [8]. Another study performed was a retrospective survey of 256 patients with MS and PD, 87 of whom were taking amantadine. Out of the group of patients taking amantadine, 5.7% developed COVID-19, while in the non-amantadine group it was 11.8% [15]. The largest observational-retrospective study of 136 855 patients with confirmed COVID-19 infection, including 319 patients taking amantadine (151 in monotherapy, 168 with antibiotics), indicated that amantadine was ineffective [14]. The use of amantadine monotherapy in COVID-19 treatment was associated with an increased risk of death in the general population. The authors pointed out that the use of amantadine is not justified. Several studies are currently being conducted worldwide to assess the potential effect of amantadine on COVID-19 infection [[11], [12], [13]].
The group of PD patients taking amantadine due to normal prescriptions (symptomatic treatment of parkinsonian symptoms or treating of choreatic peak – of - dose dyskinesias) is specifically interesting as they are already treated at the onset of infection and do not require dose titration. In other studies regarding the general population, amantadine is usually prescribed at the onset of symptoms or after PCR confirmation. It may result in substantial delay and potential decrease of effectiveness.
The results of our study conducted among PD patients indicate that there are no statistically significant differences in the incidence of COVID-19 and the severity of infection between the groups of patients taking and not taking amantadine. However, the results of the study showed that the incidence of COVID-19 is inversely proportional to the age of the respondents, H–Y stage and duration. Patients over 50 years old had a lower incidence of COVID-19 compared to younger subjects. Elderly patients in general population are at particular risk of the severe course of COVID-19 and the highest risk of death. This observation may be associated with less mobility and avoiding of social contacts during the pandemic in older PD patients, and therefore lower exposure to infection. Nevertheless, other factors have to be considered as well. Among the data collected during the study, no other significant factors were found among the PD patients that could affect the course and incidence of COVID-19 (other medications and co-morbidities) (Table 3). However, the subgroup of patients (n = 95) treated with DBS had the lowest percentage of SARS-CoV-2 infection (12%, p = 0.04), but the group was too small to make any firm conclusions and require further investigations.
The limitation of our study was the introduction of SARS-CoV-2 vaccinations during the course of the study. A part of respondents (n = 134, 24%) were vaccinated before taking part in the study. Twenty five vaccinated patients were COVID-19 positive, but this was in the pre-vaccination period. We collected data on vaccination and also did not find any significant differences. The next, but important limitation is the lack of information on mortality in this group of PD patients. It definitely would be of great value to compare this the most unfavourable disease outcome in amantadine users and not users.
5 Conclusions
Our study shows that chronic pre-treatment with amantadine does not reduce the incidence of COVID-19 and its severity among all respondents. However, patients over 50 years old, slightly more advanced and with longer duration PD had a lower incidence of SARS-CoV-2. Further studies on the effects of amantadine on COVID-19 should be continued, specially in terms of mortality.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary data
The following is the Supplementary data to this article:Multimedia component 1
Multimedia component 1
Acknowledgements
We would like to thank Paweł Przytuła from Appsilon Sp. z o.o. for his for his contribution and support in the analysis and cleaning the source data.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.parkreldis.2022.105238.
==== Refs
References
1 Update to living WHO guideline on drugs for covid-19 BMJ 378 2022 Sep 15 10.1136/bmj.o2224.PMID:36109041 o2224
2 Tipton P.W. Wszolek Z.K. What can Parkinson's disease teach us about COVID-19? Neurol. Neurochir. Pol. 54 2 2020 204 206 10.5603/PJNNS.a2020.0039 Epub 2020 Apr 23. PMID: 32323862 32323862
3 Butterworth R.F. Potential for the repurposing of adamantane antivirals for COVID-19 Drugs R 21 3 2021 Sep 267 272 10.1007/s40268-021-00351-6 Epub 2021 Jun 21. PMID: 34152583, PMCID: PMC8214976
4 Smieszek S.P. Przychodzen B.P. Polymeropoulos M.H. Amantadine disrupts lysosomal gene expression: a hypothesis for COVID19 treatment Int. J. Antimicrob. Agents 55 6 2020 Jun 106004 10.1016/j.ijantimicag.2020.106004 Epub 2020 Apr 30. PMID: 32361028, PMCID: PMC7191300
5 Cortés-Borra A. Aranda-Abreu G.E. Amantadine in the prevention of clinical symptoms caused by SARS-CoV-2 Pharmacol. Rep. 73 3 2021 Jun 962 965 10.1007/s43440-021-00231-5 Epub 2021 Feb 18. PMID: 33604795; PMCID: PMC7891470 33604795
6 Abreu G.E.A. Aguilar M.E.H. Covarrubias D.H. Durán F.R. Amantadine as a drug to mitigate the effects of COVID-19 Med. Hypotheses 140 2020 Jul 109755 10.1016/j.mehy.2020.109755 Epub 2020 Apr 25. PMID: 32361100, PMCID: PMC7182751
7 Toft-Bertelsen T.L. Jeppesen M.G. Tzortzini E. Xue K. Giller K. Becker S. Mujezinovic A. Bentzen B.H. Andreas L.B. Kolocouris A. Kledal T.N. Rosenkilde M.M. Author Correction: amantadine inhibits known and novel ion channels encoded by SARS-CoV-2 in vitro Commun Biol 4 1 2021 Dec 10 1402 10.1038/s42003-021-02940-2 PMID: 34893762; PMCID: PMC8661827 34893762
8 Rejdak K. Fiedor P. Bonek R. Goch A. Gala-Błądzińska A. Chełstowski W. Łukasiak J. Kiciak S. Dąbrowski P. Dec M. Król Z.J. Papuć E. Zasybska A. Segiet A. Grieb P. The use of amantadine in the prevention of progression and treatment of COVID-19 symptoms in patients infected with the SARS-CoV-2 virus (COV-PREVENT): study rationale and design Contemp. Clin. Trials 116 2022 May 106755 10.1016/j.cct.2022.106755 Epub 2022 Apr 4. PMID: 35390511; PMCID: PMC8978450
9 Rejdak K. Grieb P. Adamantanes might be protective from COVID-19 in patients with neurological diseases: multiple sclerosis, parkinsonism and cognitive impairment Mult Scler Relat Disord 42 2020 Jul 102163 10.1016/j.msard.2020.102163 Epub 2020 Apr 30. PMID: 32388458, PMCID: PMC7190496
10 Cortés Borra A. Does amantadine have a protective effect against COVID-19? Neurol. Neurochir. Pol. 54 3 2020 284 285 10.5603/PJNNS.a2020.0041 Epub 2020 Jun 4. PMID: 32495926 32495926
11 https://clinicaltrials.gov/ct2/show/NCT04854759
12 https://clinicaltrials.gov/ct2/show/NCT04894617
13 https://clinicaltrials.gov/ct2/show/NCT04952519
14 Mancilla-Galindo J. García-Méndez J.Ó. Márquez-Sánchez J. Reyes-Casarrubias R.E. Aguirre-Aguilar E. Rocha-González H.I. Kammar-García A. All-cause mortality among patients treated with repurposed antivirals and antibiotics for COVID-19 in Mexico City: a real-world observational study EXCLI J 20 2021 Feb 4 199 222 10.17179/excli2021-3413 PMID: 33628159; PMCID: PMC7898041 33628159
15 Kamel W.A. Kamel M.I. Alhasawi A. Elmasry S. AlHamdan F. Al-Hashel J.Y. Effect of pre-exposure use of amantadine on COVID-19 infection: a hospital-based cohort study in patients with Parkinson's disease or multiple sclerosis Front. Neurol. 2021;12(October 1 7 10.3389/fneur.2021.704186
16 Aranda-Abreu G.E. Aranda-Martínez J.D. Araújo R. Hernández-Aguilar M.E. Herrera-Covarrubias D. Rojas-Durán F. Observational study of people infected with SARS-Cov-2, treated with amantadine Pharmacol. Rep. 72 6 2020 Dec 1538 1541 10.1007/s43440-020-00168-1 Epub 2020 Oct 10. PMID: 33040252; PMCID: PMC7547815 33040252
17 Bodnar W. Aranda-Abreu G. Słaboń-Willand M. Kotecka S. Farnik M. Bodnar J. The efficacy of amantadine hydrochloride in the treatment of COVID-19 - a single-center observation study Pol. Merkur. Lek. 49 294 2021 Dec 16 389 393 PMID: 34919079
18 Tipton P.W. Wszolek Z.K. Response to "Does amantadine have a protective effect against COVID-19? Neurol. Neurochir. Pol. 54 3 2020 286 287 10.5603/PJNNS.a2020.0048 Epub 2020 Jun 25. PMID: 32583401 32583401
19 Kocaş C. Türkmen Y. Çetinkal G. Doğan S.M. Right ventricular outflow tract tachycardia after an initial dose of amantadine Turk Kardiyol. Dernegi Arsivi 43 5 2015 Jul 472 474 10.5543/tkda.2015.32885.PMID:26148081
20 Bakhati B. Sibi V.M. Mekala A.P. Ronen J.A. Mungara S.S. Amantadine-induced cardiac arrest in a patient with COVID-19 Cureus 14 1 2022 Jan 17 e21345 10.7759/cureus.21345 PMID: 35186602; PMCID: PMC8850186
21 De Clercq E. Li G. Approved antiviral drugs over the past 50 years Clin. Microbiol. Rev. 29 3 2016 Jul 695 747 10.1128/CMR.00102-15 PMID: 27281742; PMCID: PMC4978613 27281742
22 Balgi A.D. Wang J. Cheng D.Y. Ma C. Pfeifer T.A. Shimizu Y. Anderson H.J. Pinto L.H. Lamb R.A. DeGrado W.F. Roberge M. Inhibitors of the influenza A virus M2 proton channel discovered using a high-throughput yeast growth restoration assay PLoS One 8 2 2013 e55271 10.1371/journal.pone.0055271 Epub 2013 Feb 1. PMID: 23383318, PMCID: PMC3562233
23 Duwe S. Influenza viruses - antiviral therapy and resistance 5:Doc04 GMS Infect Dis 2017 Apr 25 10.3205/id000030 PMID: 30671326, PMCID: PMC6301739
24 Salamanna F. Maglio M. Landini M.P. Fini M. Body localization of ACE-2: on the trail of the keyhole of SARS-CoV-2 PMID: 33344479 Front. Med. 7 2020 Dec 3 594495 10.3389/fmed.2020.594495 PMCID: PMC7744810
25 Zhao M.M. Yang W.L. Yang F.Y. Zhang L. Huang W.J. Hou W. Fan C.F. Jin R.H. Feng Y.M. Wang Y.C. Yang J.K. Cathepsin L plays a key role in SARS-CoV-2 infection in humans and humanized mice and is a promising target for new drug development Signal Transduct. Targeted Ther. 6 1 2021 Mar 27 134 10.1038/s41392-021-00558-8 PMID: 33774649; PMCID: PMC7997800
26 Danysz W. Dekundy A. Scheschonka A. Riederer P. Amantadine: reappraisal of the timeless diamond-target updates and novel therapeutic potentials J. Neural. Transm. 128 2 2021 Feb 127 169 10.1007/s00702-021-02306-2 Epub 2021 Feb 23. PMID: 33624170, PMCID: PMC7901515 33624170
27 Fink K. Nitsche A. Neumann M. Grossegesse M. Eisele K.H. Danysz W. Amantadine inhibits SARS-CoV-2 in vitro Viruses 13 4 2021 Mar 24 539 10.3390/v13040539.PMID:33804989 PMCID: PMC8063946 33804989
28 Brenner S.R. Butterworth R.F. Repurposing of Adamantanes with Transmitter Receptor Antagonist Properties for the Prevention/Treatment of COVID-19 vol. 8 2020 4 1
| 36509028 | PMC9724557 | NO-CC CODE | 2022-12-09 23:15:04 | no | Parkinsonism Relat Disord. 2023 Jan 6; 106:105238 | utf-8 | Parkinsonism Relat Disord | 2,022 | 10.1016/j.parkreldis.2022.105238 | oa_other |
==== Front
J Genet Genomics
J Genet Genomics
Journal of Genetics and Genomics
1673-8527
1673-8527
Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and Genetics Society of China. Published by Elsevier Limited and Science Press.
S1673-8527(22)00255-7
10.1016/j.jgg.2022.11.011
Original Research
Infection of SARS-CoV-2 causes severe pathological changes in mouse testis
Chen Min abcd1
Li Shihua e1
Liu Shujun f1
Zhang Yuhang gh1
Cui Xiuhong a
Lv Limin f
Liu Bowen ad
Zheng Aihua gh∗
Wang Qihui e∗∗
Duo Shuguang f∗∗∗
Gao Fei abcd∗∗∗∗
a State Key Laboratory of Stem Cell and Reproductive Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China
b Institute for Stem Cell and Regeneration, Chinese Academy of Sciences, Beijing 100101, China
c Beijing Institute for Stem Cell and Regenerative Medicine, Beijing 100101, China
d University of Chinese Academy of Sciences, Beijing 100101, China
e CAS Key Laboratory of Pathogen Microbiology and Immunology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China
f Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China
g State Key Laboratory of Integrated Management of Pest Insects and Rodents, Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China
h CAS Center for Excellence in Biotic Interactions, University of Chinese Academy of Sciences, Beijing 100101, China
∗ Corresponding author.
∗∗ Corresponding author.
∗∗∗ Corresponding author.
∗∗∗∗ Corresponding author.
1 These authors have contributed equally to this work and share first authorship.
6 12 2022
6 12 2022
28 9 2022
23 11 2022
23 11 2022
© 2022 Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and Genetics Society of China. Published by Elsevier Limited and Science Press.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has affected more than 600 million people worldwide. Several organs including lung, intestine, and brain are infected by SARS-CoV-2. It has been reported that SARS-CoV-2 receptor angiotensin-converting enzyme-2 (ACE2) is expressed in human testis. However, whether testis is also affected by SARS-CoV-2 is still unclear. In this study, we generate a human ACE2 (hACE2) transgenic mouse model in which the expression of hACE2 gene is regulated by hACE2 promoter. Sertoli and Leydig cells from hACE2 transgenic mice can be infected by SARS-CoV-2 pseudovirus in vitro, and severe pathological changes are observed after injecting the SARS-CoV-2 pseudovirus into the seminiferous tubules. Further studies reveal that Sertoli and Leydig cells from hACE2 transgenic mice are also infected by authentic SARS-CoV-2 virus in vitro. After testis interstitium injection, authentic SARS-CoV-2 viruses are first disseminated to the interstitial cells, and then detected inside the seminiferous tubules which in turn cause germ cell loss and disruption of seminiferous tubules. Our study demonstrates that testis is most likely a target of SARS-CoV-2 virus. Attention should be paid to the reproductive function in SARS-CoV-2 patients.
Keywords
SARS-CoV-2
ACE2
Sertoli cell
Leydig cell
testis
==== Body
pmcIntroduction
Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been declared as a global pandemic by the World Health Organization (WHO) (Li et al., 2020; Zhu et al., 2020). SARS-CoV-2, as well as its close homolog SARS-CoV, which belongs to the Betacoronavirus genus, is a positive-sense single-stranded RNA virus of ∼26-32 kilobases in length (Zhou et al., 2020). SARS-CoV-2 contains four main structural proteins: spike glycoprotein (S), envelope protein, membrane protein, and nucleocapsid protein (N) (Lu et al., 2020). Spike protein, a surface protein forming a characteristic crown on virion, is indispensable for viral binding and fusion. S proteins of SARS-CoV and SARS-CoV-2 share high sequence similarity and interact with the same cellular receptor, angiotensin-converting enzyme 2 (ACE2), to enter the host cells (Hoffmann et al., 2020).
The transmission of SARS-CoV-2 from person to person mostly occurs following respiratory droplet exposure. At present, SARS-CoV-2 has been detected in respiratory fluids, saliva, blood, urine, and feces (Peng et al., 2020; Wang et al., 2020). Besides type II alveolar epithelial cells (AT2) in lungs, ACE2 is also expressed in many other organs, such as heart, kidney, intestine, liver, and testis (Clarke and Turner. 2012; Younis et al., 2020). In human testis, ACE2 is shown to be expressed in spermatogonia, Leydig and Sertoli cells, suggesting that testis is also a potential target organ of SARS-CoV-2 infection (Douglas et al., 2004; Wang and Xu. 2020). However, the studies about the effect of SARS-CoV-2 infection on male fertility are very limited. There are several studies about whether SARS-CoV-2 is present in semen of COVID-19 patients, but the results are controversial. Most of studies do not detect the virus in semen samples (Holtmann et al., 2020; Kayaaslan et al., 2020; Pan et al., 2020; Paoli et al., 2020; Rawlings et al., 2020; Song et al., 2020; Guo et al., 2021; Ma et al., 2021). Only one study reports that SARS-CoV-2 virus are detected in six semen samples of total thirty-eight COVID-19 patients, and two samples are from recovery stage and four from acute infected stage (Li et al., 2020). Among these patients, scrotal discomfort, altered semen parameters and hormone production, and spermatozoa with increased DNA fragmentation are noted. Testis injury, including reduced Leydig cells, swelling and vacuolated Sertoli cells, and inflammatory infiltration are observed in COVID-19 patients by postmortem examination (Yang et al., 2020), suggesting that SARS-CoV-2 infection has obvious adverse effects on spermatogenesis and male fertility, independent of the presence of virus in semen (Gonzalez et al., 2020; Omolaoye et al., 2021). How these adverse effects are exerted is largely unknown.
Animal models are invaluable for studying the transmission and pathogenesis of SARS-CoV-2 and for the evaluation of vaccines and antivirals. Because of the relative low conservation of ACE2 protein between human and mouse, mouse ACE2 can not effectively support the infection of SARS-CoV-2. To study the pathogenesis of SARS-CoV-2 infection, several hACE2 knock-in and transgenic mouse models were generated previously (Bao et al., 2020; Jiang et al., 2020; Sun et al., 2020). In these mouse models, the expression of hAce2 gene is under the control of mouse Ace2 promoter, tissue-specific promoter or other ubiquitous expression promoters (Bao et al., 2020; Jiang et al., 2020; Sun et al., 2020). The expression of hACE2 protein is detected in lung and interstitial pneumonia is observed after intranasally infected with live SARS-CoV-2. However, the expression pattern of ACE2 and the SARS-CoV-2 susceptibility, as well as the SARS-CoV-2 induced pathology in testis are largely unknown. In this study, we generated a transgenic mouse model in which the expression of hACE2 is under the control of hACE2 promoter. We found that hACE2 was expressed in testis and other organs including lung, intestine, and heart, etc. Seminiferous tubule injection of VSV-based SARS-CoV-2 pseudovirus (rVSV-SARS-CoV-2) resulted in severe pathological changes and disruption of seminiferous tubules in testis. In vitro studies showed that Leydig cells and Sertoli cells could be infected by both SARS-CoV-2 pseudovirus and authentic SARS-CoV-2 virus. We also found that authentic SARS-CoV-2 virus first damaged the Leydig cells after injected into testis interstitium, and then the expression of viral protein was detected in the seminiferous tubules which in turn caused germ cell loss and disruption of seminiferous tubules.
Results
Generation of hACE2 transgenic mice
As shown in the schematic diagram of hACE2 transgenic construct in Fig. 1 A, the expression of hACE2 gene was under the control of hACE2 promoter. hACE2 transgenic mice were generated by pronuclear microinjection. PCR genotyping of hACE2 transgenic mouse was shown in Fig. 1B. The expression of hACE2 protein was detected in multiple organs, including heart, intestine, liver, lung, spleen, and brain (Fig. 1C–1O). High level of hACE2 protein was also detected in testes (Fig. 1C). Immunohistochemistry analysis showed hACE2 was mainly expressed in alveolar epithelial cells of lung (Fig. 1E). In testis, hACE2 was expressed in Leydig cells, spermatogenic cells, and Sertoli cells (Fig. 1G).Fig. 1 Generation of hACE2 transgenic mouse model. A: Schematic diagram of hACE2 transgenic construct. B: PCR genotyping of hACE2 transgenic mouse. C: Western blot analysis of hACE2 protein expression in different tissues. D–O: Immunohistochemistry analysis of hACE2 protein expression in lung (D and E), testis (F and G), heart (H and I), liver (J and K), brain (L and M) and spleen (N and O) of wildtype and hACE2 transgenic mice.
Fig. 1
The testes of hACE2 transgenic mice were infected by rVSV-SARS-CoV-2 pseudovirus
To determine whether the testis of our hACE2 mouse model is susceptible to SARS-CoV-2 infection, we used a replication-competent SARS-CoV-2 pseudotype-rVSV-SARS-CoV-2 with a luciferase reporter (rVSV-luciferase-SARS-CoV-2) or an eGFP reporter (rVSV-eGFP-SARS-CoV-2). rVSV-SARS-CoV-2 encoded the SARS-CoV-2 spike in place of the original glycoprotein in VSV backbone which served as a powerful tool to study cell tropism (Li et al., 2020). The rVSV-luciferase-SARS-CoV-2 was microinjected into the seminiferous tubules of the testes in wildtype and hACE2 transgenic mice. The mice were analyzed for luminescence 24 hrs later using IVIS Spectrum. Strong luminescence signal was detected in the testes of hACE2 mice with pseudovirus injection, but no signal was detected in wildtype testes (Fig. 2 A). We further analyzed the histology of the testes at 3 and 18 days post infection (Figs. 2B–2Q, S1A–S1H). In wildtype mice, no obvious histological change was noted in the testes (Figs. 2C, 2G, 2K, 2O, S1B, S1F). In hACE2 transgenic mice, numerous large vacuoles were observed in the seminiferous tubules at 3 days post infection (Figs. 2E and S1D, arrowheads). On day 18, massive germ cell loss was noted in the testes of hACE2 transgenic mice (Figs. 2M and S1H, asterisks). A large number of CD3-positive inflammatory cells were also observed in the interstitium and seminiferous tubules of pseudovirus injected hACE2 transgenic mice (Fig. 2Q, arrows). These results indicated that the testis of hACE2 mice was susceptible to SARS-CoV-2 through the receptor hACE2.Fig. 2 The testis of hACE2 transgenic mice was susceptible to rVSV-SARS-CoV-2 infection. A: Accumulation of the virus in testis was observed in hACE2 transgenic mice (arrows), but not in wildtype mice at 24 hrs after seminiferous tubules injection of rVSV-luciferase-SARS-CoV-2 as measured by luminescence. B–Q: The expression of MVH and CD3 in the testes of wildtype and hACE2 transgenic mice was examined by immunohistochemistry. MVH was specifically expressed in germ cells (B–E, J–M, arrows), large vacuoles were observed in the seminiferous tubules at D3 (E, arrowheads) and massive germ cells loss was observed at D18 after injection in hACE2 transgenic mice (Q, asterisks). There was no CD3-positive T lymphocyte in wildtype mice with (G and O) or without infection (F and N). A large number of CD3-positive T lymphocytes were detected in testis of hACE2 mice at D18 after injection (Q, arrows).
Fig. 2
To test the cellular tropism of SARS-CoV-2 in the testis, Leydig cells and Sertoli cells were isolated from wildtype and hACE2 mice, cultured in vitro and subjected to rVSV- eGFP-SARS-CoV-2 (Li et al., 2020). Infectivity was quantified by counting GFP positive cells. Both Sertoli and Leydig cells from hACE2 transgenic mice showed susceptibility to rVSV-eGFP-SARS-CoV-2 (Figs. S2–S4). In contrast, no GFP signal was noted in Sertoli and Leydig cells from wildtype mice (Figs. S2–S4). These results indictaed that Sertoli and Leydig cells from hACE2 transgenic mice were the SARS-CoV-2 susceptible cells in the testis.
hACE2 transgenic mice were susceptible to authentic SARS-CoV-2 virus infection
The susceptibility of SARS-CoV-2 in the hACE2 transgenic mice was further valuated by authentic SARS-CoV-2. hACE2 transgenic mice were intranasally infected with authentic SARS-CoV-2. The viral RNA in lung and testis was analyzed by quantitative reverse transcription PCR (qRT-PCR) at 3, 6, and 9 days post infection. As shown in Fig. 3 A, SARS-CoV-2 RNA could be detected in lungs and testes of hACE2 mice at 3, 6 and 9 days post infection. No viral RNA was found in control group. The histology of lungs and testes of hACE2 mice was further analyzed with H&E staining (Fig. 3B–3E). At 6 days post infection, infiltration of immune cells and thickened alveolar walls were found in lungs of hACE2 mice (Fig. 3C, arrows). However, no obvious change could be observed in testes after SARS-CoV-2 infection (Fig. 3E). The structure of seminiferous tubules was intact and there was no infiltration of inflammatory cells. These results indicated that the lung of hACE2 mice model was susceptible for SARS-CoV-2 infection, but the infection via intranasal route did not cause evident pathological changes in testes. Nucleoprotein (N) expression of SARS-CoV-2 in lung and testis was also analyzed (Fig. 3F–3I). The results showed that nucleoprotein could be detected in lung (Fig. 3G, arrows), but not in testes (Fig. 3I) of hACE2 mice after infection.Fig. 3 Effect of SARS-CoV-2 virus infection intranasally on lung and testis of hACE2 mice. A: Real-time RT-PCR analysis of viral RNA in lungs and testes. hACE2 mice were intranasally inoculated with live SARS-CoV-2 or PBS. At 3, 6, and 9 days after inoculation, the viral RNA in lungs and testes was extracted and analyzed using RT-qPCR. Viral burden was expressed on a log10 scale as viral RNA copies per g after comparison with a standard curve. B–E: Hematoxylin and eosin staining of lungs and testes in hACE2 mice at 6 days after inoculation. F–I: Immunohistochemistry analysis of nucleoprotein of SARS-CoV-2 in lungs and testes in hACE2 mice at 6 days after inoculation. Nucleoprotein expression was evident in lungs (G, arrows), but it could not be detected in testes of hACE2 mice (I).
Fig. 3
Thus, we try to infect the testis by direct injection of the authentic SARS-CoV-2 virus into the testis interstitium. The histology of the testes at 3 days and 6 days post injection was analyzed and shown in Figs. 4 and S5. At D3 after virus injection, the nucleoprotein of SARS-CoV-2 was detected in most testicular interstitial cells of hACE2 mice (Fig. 4 D, arrows), but not in wildtype mice (B). However, no infection was observed in the seminiferous tubules and the structure of seminiferous tubules was not affected in hACE2 mice at D3 post injection (Figs. 4H and S5D). At D6, most virus in the intersitium was eliminated. On the contrary, the N protein was detected in the seminiferous tubules of hACE2 mice (Fig. 4L, asterisk). The dissemination of virus from interstitium to the surrounding seminiferous tubules was observed in some areas of testes in hACE2 mice (Fig. 4L, arrows). The infection resulted in infiltration of inflammatory cells in the seminiferous tubules and the interstitium (Fig. S5L, arrows, CD3+). We also found that a large number of seminiferous tubules were disrupted with massive germ cell loss in hACE2 mice at D6 after virus injection (Figs. 4P, S5H, S5L, asterisks). These results indicated that the interstitium of the testis could be infected by SARS-CoV-2 virus and the virus further disseminated from interstitium to the seminiferous tubules which in turn led to the disruption of seminiferous tubules.Fig. 4 The testis of hACE2 transgenic mice was susceptible to live SARS-CoV-2 infection. A–P: Immunohistochemistry analysis of MVH and Nucleoprotein of SARS-CoV-2 at D3 (A–H) and D6 (I–P) in testes after SARS-CoV-2 was injected into the testes interstitium of wildtype and hACE2 mice. Nucleoprotein of the virus was detected in the interstitial cells of testes in hACE2 mice (D, arrows), not in wildtype mice (B) at D3 after injection. At D6, most Nucleoprotein signal in the interstitium disappeared, instead, Nucleoprotein was observed in some seminiferous tubules (L, asterisk) in hACE2 mice. The invasion of the SARS-CoV-2 from interstitium to the surrounding seminiferous tubules was also observed (L, arrows). The infection caused germ cell loss from the seminiferous tubules (P, asterisks).
Fig. 4
Sertoli cells and Leydig cells were further isolated from wildtype and hACE2 mice and incubated with SARS-CoV-2 virus. After incubation for 24 hrs, Leydig cells and Sertoli cells were fixed and processed for immunofluorescence analysis. SARS-CoV-2 Nucleoprotein/3β-HSD double staining for Leydig cells and SARS-CoV-2 Nucleoprotein/SOX9 for Sertoli cells was shown in Fig. 5 A–5F. Nucleoprotein/3β-HSD and Nucleoprotein/SOX9 double positive cells were observed in Leydig cells (Fig. 5C, arrows) and Sertoli cells (Fig. 5F, arrows) from hACE2 mice respectively. In contrast, no Nucleoprotein signal (green) was detected in 3β-HSD-positive Leydig cells (Fig. 5A) and SOX9-positive Sertoli cells (Fig. 5D,) from wildtype mice.Fig. 5 Leydig cells and Sertoli cells of hACE2 transgenic mice were susceptible to live SARS-CoV-2 infection. Primary Leydig and Sertoli cells were isolated from the testes of wildtype and hACE2 transgenic mice and infected with live SARS-CoV-2. Nucleoprotein and 3β-HSD (A–C), Nucleoprotein and SOX9 (D–F) double staining experiments were performed. In wildtype cells, 3β-HSD was expressed in cytoplasm of Leydig cells (A, red) and SOX9 was expressed in nuleus of Sertoli cells (D, red). No Nucleoprotein of SARS-CoV-2 was detected in wildtype cells after infection (A and D) and cells from hACE2 transgenic mice without infection (B and E). Nucleoprotein/3β-HSD, Nucleoprotein/SOX9 double positive cells were evident in Leydig cells (C, arrows) and Sertoli cells (F, arrows) from hACE2 mice after incubation with SARS-CoV-2. Nuclei were stained blue with DAPI.
Fig. 5
Discussion
Several mouse models expressing hACE2 have been generated to study the pathogenesis and therapeutics of SARS-CoV-2, including transgenic, knock-in, and adenovirus transduced mouse models (Munoz-Fontela et al., 2020; Rathnasinghe et al., 2020). There are several types of transgenic mouse models, in which the expression of hACE2 is under the control of different promoters, such as cytokeratin epithelial cell promoter (K18), synthetic CAG composite promoter, lung ciliated epithelial cell promoter (HFH4), and human ACE2 promoter (Munoz-Fontela et al., 2020; Rathnasinghe et al., 2020). Whether hACE2 is expressed in the testis of these mice models is not clear. In human, ACE2 is reported to be expressed in spermatogonia, Leydig and Sertoli cells of testis, suggesting the testis is a potential target of SARS-CoV-2 infection (Wang and Xu. 2020). To mimic the expression pattern of ACE2 in human, we generated a mouse model in which the expression of human ACE2 cDNA was under the control of the promoter of human ACE2 gene. In this model, ACE2 protein was expressed in lung, intestine, brain which was consistent with other mouse models. Most importantly, ACE2 protein was also detected in testes.
In this study, we found that in vitro cultured Sertoli cells and Leydig cells could be infected by both pseudotyped and authentic SARS-CoV-2 virus, suggesting that these cells could be the targets of SARS-CoV-2 infection in vivo. The expression of N proteins was first detected in Leydig cells after interstitial injection of authentic SARS-CoV-2 virus, and the viruses further spread into seminiferous tubules, suggesting that SARS-CoV-2 virus can break through the blood-testis barrier (BTB) and enter the seminiferous tubules. BTB is one of the tightest blood-tissue barriers formed by cell junctions between adjacent Sertoli cells. BTB provides a physical and immunoprivileged microenvironment for germ cell development which sequesters germ cells in the adluminal compartment from harmful toxicants and against the body’s immune system. It has been reported that several viruses can penetrate the barriers and induce testicular dysfunctions, such as Mumps, HIV, Zika, Ebola (Seymen. 2021).
A recent study using hamster found intranasal SARS-CoV-2 challenge caused testicular damage and N protein of SARS-CoV-2 could be detected in testes, although viral loads and N protein expression were markedly lower in testis than lung (Li et al., 2022). Another study detected SARS-CoV-2 RNA in testes but histophthological changes was not found after infecting hamsters intranasally (Campos et al., 2021). In our study, although virus RNA was detected in testes of hACE2 mice after intranasal exposure of SARS-CoV-2 virus, no pathological change was observed and N protein of SARS-CoV-2 virus was not detected in testes, suggesting the amount of virus in testes after intranasal exposure is very few which could not cause pathological changes in testis. We noticed that although infiltration of immune cells was observed in lung of hACE2 mice after intranasal infection, the pathological change in this mouse model was not as severe as previously reported (Bao et al., 2020; Jiang et al., 2020). We speculated that the infection of testes was secondary to the infection of lung. The infection of testes probably needs high viral load in blood. Higher viral load is reported to be associated with more severe disease symptoms (Moshrefi et al., 2021). The expression level of hACE2 is probably another factor that determines whether the testes could be infected and the degree of damage with intranasal exposure of SARS-CoV-2.
Although most studies report the absence of SARS-CoV-2 in semen of COVID-19 patients, it is evident that there is testis injury as scrotal discomfort, altered semen parameters and hormone production in many patients (Gonzalez et al., 2020; Yang et al., 2020; Omolaoye et al., 2021). Pathological findings such as injury in seminiferous tubules and Leydig cells, and lymphocyte inflammation is also observed in autopsy samples (Yang et al., 2020). How SARS-CoV-2 affects testes? The indirect effect after infection such as fever and drugs (corticosteroids) may cause testicular damage (Sheikhzadeh Hesari et al., 2021). The direct effect mediated through the binding of spike protein of SARS-CoV-2 with hACE2 receptor should not be neglected. At present, the number of semen samples collected in clinical studies is still limited and most samples were collected at the recovery stage. More studies need to be performed at the acute stage of COVID-19 to confirm the presence of SARS-CoV-2 virus in semen. Our study demonstrated that SARS-CoV-2 virus could infect Sertoli and Leydig cells through its receptor hACE2. The replication of virus in these cells caused testis damage. However, no pathological changes were detected in testis after intranasal exposure of SARS-CoV-2. Based on these results, we could not exclude the possibility of testis damage after COVID-19 infection in human. It is possible that high viral load in blood and high expression level of hACE2 may cause testis infection and damage and this needs further investigation.
In summary, we constructed a humanized ACE2 transgenic mice model in which ACE2 expression pattern resembled to that in human. Injection of SARS-CoV-2 virus into the testis interstitium could infect the testicular cells and cause disruption of seminiferous tubules. Our study demonstrates that testis is most likely also a target of SARS-CoV-2 virus. Attention should be paid to the testis function in SARS-CoV-2 patients. Our mouse model is helpful for understanding the pathogenesis of SARS-CoV-2 infection, studying the influence of SARS-CoV-2 virus on reproductive system, and evaluating vaccines and therapeutics. More studies are needed to investigate whether the virus can be transmitted to the testis in mice which developed severe interstitial pneumonia.
Material and methods
Ethics Statement
Recombinant vesicular stomatitis virus (rVSV) pseudovirus used in this srtudy were conducted in a biosafety level 2 (BSL2) facility. All experiments with authentic viruses and animals were performed in a biosafety level 3 (BSL3) laboratory. The study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Institute of Microbiology, Chinese Academy of Sciences (IMCAS) Ethics Committee, and all experiments conformed to the relevant regulatory standards. The experiments and protocols were approved by the Committee on the Ethics of Animal Experiments of IMCAS (Permit Number: APIMCAS2022073). All animal experiments were conducted under isoflurane anesthesia to minimize animal suffering.
Mice
Animal experiments were carried out in accordance with the protocols approved by the Institutional Animal Care and Use Committee (IACUC) of the Institute of Zoology, Chinese Academy of Sciences (CAS) (SYXK 2018-0021). To construct hACE2 transgenic mice, human ACE2 promoter and coding sequence were amplified by PCR and inserted into pUC19L vector. The primers were: ACE2 promoter F: CTCGGTACTGCTGTCCCAGGCTC, ACE2 promoter R: ATGGTGGCCGTCCCCTGTGAGCC; ACE2 CDs F: AGGGGACGGCCACCATGTCAAGC, ACE2 CDs R: CCGCTGCCCTTGTCATCGTCATCCT. The plasmid was microinjected into fertilized ova from C57BL/6 mice, which were then transplanted into pseudo-pregnant mice. The offspring were genotyped using the following primers: F: AGTCACGACGTTGTAAAACGACGGCCAGTG, R: ATTTCACACAGGAAACAGCTATGACCATGATTACGCCAAG. The positive offspring were backcrossed with C57BL/6 mice to produce generation F1.
Pseudoviruses construction and infection
The replication-competent vesicular stomatitis virus (VSV) based recombinant SARS-CoV-2 pseudovirus (rVSV-SARS-CoV-2) was described previously (Li et al., 2020). Briefly, glycoprotein coding sequence of VSV was replaced with the codon-optimized spike (S) of SARS-CoV-2 Wuhan-Hu-1 strain (GeneBank: YP_009724390.1) and an eGFP or luciferase reporter was inserted in front of N. For mice injection, 8-week-old male wildtype and hACE2 transgenic mice were used. After anesthesia, 10 μL of rVSV pseudovirus solution (mixture of rVSV pseudovirus with trypan blue, 5 × 106 focus-forming units per ml) was injected into the seminiferous tubules through rete testis using a glass capillary under a stereomicroscope. For no-infection group, 10 μL PBS was injected. Twenty-four hrs later, the mice were analyzed for luminescence using IVIS Spectrum (PerkinElmer IVIS Spectrum, 5717). At D3 and D18 after injection, the testes were collected for further histological analysis.
For cell infection, the viral titers of rVSV-eGFP-SARS-CoV-2 on Sertoli and Leydig cells were determined by a focus-forming assay (FFA). Cells (2 × 104/well) were seeded in 96-well plates at 24 hrs before infection. Viruses were serially diluted at 1:10 dilution in DMEM with 2% FBS and incubated with cells for 4 hours at 37°C. Then, cells were washed once and incubated at 28°C in fresh medium (with 20 mM NH4Cl). GFP-positive cells were counted under a fluorescent microscope at 20 hrs post infection.
Authentic SARS-CoV-2 infection
SARS-CoV-2 strain hCoV-19/China/CAS-B001/2020 (National Microbiology Data Center NMDCN0000102-3, GISAID databases EPI_ISL_514256-7) was kindly provided by Prof. Yuhai Bi. The viruses were amplified and titered on Vero-E6 cells. For intranasal infection, 8-week-old male wildtype mice and hACE2 transgenic mice were anesthetized with isoflurane (5% initial and 1.0%–1.5% for maintenance) though intranasal route, and then intranasally infected with 5×105 TCID50/50 μL of SARS-CoV-2 respectively. For testis injection, a dose of 105 TCID50/mL in 10 μL SARS-CoV-2 was injected into the testis interstitium of 8-week-old male wildtype mice and hACE2 transgenic mice. For no-infection group, 10 μL PBS was injected into the testis interstitium. At indicated days after infection, the testis and lung were harvested for virus RNA and histological analysis.
For cell infection, Sertoli cells and Leydig cells were infected with SARS-CoV-2 at a MOI of 1 in 12-well plates. Twenty-four hours later, the cells were fixed with 4% paraformaldehyde for 48 hrs and processed for immunofluorescence analysis.
Isolation of viral RNA and real-time RT-PCR
For viral RNA detection, all tissues were weighed and homogenized with zirconia beads in the Bead Ruptor Elite instrument (Omni International, Kennesaw, GA, USA) in 1 ml PBS. All homogenized tissues from infected mice were stored at –80°C until virus titration. Viral RNAs were extracted with RNeasy Mini Kit (QIAGEN) and determined by real-time RT-PCR targeting the ORF1ab gene of SARS-CoV-2 on an ABI QuantStudio 7 instrument (Applied Biosystems, CA, USA) using One Step PrimeScript RT-PCR kit (TaKaRa, Japan). Viral burden was expressed on a log10 scale as viral RNA copies per g after comparison with a standard curve.
Isolation and culture of Sertoli cell
Sertoli cells were isolated from 4-week-old mice as previously described (Chen et al., 2014). Briefly, the testes were decapsulated and sequentially digested in 3 steps of enzymes: 1 mg/mL collagenase IV (VETEC, V900893) and 1 mg/mL DNase I (AppliChem, A37780500) in DMEM for 15 min, then 1 mg/ml collagenase I, 1 mg/mL DNase I and 1 mg/ml hyaluronidase (SIGMA, SIAL-H3506) for 10 min, and then 1 mg/ml collagenase IV, 1 mg/mL hyaluronidase, 2.5 mg/ml trypsin and 1 mg/mL DNase I for 20 min. FBS was added to terminate the digestion and cell suspension was filtered through a 40 μm filter. Cells were centrifuged, washed, and resuspended in DMEM/F12 supplemented with 10% FBS. Forty-eight hrs later, the cells were subjected to hypotonic treatment with 20 mM Tris (pH 7.4) at room temperature for 2.5 min to remove the residual spermatogonia and the unlysed cells were ready for virus infection.
Isolation and culture of Leydig cell
Percoll-purified Leydig cells from 4-week-old mice were prepared as previously described (Chen et al., 2017). Briefly, the testes were decapsulated and incubated in PBS containing 1 mg/mL collagenase IV in a water bath with circular agitation (100 rpm) at 37°C for 15 min. Tubules were allowed to settle, and the cell suspension was filtered through a 40 μm filter. After washing in DMEM/F12, the cell suspension was separated in a discontinuous Percoll (GE) gradient of 30%, 40%, 50% and 60% at 800 × g for 30 min. The gradient fraction containing Leydig cells between 50% and 60% layers (1.067–1.077 g/ml) was collected and washed in DMEM/F12 and then resuspended in DMEM/F12 supplemented with 10% FBS. Forty-eight hrs later, when cells were approximately 70% confluent, Leydig cells were ready for virus infection.
Immunohistochemistry and immunofluorescence analysis
Immunohistochemistry and immunofluorescence analysis were performed as described previously (Chen et al., 2021). After rehydration and antigen retrieval, the 5-μm sections were blocked with 5% BSA, incubated with the primary antibody for 1 hr and the corresponding secondary antibody for 1 hr. The following primary antibodies were used: MVH (Abcam, ab13840), SOX9 (Millipore, AB5535; Sigma, AMAB90795), Nucleoprotein (Sino biological, 40143-R001), CD3 (Abcam, ab11089), 3β-HSD (Santa Cruz, sc-30820). The secondary antibodies were: FITC-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch, 711-095-152), TRITC-conjugated donkey anti-mouse IgG (Jackson ImmunoResearch, 715-025-151), TRITC-conjugated donkey anti-goat IgG (Jackson ImmunoResearch, 705-025-147). For immunohistochemistry, staining was visualized using a diaminobenzidine substrate kit, examined with a Nikon microscope, and images were captured by a Nikon DS-Ri1 CCD camera. For immunofluorescence, the sections were examined with a confocal laser scanning microscope (Carl Zeiss Inc, Thornwood, NY)
Western blot analysis
Tissues were lysed with RIPA buffer (50 mM Tris–HCl [pH 7.5], 150 mM NaCl, 1% NP-40, 0.1% SDS, 1% sodium deoxycholate, 5 mM EDTA) supplemented with protease inhibitors cocktail (Roche) and 1 mM PMSF. Equal amounts of total protein were separated by SDS/PAGE gels, transferred to nitrocellulose membrane, and probed with the primary antibodies. The images were captured with the ODYSSEY Sa Infrared Imaging System (LI-COR Biosciences, Lincoln, NE). The antibodies used were ACE2 (Abcam, ab108209), GAPDH (Boao, ab1039t).
Statistical analysis
For mice infection experiments, five mice for each group at each time point were used for immunostaining or quantitative experiments. For immunostaining, one representative picture of similar results from five mice at each time point is presented. All Sertoli and Leydig cell culture experiments were repeated at least three times by using three different cell preparations. The quantitative results are presented as the mean ± SEM. Statistical analyses were conducted using GraphPad Prism version 9.0.0. Data were analyzed using one-way ANOVA. P-values < 0.05 were considered to indicate significance.
CRediT authorship contribution statement
Min Chen, Shihua Li: Conceptualization, Methodology, Data curation, Investigation, Validation, Writing - Original draft, Review & Editing. Shujun Liu, Yuhang Zhang: Methodology, Data curation, Investigation, Validation. Xiuhong Cui, Limin Lv, Bowen Liu: Methodology, Data curation. Aihua Zheng, Qihui Wang, Shuguang Duo: Conceptualization, Methodology, Data curation, Investigation, Validation, Writing - Review & Editing. Fei Gao: Conceptualization, Methodology, Data curation, Investigation, Validation, Writing - Original draft, Review & Editing.
Uncited reference
Li et al., 2022a, Li et al., 2020b, Li et al., 2020c, Li et al., 2020d.
Declaration of Competing Interest
The authors declare that they have no conflict of interest.
Appendix A Supplementary data
The following is the Supplementary data to this article:
Acknowledgments
We appreciate the staff of the biosafety level 3 (BSL3) laboratory in the institute of Microbiology, Chinese Academy of Sciences. This work was supported by National key R&D program of China (2018YFA0107700), the National Natural Science Foundation of China (32170855, 31970785), and Biological Resources Program of Chinese Academy of Sciences (KFJ-BRP-005).
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.jgg.2022.11.011.
==== Refs
References
Bao L. Deng W. Huang B. Gao H. Liu J. Ren L. Wei Q. Yu P. Xu Y. Qi F. The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice Nature 583 2020 830 833 32380511
Campos R.K. Camargos V.N. Azar S.R. Haines C.A. Eyzaguirre E.J. Rossi S.L. SARS-CoV-2 Infects Hamster Testes Microorganisms 9 2021 1318 34204370
Chen M. Dong F. Chen M. Shen Z. Wu H. Cen C. Cui X. Bao S. Gao F. PRMT5 regulates ovarian follicle development by facilitating Wt1 translation Elife 10 2021 e68930
Chen M. Wang X. Wang Y. Zhang L. Xu B. Lv L. Cui X. Li W. Gao F. Wt1 is involved in leydig cell steroid hormone biosynthesis by regulating paracrine factor expression in mice Biol. Reprod 90 2014 71 24571983
Chen M. Zhang L. Cui X. Lin X. Li Y. Wang Y. Wang Y. Qin Y. Chen D. Han C. Wt1 directs the lineage specification of sertoli and granulosa cells by repressing Sf1 expression Development 144 2017 44 53 27888191
Clarke N.E. Turner A.J. Angiotensin-converting enzyme 2: the first decade Int. J. Hypertens 2012 2012 307315 22121476
Douglas G.C. O'Bryan M.K. Hedger M.P. Lee D.K. Yarski M.A. Smith A.I. Lew R.A. The novel angiotensin-converting enzyme (ACE) homolog, ACE2, is selectively expressed by adult Leydig cells of the testis Endocrinology 145 2004 4703 4711 15231706
Gonzalez D.C. Khodamoradi K. Pai R. Guarch K. Connelly Z.M. Ibrahim E. Arora H. Ramasamy R. A Systematic Review on the Investigation of SARS-CoV-2 in Semen Res. Rep. Urol. 12 2020 615 621 33294423
Guo L. Zhao S. Li W. Wang Y. Li L. Jiang S. Ren W. Yuan Q. Zhang F. Kong F. Absence of SARS-CoV-2 in semen of a COVID-19 patient cohort Andrology 9 2021 42 47 32598557
Hoffmann M. Kleine-Weber H. Schroeder S. Kruger N. Herrler T. Erichsen S. Schiergens T.S. Herrler G. Wu N.H. Nitsche A. SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor Cell 181 2020 271 280 32142651
Holtmann N. Edimiris P. Andree M. Doehmen C. Baston-Buest D. Adams O. Kruessel J.S. Bielfeld A.P. Assessment of SARS-CoV-2 in human semen-a cohort study Fertil Steril 114 2020 233 238 32650948
Jiang R.D. Liu M.Q. Chen Y. Shan C. Zhou Y.W. Shen X.R. Li Q. Zhang L. Zhu Y. Si H.R. Pathogenesis of SARS-CoV-2 in Transgenic Mice Expressing Human Angiotensin-Converting Enzyme 2 Cell 182 2020 50 58 32516571
Kayaaslan B. Korukluoglu G. Hasanoglu I. Kalem A.K. Eser F. Akinci E. Guner R. Investigation of SARS-CoV-2 in Semen of Patients in the Acute Stage of COVID-19 Infection Urol. Int. 104 2020 678 683 32781456
Li C. Ye Z. Zhang A.J. Chan J.F. Song W. Liu F. Chen Y. Kwan M.Y. Lee A.C. Zhao Y. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections by intranasal or testicular inoculation induces testicular damage preventable by vaccination in golden Syrian hamsters Clin. Infect. Dis. 75 2022 e974 e990 35178548
Li D. Jin M. Bao P. Zhao W. Zhang S. Clinical characteristics and results of semen tests among men with coronavirus disease 2019 JAMA Netw. Open 3 2020 e208292
Li H. Zhao C. Zhang Y. Yuan F. Zhang Q. Shi X. Zhang L. Qin C. Zheng A. Establishment of replication-competent vesicular stomatitis virus-based recombinant viruses suitable for SARS-CoV-2 entry and neutralization assays Emerg. Microbes Infect. 9 2020 2269 2277 32990161
Li Q. Guan X. Wu P. Wang X. Zhou L. Tong Y. Ren R. Leung K.S.M. Lau E.H.Y. Wong J.Y. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia N. Engl. J. Med. 382 2020 1199 1207 31995857
Lu R. Zhao X. Li J. Niu P. Yang B. Wu H. Wang W. Song H. Huang B. Zhu N. Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding Lancet 395 2020 565 574 32007145
Ma L. Xie W. Li D. Shi L. Ye G. Mao Y. Xiong Y. Sun H. Zheng F. Chen Z. Evaluation of sex-related hormones and semen characteristics in reproductive-aged male COVID-19 patients J. Med. Virol. 93 2021 456 462 32621617
Moshrefi M. Ghasemi-Esmailabad S. Ali J. Findikli N. Mangoli E. Khalili M.A. The probable destructive mechanisms behind COVID-19 on male reproduction system and fertility J. Assist. Reprod. Genet. 38 2021 1691 1708 33977466
Munoz-Fontela C. Dowling W.E. Funnell S.G.P. Gsell P.S. Riveros-Balta A.X. Albrecht R.A. Andersen H. Baric R.S. Carroll M.W. Cavaleri M. Animal models for COVID-19 Nature 586 2020 509 515 32967005
Omolaoye T.S. Adeniji A.A. Cardona Maya W.D. du Plessis S.S. SARS-COV-2 (Covid-19) and male fertility: Where are we? Reprod Toxicol 99 2021 65 70 33249233
Pan F. Xiao X. Guo J. Song Y. Li H. Patel D.P. Spivak A.M. Alukal J.P. Zhang X. Xiong C. No evidence of severe acute respiratory syndrome-coronavirus 2 in semen of males recovering from coronavirus disease 2019 Fertil Steril 113 2020 1135 1139 32482249
Paoli D. Pallotti F. Colangelo S. Basilico F. Mazzuti L. Turriziani O. Antonelli G. Lenzi A. Lombardo F. Study of SARS-CoV-2 in semen and urine samples of a volunteer with positive naso-pharyngeal swab J. Endocrinol. Invest. 43 2020 1819 1822 32329026
Peng L. Liu J. Xu W. Luo Q. Chen D. Lei Z. Huang Z. Li X. Deng K. Lin B. SARS-CoV-2 can be detected in urine, blood, anal swabs, and oropharyngeal swabs specimens J Med. Virol. 92 2020 1676 1680 32330305
Rathnasinghe R. Strohmeier S. Amanat F. Gillespie V.L. Krammer F. Garcia-Sastre A. Coughlan L. Schotsaert M. Uccellini M.B. Comparison of transgenic and adenovirus hACE2 mouse models for SARS-CoV-2 infection Emerg. Microbes Infect 9 2020 2433 2445 33073694
Rawlings S.A. Ignacio C. Porrachia M. Du P. Smith D.M. Chaillon A. No Evidence of SARS-CoV-2 seminal shedding despite SARS-CoV-2 persistence in the upper respiratory tract Open Forum. Infect. Dis. 7 2020 ofaa325
Seymen C.M. The other side of COVID-19 pandemic: Effects on male fertility J. Med. Virol. 93 2021 1396 1402 33200417
Sheikhzadeh Hesari F. Hosseinzadeh S.S. Asl Monadi Sardroud M.A. Review of COVID-19 and male genital tract Andrologia 53 2021 e13914
Song C. Wang Y. Li W. Hu B. Chen G. Xia P. Wang W. Li C. Diao F. Hu Z. Absence of 2019 novel coronavirus in semen and testes of COVID-19 patientsdagger Biol. Reprod 103 2020 4 6 32297920
Sun S.H. Chen Q. Gu H.J. Yang G. Wang Y.X. Huang X.Y. Liu S.S. Zhang N.N. Li X.F. Xiong R. A mouse model of SARS-CoV-2 infection and pathogenesis Cell Host Microbe 28 2020 124 133 32485164
Wang W. Xu Y. Gao R. Lu R. Han K. Wu G. Tan W. Detection of SARS-CoV-2 in different types of clinical specimens JAMA 323 2020 1843 1844 32159775
Wang Z. Xu X. scRNA-seq profiling of human testes reveals the presence of the ACE2 receptor, a target for SARS-CoV-2 infection in spermatogonia, leydig and sertoli cells Cells 9 2020 920 32283711
Yang M. Chen S. Huang B. Zhong J.M. Su H. Chen Y.J. Cao Q. Ma L. He J. Li X.F. Pathological findings in the testes of COVID-19 patients: clinical implications Eur. Urol. Focus 6 2020 1124 1129 32563676
Younis J.S. Abassi Z. Skorecki K. Is there an impact of the COVID-19 pandemic on male fertility? The ACE2 connection Am. J. Physiol. Endocrinol. Metab. 318 2020 E878 E880 32421367
Zhou P. Yang X.L. Wang X.G. Hu B. Zhang L. Zhang W. Si H.R. Zhu Y. Li B. Huang C.L. A pneumonia outbreak associated with a new coronavirus of probable bat origin Nature 579 2020 270 273 32015507
Zhu N. Zhang D. Wang W. Li X. Yang B. Song J. Zhao X. Huang B. Shi W. Lu R. A novel coronavirus from patients with pneumonia in China, 2019 N. Engl. J. Med. 382 2020 727 733 31978945
| 36494057 | PMC9724560 | NO-CC CODE | 2022-12-07 23:20:14 | no | J Genet Genomics. 2022 Dec 6; doi: 10.1016/j.jgg.2022.11.011 | utf-8 | J Genet Genomics | 2,022 | 10.1016/j.jgg.2022.11.011 | oa_other |
==== Front
New Microbes New Infect
New Microbes New Infect
New Microbes and New Infections
2052-2975
The Authors. Published by Elsevier Ltd.
S2052-2975(22)00109-3
10.1016/j.nmni.2022.101057
101057
Letter to the Editor
Monkeypox exposure in patients living with HIV (PLWH): Epidemiological and clinical significance and a summary of response recommendations
Semnani Farbod ∗1
Rayati Damavandi Amirmasoud 1
Aarabi Seyed Sahab 1
Students' Scientific Research Center, Tehran University of Medical Sciences, Tehran, Iran
School of Public Health, Tehran University of Medical Sciences, Tehran, Iran
∗ Corresponding author. Students' Scientific Research Center, Tehran University of Medical Sciences, Tehran, 19836-33415, Iran.
1 These authors had equal contributions.
6 12 2022
6 12 2022
10105730 10 2022
18 11 2022
21 11 2022
© 2022 The Authors. Published by Elsevier Ltd.
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
==== Body
pmcThe recent monkeypox outbreak in non-endemic areas, a potentially global epidemic raised by the World Health Organization (WHO) on May 21, 2022, has inflicted nearly 80 thousand patients as of November 10, 2022.
Although it has not been considered a sexually transmitted disease and almost anyone can contract the disease through mostly close contact, homosexuals, bisexuals, and MSM comprise most of the monkeypox cases. On the other hand, the Human immunodeficiency virus (HIV) is a common co-infection among this group. Hence, there is a need for public health policymakers to consider these risk factors and engage in risk communication [1].
In previous outbreaks, children, pregnant women, and immunocompromised individuals such as PLWH with poorly controlled diseases are at higher risks of disease progression and more severe outcomes. Nevertheless, more recent studies on emerging monkeypox cases among HIV-positive cases under effective ART treatment in Europe do not report a higher rate of hospitalization and mortality compared to the cases without HIV infection, suggesting the role of HIV-induced immunosuppression on patients' prognosis.
The epidemiologic shift in the recent outbreak in non-endemic countries might imply a paradigm shift in our perception of typical and atypical monkeypox presentations. The emergence of herald skin lesions in anogenital areas without the prodromal phase may strongly suggest sexual transmission via local inoculation of skin-to-skin contact and may not be associated with HIV infection. Hence, the monkeypox case definition and transmission route may need reevaluation as the outbreak continues.
Summary of clinical recommendations
Herein, we propose a short summary of the recommendations released for monkeypox management, i.e. CDC, The British HIV Association (BHIVA), and the World Health Organization (WHO), with a focus on PLWH with suspected monkeypox exposure (Fig. 1 ). If a new diagnosis of HIV infection is established, ART should be started as soon as possible and best within the first week [2]. Should the diagnosis of monkeypox and HIV be made simultaneously, these patients must be prioritized for the most urgent treatments [2]. The clinicians should suspect infection with monkeypox in a person living with HIV with a characteristic rash, even without apparent epidemiological reference. Diagnosis is confirmed by nucleic acid amplification testing (conventional or real-time PCR) taken from any suspicious skin or mucosal lesion. After the diagnosis is confirmed, 21-day isolation is suggested for all patients [2]. However, as immunocompromised PLWH with poorly controlled illness (CD4 < 200 cell/mL or HIV RNA >200 copies/mL) or individuals with AIDS diagnosis in the past six months may have prolonged virus shedding from the upper respiratory tract, a further clinical assessment is required regarding the termination of the isolation [2,3]. Nonetheless, due to data scarcity, strict thresholds for CD4 count or HIV RNA copies should not be considered alone and the overall risk for immunosuppression should be clinically evaluated [3,4]. Moreover, after the monkeypox diagnosis is established, clinicians should consider the hospitalization of these groups for closer monitoring, even if they remain asymptomatic [2].Fig. 1 A summary of response recommendations for people living with HIV with monkeypox exposure, MSM = men having sex with men, MVA-BN= Modified Vaccinia Ankara - Bavarian Nordic.
Fig. 1
Post-exposure vaccination with ACAM2000, a live replication-competent FDA-approved vaccine for monkeypox, is recommended against for PLWH due to adverse events like progressive vaccinia. MVA-BN (JYNNEOS), a third-generation replication-deficient smallpox vaccine recently authorized in the USA, Canada, and the EU, should be prioritized for these patients. Regarding treatment, tecovirimat, vaccinia immune globulin (VIG), cidofovir, and brincidofovir can be considered for HIV + patients who are severely ill [5]. Tecovirimat, an inhibitor of the viral envelope protein VP37, may be considered the first-line option for treating symptomatic immunosuppressed HIV + monkeypox patients regardless of disease severity [5]. Tecovirimat is approved by the European Medicines Agency for monkeypox and in the USA, Canada, and Europe for human smallpox disease. Nonetheless, it is still not FDA-approved for monkeypox treatment and may be administered through non-research expanded access Investigational New Drug (EA-IND) protocol. Although proven safe, its efficacy has not been evaluated in human trials. Tecovirimat and brincidofovir have potential drug-drug interactions with antiretrovirals and should be administered cautiously. However, none of these considerations should hinder the co-administration of ART and monkeypox-specific treatments [4].
The evidence is still patchy regarding monkeypox in possibly high-risk groups, including PLWH, without receiving appropriate ART or immunocompromised. Therefore, the figure provided may help clinicians and policymakers to benefit from a more coherent illustration of the current recommendations proposed by different sources.
Ethics approval and consent to participate
Not applicable.
Consent for publication
A signed license to publish is provided.
Availability of data and materials
Not applicable.
Funding
Not applicable.
All authors participated in preparing the final draft of the manuscript, revised the manuscript, and critically assessed the academic content. All authors have read and approved the manuscript and confirmed the accuracy or integrity of any part of the work.
CRediT authorship contribution statement
Farbod Semnani: Conceptualization, Data curation, Investigation, Methodology, Supervision, Validation, Writing – original draft, Writing – review & editing, All authors participated in preparing the final draft of the manuscript, revised the manuscript, and critically assessed the academic content. All authors have read and approved the manuscript and confirmed the accuracy or integrity of any part of the work. Amirmasoud Rayati Damavandi: Conceptualization, Data curation, Investigation, Methodology, Validation, Writing – original draft, Visualization, All authors participated in preparing the final draft of the manuscript, revised the manuscript, and critically assessed the academic content. All authors have read and approved the manuscript and confirmed the accuracy or integrity of any part of the work. Seyed Sahab Aarabi: Conceptualization, Data curation, Investigation, Resources, Methodology, Validation, Writing – original draft, All authors participated in preparing the final draft of the manuscript, revised the manuscript, and critically assessed the academic content. All authors have read and approved the manuscript and confirmed the accuracy or integrity of any part of the work.
Declaration of competing interest
The authors do not declare a conflict of interests.
Acknowledgments
Not applicable.
==== Refs
References
1 Sah R. Reda A. Abdelaal A. Mohanty A. Siddiq A. Alshahrani N.Z. A potential monkeypox pandemic: are we making the Same mistakes as COVID-19? New Microbe. New Infect 49 50 2022 Nov 101030
2 Clinical management and infection prevention and control for monkeypox: interim rapid response guidance 10 June 2022 [Internet]. [cited 2022 Oct 2]. Available from: https://www.who.int/publications/i/item/WHO-MPX-Clinical-and-IPC-2022.1
3 BHIVA rapid guidance on monkeypox virus [Internet]. [cited 2022 Oct 2]. Available from: https://www.bhiva.org/BHIVA-rapid-guidance-on-monkeypox-virus
4 O’Shea J. Filardo T.D. Morris S.B. Weiser J. Petersen B. Brooks J.T. Interim guidance for prevention and treatment of monkeypox in persons with HIV infection — United States MMWR Morb Mortal Wkly Rep [Internet]. 2022 Aug 12 [cited 2022 Oct 2];71(32):1023–8. Available from: http://www.cdc.gov/mmwr/volumes/71/wr/mm7132e4.htm?s_cid=mm7132e4_w August 2022
5 Guidance for tecovirimat use under expanded access investigational new drug protocol during 2022 U.S. Monkeypox outbreak | monkeypox | poxvirus | CDC [Internet]. [cited 2022 Oct 2]. Available from: https://www.cdc.gov/poxvirus/monkeypox/clinicians/Tecovirimat.html
| 36504594 | PMC9724564 | NO-CC CODE | 2022-12-11 23:16:20 | no | New Microbes New Infect. 2022 Dec 6 November-December; 49:101057 | utf-8 | New Microbes New Infect | 2,022 | 10.1016/j.nmni.2022.101057 | oa_other |
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00341-0
10.1016/j.jiph.2022.12.002
Editorial
The current monkeypox outbreak: Knowledge gaps and research priorities
Asaad Ahmed
El-Sokkary Rehab ⁎
Department of Medical Microbiology, Faculty of Medicine, Zagazig University, Zagazig, Egypt
⁎ Corresponding author.
6 12 2022
6 12 2022
13 10 2022
29 11 2022
4 12 2022
© 2022 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
Keywords
equity
infection control
research question
study design
==== Body
pmcSince May 2022, the world has witnessed the resurgence of the largest outbreak of monkeypox (MPX) after its initial diagnosis among a group of children from the Democratic Republic of the Congo (DRC) in 1970 [1]. This continuous surge could be explained by several factors. First, the cessation of smallpox vaccination after eradication in 1980 has been associated with an increased population lacking orthopoxvirus cross-immunological defense. Second, advances in civilization and rapid global travel facilitate previous sporadic cases and localized clusters to quickly shift as global epidemics. Finally, the possibility of mutations in the viral genome and their implication in MPX transmission and pathogenesis would not be ignored [2]. To secure everyone's health, the necessity for health equity and global connectedness has been highlighted. The current situation imposes taking a broader perspective on monkeypox in relation to emerging infectious diseases and pandemic preparedness. Highlighting current research priorities support the worldwide interdisciplinary collaborations to control MPX.
Infection control perspective. Lessons from previous global epidemics of Severe Acute Respiratory Syndrome (SARS), Middle East Respiratory Syndrome Coronavirus (MERS-CoV), and recently Coronavirus disease 19 (COVID 19) showed that a comprehensive preventive plan should be urgently considered by clinicians, laboratorians, epidemiologists, and health decision-makers in all countries. Besides, the complex epidemiological profile of the current MPX outbreak in non-endemic countries [1] warrants effective preventive strategies to include both healthcare and community settings. In healthcare facilities, accurate and rapid diagnosis of active cases seems paramount. This task needs well-trained staff supported with all clinical and laboratory diagnostic tools to avoid missing confirmed, possible, or probable cases. In parallel, the protection of healthcare workers should be emphasized through strict adherence to infection control procedures including standard, contact, and droplet precautions for both inpatient wards and outpatient clinics [3], [4]. It is worth noting that hand hygiene and personal protective equipment (PPE) remain the cornerstone for control of MPX transmission and preventing cross-infection. In community settings, the role is to stop MPX transmission by breaking the infection chain. Intense surveillance activities with active notification measures would help improve the quality of patient care and detect and manage household contacts to prevent further virus transmission [4]. The role of animal reservoirs and how to prevent spillover is questionable. Education in a one-health context can protect against many pathogens [5]. Social science considerations can play a critical role in messaging and in interrupting transmission, reducing stigmatization and misinformation, and improving access to care [6].
Although the smallpox vaccine yields immunologic protection against MPX, its utility in the current 2022 outbreak is unlikely. This could be explained by the cessation of smallpox vaccination programs in the past 50 years, and the lack of vaccine supply in limited-resource countries. Smallpox and MPX vaccines can be used as pre- and post-exposure measures. Pre-exposure vaccination is best administered with second or third-generation vaccines such as ACAM2000, LC16m18, and JYNNEOS vaccines to protect contacts and susceptible individuals at high risk. Post-exposure vaccination can be administered within 4 – 14 days after exposure to ameliorate the infection and decrease the severity of the disease [7], [8]. Earlier experience with COVID 19 shows that heterologous vaccine programs utilizing vaccines from different platforms to augment vaccine-induced immunity are highly warranted to stimulate robust immunity which could eventually prevent viral infection and its transmission [9]. However, public confidence in vaccines should be reinforced [10]. The following research questions need to be addressed: How vaccines can and should be used, safety data, Cost, vaccine delivery strategies, and duration of vaccine effect. Could vaccines be used to generate herd immunity? Could asymptomatic vaccinated people transmit it? [5]. Following and upon availability, a multidimensional approach to increase the vaccine acceptability rate will be required [11].
Diagnosis and management perspective: Equity in the accessibility of laboratory diagnoses encounter the currently exerted efforts. The need for a biosafety level 3 for sample processing and laboratory work is a challenge, especially in limited-resource settings. The best testing techniques for finding cases are still not adequately addressed. Data on antivirals is evolving, and their effectiveness is still being investigated [12], [13]. The Food and Drug Administration (FDA) and European Medicines Agency (EMA) highly support randomized trials. It is crucial to specify who needs to be treated [5].
In conclusion, there is still much to know about monkeypox research priorities that should be promoted on both national and international levels. Sharing best practices is essential. Global research coordination is required to promote equity, generalizability, timeliness, and local competency. Interdisciplinary research teams could be of help.
==== Refs
References
1 World Health Organization. Multi-Country Monkeypox Outbreak in Non-Endemic Countries. 〈https://www.who.int/emergencies/disease-outbreak-news/item/2022-DON385〉; 2022 [Accessed 24 Novemober 2022].
2 Bunge E.M. Hoet B. Chen L. The changing epidemiology of human monkeypox-A potential threat? A systematic review PLoS Negl Trop Dis 16 2022 e0010141
3 Daskalakis D. McClung R.P. Mena L. Mermin J. Centers for Disease Control and Prevention’s Monkeypox Response Team. Monkeypox: Avoiding the Mistakes of Past Infectious Disease Epidemics Ann. Intern. Med. 2022
4 Di Gennaro F. Veronese N. Marotta C. Shin J.I. Koyanagi A. Silenzi A. Antunes M. Human Monkeypox: A Comprehensive Narrative Review and Analysis of the Public Health Implications Microorganisms 10 2022 1633 36014051
5 World Health Organization. WHO consultation sets research priorities for monkeypox. 〈https://www.who.int/news/item/03-06-2022-who-consultations-sets-research-priorities-for-monkeypox〉; 2022 [Accessed 24 November 2022].
6 Farahat R.A. Umar T.P. Khan S.H. Shrestha A.B. Kamran A. Essar M.Y. El-Sokkary R.H. Preparedness of Eastern Mediterranean countries in view of monkeypox emergence during the COVID-19 pandemic: A call for action Int J Surg 105 2022 Sep 106878
7 Weltzin R. Liu J. Pugachev K.V. Clonal vaccinia virus grown in cell culture as a new smallpox vaccine Nat Med 9 2003 1125 1130 12925845
8 Poland G.A. Kennedy R.B. Tash P.K. Prevention of monkeypox with vaccines: a rapid review Lancet Infect Dis 2022 10.1016/S1473-3099(22)00574-6
9 Goldblatt D. SARS-CoV-2: from herd immunity to hybrid immunity Nat Rev Immunol 22 2022 333 334 10.1038/s41577-022-00725-0 35440758
10 Al-Hanawi M.K. Alshareef N. El-Sokkary R.H. Relief After COVID-19 Vaccination: A Doubtful or Evident Outcome? Front Med (Lausanne) 8 2022 Jan 10 800040
11 El-Sokkary R.H. El Seifi O.S. Hassan H.M. Mortada E.M. Hashem M.K. Gadelrab M.R.M.A. Tash R.M.E. Predictors of COVID-19 vaccine hesitancy among Egyptian healthcare workers: a cross-sectional study BMC Infect Dis 21 1 2021 Aug 5 762 34353279
12 Stittelaar K.J. Neyts J. Naesens L. van Amerongen G. van Lavieren R.F. Holý A. De Clercq E. Niesters H.G. Fries E. Maas C. Mulder P.G. van der Zeijst B.A. Osterhaus A.D. Antiviral treatment is more effective than smallpox vaccination upon lethal monkeypox virus infection Nature 439 7077 2006 Feb 9 745 748 16341204
13 Cheema A.Y. Ogedegbe O.J. Munir M. Alugba G. Cureus Ojo T.K. Monkeypox: A Review of Clinical Features, Diagnosis, and Treatment 14 7 2022 Jul 11 e26756 10.7759/cureus.26756
| 36495815 | PMC9724565 | NO-CC CODE | 2022-12-08 23:19:01 | no | J Infect Public Health. 2023 Jan 6; 16(1):78-79 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.12.002 | oa_other |
==== Front
Travel Med Infect Dis
Travel Med Infect Dis
Travel Medicine and Infectious Disease
1477-8939
1873-0442
Published by Elsevier Ltd.
S1477-8939(22)00263-0
10.1016/j.tmaid.2022.102517
102517
Article
Incidence, geographic distribution, clinical characteristics, and socioeconomic and demographic determinants of monkeypox in Brazil: A nationwide population-based ecological study
Martins-Filho Paulo Ricardo ∗
Investigative Pathology Laboratory, Federal University of Sergipe, Aracaju, SE, Brazil
Nicolino Rafael Romero
Department of Preventive Veterinary Medicine, Veterinary School, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil
da Silva Kelly
Graduate Program in Applied Health Sciences, Federal University of Sergipe, Lagarto, SE, Brazil
∗ Corresponding author. Universidade Federal de Sergipe, Hospital Universitário, Laboratório de Patologia Investigativa, Rua Cláudio Batista, s/n. Sanatório, Aracaju, Sergipe, CEP: 49060-100, Brazil.
6 12 2022
6 12 2022
10251730 11 2022
5 12 2022
© 2022 Published by Elsevier Ltd.
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
==== Body
pmcHuman monkeypox is a viral disease caused by an orthopoxvirus that is endemic in some African countries, including the Democratic Republic of the Congo and Nigeria. However, a sudden increase in the number of cases worldwide has been reported since May 2022, and over 80,000 people have been diagnosed with monkeypox during the current outbreak. Studies have shown that most cases of disease have occurred predominantly among men who have sex with (MSM) with a high prevalence of lesions in the anogenital region [1,2]. Furthermore, there is growing evidence that close skin-to-skin contact is the primary route of monkeypox transmission [3].
Although Brazil accounts for approximately 12% of cases worldwide, epidemiological studies on monkeypox in the country are lacking. This study aimed to (1) report the incidence and geographic distribution of monkeypox in Brazil at the state level; (2) describe the clinical characteristics of patients with the disease; and (3) examine the relationship between incidence rates and socioeconomic and demographic determinants.
This nationwide population-based ecological study was carried out using open data from state health department bulletins in Brazil, a country with approximately 8.5 million km2, an estimated population of ∼213 million people, and a population density of ∼25 inhabitants per km2. In addition, Brazil is geopolitically divided into five regions (Northeast, North, Central-West, Southeast and South), has 26 states and one federal administrative district, 5570 municipalities, and a human development index (HDI) of 0.765.
To estimate monkeypox incidence rates in Brazil at the regional and state levels, we extracted data on confirmed cases from official bulletins from June 9 (first case report) to November 23, 2022. Furthermore, the following clinical variables were extracted: sex, age, race/ethnicity, sexual orientation, potential transmission route, and signs and symptoms. Other variables of interest included HDI and population density for each Brazilian state.
The monkeypox incidence rates per 100,000 inhabitants were calculated using the population estimate provided by the Brazilian Institute of Geography and Statistics (IBGE, acronym in Portuguese) (https://www.ibge.gov.br). The geographic distribution of monkeypox incidence was modelled using the Jenks natural breaks classification method in the ArcGis 10.3 software. Descriptive statistics were provided for the distribution of cases by sex, gender, race/ethnicity, sexual orientation, potential transmission route, and the presence of signs and symptoms. The relationship between monkeypox incidence rates and socioeconomic (HDI) and demographic (population density) determinants in a state-level analysis was carried out using a linear regression model with a backward selection process in the JASP software version 0.13 (JASP Team, Amsterdam, Netherlands). The significance level in the final model was set at 5%. Because all data were obtained from a public domain database and were deidentified, no institutional review board approval or informed consent were required.
A total of 9729 cases and 12 deaths associated with monkeypox were confirmed until November 23, 2022. The overall incidence rate was 4.6 cases per 100,000 inhabitants, with the higher rates found in the central-west (7.0 cases per 100,000) and southwest (6.8 cases per 100,000) regions. At the state-level, the higher incidence rates were found in Distrito Federal (DF) (10.8 cases per 100,000), São Paulo (SP) (8.9 cases per 100,000), Goiás (GO) (7.6 cases per 100,000), and Rio de Janeiro (RJ) (7.3 cases per 100,000) (Fig. 1 ). The states with the highest HDI (β = 31.42, standard error [SE] = 9.30; p = 0.002) and population density (β = 0.01, SE = 0.01; p = 0.035) registered the highest incidence rates of the disease (model fit: R2 = 0.606; Durbin-Watson statistics = 1.901).Fig. 1 Geographic distribution of monkeypox incidence at state level in Brazil.
Fig. 1
Most cases were diagnosed among men (92.2%), aged 20–39 years (73.6%), whites (44.9%), and self-identified homo- or bisexual (67.5%). Approximately 5% of cases were reported in children and adolescents. The most common systemic features included fever (57.9%), lymphadenopathy (41.1%), headache (39.6%), and myalgia (36.9%). Cutaneous lesions were reported in 92.4% of cases, particularly in the anogenital region (61.5%). Oral lesions were reported in about 9% of patients, and proctitis in 2.1%. A limited number of patients have described the potential transmission route of the disease, and sexual exposure before the onset of signs and symptoms has been reported in 40% of cases (Table 1 ).Table 1 Geographic distribution and clinical characteristics of monkeypox cases in Brazil.
Table 1Geographic distribution Cases Incidence rate (per 100,000 inhabitants)
Brazil 9729 4.6
Geographic regions
Central-West 1161 7.0
Southwest 6106 6.8
South 927 3.1
North 403 2.1
Northeast 1132 2.0
Clinical characteristics Cases Percentage (%)
Sex (n = 9100)
Male 8386 92.2
Female 714 7.8
Age (years) (n = 8546)
0–19 411 4.8
20–39 6287 73.6
≥40 1835 21.5
Missing data 13 0.1
Race/ethnicity (n = 6124)
White (“branco”) 2748 44.9
Brown (“pardo”) 1649 26.9
Black (“preto”) 667 10.9
Yellow (“amarelo” or of Asian ancestry) 78 1.3
Indigenous (“indígena”) 7 0.1
Missing data 975 15.9
Sexual orientation (n = 1753)
Homosexuality 1012 57.7
Heterosexuality 324 18.5
Bisexuality 172 9.8
Other 21 1.2
Missing data 224 12.8
Sexual exposure (n = 1457) 579 40.0
Signs and symptoms (n = 7518)
Cutaneous lesions 6944 92.4
Genital and/or anal lesions 4622 61.5
Fever 4353 57.9
Lymphadenopathy 3088 41.1
Headache 2980 39.6
Myalgia 2773 36.9
Asthenia 2511 33.4
Sore throat 807 10.7
Oral lesions 662 8.8
Arthralgia 227 3.0
Proctitis 158 2.1
Photosensitivity 73 1.0
Conjunctivitis 36 0.5
This is the first study to describe the incidence rates, geographic distribution, and clinical characteristics of monkeypox in Brazil. During the first months of the current outbreak, Brazil experienced high transmission rates and a significant increase in confirmed monkeypox cases [4]. We estimated an accumulated incidence similar to the found in Germany (4.4 cases per 100,000) and United Kingdom (5.5 cases per 100,000), but lower than that reported in Spain (15.6 cases per 100,000) and United States (8.6 cases per 100,000), and higher than that registered in Italy (1.6 cases per 100,000) (https://ourworldindata.org/monkeypox). Differences in disease burden between countries can be explained by a variety of factors, including geographic, socioeconomic, educational, cultural, infrastructure, and public health policy aspects.
In Brazil, a country with continental dimensions, regional differences are also noticeable. As a result, it is to be expected that the spread of monkeypox is not geographically uniform. In this study, we found higher disease incidence rates in states with better socioeconomic indicators and higher population density. Similar findings have also been demonstrated in relation to COVID-19 [5,6] and may reflect better access to healthcare in more developed urban communities. In addition, it was shown that human population density is an important variable in monkeypox distribution in endemic countries, which may be caused by increased human-environment interaction [7].
Individuals with monkeypox in Brazil appear to have clinical features similar to those reported in other non-endemic countries [1] and most of them are gay/bisexual male young adults. However, a significant number of cases have been described in women and the pediatric population, which may indicate an incipient change in the distribution pattern of the disease [8]. Although the route of transmission is poorly reported by official bulletins in Brazil, there is evidence of a high risk of infection from close contact with skin lesions [3]. Furthermore, the high prevalence of lesions in the anogenital region suggests high-risk sexual behavior among these individuals and the need to implement sex education policies and programs. Importantly, a recent global case series study showed that approximately 75% of women with monkeypox have anogenital lesions and 25% have oral lesions, with sexual contact being the most likely route of transmission [9]. These findings add to the growing body of evidence linking sexual practices to clinical lesions also present in women. Unfortunately, knowledge about disease transmission chains among children and adolescents in the current outbreak is limited.
Despite the potential for underreporting, state-level findings are the best publicly available data in the country for investigating regional differences in monkeypox estimates. Our study showed that most cases of monkeypox in Brazil have been registered in states with better socioeconomic indicators and higher population density. In addition, monkeypox in Brazil has been characterized by a high incidence of the disease in the anogenital region among young MSM. Strategies for improving diagnosis in vulnerable socioeconomic populations, including people from racial and ethnic minority groups, must be implemented.
Authors contributions
All authors contributed equally to this manuscript.
Financial source
There is no funding source.
Declaration of competing interest
The authors declare they have no conflicts of interest.
==== Refs
References
1 Martins-Filho P.R. Tanajura D.M. Vecina-Neto G. Multi-country monkeypox outbreak: a quantitative evidence synthesis on clinical characteristics, potential transmission routes, and risk factors Eur J Intern Med 2022 10.1016/j.ejim.2022.09.013
2 Thornhill J.P. Barkati S. Walmsley S. Rockstroh J. Antinori A. Harrison L.B. Monkeypox virus infection in humans across 16 countries — april–june 2022 N Engl J Med 387 2022 679 691 10.1056/NEJMoa2207323 35866746
3 Martins-Filho P.R. Tanajura D.M. Alves dos Santos C. Polymerase chain reaction positivity and cycle threshold values in biological samples from patients with monkeypox: a meta-analysis Trav Med Infect Dis 50 2022 102448 10.1016/j.tmaid.2022.102448
4 Schrarstzhaupt I.N. Fontes-Dutra M. Diaz-Quijano F.A. Early estimates of the incidence trend and the reproductive number of the monkeypox epidemic in Brazil Trav Med Infect Dis 50 2022 102484 10.1016/j.tmaid.2022.102484
5 Martins-Filho P.R. de Souza Araújo A.A. Quintans-Júnior L.J. Santos V.S. COVID-19 fatality rates related to social inequality in Northeast Brazil: a neighbourhood-level analysis J Trav Med 27 2020 10.1093/jtm/taaa128
6 Martins-Filho P.R. Relationship between population density and COVID-19 incidence and mortality estimates: a county-level analysis J Infect Public Health 14 2021 1087 1088 10.1016/j.jiph.2021.06.018 34245973
7 Arotolu T.E. Afe A.E. Wang H. Lv J. Shi K. Huang L. Spatial modeling and ecological suitability of monkeypox disease in Southern Nigeria PLoS One 17 2022 e0274325 10.1371/journal.pone.0274325
8 Martins-Filho P.R. de Souza M.F. Oliveira Góis M.A. Bezerra G.V.B. Gonçalves C.C.A. dos Santos Nascimento É.R. Unusual epidemiological presentation of the first reports of monkeypox in a low-income region of Brazil Trav Med Infect Dis 50 2022 102466 10.1016/j.tmaid.2022.102466
9 Thornhill J.P. Palich R. Ghosn J. Walmsley S. Moschese D. Cortes C.P. Human monkeypox virus infection in women and non-binary individuals during the 2022 outbreaks: a global case series Lancet 2022 10.1016/S0140-6736(22)02187-0
| 36493982 | PMC9724566 | NO-CC CODE | 2022-12-12 23:20:28 | no | Travel Med Infect Dis. 2023 Dec 6 March-April; 52:102517 | utf-8 | Travel Med Infect Dis | 2,022 | 10.1016/j.tmaid.2022.102517 | oa_other |
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00339-2
10.1016/j.jiph.2022.11.032
Original Article
Human Monkeypox and Preparedness of Bangladesh: A Knowledge and Attitude Assessment Study among Medical Doctors
Hasan Mehedi ab
Hossain Mohammad Ali ac
Chowdhury Sreshtha ab
Das Pranta d
Jahan Ishrat a
Rahman Md. Ferdous e
Haque Miah Md. Akiful b
Rashid Md Utba ae
Khan Md Abdullah Saeed af
Hossian Mosharop ab
Nabi Mohammad Hayatun a
Hawlader Mohammad Delwer Hossain a1⁎
a Department of Public Health, North South University, Dhaka 1229, Bangladesh
b Public Health Professional Development Society (PPDS), Dhaka 1215, Bangladesh
c Ibn Sina Medical College Hospital, Kallyanpur, Dhaka 1216, Bangladesh
d Department of Statistics, University of Dhaka, Dhaka 1000, Bangladesh
e International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b), Mohakhali, Dhaka 1212, Bangladesh
f Pi Research Consultancy Center, Lalbagh, Dhaka 1211, Bangladesh
⁎ Corresponding author.
1 https://orcid.org/0000-0002-1443-6257
6 12 2022
6 12 2022
13 7 2022
28 11 2022
30 11 2022
© 2022 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
Objective
The recent increasing incidence of human monkeypox cases highlights the necessity of early detection, prompt response and preventive management to stop it in its tracks, and healthcare workers play the most crucial role here. This study aims at assessing the preparedness of Bangladeshi medical doctors by assessing their knowledge and attitude regarding monkeypox.
Methodology
This cross-sectional study was conducted among the practicing medical doctors all over Bangladesh. The data was collected from 26th May to 4th June of 2022 using a semi-structured and self-administered questionnaire which was distributed through the internet, and a total of 389 data was collected. The cut-off points for defining good knowledge and positive attitude towards human monkeypox were considered as 70% and 80% of total values, respectively. Multivariable logistic regression analyses were carried out to identify the factors associated with good knowledge and a positive attitude. Statistical software R version 4.2.0 was used for data analysis.
Result
Of all, 330 (84.83%) doctors displayed a positive attitude towards preventive practices, but only 119 (30.59%) participants had good knowledge regarding monkeypox. In multivariable logistic regression analysis, getting any information about monkeypox in the medical curriculum and learning about monkeypox within the last one month had a significant association with good knowledge. Apart from the participant's age, no other variables revealed any significant association with a positive attitude toward preventive practices. Good knowledge showed a significant association with positive attitude (p<0.05).
Conclusion
Knowledge regarding human monkeypox among medical doctors in Bangladesh was comparatively lower than the attitude towards its preventive measures. Developing and implementing practical sessions regarding the virus to enhance the knowledge and capacity of the medical doctors could be an effective strategy to get prepared for the monkeypox outbreak in Bangladesh.
Keywords
Monkeypox
Medical Doctors
Bangladesh
Viral Outbreak
Pandemic
==== Body
pmc1 Introduction
A zoonotic infection with smallpox-like signs and symptoms, human monkeypox (HMPX) is caused by the monkeypox virus (MPXV), which is a member of the genus orthopoxvirus. It is found predominantly in Central and Western Africa [34], [35], [36]. Timing of illness onset, rash distribution, and timing of rash occurrence are comparable to smallpox [28], [33], [34], [35], [36]. The first case of HMPX was recorded in the Democratic Republic of the Congo in 1970, and since then, multiple outbreaks and sporadic cases have been documented throughout the Central as well as the Western African nations [5]. The first human monkeypox cases reported outside of Africa occurred in 2003 in the United States of America [17]. The number of HMPX cases has increased dramatically over the past two decades [10], [28], and as a result, it is currently regarded as the most relevant orthopoxvirus from a public health view [37]. Republic of Congo, Democratic Republic of Congo, Central African Republic, Liberia, Nigeria, and Cameroon have reported recent HMPX outbreaks [24], [37], [5]. Although the case fatality rate (CFR) for the 2017–2018 outbreak in Nigeria was 6%, recent reports mention of a CFR of 2.6% in African Union member countries and globally 0.04% in 2022 which is much lower compared to the historical data [1], [19], [6], [7], [8], [37].
Recently the virus was reported to spread outside African continent and the United States [17], [34], [35], [36], the United Kingdom [32], and Israel [11], encountered HMPX transferred from other countries. In May of 2019, Singapore reported the first verified HMPX case in Asia; the patient was a Nigerian citizen who attended a training in Singapore [22]. About 100 countries have reported cases of monkeypox since May 2022, with over 62,000 confirmed cases [17]. As of yet, no confirmed case has been found in Bangladesh (The Daily Star 2022). In May of 2022, however, Bangladesh issued a health warning due to an outbreak of monkeypox worldwide [3].
The Democratic Republic of the Congo (DRC) performs routine surveillance for the monkeypox as it is endemic in this region [25], [4]. In DRC guidelines mandate for biweekly notification and incidence reporting and case investigations for suspected cases include specimen diagnosis and case report submission ([4]). WHO recommends reporting suspicious cases to public health authorities immediately, and case investigation should include a clinical assessment, possible sources identification and laboratory examination. Once a case is suspected, contact identification, contact tracing should be conducted immediately[34], [35], [36]
Monkeypox virus has been found to spread through direct contact with infectious rash, scabs or fluid from sores. Transmission through respiratory secretions and saliva during prolonged intimate physical contact including sex has been reported [6], [7], [8]. The secondary attack rate among unvaccinated contacts could be as high as 11% [5]. Healthcare-associated transmissions of MPXV were reported from central and west African countries (Yinka-Ogunleye et al. 2003; [21]; Lakhani et al. 2019; [37]). Enhancing the ability of healthcare professionals to identify cases and improve patient management is one of the most essential features of the surveillance system [4]. Healthcare professionals, especially medical doctors, should be familiar with the clinical symptoms of monkeypox in order to promptly detect, report, and treat new cases to prevent their spread.
A prior study in Indonesia, indicated that the understanding of HMPX among general practitioners (GPs) was quite low and approximately only 10% of them had a decent knowledge [13], [14]. Identified causative factors were: (a) no HMPX instances have been documented in Indonesia; and (b) the disease is not required to be taught in medical schools in the country according to the Indonesian Standard of Medical Doctor Competency [SKDI]. Due to these reasons the healthcare workers might not be able to handle an outbreak of HMPX [22], [13], [14]. Another report showed that a lack of knowledge about monkeypox, especially among healthcare workers, was one of the things that made it hard to stop the disease from coming back [33]. The recent growing incidence of human monkeypox cases necessitates prevention, early detection, prompt response and management by healthcare providers. Therefore, it is crucial for healthcare workers to be knowledgeable and prepared for monkeypox cases. No study so far has been conducted to assess the status of knowledge and attitude regarding monkeypox among Bangladeshi medical doctors. Hence, the present study was designed and undertaken to assess the knowledge regarding monkeypox and attitude towards its prevention among Bangladeshi medical doctors.
2 Methodology
2.1 Study Design
This was a cross-sectional study among practicing medical doctors all over Bangladesh. Data was collected between 26th May 2022 to 4th June. The researchers developed a semi-structured questionnaire. The questionnaire was circulated via the internet in the professional forum of Bangladeshi doctors. Those who gave consent participated in this study and submitted the questionnaire in Google form. The questionnaire was developed in English. Before circulating the questionnaire to the target population, it was pre-tested among 10 medical doctors. Based on their feedback, the final version was created and data collection was done.
2.2 Measures
One of the main purposes of the study was to assess the knowledge and attitude toward Human Monkeypox among Bangladeshi physicians. Based on the goal and to collect potential explanatory variables, it was divided into two sections, a) knowledge, and b) attitude. Knowledge is the understanding of facts and processes, which is acquired though information and experiences. And attitude is the feeling towards someone or something, which results readiness to act or behave in a certain way [26]. The explanatory variables were (a) Sociodemographic items; (b) Workplace characteristics; (c) Characteristics of the medical professional; and (d) Exposure to monkeypox-related information. There were 24 questions in the knowledge section and 6 questions in the attitude section. Questions were adapted from previous studies [13], [14]. The correct answers were determined based on guidelines and factsheets for monkeypox [34], [35], [36], [6], [7], [8]. For the knowledge domain, correct answer received a score of 1 and incorrect answers received a score of zero. The scores were added together to give a total knowledge score that ranges from 0 to 24, where a higher score indicates better knowledge. Regarding attitude, the questions were scored on a 3-point Likert scale (disagree, neither agree nor disagree, agree). The responses are given as 1 for ‘disagree’, 2 for ‘neither disagree or agree’ and 3 for ‘agree’. Scores were added to give an overall score of 6 to 18 for attitudes, where a higher score indicates a more positive attitude towards the monkeypox preventive measures.
2.3 Statistical Analysis
Respectively, 70% and 80% of total values were considered as the cut-off point for defining good knowledge and positive attitude. Categorical and continuous variables were presented as frequency (percentage) and mean ± standard deviation (SD), respectively. Univariate followed by multivariable logistic regression analyses were performed to identify factors associated with good knowledge and positive attitude. The outcomes were presented as odds ratios (ORs) with corresponding 95% confidence intervals (CIs). P-values < 0.05 were considered statistically significant, and statistical software R version 4.2.0 was used for data analysis.
2.4 Ethical Consideration
Ethical approval was obtained from the Institutional Review Board (IRB)/Ethical Review Committee (ERC) of North South University (2022/0R-NSU/IRB/0501). The purpose of the study was elaborately explained in the online questionnaire form and a consent script was included. Those who consented participated in this study. Participants’ privacy and data safety were strictly maintained according to the IRB instruction and ethical standards of the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards were followed wherever applicable.
3 Result
A total of 389 respondents who met the inclusion criteria voluntarily took part in this survey and were included in the final statistical analysis. The majority of our participants were female medical doctors (52.7%), aged less than 30 years (52,7%), graduated from government institutes (50.13%), worked in private hospitals (57.58%), with no post-graduate degree (83.55%) and practicing for less than 5 years (51.41%). Only 29.05% of doctors admitted learning information on monkeypox in their medical curriculum, and 82.26% learned about monkeypox within the last month prior to the survey. Among all, 119 (30.59%) doctors had good knowledge of monkeypox disease, its causation, symptoms, transmissions, and treatment. However, over four fifths (n=330, 84.83%) of the participants displayed a positive attitude ( Fig. 1 & Table 1). All the responses to knowledge and attitude questions with their responses are provided in the supplementary file (Table S1, Table S2).Fig. 1 Distribution of knowledge and attitude level regarding monkeypox (Good knowledge was defined as having >70% score and good attitude was defined as having >80% score).
Fig. 1
Table 1 Knowledge and attitude of participants in terms of sociodemographic and professional characteristics.
Table 1Variable Frequency n (%) Knowledge Attitude
Poor Good Negative Positive
Age
<30 205 (52.7) 146 (71.22) 59 (28.78) 41 (20) 164 (80)
≥30 184 (47.3) 124 (67.39) 60 (32.61) 18 (9.78) 166 (90.22)
Gender
Female 205 (52.7) 144 (70.24) 61 (29.76) 30 (14.63) 175 (85.37)
Male 184 (47.3) 126 (68.48) 58 (31.52) 29 (15.76) 155 (84.24)
Institute
Govt. Institute 195 (50.13) 136 (69.74) 59 (30.26) 28 (14.36) 167 (85.64)
Private Institute 194 (49.87) 134 (69.07) 60 (30.93) 31 (15.98) 163 (84.02)
Workplace
Govt. Hospital 140 (35.99) 94 (67.14) 46 (32.86) 18 (12.86) 122 (87.14)
Only Private Chamber/GP 25 (6.43) 19 (76) 6 (24) 6 (24) 19 (76)
Private Hospital 224 (57.58) 157 (70.09) 67 (29.91) 35 (15.63) 189 (84.37)
Education
Graduate (MBBS) 325 (83.55) 233 (71.69) 92 (28.31) 54 (16.62) 271 (83.38)
Postgraduate (Completed) 64 (16.45) 37 (57.81) 27 (42.19) 5 (7.81) 59 (92.19)
Got information about monkeypox
No 276 (70.95) 207 (75) 69 (25) 41 (14.86) 235 (85.14)
Yes 113 (29.05) 63 (55.75) 50 (44.25) 18 (15.93) 95 (84.07)
Learned about monkeypox
Before last 1 month 69 (17.74) 33 (47.83) 36 (52.17) 11 (15.94) 58 (84.06)
Within last 1 month 320 (82.26) 237 (74.06) 83 (25.94) 48 (15) 272 (85)
Duration of medical practice
<5 years 200 (51.41) 132 (66) 68 (34) 35 (17.5) 165 (82.5)
≥5 years 189 (48.59) 138 (73.02) 51 (26.98) 24 (12.7) 165 (87.3)
Univariate regression analysis identified level of education, getting information about monkey pox in medical curriculum and the time of learning about monkeypox to be significantly associated with good knowledge about monkey pox. Those that completed post-graduation degree, got information about monkey pox in medical curriculum had higher odds of having good knowledge regarding monkey pox (OR=1.85 and OR=2.38, respectively). And those that learned about monkeypox within the last month had lower odds ratio of 0.32 (95% CI= 0.19, 0.55) ( Table 2: Univariate logistic regression analysis exploring factors associated with good knowledge and attitude about monkeypox among medical doctors). In case of attitude, older participants displayed higher odds of having positive attitude (OR=2.31, 95% CI=1.29, 4.27)Table 2 Univariate logistic regression analysis exploring factors associated with good knowledge and attitude about monkeypox among medical doctors.
Table 2Model for Knowledge Attitude
Variables OR 95% Confidence Interval p-value OR 95% Confidence Interval p-value
Lower Upper Lower Upper
Age
<30 years Ref. Ref.
≥30 years 1.20 0.78 1.85 0.414 2.31 1.29 4.27 0.006
Gender
Female Ref. Ref.
Male 1.09 0.71 1.67 0.706 0.92 0.53 1.60 0.757
Institute
Govt. Institute Ref. Ref.
Private Institute 1.03 0.67 1.60 0.886 0.88 0.50 1.54 0.656
Workplace
Govt. Hospital Ref. Ref.
Only Private chamber/GP 0.65 0.22 1.64 0.383 0.47 0.17 1.42 0.153
Private Hospital 0.87 0.55 1.38 0.555 0.80 0.42 1.45 0.467
Education
Graduate (MBBS) Ref. Ref.
Post-graduate (Completed) 1.85 1.06 3.2 0.029 2.35 0.99 6.97 0.080
Got information about monkeypox
No Ref. Ref.
Yes 2.38 1.50 3.78 <0.001 0.92 0.51 1.72 0.789
Learned about monkeypox
Before last 1 month Ref. Ref.
Within last 1 month 0.32 0.19 0.55 <0.001 1.07 0.50 2.13 0.843
Duration of medical practice
<5 years Ref. Ref.
≥5 years 0.72 0.46 1.11 0.134 1.46 0.84 2.58 0.189
Note: Reference category for both knowledge and attitude is Poor and p value <0.05 indicates significant
Multivariate logistic regression analysis revealed a statistically significant association of whether they got any information about monkeypox in the medical curriculum or not and the time of learning about monkeypox with good knowledge regarding monkeypox ( Table 3) when adjusted for other covariates. Getting information about monkeypox at any point in their medical academic curriculum were associated with a significant aOR value of 1.83 (95% CI=1.11, 3.01). However, learning about monkeypox within the last month showed lower odds ratio (aOR=0.43) of possessing good knowledge compared to those who learned about it before one month. Although in the univariate model education of the participant showed significant association, in multivariate model it was observed to be not significant (aOR=1.67, 95% CI=0.94, 2.96) when adjusted for other covariates. Only age of a participant showed a significant statistical association with positive attitude in the logistics regression model (aOR=2.31, 95% CI=1.29, 4.27) (Table 3).Table 3 Multivariate logistic regression analysis exploring factors associated with good knowledge and attitude about monkeypox among medical doctors.
Table 3Model for Variables aOR 95% Confidence Interval p-value
Lower Upper
Knowledge Education
Graduate (MBBS) Ref.
Post-graduate (Completed) 1.67 0.94 2.96 0.078
Got information about monkeypox
No Ref.
Yes 1.83 1.11 3.01 0.017
Learned about monkeypox
Before last 1 month Ref.
Within last 1 month 0.43 0.24 0.76 0.004
Attitude Age
<30 years Ref.
≥30 years 2.31 1.29 4.27 0.006
Note: Reference category for both knowledge and attitude is Poor and p value <0.05 indicates significant
Statistically significant association was found between knowledge and attitude of a participant; out of 119 doctors scored good in knowledge, 110 (92.44%) scored 80% or more in the attitude domain ( Table 4).Table 4 Association between knowledge and attitude related to monkeypox among medical doctors.
Table 4Knowledge Attitude p-value
Positive Negative
Good 110 9 0.009
Poor 220 50
Note: p-value < 0.05 indicates significant.
Overall, participants had a better attitude despite the overwhelming poor knowledge of monkeypox.
4 Discussion
After overcoming the COVID-19 pandemic for two years, new zoonotic monkeypox virus transmission has recently been reported in many non-endemic nations. Although it is an unusual, self-limiting infection that is generally milder than smallpox and is not a cause for alarm at this time, early discovery and prompt action are crucial for disease control [31]. To effectively combat this epidemic, all governments and their respective public health departments should be on high alert and work together closely. People must be informed about the disease's treatment options and preventative and risk factors via massive awareness initiatives. The frontline medical professionals need to be educated on handling confirmed or suspected cases and protecting themselves while doing so. It is essential for medical professionals and physicians, in particular, to have a sufficient grasp of this condition to completely detect, diagnose, and treat instances of it. Only then can the disease be managed effectively.
According to our study's findings, the knowledge level about monkeypox among doctors in Bangladesh is inadequate. Approximately 31% were capable of providing accurate responses to 70% or more of the 24 questions. In a similar study, general practitioners in Indonesia knew relatively little about the emerging infection known as monkeypox [13], [14]. In contrast, regarding endemic diseases in Bangladesh, the level of knowledge is rather high; most doctors are well-versed in diseases such as "dengue," which are transmitted by Aedes aegypti [18]. It should not come as a surprise that participants in general lack awareness about monkeypox since it is a re-emerging infectious disease, and instances of it have never been documented in Bangladesh. Another possible reason behind this fact is that, due to time constraints, medical students in Bangladesh may be mainly focused on common diseases. Because of this, medical graduates are not expected to be able to treat or manage this condition altogether.
The study also reveals a positive attitude towards monkeypox and its prevention among doctors as more than 84% had a good score in the attitude domain. A similar finding was observed in a study conducted in Conakry, Guinea, during the Ebola virus outbreak [2]. The reason behind this positive attitude is that the physicians are already worn down by their experiences with COVID-19 and are aware of the repercussions of a pandemic. Because of this, they exhibited a positive attitude toward learning about this virus, and they are highly willing to take the necessary steps to avoid the virus's spread and to follow the treatment protocol. According to the findings of this research, education was not found to be significantly associated with good knowledge or positive attitude. However, older age was found to be associated with positive attitude. This is because as one gets older, they can amass more experience, increase their depth of knowledge and increase positivity in their attitude. However, the results of this research demonstrated that a longer period of medical practice was associated with a decreased likelihood of having good knowledge. Similarly, a systematic review of 12 research published between 1966 and 2004 indicated that more years of experience in medical practice were associated with lower knowledge levels [9]. In this case, it can be said that, because of long-term clinical practice, medical professionals gain expertise in diagnosing and treating prevalent illnesses, but they become less aware of new diseases or infections.
We found that individuals who got information about monkeypox in curriculum of medical school and individuals who learned about monkeypox in the last one month became more knowledgeable about the disease. However, this is contrary to the findings in Indonesia, where there was no significant difference found in knowledge between physicians who received information on monkeypox as part of their medical education and those who did not [13], [14].
A statistically significant relationship was observed in our investigation between the individuals' knowledge levels and attitudes. These findings are consistent with the results reported studies carried out in Pakistan and Indonesia [12], [15], [16]. The conclusion that can be drawn from this is that one's level of knowledge can always influence one's attitude and that having a better comprehension of anything may lead to having a more positive attitude.
The study will be the first research of its kind to come out of Bangladesh, and it will help us gauge the scope of the status of knowledge and attitude regarding monkeypox in Bangladesh, and guide us to take necessary strategies to enhance KAP on further aspects of the issue. Survey results will inform an intervention approach reflective of local conditions and cultural influences; this will allow us to tailor our efforts to the target group's needs. However, it is crucial to acknowledge some of our study's limitations and the approaches we took to address them. Due to our study’s cross-sectional nature, we cannot infer causality for the associations we presented in this paper. Furthermore, in non-random studies, selection bias is always a possible drawback. Also, the fact that the questionnaire was self-administered is another constraint: our research findings are not generalizable since we used a convenience sample.
5 Conclusion
The attitude towards human monkeypox prevention among medical doctors in Bangladesh was quite good. However, the knowledge level regarding the virus was low. The study identified that presence of information regarding monkeypox in the medical curriculum and when a participant learned about monkeypox had a significant association with knowledge level. On the other hand, the participants’ attitude towards human monkeypox was satisfactory and only age of a participant had a significant association with attitude. Developing and implementing practical knowledge-sharing sessions to enhance the capacity of doctors regarding human monkeypox could be an effective strategy towards improving the current condition and preparing for any future outbreak.
CRediT authorship contribution statement
MH and MEH did the literature search. DHH, HTN and MASK conceived and designed the study. DHH, IJ and PD oversaw its implementation, analysis, and write-up. MASK planned the statistical analyses. MAH, MH and MEH outlined the data collection procedure. SC, MFR, MAH, MEH and IJ contributed to the field implementation of the study and did data entry. MASK, HTN and MAH verified the underlying data. PD did the statistical analyses. IJ, MFR, MEH, SC wrote the first draft of the manuscript. All authors read and approved the manuscript.
Uncited references
[20], [23], [27], [29], [30]
Conflict of Interest
We have no conflict of interest to declare.
Appendix A Supplementary material
Supplementary material
Acknowledgment
The authors in this study express their heartfelt gratitude for the those that participated in this study
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jiph.2022.11.032.
==== Refs
References
1 Africa CDC. Outbreak brief 11: Monkeypox in Africa union member states. [accessed 2022, October 08]. AFRICA CDC. 〈https://africacdc.org/disease-outbreak/outbreak-brief-11-monkeypox-in-africa-union-member-states/〉
2 Alioune C. Knowledge, attitudes and practices of health care workers on Ebola in hospital towards Ebola virus disease, Conakry, guinea, 2016 Cent Afr J Publ Health [Internet] 4 1 2018 1 10.11648/j.cajph.20180401.11
3 Bangladesh declares health alert amid monkeypox outbreak. 2022. ANI News [Internet]. [accessed 2022 Jul 3]. 〈https://www.aninews.in/news/world/asia/bangladesh-declares-health-alert-amid-monkeypox-outbreak20220522183345/〉
4 Bass J. Tack D.M. McCollum A.M. Kabamba J. Pakuta E. Malekani J. Nguete B. Monroe B.P. Doty J.B. Karhemere S. Enhancing health care worker ability to detect and care for patients with monkeypox in the Democratic Republic of the Congo Int Health [Internet] 5 4 2013 237 243 10.1093/inthealth/iht029 24246742
5 Beer E.M. Rao V.B. A systematic review of the epidemiology of human monkeypox outbreaks and implications for outbreak strategy PLoS Negl Trop Dis [Internet] 13 10 2019 e0007791 10.1371/journal.pntd.0007791
6 CDC. 2022a. Monkeypox. Centers for Disease Control and Prevention [Internet]. [accessed 2022 Jul 3]. 〈https://www.cdc.gov/poxvirus/monkeypox/index.html〉
7 CDC. (2022b, September 1). 2022 Monkeypox outbreak global map. Centers for Disease Control and Prevention. 〈https://www.cdc.gov/poxvirus/monkeypox/response/2022/world-map.html〉
8 CDC. (2022c, September 16). How it spreads. Centers for Disease Control and Prevention. 〈https://www.cdc.gov/poxvirus/monkeypox/if-sick/transmission.html〉
9 Choudhry N.K. Fletcher R.H. Soumerai S.B. Systematic review: the relationship between clinical experience and quality of health care Ann Intern Med [Internet] 142 4 2005 260 273 10.7326/0003-4819-142-4-200502150-00008 15710959
10 Durski K.N. McCollum A.M. Nakazawa Y. Petersen B.W. Reynolds M.G. Briand S. Djingarey M.H. Olson V. Damon I.K. Khalakdina A. Emergence of Monkeypox - west and central Africa, 1970-2017 MMWR Morb Mortal Wkly Rep [Internet] 67 10 2018 306 310 10.15585/mmwr.mm6710a5 29543790
11 Erez N. Achdout H. Milrot E. Schwartz Y. Wiener-Well Y. Paran N. Politi B. Tamir H. Israely T. Weiss S. Diagnosis of imported Monkeypox, Israel, 2018 Emerg Infect Dis [Internet] 25 5 2019 980 983 10.3201/eid2505.190076 30848724
12 Ul Haq N. Hassali M.A. Shafie A.A. Saleem F. Farooqui M. Aljadhey H. A cross sectional assessment of knowledge, attitude and practice towards Hepatitis B among healthy population of Quetta, Pakistan BMC Public Health [Internet] 12 1 2012 692 10.1186/1471-2458-12-692 22917489
13 Harapan H. Setiawan A.M. Yufika A. Anwar S. Wahyuni S. Asrizal F.W. Sufri M.R. Putra R.P. Wijayanti N.P. Salwiyadi S. Razi Maulana Khusna A. Nusrina I. Shidiq M. Fitriani D. Muharrir M. Husna C.A. Yusri F. Reza Maulana Andalas M. Knowledge of human monkeypox viral infection among general practitioners: a cross-sectional study in Indonesia Pathog Glob Health [Internet] 114 2 2020 68 75 10.1080/20477724.2020.1743037 32202967
14 Harapan H. Setiawan A.M. Yufika A. Anwar S. Wahyuni S. Asrizal F.W. Sufri M.R. Putra R.P. Wijayanti N.P. Salwiyadi S. Razi Maulana Khusna A. Nusrina I. Shidiq M. Fitriani D. Muharrir M. Husna C.A. Yusri F. Reza Maulana Utomo P.S. Confidence in managing human monkeypox cases in Asia: A cross-sectional survey among general practitioners in Indonesia Acta Trop [Internet] 206 105450 2020 105450 10.1016/j.actatropica.2020.105450
15 Harapan H. Rajamoorthy Y. Anwar S. Bustamam A. Radiansyah A. Angraini P. Fasli R. Salwiyadi S. Bastian R.A. Oktiviyari A. Akmal I. Iqbalamin M. Adil J. Henrizal F. Darmayanti D. Pratama R. Setiawan A.M. Mudatsir M. Hadisoemarto P.F. Müller R. Knowledge, attitude, and practice regarding dengue virus infection among inhabitants of Aceh, Indonesia: a cross-sectional study BMC Infectious Diseases 18 1 2018 10.1186/s12879-018-3006-z
16 Harapan H. Rajamoorthy Y. Utomo P.S. Anwar S. Setiawan A.M. Alleta A. Bambang A. Ramadana M.R. Ikram I. Wahyuniati N. Maulana R. Ichsan I. Indah R. Wagner A.L. Kuch U. Groneberg D.A. Rodríguez-Morales A.J. Andalas M. Müller R. Imrie A. Knowledge and attitude towards pregnancy-related issues of Zika virus infection among general practitioners in Indonesia BMC Infectious Diseases 19 1 2019 693 10.1186/s12879-019-4297-4 31387537
17 Ilic I. Zivanovic Macuzic I. Ilic M. Global outbreak of human Monkeypox in 2022: Update of epidemiology Tropical Medicine and Infectious Disease 7 10 2022 264 10.3390/tropicalmed7100264 36288005
18 Koonisetty K.S. Aghamohammadi N. Urmi T. Yavaşoglu S.İ. Rahman M.S. Nandy R. Haque U. Assessment of knowledge, attitudes, and practices regarding dengue among physicians: A web-based cross-sectional survey Behavioral Sciences 11 8 2021 105 10.3390/bs11080105 34436095
19 Kozlov M. How deadly is monkeypox? What scientists know Nature 609 7928 2022 663 10.1038/d41586-022-02931-1 36100744
20 Learned L.A. Reynolds M.G. Wassa D.W. Li Y. Olson V.A. Karem K. Stempora L.L. Braden Z.H. Kline R. Likos A. Extended interhuman transmission of monkeypox in a hospital community in the Republic of the Congo, 2003 Am J Trop Med Hyg [Internet] 73 2 2005 428 434 10.4269/ajtmh.2005.73.428 16103616
21 Nakoune E. Lampaert E. Ndjapou S.G. Janssens C. Zuniga I. Van Herp M. Fongbia J.P. Koyazegbe T.D. Selekon B. Komoyo G.F. A nosocomial outbreak of human Monkeypox in the Central African Republic Open Forum Infect Dis [Internet] 4 4 2017 10.1093/ofid/ofx168
22 Ng O.T. Lee V. Marimuthu K. Vasoo S. Chan G. Lin R.T.P. Leo Y.S. A case of imported Monkeypox in Singapore Lancet Infect Dis [Internet] 19 11 2019 1166 10.1016/S1473-3099(19)30537-7
23 No monkeypox case detected in Bangladesh yet, BSMMU says. 2022. The Daily Star [Internet]. [accessed 2022 Jul 3]. 〈https://www.thedailystar.net/health/disease/news/no-monkeypox-case-detected-bangladesh-yet-bsmmu-says-3030726〉
24 Ogoina D. Izibewule J.H. Ogunleye A. Ederiane E. Anebonam U. Neni A. Oyeyemi A. Etebu E.N. Ihekweazu C. The 2017 human monkeypox outbreak in Nigeria-Report of outbreak experience and response in the Niger Delta University Teaching Hospital, Bayelsa State, Nigeria PLoS One [Internet] 14 4 2019 e0214229 10.1371/journal.pone.0214229
25 Osadebe L. Hughes C.M. Shongo Lushima R. Kabamba J. Nguete B. Malekani J. Pukuta E. Karhemere S. Muyembe Tamfum J.-J. Wemakoy Okitolonda E. Enhancing case definitions for surveillance of human monkeypox in the Democratic Republic of Congo PLoS Negl Trop Dis [Internet] 11 9 2017 e0005857 10.1371/journal.pntd.0005857
26 Pradhan H.B. A Textbook of Health Education (Philosophy and Principles) 2nd ed. 2017 Educational Publishing House Kathmandu
27 Petersen B.W. Kabamba J. McCollum A.M. Lushima R.S. Wemakoy E.O. Muyembe Tamfum J.-J. Nguete B. Hughes C.M. Monroe B.P. Reynolds M.G. Vaccinating against monkeypox in the Democratic Republic of the Congo Antiviral Res [Internet] 162 2019 171 177 10.1016/j.antiviral.2018.11.004 30445121
28 Petersen E. Kantele A. Koopmans M. Asogun D. Yinka-Ogunleye A. Ihekweazu C. Zumla A. Human Monkeypox: Epidemiologic and clinical characteristics, diagnosis, and prevention Infect Dis Clin North Am [Internet] 33 4 2019 1027 1043 10.1016/j.idc.2019.03.001 30981594
29 Reed K.D. Melski J.W. Graham M.B. Regnery R.L. Sotir M.J. Wegner M.V. Kazmierczak J.J. Stratman E.J. Li Y. Fairley J.A. The detection of monkeypox in humans in the Western Hemisphere N Engl J Med [Internet] 350 4 2004 342 350 10.1056/NEJMoa032299 14736926
30 Sejvar J.J. Chowdary Y. Schomogyi M. Stevens J. Patel J. Karem K. Fischer M. Kuehnert M.J. Zaki S.R. Paddock C.D. Human monkeypox infection: a family cluster in the midwestern United States J Infect Dis [Internet] 190 10 2004 1833 1840 10.1086/425039 15499541
31 Shanmugaraj B. Phoolcharoen W. Khorattanakulchai N. Emergence of monkeypox: Another concern amidst COVID-19 crisis Asian Pac J Trop Med [Internet] 15 5 2022 193 10.4103/1995-7645.346081
32 Vaughan A. Aarons E. Astbury J. Balasegaram S. Beadsworth M. Beck C.R. Chand M. O’Connor C. Dunning J. Ghebrehewet S. Two cases of monkeypox imported to the United Kingdom, September 2018 Euro Surveill [Internet] 23 38 2018 10.2807/1560-7917.ES.2018.23.38.1800509
33 World Health Organization. Informal Consultation on Monkeypox 2017 [Internet]. [accessed 2022 Jul 3]. 〈https://apps.who.int/iris/bitstream/handle/10665/272620/WHO-WHE-IHM-2018.3-eng.pdf〉
34 World Health Organization. Monkeypox. Who.int [Internet]. [accessed 2022a Jul 3]. 〈https://www.who.int/news-room/questions-and-answers/item/monkeypox〉
35 World Health Organization. Multi-country monkeypox outbreak: situation update. (n.d.). Who.int. [accessed October 8, 2022b] from 〈https://www.who.int/emergencies/disease-outbreak-news/item/2022-DON396〉
36 World Health Organization. Surveillance, case investigation and contact tracing for monkeypox: interim guidance, [accessed 25 August 2022c]. Who.int; 〈https://www.who.int/publications/i/item/WHO-MPX-Surveillance-2022.3〉
37 Yinka-Ogunleye A. Aruna O. Dalhat M. Ogoina D. McCollum A. Disu Y. Mamadu I. Akinpelu A. Ahmad A. Burga J. Outbreak of human monkeypox in Nigeria in 2017-18: a clinical and epidemiological report Lancet Infect Dis [Internet] 19 8 2019 872 879 10.1016/S1473-3099(19)30294-4 31285143
| 36508945 | PMC9724567 | NO-CC CODE | 2022-12-09 23:15:04 | no | J Infect Public Health. 2023 Jan 6; 16(1):90-95 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.11.032 | oa_other |
==== Front
Virol Sin
Virol Sin
Virologica Sinica
1674-0769
1995-820X
The Authors. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.
S1995-820X(22)00195-X
10.1016/j.virs.2022.12.001
Letter
Development of rapid nucleic acid assays based on the recombinant polymerase amplification for monkeypox virus
Li Yuchang a1
Gao Yanhong b1
Tang Ying a
Li Jing a
Zhang Sen a
Jiang Tao a∗∗
Kang Xiaoping a∗
a State Key Laboratory of Pathogen and Biosecurity, The Academy of Military Medical Science, Institute of Microbiology and Epidemiology, Beijing 100071, China
b Laboratory Department of the First Medical Center, Chinese PLA General Hospital, Beijing 100850, China
∗ Corresponding author. (X. Kang)
∗∗ Corresponding author. (T. Jiang)
1 Yuchang Li and Yanhong Gao Contributed equally to this work.
6 12 2022
6 12 2022
26 8 2022
2 12 2022
© 2022 The Authors
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
==== Body
pmcDear Editor,
Monkeypox virus (MPXV), an enveloped double-stranded DNA virus with 190 open-reading frames and a genome length of about 200 kb, belongs to the genus Orthomyxovirus (OPXV; subfamily Chordopoxvirinae, family Poxviridae), which causes a disease with symptoms similar to, but less severe than, smallpox. MPXV is subdivided into two clades: clade I for the former Congo Basin clade and clade II for the former West African, and the clade I is more pathogenic. The clade II consists two subclades, clade IIa and clade IIb, with the latter referring primarily to the group of variants largely circulating in the 2022 global outbreak (ICTV, 2020; Bunge et al., 2022; Mauldin et al., 2022).
Monkeypox can be transmitted from animals to humans. Cases often occur close to tropical rainforests. Though human-to-human transmission of MPXV is limited, it can occur through direct contact with infectious skin or mucocutaneous lesions. The incubation period of monkeypox has historically ranged from 5–21 days. Typically, the prodromal phase of clinical illness lasts 1–5 days, during which time patients may experience fever, headache, back pain, muscle aches, and lymphadenopathy. The second phase which occurs typically after the fever subsides, and is characterized by a rash that may involve the oral mucous membranes, conjunctiva, cornea and/or genitalia (Adler et al., 2022; Miura et al., 2022).
Since 1 January and of 7 Oct 2022, 71,237 confirmed Monkeypox-infected cases in 107 countries have been reported to World Health Organization (WHO), and total deaths reached 26. Globally, the overall risk is assessed as moderate, considering this is the first time that cases and clusters are reported concurrently in five WHO Regions. The risk in the European Region is considered to be high due to its report of a geographically widespread outbreak involving several newly affected countries, as well as a somewhat atypical clinical presentation of the cases. In other WHO Regions, the risk is considered moderate with consideration for epidemiological patterns, possible risk of importation of cases and capacities to detect cases and respond to the outbreak. In newly affected countries, this is the first time that cases have mainly, but not exclusively, been confirmed among men who have had recent sexual contact with a new or multiple male partners (WHO, 2022a; 2022b).
For MPXV test confirmation, the best diagnostic specimens are directly from the rash-skin, fluid or crusts, or biopsy where feasible (Erez et al., 2019; Thornhill et al., 2022). Methods that detect antigens and antibody may not be useful as they do not distinguish between orthomyxoviruses (Stern et al., 2016). At present, detecting viral DNA by quantitative polymerase chain reaction (qPCR) is the preferred laboratory test for monkeypox; indeed, a positive PCR result is part of case definition, regardless of associated symptoms or their absence (WHO, 2022a; 2022b; Erez et al., 2019). However, qPCR method is time consuming (>1 hours) and requires complex and expensive equipment. Therefore, there is a need for a faster and simpler method for MPXV testing.
Recombinant polymerase isothermal amplification (RPA) technology utilizes a recombinant enzyme obtained from bacteria or fungi. At room temperature, the recombinant enzyme binds tightly to the primer DNA to form a polymer. First, the primer recognizes a complementary sequence in the template DNA. Next, with the help of a single-stranded DNA binding protein, the double-stranded structure of the template DNA is unwounded and opened by DNA polymerase. A new DNA strand is formed, and the amplification product grows exponentially. The RPA technology can amplify a target gene within 15–20 min at room temperatures (25–42 °C), meaning that the amplification reaction is faster than qPCR; also, and the equipment is both compact and cheap, making it more suitable than qPCR for application in local unit (Jia et al., 2019; Wang et al., 2022; Davi et al., 2019; Tang et al., 2022).
In this study, we developed two RPA methods to detect MPXV, fluorescence-RPA (F-RPA) and vertical flow strip RPA (VF-RPA). F-RPA is performed in a small portable fluorescence-RPA device, and the fluorescence curve can be observed in real time, which is suitable for local medical unit; VF-RPA requires only a small thermostatic bath; the VF-RPA results can be visualized on a strip, making the assay sensitive, easy to operate and low-cost. Thus, the VF-RPA is of potentially valuable for MPXV detection in local units. One concern, however, is avoiding DNA contamination during tube opening and sampling. In this study, the VF-RPA reaction takes place within a simple disposable vertical-flow strip cassette, which is designed for instrument-free, cross-contamination-proof detection of amplicons, enabling rapid identification of MPXV in field. Therefore, both of F-RPA and VF-RPA would be potentially useful in local medical unit for rapid MPXV infection identification.
To develop the F-RPA and VF-RPA, we first designed the primer and probe sequences for MPXV-RPA assay. MPXV genome sequences used for sequence alignment and primer/probe design (including both the West African clade and central African clade, as well as the emerging MPXV strains) were obtained from the NCBI websites. The downloaded MPXV strains and the gene accession numbers are shown in Supplementary Figure S1.
The sequences were aligned using Meg Align and a 399 nucleotide (nt) fragment of B7R (168791–169189, sequence ID: KP849470.1) was selected as target gene. This fragment was synthesized and cloned into plasmid pUNC57 via the KpnI and HindIII restriction sites (Sango Bio Inc., Shanghai, China). The recombinant plasmid pUNC57-B7R was used as the template for the MPXV-RPA. The RPA primers and probe were also designed and synthesized by Sango Bio Inc. (Shanghai, China). The primer and probe sequences are listed in Supplementary Table S1 and partial alignments of oligonucleotide binding regions are in Supplementary Figure S1. The sequence of the candidate primer and probe are identical to that of MPXV Clade I and IIa, and have only a single base mutation when compared with the strain circulating in 2022 (MPXV Clade IIb), this mutation occurs immediately after the THF cleavage site in the probe. In theory, the single base mutation will not affect target gene detection.
To screen the primer sets, two forward primers, two reverse primers and one probe were designed; thus, four candidate primer sets could be used to amplify the serially diluted plasmid containing target genes (102–106 copies/μL) to screen the most sensitive primer sets for MPXV-RPA assay. The F-RPA reaction was performed to screen the primer sets screening. The protocol is as follows: 45.5 μL of a master mixture comprising 2 .1 μL of forward prime B7R-F1 (10 μmol/L), 2.1 μl of reverse primer B7R-R1, 0.6 μL of B7R-Probe (10 μmol/L), and 40.7 μL of rehydration buffer A (Zhongce Ltd., Hangzhou, China) was placed in a 0.2 mL reaction tube containing a dried enzyme pellet (Zhongce Ltd., Hangzhou, China). The tube was vortexed to dissolve the dried enzyme pellet. Next, 2 μL of DNA template was added, and 2.5 μL of magnesium acetate (MgAc; 280 mmol/L) was pipetted onto the inside of tube lids, the total reaction volume was 50 μL. Next, the lids were closed, and the RPA tubes were centrifuged using a mini-spin centrifuge to deposit the MgAc; the tubes were immediately placed into an isothermal fluorescence scanner (Qitian, China) at 42 °C for 20 min since the reaction will proceed immediately after the addition of magnesium acetate. The real-time fluorescence signals were measured over time. The fluorescence signal threshold was set as three standard deviations above the background signal determined within one minute; the slope (mV/time) was calculated by F1620 software. Signals were interpreted using combined threshold and signal slope analysis. A positive reading was set as a slope greater than 15 mV/min, with a fluorescence signal > 500.
The results showed that among the four primer sets (F1R1, F1R2, F2R1 and F2R2), the limit of detection for F2R1 and F2R2 was about 104 copies/μL, which was significantly higher than that of F1R1 and F1R2 (102 copies/μL), therefore, the latter pairs were discarded first. The LODs of both F1R1 and F1R2 reached 100 copies/μL (Fig. 1 A). To select the most sensitive primer set, further dilutions (100, 50, 25, 12.5, 6.25 copies/μL) of the template were tested using F1R1 and F1R2, the F-RPA results showed that F1R1 detected a minimum concentration of 12.5 copies/μL, while F1R2 detected only 50 copies/μL, indicating that F1R1 has higher sensitivity (Fig. 1A). Therefore, we selected F1R1 as the primer pair for monkeypox detection.Fig. 1 Establishment and evaluation of the fluorescence-RPA (F-RPA) for monkeypox virus (MPXV). A Sensitivity of the four candidate primer sets. A-1, Sensitivity of the F1R1 primer set; A-2, Sensitivity of the F1R2 primer set; left: the concentration of the template ranging from10–106 copies/μL, right: the concentration of the template ranging from 100–6.25 copies/μL; A-3, Sensitivity of the F2R1 primer set; A-4, Sensitivity of the F2R2 primer set. Primer set F1R1 showed the strongest signals and highest sensitivity. A positive reading was set as a slope (mV/time) greater than 15 mV/min, with the fluorescence signal > 500. The dotted/dashed lines represent the cutoff value the fluorescence signal. B Limit of detection (LOD) of the F-RPA assay for MPXV Clade IIa. C Limit of detection (LOD) of the F-RPA assay for MPXV Clade IIb. The inner line is the probit curve (dose-response rule). The outer dotted/dashed lines are the 95% confidence intervals. D Specificity of the F-RPA. DNA/RNA samples from fever- or rash-associated inactivated virus cultures were tested. Samples included vaccinia virus (VACV), cowpox virus (CPV), Ectromelia virus (ECTV), measles virus (MV), rubella virus (RV), vesicular stomatitis virus (VSV), human herpes virus 2 (HSV-2), adenovirus 55 subtype (ADV55), yellow fever virus (YF17D), Japanese encephalitis virus (JEV), west Nile virus (WNV), H1N1 influenza virus (H1N1), reovirus (REOV), and smallpox virus (SPV). No positive signal appeared in any of these samples. POS, positive sample; NEG, negative sample. E F-RPA test conducted using ten MPXV-positive samples.
Fig. 1
First, we used primer set F1R1 to develop the MPXV F-RPA assay. The limit of detection (LOD) determination experiment was conducted as follows: seven different concentrations (1–106 copies/μL) of synthesis plasmid pUNC57-B7R (containing the target gene) were used as templates. All DNAs were tested in duplicate and in three independent runs. The results revealed that F-RPA generated positive results in test samples containing 100 copies/μL or more; not all samples containing 10 copies/μL produced positive results. Next, the template DNAs were serially diluted 2-fold (100–6.25 copies/μL) to yield five different concentrations. These were applied to the F-RPA in three individual replicate reactions (eight replicates for each sample at each experiment). The positive percentage was calculated for each dilution, and the LOD with 95% confidence interval was calculated with nonlinear probit regression analysis (SPSS 2.0 software). Based on the probit regression average, the LOD for MPXV F-RPA was calculated as 15.32 copies/μL, with a 95% CI of 11.42–26.60 copies/μL (Fig. 1B).
To test the utility of the RPA for MPXV Clade IIb, we synthesized the corresponding B7R gene fragment in the MPXV/Germany/2022/ON/RKI169 strain (sequence ID: OP0188588.1; nt166236-166634), the LOD reached 20.62 copies/μL, which was comparable with that for the previous MPXV virus gene, indicating that this RPA system could not only detect the previously prevalent monkeypox virus, but can also be used to detect the current monkeypox virus strain (Fig.1C).
The specificity of the F-RPA assay was determined using a representative panel of inactivated viruses, including vaccinia virus (VACV), cowpox virus (CPV), Ectromelia virus (ECTV), measles virus (MV), rubella virus (RV), vesicular stomatitis virus (VSV), human herpes virus 2 (HSV-2), adenovirus 55 subtype (ADV55), yellow fever virus (YF17D), Japanese encephalitis virus (JEV), west Nile virus (WNV), H1N1 influenza virus (H1N1), reovirus (REOV), and smallpox virus (SPV) synthetic plasmid, all of the concentration of virus were higher than 106 pfu/mL. No positive signals were observed in any of these samples (Fig.1D).
Next, we combined the RPA reaction with a disposable sealed nucleic acid detection cassette to generate the VF-RPA. The schematic diagram of VF-RPA is shown in Fig. 2 A. To carry out the VF-RPA protocol, the VF-RPA reaction was performed first; the protocol is similar to the F-RPA, but the method of labeling the reverse primer and probe is different. The concentration of the primers and the reaction time were optimized to achieve higher sensitivity and specificity. The optimized VF-RPA was conducted as following: The final volume (50 μL) comprised 1.4 μL of forward primer B7R-F1 (10 μmol/L), 1.4 μL of biotin labelled reverse primer B7R-R1-VF, 0.4 μL of FAM labelled B7R-Probe-VF (10 μmol/L) (see Supplementary Table S1 for sequences and labeling description), 2 μL of template DNA, 2.5 μL of magnesium acetate and 42.3 μL of rehydration buffer. After incubating at 42 °C for 15 min, the positive VF-RPA reaction generated a dual-labeled hybrid fragment: a FAM-labeled B7R -probe -VF and a biotin-labeled amplicon. Next, the tubes were placed in a type 3 disposable nucleic acid detection cassette (USTAR BioTech Corporation, Hangzhou, China), at room temperature for 5–10 min for vertical flow strip detection. Pushing the handle cuts the tube, allowing the RPA product to flow into the strip. In the cassette, a strip with a test (T) line (which captures the dual-labeled probe-amplicon) and a control (C) line (which captures gold-labeled rabbit anti-FAM antibody). A positive result occurs when both the T line and the C line were visible through the detection window. A negative read occurs when only the C line is displayed. The assay is regarded as invalid if neither the T line nor the C line is displayed.Fig. 2 Evaluation of the vertical flow strip RPA (VF-RPA) for monkeypox virus (MPXV).
A Schematic illustration of the VF-RPA. A-a, RPA reaction; A-b, Hybridization of the VF-probe with the biotin-labeled amplicons; A-c, Color development in the type 3 disposable nucleic acid detection cassette. B VF-RPA results for DNA samples (diluted 10-fold; 100–106 copies/μL). C Limit of detection (LOD) of the VF-RPA assay for MPXV. The inner line is the probit curve (dose-response rule). The outer dotted/dashed lines are the 95% confidence intervals. D Specificity of the VF-RPA. DNA/RNA samples from fever- or rash-associated inactivated virus cultures were tested. Viruses included vaccinia virus (VACV), cowpox virus (CPV), Ectromelia virus (ECTV), measles virus (MV), rubella virus (RV), vesicular stomatitis virus (VSV), human herpes virus 2 (HSV-2), adenovirus 55 subtype (ADV55), yellow fever virus (YF17D), Japanese encephalitis virus (JEV), west Nile virus (WNV), H1N1 influenza virus (H1N1), reovirus (REOV), and smallpox virus (SPV). No positive result appeared in any of these samples. E VF-RPA test conducted using ten MPXV-positive samples.
Fig. 2
The LOD of the VF-RPA was determined using serial dilutions of the synthetic plasmid as templates, the method is similar with the F-RPA, based on the nonlinear probit regression analysis by SPSS 2.0 software, the LOD of the MPXV VF-RPA was calculated as 8.53 copies/μL (95% CI: 6.69–13.85 copies/μL) (Fig. 2B and 2C), which is a little lower than that of the F-RPA (15.32 copies/μL), meaning that VF-RPA is more sensitive. Actually, the sequences of the two sets of primers and probes are the same, which means that the amplification efficiency of the target gene is similar for the F-RPA and VF-RPA. The main differences between F-RPA and VF-RPA are the way in which the primers and probes are labelled, and the way in which the results are presented. For the F-RPA, the probe was labelled with FAM, and the results were based mainly on fluorescence intensity. In the VF-RPA, the probe was labelled with FAM, and the reverse primer was labelled with biotin. Therefore, the amplicon was captured by the specific binding to avidin on the test strip, and colloidal gold aggregation and colour development were performed simultaneously by the specific binding of FAM to anti-FAM (conjugated with colloidal gold); thus, the signalling cascade in VF-RPA is stronger than in the F-RPA, resulting in higher sensitivity. Since the signal in the VF-RPA is amplified, the risk of a false positive via nonspecific amplification is increased. Therefore, to reduce the nonspecific signal, we shortened the amplification reaction to 15 min. The specificity of the VF-RPA assay was also determined as for the F-RPA; no positive signals were observed, (Fig.2D), meaning that both the F-RPA and VF-RPA are specific for MPXV.
To date, there are no clinical samples of monkeypox infection or MPXV strains in China up to now, which led to great difficulty with respect to clinical validation of the F-RPA/VF-RPA. Fortunately, in 2018 and 2019, we participated in the international viral External Quality Assurance Exercise (EQAEs) for Orthomyxovirus samples identification test, which was sponsored by the Robert Koch Institute, Germany; therefore, we used EQAE samples for F-RPA/VF-RPA validation. These included ten inactivated MPXV samples (four MPXV Clade IIa and six Clade I), two vaccinia virus samples, four cowpox virus samples, and five negative matrix samples (a total of 21 samples). The CT value of the ten MPXV samples in the qPCR assay varied from 28.59 to 37.02 (The qPCR protocol was in supplemental material and the CT value for the ten MPXV samples was demonstrated in Supplementary Table S2), indicating that the concentration of MPXV gene varied from high to low. Although the species of the tested samples were revealed later by the sponsor, the actual concentration of the ten MPXV was unknown. About 22 measles samples from Hebei Children’s Hospital and 23 common fever samples from Chinese PLA General Hospital, were used as negative samples; in total, we evaluated 10 positive MPXV samples and 56 negative samples. Both of the F-RPA and VF-RPA provided accurate result; the Kappa value reached 1.00 by consistency test with qPCR using SPSS 2.0 software, (Fig. 1E, Fig. 2E and Table 1 ), demonstrating that both the F-RPA and VF-RPA are applicable for detection of MPXV nucleic acid.Table 1 Result of the F-RPA/VF-RPA Validation.
Table 1 F-RPA P-RPA qPCR
Sample (no.) Negative Positive Negative Positive Negative Positive
MPXV (10) 0 10 0 10 0 10
CPV (4) 4 0 4 0 4 0
VCV (2) 2 0 2 0 2 0
Matrix (5) 5 0 5 0 5 0
MV (22) 22 0 22 0 22 0
Fever (23) 23 0 23 0 23 0
Total 56 10 56 10 56 10
MPXV: monkeypox virus samples; CPV: cowpox virus samples; VCV: vaccinia virus samples; MV: swab samples from measles virus infected patients; Fever: throad swab samples from common fever patients. The consistency of F-RPA/VF-RPA with qPCR results was analyzed by SPSS 2.0 software, the Kappa value was 1. 00.
In nucleic acid detection methods, target gene selection greatly affects the sensitivity and specificity. The abundance of different target genes varies between samples, which affects sensitivity, whereas the similarity of the target gene sequence to that of other species affects the specificity. With respect to primers and probe, formation of dimers and hairpins can have a marked effect on amplification efficiency, again affecting the sensitivity and specificity of an assay. In 2019, Davi et al. used G12R gene as the detection target and established a fluorescence-RPA assay for monkeypox virus; the LOD of the assay was 12 copies/μL (Davi et al., 2019). In this study, we used B7R gene as target and the sensitivity of F-RPA/VF-RPA established herein is comparable with that, suggesting that both G12R and B7R are suitable targets for MPXV RPA assays. The genome of monkeypox virus is as long as 200 kb, therefore, there will be other suitable target genes and primers (in addition to G12R and B7R) for RPA, resulting in more options for rapid detection of MPXV.
The VF-RPA/F-RPA described herein requires prior extraction of DNA, which may limit its use for in-field MPXV detection. In a previous study, we developed an integrated RPA assay for SARS-CoV-2, which can amplify the target gene directly from a sample without nucleic acid extraction; this method should be applicable to other viruses such as MPXV. However, because no MPXV strain is available in China, we cannot optimize and validate the integrated RPA for MPXV detection (Tang et al. 2022). Once we obtain MPXV, we will develop an integrated RPA assay, thereby providing a more convenient and powerful detection method for prevention and control of monkeypox infection.
Footntes
This work was funded by the National Key Research and Development Plan of China (2021YFC2300200-02) and the State Key Laboratory of Pathogen and Biosecurity (Academy of Military Medical Science, SKLPBS2111). The authors declare that they have no conflict of interest. The informed consent have been obtained from all participants and the studies have been approved by the ethics committee of the Academy of Military Medical Science.
All the data generated during the current study are included in the manuscript.Supplementary data to this article can be found online at https://doi.org/10.1016/j.virs.####
Uncited References
The International Committee on Taxonomy of Viruses, 2022; World Health Organization, 2022a; World Health Organization, 2022b.
Appendix A Supplementary data
The following are the Supplementary data to this article:
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.virs.2022.12.001.
==== Refs
References
The International Committee on Taxonomy of Viruses (ICTV), 2022. Virus Taxonomy: 2020 Release. https://talk.ictvonline.org/taxonomy
Bunge E.M. Hoet B. Chen L. Lienert F. Weidenthaler H. Baer L.R. Steffen R. The changing epidemiology of human monkeypox-A potential threat? A systematic review PLoS Negl Trop Dis 16 2022 e0010141
Mauldin M.R. McCollum A.M. Nakazawa Y.J. Mandra A. Whitehouse E.R. Davidson W. Zhao H. Gao J. Li Y. Doty J. Yinka-Ogunleye A. Akinpelu A. Aruna O. Naidoo D. Lewandowski K. Afrough B. Graham V. Aarons E. Hewson R. Vipond R. Dunning J. Chand M. Brown C. Cohen-Gihon I. Erez N. Shifman O. Israeli O. Sharon M. Schwartz E. Beth-Din A. Zvi A. Mak T.M. Ng Y.K. Cui L. Lin R.T.P. Olson V.A. Brooks T. Paran N. Ihekweazu C. Reynolds M.G. Exportation of Monkeypox Virus From the African Continent J Infect Dis 225 2022 1367 1376 32880628
Adler H. Gould S. Hine P. Snell L.B. Wong W. Houlihan C.F. Osborne J.C. Rampling T. Beadsworth M.B. Duncan C.J. Dunning J. Fletcher T.E. Hunter E.R. Jacobs M. Khoo S.H. Newsholme W. Porter D. Porter R.J. Ratcliffe L. Schmid M.L. Semple M.G. Tunbridge A.J. Wingfield T. Price N.M. NHS England High Consequence Infectious Diseases (Airborne) Network. Clinical features and management of human monkeypox: a retrospective observational study in the UK Lancet Infect. Dis. 22 2022 1153 1162 35623380
Miura F. van Ewijk C.E. Backer J.A. Xiridou M. Franz E. Op de Coul E. Brandwagt D. van Cleef B. van Rijckevorsel G. Swaan C. van den Hof S. Wallinga J. Estimated incubation period for monkeypox cases confirmed in the Netherlands, May 2022 Euro surveilL 27 2022 2200448
World Health Organization (WHO), 2022a. Multi-country monkeypox outbreak: situation update. https://www.who.int/emergencies/disease-outbreak-news/item/2022-DON393
World Health Organization (WHO), 2022b. Surveillance, case investigation and contact tracing for monkeypox Interim guidance, 24 June 2022. https://apps.who.int/iris/handle/10665/357186
Erez N. Achdout H. Milrot E. Schwartz Y. Wiener-Well Y. Paran N. Politi B. Tamir H. Israely T. Weiss S. Beth-Din A. Shifman O. Israeli O. Yitzhaki S. Shapira S.C. Melamed S. Schwartz E. Diagnosis of Imported Monkeypox, Israel, 2018 Emerg Infect Dis 25 2019 980 983 30848724
Thornhill J.P. Barkati S. Walmsley S. Rockstroh J. Antinori A. Harrison L.B. Palich R. Nori A. Reeves I. Habibi M.S. Apea V. Boesecke C. Vandekerckhove L. Yakubovsky M. Sendagorta E. Blanco J.L. Florence E. Moschese D. Maltez F.M. Goorhuis A. Pourcher V. Migaud P. Noe S. Pintado C. Maggi F. Hansen A.E. Hoffmann C. Lezama J.I. Mussini C. Cattelan A. Makofane K. Tan D. Nozza S. Nemeth J. Klein M.B. Orkin C.M. SHARE-net Clinical Group Monkeypox Virus Infection in Humans across 16 Countries - April-June 2022 N Engl J Med 387 2022 679 691 35866746
Stern D. Olson V.A. Smith S.K. Pietraszczyk M. Miller L. Miethe P. Dorner B.G. Nitsche A. Rapid and sensitive point-of-care detection of Orthopoxviruses by ABICAP immunofiltration Virol J 13 2016 207 27938377
Jia J. Li Y. Wu X. Zhang S. Hu Y. Li J. Jiang T. Kang X. Reverse Transcription Recombinase Polymerase Amplification Assays for Rapid Detection of Tick-Borne Encephalitis Virus Infection Virol Sin 34 2019 338 341 30941699
Wang S. Li Y. Zhang F. Jiang N. Zhuang Q. Hou G. Jiang L. Yu J. Yu X. Liu H. Zhao C. Yuan L. Huang B. Wang K. Reverse transcription recombinase-aided amplification assay for H5 subtype avian influenza virus Virol J 19 2022 129 35907986
Davi S.D. Kissenkötter J. Faye M. Böhlken-Fascher S. Stahl-Hennig C. Faye O. Faye O. Sall A.A. Weidmann M. Ademowo O.G. Hufert F.T. Czerny C.P. Abd El Wahed A. Recombinase polymerase amplification assay for rapid detection of Monkeypox virus Diagn Microbiol Infect Dis 95 2019 41 45 31126795
Tang Y. Wang Y. Li Y. Zhao H. Zhang S. Zhang Y. Li J. Chen Y. Wu X. Qin C. Jiang T. Kang X. An integrated rapid nucleic acid detection assay based on recombinant polymerase amplification for SARS-CoV-2 Virol Sin 37 2022 138 141 35234627
| 36494078 | PMC9724568 | NO-CC CODE | 2022-12-08 23:18:15 | no | Virol Sin. 2022 Dec 6; doi: 10.1016/j.virs.2022.12.001 | utf-8 | Virol Sin | 2,022 | 10.1016/j.virs.2022.12.001 | oa_other |
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00340-9
10.1016/j.jiph.2022.11.033
Original Article
Designing multi-epitope monkeypox virus-specific vaccine using immunoinformatics approach
Zaib Sumera a⁎
Rana Nehal a
Areeba a
Hussain Nadia ef
Alrbyawi Hamad c
Dera Ayed A. bd
Khan Imtiaz b⁎
Khalid Mohammad g
Khan Ajmal h
Al-Harrasi Ahmed h⁎
a Department of Basic and Applied Chemistry, Faculty of Science and Technology, University of Central Punjab, Lahore 54590, Pakistan
b Manchester Institute of Biotechnology, The University of Manchester, 131 Princess Street, Manchester M1 7DN, United Kingdom
c Pharmaceutics and Pharmaceutical Technology Department, College of Pharmacy, Taibah University, Medina 42353, Saudi Arabia
d Department of Clinical Laboratory Sciences, College of Applied Medical Sciences, King Khalid University, Abha, Saudi Arabia
e Department of Pharmaceutical Sciences, College of Pharmacy, Al Ain University, Al Ain, UAE
f AAU Health and Biomedical Research Center, Al Ain University, Abu Dhabi, UAE
g Department of Pharmaceutics, College of Pharmacy, King Khalid University, Asir-Abha 61421, Saudi Arabia
h Natural and Medical Sciences Research Center, University of Nizwa, Nizwa 616, Oman
⁎ Corresponding authors.
6 12 2022
1 2023
6 12 2022
16 1 107116
19 9 2022
15 11 2022
30 11 2022
© 2022 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
Background
Monkeypox virus is an enveloped DNA virus that belongs to Poxviridae family. The virus is transmitted from rodents to primates via infected body fluids, skin lesions, and respiratory droplets. After being infected with virus, the patients experience fever, myalgia, maculopapular rash, and fluid-filled blisters. It is necessary to differentiate monkeypox virus from other poxviruses during diagnosis which can be appropriately envisioned via DNA analysis from swab samples. During small outbreaks, the virus is treated with therapies administered in other orthopoxviruses infections and does not have its own specific therapy and vaccine. Consequently, in this article, two potential peptides have been designed.
Methods
For the purpose of designing a vaccine, protein sequences were retrieved followed by the prediction of B- and T-cell epitopes. Afterward, vaccine structures were predicted which were docked with toll-like receptors. The docked complexes were analyzed with iMODS. Moreover, vaccine constructs nucleotide sequences were optimized and expressed in silico.
Results
COP-B7R vaccine construct (V1) has antigenicity score of 0.5400, instability index of 29.33, z-score of − 2.11-, and 42.11% GC content whereas COP-A44L vaccine construct (V2) has an antigenicity score of 0.7784, instability index of 23.33, z-score of − 0.61, and 48.63% GC content. It was also observed that COP-A44L can be expressed as a soluble protein in Escherichia coli as compared to COP-B7R which requires a different expression system.
Conclusion
The obtained results revealed that both vaccine constructs show satisfactory outcomes after in silico investigation and have significant potential to prevent the monkeypox virus. However, COP-A44L gave better results.
Graphical Abstract
ga1
Abbreviations
MPXV, Monkeypox virus
IID, Initial Intrusion Duration
CDC, Centers for Disease Control and Prevention
VIG, Vaccinia Immunoglobulins
IEDB, Immune Epitope Database
TLR3, Toll-like receptor 3
NMA, Normal Mode Analysis
CAI, Codon Adaptation Index
II, Instability Index
Keywords
Monkeypox virus
Vaccine
Codon optimization
Epitope prediction
Molecular docking
Expression analysis
==== Body
pmc1 Introduction
The monkeypox is a zoonotic infection caused by a brick-shaped enveloped monkeypox virus (MPXV) that belongs to the family of ancient viruses, namely Poxviridae, characterized by a linear double-stranded DNA genome [1], [2]. It was first reported in 1959 as a pox-like disease in monkey colonies that were confined for study in Denmark. Subsequently, several outbreaks were also reported in countries of Central and West Africa and even in the United States (U.S.) with mortality rates ranging from 1% to 10% [3], [4]. Moreover, several cases have also been reported across the borders of Africa in Singapore, and South Korea, together with the most recently reported in Taiwan (June 24, 2022) [5]. The virus is primarily transmitted from wild animals (rodents and primates) to humans; however, human-to-human transmission is frequent. The major routes of transmission of MPXV among humans involve contact with contaminated items, infected body fluids, skin lesions on body of patients, and respiratory droplets [1], [3]. Furthermore, sexual route of transmission among bisexuals has also been revealed in the recent outbreak [6].
The transmission of MPXV is followed by an entry into cells which depends on its ability to evade antiviral immune responses and on the presence of ten viral accessory genes [7]. These viral accessory genes are also known as host range genes, and determine divergence in the host range and impede various aspects of cellular innate responses [8], [9]. The induction of type I interferon-mediated antiviral signaling is suppressed by viral protein B16. Moreover, Tumor Necrosis Factor alpha (TNF-α) and interferon-stimulated genes remain silent and are not expressed during infection with MPXV [10]. Inside the host, MPXV undergoes an incubation period of 5–21 days among which the first five days depict initial intrusion duration (IID). During IID, the patient experiences fever, lymph node inflammation, myalgia, severe headache, asthenia, and backache as the main symptoms. After 1–3 days of fever, the patient also experiences maculopapular rashes which subsequently develop into pus-containing fluid-filled blisters and burst in ten subsequent days [11].
Analysis of viral DNA extracted from swab samples obtained from the crust of vesicles represents the most suitable diagnosis procedure for the identification of monkeypox infection [12]. To date, no particular treatment for MPXV is available according to the Centers for Disease Control and Prevention (CDC). However, patients are being treated with other orthopoxvirus therapies such as cidofovir, brincidofovir and tecovirimat [6], [13]. Moreover, until now, small outbreaks are controlled by administering vaccinia immunoglobulins (VIG), and smallpox vaccines such as jynneos and LC16 m8, although not specific for MPXV [6], [14].
Therefore, the fundamental aim of this research centers around the design of multi-epitope MPXV-specific vaccine using immunoinformatics approach that can help in the eradication of viral infection. Two virulent genes, namely COP-A44L and COP-B7R, have been selected for vaccine development. COP-B7R is absent in variola virus but present in MPXV; whereas, COP-A44L encodes a protein comprising 140 amino acids shorter in variola virus as compared to MPXV [15]. COP-A44L protein product (3-β-hydroxysteroid dehydrogenase) catalyzes the conversion of pregnenalone to androstendione and is mandatory for the formation of immunosuppressive steroid hormones [16]. Similarly, COP-B7R is endoplasmic reticulum residing viral protein whose virulence mechanism is unknown. However, it may affect apoptotic mechanisms or may be expressed on cell surface and is involved in immune responses [17].
2 Methodology
2.1 Retrieval of protein sequences
The genome of poxvirus is around 200 Kb in size and contains 200 proteins. It has a linear double-stranded DNA genome having covalently closed hairpin end and 10 Kb inverted terminal repeats at each end. The protein selected was COP-A44L, a full length protein encoding 140 amino acids while other protein was COP-B7R, a resistant protein in monkeypox virus. The whole genome sequence was retrieved from GenBank (NCBI) (https://www.ncbi.nlm.nih.gov/). In addition, the protein sequence was taken from GenBank. However, Expasy (https://www.expasy.org/) and PSIPRED tools (http://bioinf.cs.ucl.ac.uk/psipred/) were used to determine the physiochemical properties, and secondary structure [18]. The number of amino acid and cysteine residues were accessed through DiANNA 1.1 web server (http://clavius.bc.edu/∼clotelab/DiANNA/) [19]. The online server VaxiJen v2.0 (http://www.ddg-pharmfac.net/vaxijen/VaxiJen/VaxiJen.html) was used to retrieve antigenicity score whereas number of TM helices was determined using TMHMM v2.0 server (https://services.healthtech.dtu.dk/service.php?TMHMM-2.0) [20], [21]. The prediction of allergenicity was determined through AllerTOP v. 2.0 (https://ddgpharmfac.net/AllergenFP/) [22].
2.2 B- and T-cell epitope prediction
For B- and T-cell epitope prediction of COP-A44L and COP-B7R, the immune epitope database (IEDB) (https://www.iedb.org/) analysis resource was used [23], [24], [25]. For this purpose, protein sequence was retrieved from the NCBI (https://www.ncbi.nlm.nih.gov/) and their epitopes were predicted separately. Subsequently, the same epitopes with IC50 value ≤ 100 based on their antigenicity and allergenicity were extracted and analyzed for population coverage using Population Coverage – IEDB Analysis Resource [26].
2.3 Prediction of 2D and 3D structures
A 2D structure represents covalent bonds in the molecule and PSIPRED tools (http://bioinf.cs.ucl.ac.uk/psipred/) were used for analysis [18]. However, 3D structures are three-dimensional coordinates of a molecule determined by an appropriate approach. The 3D structures for COP-A44L and COP-B7R were determined through SCRATCH Protein Predictor (https://scratch.proteomics.ics.uci.edu/) and I-TASSER (https://zhanggroup.org/I-TASSER/), respectively [27], [28]. Furthermore, refinement of the best model was achieved through Galaxy web server (https://galaxy.seoklab.org/) which showed the five models with Ramachandran plot, MolProbity, clash score RMSD and GDT-HA [29], [30]. Furthermore, the Ramachandran plot for the first model was predicted via RAMPAGE (https://zlab.umassmed.edu/bu/rama/) [31].
2.4 Molecular docking of designed chimeric protein with toll-like receptor
The interaction between antigenic molecule and specific immune receptor leads to the immune response. Toll-like receptor 3 (TLR3) was used as a receptor for determining the interaction of refined protein through a ClusPro server (https://cluspro.bu.edu/) [32]. This server is used to find the native sites within the protein and helps in protein-protein docking by providing various results. Molecular docking of multi-epitope vaccine peptide with TLR3 receptor was determined using ClusPro server to obtain the best docked model. Later, iMODS server (https://imods.iqfr.csic.es/) was used to perform the normal mode analysis (NMA) in internal coordinates of nucleic acid and protein atomic structure [33].
2.5 In silico cloning optimization of designed vaccine candidate
To perform reverse translation and codon optimization, EMBOSS Backtranseq (https://www.ebi.ac.uk/Tools/st/emboss_backtranseq/), and NovoPro (https://www.novoprolabs.com/tools/codon-optimization) were used. The acquired results showed output of codon adaptation index (CAI) and percentage of GC content, used to access protein expression level [34]. CAI helps to provide codon information and> 0.8 is considered as a good score, while GC content should be in the range of 30–70%. Furthermore, restriction enzyme cloning was performed via SnapGene that ensures the expression of vaccine construct. Vectors pET-28c (+) and pET-21a (+) were used to clone optimized gene sequences of final vaccine constructs [35]. A brief overview of the implemented methodology is shown in Fig. 1.Fig. 1 Diagrammatic illustration of envisioned methodology. All steps and software used for vaccine construction are shown in the flow chart.
Fig. 1
3 Results and discussion
3.1 Analysis of protein sequence
Two protein sequences of COP-A44L and COP-B7R were retrieved from the NCBI and were used for the preparation of multi-epitope vaccine against the MPXV. Expasy-ProtParam (https://www.expasy.org/) was used for the physicochemical analysis. The physicochemical analysis of COP-B7R suggested the presence of 182 amino acids in the structure with molecular weight of 21,390.08 Da and theoretical pI of 5.65. Additionally, the half-life of the protein was 30 h (mammalian reticulocytes, in vitro),> 20 h (yeast, in vivo) and> 10 h (Escherichia coli, in vivo), whereas, for COP-A44L, whose molecular weight was found to be 39,338.36 Da with a sequence length of 346 amino acids, theoretical pI of 7.06, and stability index of 34.74 (stable protein). Furthermore, the estimated half-life was 30 h,> 20 h and> 10 h in mammalian reticulocytes (in vitro), yeast (in vivo) and E. coli (in vivo), respectively.
VaxiJen v2.0 online server (http://www.ddgpharmfac.net/vaxijen/VaxiJen/VaxiJen.html) showed that both the protein sequence were probable ANTIGEN (COP-B7R: 0.4395; COP-A44L: 0.4016), while, AllerTOP v. 2.0 (https://www.ddg-pharmfac.net/AllerTOP/) rendered both as non-allergic. Later, the functional sequence of the protein was subjected to linear B- and T-cell epitope prediction.
3.2 B- and T-cell epitope prediction analysis
Immune Epitope Database (IEDB) Analysis Resource (https://www.iedb.org/) is a repository website of computational tools used for analysis and prediction of B- and T-cell epitopes. B-cell epitopes can be linear or non-linear (discontinuous). The prediction of linear B-cell epitopes is more advanced and practical as compared to discontinuous B-cell epitopes. It is because linear B-cell epitopes are derived from the sequence of proteins and consist of sequential residues which instantly displace antigens for the formation of antibodies. Contrarily, discontinuous B-cell epitopes are derived from various patches of input protein in a non-sequential method and require three-dimensional structure of the protein. Additionally, the production of a selective antibody requires a suitable scaffold for grafting epitopes, which is feasible when linear B-cell epitopes are utilized [36]. Therefore, BepiPred-2.0 which is a web server for predicting linear B-cell epitopes from antigen sequences, is used in this research. BepiPred-2.0 is based on a random forest algorithm trained on epitope data derived from Protein Data Bank (PDB) and shows improved prediction of linear B-cell epitopes [37]. Subsequent to the retrieval of peptide sequences from the IEDB (http://tools.iedb.org/bcell/), epitopes having antigenicity score> 0.5 for COP-B7R and> 0.7 for COP-A44L were selected and subjected to vaccine construction. Additionally, these epitopes were also accessed by Population Coverage – IEDB Analysis Resource (http://tools.iedb.org/population/) to analyze percentage of the world population and was predicted to be 68.44% (MHC-I) and 50.93% (MHC-II) for COP-A44L.
3.3 Construction of multi-epitope subunit vaccine
For COP-B7R, the total number of predicted epitopes used to design chimera was 9 B-cell epitopes, 19 MHC-I epitopes, and 14 MHC-II epitopes while vaccine construct for COP-A44L constitutes 14 B-cell epitopes, 18 MHC-I epitopes and 8 MHC-II epitopes as shown in Fig. 2. These predicted epitopes were merged to form a continuous sequence via specific linkers. Both B- and T-cells were joined through a linker GPGPG and AAV. The adjuvant used was the TLR3 (PDB ID: 1ZIW) agonist, 50 S ribosomal L7/L12 (Locus RL7_MYCTU) with accession number P9WHE3 and added to amino terminus of vaccine peptide through EAAAK linker for specific immune response. Lastly, a 6xHis-tag was added to the C-terminal for protein purification and identification. Peptides with 494 amino acids (COP-B7R) and 366 amino acids (COP-A44L) were generated at the end. The COP-B7R vaccine construct will be denoted as V1 while vaccine construct of COP-A44L will be denoted as V2 throughout the article.Fig. 2 Vaccine constructs. A represents the V1 while V2 structure is depicted in B. The vaccine construct consists of B-cell epitopes, HTL and CTL epitopes, linkers and adjuvant.
Fig. 2
3.4 Prediction of the antigenicity and allergenicity of the vaccine candidate
The allergenicity and antigenicity of final sequence of COP-B7R vaccine construct (V1) (with adjuvant sequence) was predicted by VaxiJen v2.0 and AllerTOP v. 2.0 that showed 0.5400 (antigenicity) and non-allergen. The results showed that the generated sequence was antigenic and non-allergic in nature. The COP-A44L vaccine construct (V2) was likely to be non-allergic with an antigenic score of 0.7784. The predicted antigenic scores are higher than that of the reported vaccine constructs (in silico) which were 0.44–0.47 [38] and 0.5311 [39].
3.5 Physicochemical properties and solubility prediction
The predicted molecular weight of the final protein of V1 was 57,803.24 Dalton with theoretical isoelectric point value of 6.06. Additionally, the protein half-life was 1 h (mammalian reticulocytes, in vitro), 30 min (yeast, in vivo) and> 10 h (Escherichia coli, in vivo). However, solubility of V1 was determined through SOLUPROT (https://loschmidt.chemi.muni.cz/soluprot/) which showed that V1 cannot be expressed in E. coli. The instability index was 29.33 which classified the protein as stable. Furthermore, the predicted aliphatic index was 79.90 with a grand average of hydropathicity of − 0.416. The corresponding negative sign indicates the hydrophilic nature of protein and ability to interact with other water molecules.
After physicochemical evaluation, the molecular weight of V2 was predicted to be 41,377.02 Dalton having 366 residues. Furthermore, the solubility of V2 was predicted to be 0.522 by SOLUPROT, indicating soluble expression in E. coli. Moreover, its theoretical pI, instability index (II) and hydropathicity were 9.14, 23.23 and − 0.277, respectively.
3.6 2D and 3D structure analysis
3.6.1 Secondary structure prediction
According to PSIPRED results, V1 encompasses 3 helices in the secondary structure with a number of strands and coils. Similarly, V2 constitutes six helical conformations and numerous strands and coils in the predicted secondary structure. Moreover, MEMSAT analysis predicts the extracellular, transmembrane, and cytoplasmic domains of constructed vaccine peptide sequences. The V2 contains a pore lining domain from 173 to 188 amino acids while the N-terminal domain is designated as extracellular domain and the C-terminal domain as intracellular domain (Fig. S1).
3.6.2 Tertiary structure modeling
I-TASSER was used for the prediction of 3D structure of the V1 final proteins. The results showed 10 threading template with five best models of 1d2pA, 7w6bA, 3holA, 7w7iA and 5nxkA. All the 10 templates showed good alignment and z-score, ranging from 1.1 to 1.4. Among the five models, the one with high C-score was selected for refinement ( Fig. 3A) as the C score ranges from − 2.53 to − 4.76. The selected model had an estimated TM-score of 0.42 ± 0.14 and RMSD of 13.5 ± 4.0 Å.Fig. 3 Modeling and refinement of 3-dimensional (3D) structure of V1 and V2. (A) V1 3D model has been generated by homology modeling via I-TASSER. (3B) V2 3D model is generated by SCRATCH Protein Predictor as the amino acid sequence was less than 400 residues. (C) represents the refined model obtained by GalaxyRefine. (D) GalaxyRefine generated 3D model having Rama favored score of 96.4.
Fig. 3
SCRATCH Protein Predictor was used for the prediction of 3D structure of the V2. It was used because SCRATCH Protein Predictor only predicts the protein sequence less than 400 amino acids with high authenticity in less time (Fig. 3B).
3.6.3 Tertiary structure refinement
The term refinement is used for the vaccine model to refine 3D predicted structure of the I-TASSER and SCRATCH Protein Predictor. In this case, Galaxy web server was used to refine model 1 retrieved from I-TASSER and SCRATCH Protein Predictor. GalaxyRefine server showed the five models with Ramachandran plot, MolProbity, RMSD and GDT-HA. Model 4 was found to be the best among all models due to its parameters, including GDT-HA (0.9160), RMSD (0.516), and MolProbity (3.034). In addition, the clash score was 37.5 with poor rotamers score of 2.2, and the Ramachandran plot score of 80.9. This model was further analyzed for vaccine construction (Fig. 3C). Contrarily, model 1 obtained via GalaxyRefine having GDT-HA 0.8854, RMSD 0.561, MolProbity 1.854, clash score 12.8, poor rotamers 0.6 and 96.4 Rama favored score was selected for further predictions and analyses of V2 (Fig. 3D).
The Ramachandran plot for V1 showed highly preferable area as green crosses with a percentage of 90.6. However, preferred observations are shown as brown triangles (7.3%) and red spots show questionable observation (1.9%) ( Fig. 4A). Subsequently, the Ramachandran plot predicted for V2 showed 100% residues (328 amino acids) in highly preferred region which are represented as green crosses, with no residues in additional preferred and questionable regions (Fig. 4B). The refined models of V1 and V2 were further validated via ProSA-web (https://prosa.services.came.sbg.ac.at/prosa.php) which showed a z-score of − 2.11 and − 0.61, respectively (Fig. 4C and D). The higher the z-score, the higher will be the model quality.Fig. 4 Validation of V1 and V2. (A) Ramachandran plot of V1 for the refined model. The green crosses are the amino acid residues in most favored/allowed region, the brown triangles are the amino acid residues in favored region and the red dots are the residues in disallowed region. (B) Ramachandran plot of V2 shows all the amino acid residues in highly preferred region with green crosses (100%). There are no brown triangles (0.0%) as well as no questionable observations (red circles). (C) This graph is generated by ProSA-web for V1 showing the quality of 3D model on the basis of z-score. (D) This graph predicts the quality of 3D refined model of V2 generated via ProSA-web on the basis of z-score (z-score = −0.61).
Fig. 4
3.7 Molecular docking and dynamics simulation of subunit vaccine with immune receptor (TLR3)
The ClusPro was used for determining the protein binding and hydrophobic interaction sites on the proteins surface ( Fig. 5). iMODS server was used for the analysis of protein-protein docking and explains the deformability in the main chain and the deformed nature of each residue. According to the graphs as shown in Fig. 6A and B, the peaks are quite higher in 6 A depicting greater deformation in V1 as compared to V2 for which only few peaks are higher. Moreover, the eigenvalue is also predicted which is linked to the normal mode representing the stiffness of the model. It represents energy value needed to deform the structure. The lower eigenvalue causes the easy deformation. So according to the graphs in Fig. 6C and D, the eigenvalues of V1 and V2 are 7.088347e-06 and 8.277109e-07, respectively. These values interpret that the complex of V2 with TLR3 is easier to deform.Fig. 5 Molecular docking. (A) Docking complex of V1 with TLR3. (B) Docking complex of V2 (blue) with TLR3 (pink color). Here the red colored regions show the active site of TLR3 with which the vaccine construct interacts.
Fig. 5
Fig. 6 iMODS analysis of V1 and V2. (A) This graph represents the deformability potential of each residue in V1. Higher the peak, the higher will be the deformability. (B) This graph is visual representation of deformability potential of atoms in V2. In this graphical representation, most of the peaks have lower deformability. (C) This graph shows the eigenvalue for V1 which interprets the ease to deform the structure. Lower the eigenvalue, the higher the chance of deformation. (D) This graph for V2 interprets the susceptibility of a molecule to deform based on the eigenvalue. Here the eigenvalue is 8.277109e-07.
Fig. 6
The covariance matrix showed interaction between the different pair of residues explained in Fig. 7A and B. The motion of atoms in V1 are not very correlated because blue and white regions are more obvious.On the other hand, the covariance matrix for V2 contains more red region, less blue region and nearly no white region showing correlated motion of atoms. However, Fig. 7C and D explain the elastic network model representing the atom pairs connected by strings. Each dot in graph exhibits the one spring between the corresponding atom pairs. These dots represent the stiffness and darker gray shows the stiffer spring.Fig. 7 V1 and V2 analysis using iMODS. (A) This graph elucidates covariance for V1. As most of the region is red so it shows correlated motion of residues; whereas, blue region shows uncorrelated motion between residues. (B) For V2, most of the colored region in graph is red so most of the residues have correlated motion; others (blue) have uncorrelated motion, while fewer have anti-correlated motion. (C) This is the elastic model network for V1 which illustrates spring formation between corresponding pair of atoms. The gray color dots show spring formation; the thicker the gray color, stronger is the spring between corresponding pairs. In this graph, most of the atoms form stronger springs. (D) According to elastic model network for V2, the atoms up to 600 are showing more strong spring formation as evident from the stronger gray color. Other atoms are also forming spring but they are not as dense as the former.
Fig. 7
3.8 Optimization of the codon of final vaccine construct
For maximum protein expression, NovoPro was used to optimize the codon sequence generated by EMBOSS Backtranseq which converts the protein residues into a nucleotide sequence. The length of the optimized codon sequence for V1 was 1482 base pairs with a codon adaptation index (CAI) of 0.84 and average GC content of 42.11% that exhibited the good possibility of showing expression in the E. coli (strain K12) vector. Finally, adapted codon sequence was inserted into pET-21a (+) vector for designing a recombinant plasmid using SnapGene software.
For in silico restriction enzyme cloning of V2, the gene sequence generated by EMBOSS Backtranseq was optimized by NovoPro to be expressed in E. coli (strain K12). Consequently, the codon optimization index (CAI) was improved from 0.65 to 0.80 with 48.63% GC content in optimized sequence (1098 base pairs). Subsequently, the analysis for restriction fragments indicates PpuMI (at 18 bp) and ClaI (at 1040 bp) as the most favorable restriction sites. Resultantly, the optimized sequence was auspiciously ligated in pET-28c (+) plasmid for obtaining expression in E. coli (Figs. S2-S5).
After analysis of results and comparison between V1 and V2 as shown in Table 1, COP-A44L was found to possess more antigenic score and stable tertiary structure (predicted by higher z-score and Rama favored score).Table 1 A comparison between V1 and V2 based on their significant attributes.
Table 1Significant Attributes COP-B7R Vaccine Construct (V1) COP-A44L Vaccine Construct (V2)
Allergenicity Non-Allergic Non-Allergic
Antigenicity Score 0.5400 0.7784
Expression inEscherichia coli No Yes (as soluble protein)
Instability Index (II) 29.33 23.23
Hydropathicity -0.416 -0.277
Number of helical conformations in secondary structure 3 6
Rama favored score of tertiary structure 80.9 96.4
z-score -2.11 -0.61
Codon Optimization Index (CAI) 0.84 0.80
GC content 42.11% 48.63%
Expression Vector pET-21a (+) pET-28c (+)
4 Conclusion
Monkeypox virus outbreak have been reported number of times in Central and West African countries. Recently, these outbreaks have also occurred in countries outside Africa. In the present study, we have developed two novel vaccine constructs for providing specific immunity against the virus. The vaccine constructs were analyzed for physicochemical properties, allergenicity, antigenicity, molecular dynamic simulation and cloning via restriction enzymes. Both the vaccine constructs namely, V1 (COP-B7R) and V2 (COP-A44L), contained epitopes with high population coverage, linked with stable linkers and adjuvant, showed antigenic potential and were found as non-allergic. The construct V2 can be expressed in E. coli and shows higher z-score than V1. In a nutshell, V2 showed better results than V1 by applying bioinformatics approaches.
Future perspectives
Owing to the frequent outbreaks of MPXV, the development of preventive measures is the need of time. In this article, two potential vaccine constructs have been designed by in silico methods, however, both the vaccines have to be validated by in vivo research and clinical trials before its commercialization.
Ethics approval
Not applicable because there are no animals and human used in this study.
Funding
The project was supported by grant from the Oman Research Council (TRC) through the funded project (BFP/RGP/EBR/22/021).
Competing interests
There are no conflicts to declare.
Appendix A Supplementary material
Supplementary material
.
Supplementary material
.
Supplementary material
.
Acknowledgments
The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work through Large Groups under grant number (RGP.2/244/43).
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Authors’ contributions
The manuscript was written through contribution of all authors. All authors have given approval to the final version of the manuscript.
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jiph.2022.11.033.
==== Refs
References
1 Alakunle E. Moens U. Nchinda G. Okeke M.I. Monkeypox virus in Nigeria: infection biology, epidemiology, and evolution Viruses 12 11 2020 1257 33167496
2 Moore M.J. Rathish B. Zahra F. Monkeypox StatPearls [Internet] 2022 StatPearls Publishing, Treasure Island (FL)
3 Peter O.J. Kumar S. Kumari N. Oguntolu F.A. Oshinubi K. Musa R. Transmission dynamics of Monkeypox virus: a mathematical modelling approach Model Earth Syst Environ 8 3 2022 3423 3434 34667829
4 CDC, 2003, What you should know about monkeypox. 〈(https://www.cdc.gov/poxvirus/monkeypox/)〉.
5 Yang Z.S. Lin C.Y. Urbina A.N. Wang W.H. Assavalapsakul W. Tseng S.P. Lu P.L. Chen Y.H. Yu M.L. Wang S.F. The first monkeypox virus infection detected in Taiwan-the awareness and preparation Int J Infect Dis S1201–9712 22 2022 00445-3
6 Rizk J.G. Lippi G. Henry B.M. Forthal D.N. Rizk Y. Prevention and treatment of monkeypox Drugs 82 9 2022 957 963 35763248
7 Xiang Y. White A. Monkeypox virus emerges from the shadow of its more infamous cousin: family biology matters Emerg Microbes Infect 11 1 2022 1768 1777 35751396
8 Werden S.J. Rahman M.M. McFadden G. Poxvirus host range genes Adv Virus Res 71 2008 135 171 18585528
9 Oliveira G.P. Rodrigues R.A.L. Lima M.T. Drumond B.P. Abrahão J.S. Poxvirus host range genes and virus-host spectrum: a critical review Viruses 9 11 2017 331 29112165
10 Hutson C.L. Gallardo-Romero N. Carroll D.S. Clemmons C. Salzer J.S. Nagy T. Hughes C.M. Olson V.A. Karem K.L. Damon I.K. Transmissibility of the monkeypox virus clades via respiratory transmission: investigation using the prairie dog-monkeypox virus challenge system PLoS One 8 2 2013 e55488
11 Hutson C.L. Gallardo-Romero N. Carroll D.S. Clemmons C. Salzer J.S. Nagy T. Hughes C.M. Olson V.A. Karem K.L. Damon I.K. Transmissibility of the monkeypox virus clades via respiratory transmission: investigation using the prairie dog-monkeypox virus challenge system PLoS One 8 2 2013 e55488
12 Durski K.N. McCollum A.M. Nakazawa Y. Emergence of monkeypox in West Africa and Central Africa, 1970–2017 Wkly Epidemiol Rec 93 11 2018 125 132 29546750
13 Grothe J.H. Cornely O.A. Salmanton-García J. VACCELERATE consortium. Monkeypox diagnostic and treatment capacity at epidemic onset: A VACCELERATE online survey J Infect Public Health 15 10 2022 1043 1046 36049256
14 Gruber M.F. Current status of monkeypox vaccines NPJ Vaccin 7 1 2022 94
15 Weaver J.R. Isaacs S.N. Monkeypox virus and insights into its immunomodulatory proteins Immunol Rev 225 2008 96 113 18837778
16 Moore J.B. Smith G.L. Steroid hormone synthesis by a vaccinia enzyme: a new type of virus virulence factor EMBO J 11 5 1992 1973 1980 1582424
17 Price N. Tscharke D.C. Hollinshead M. Smith G.L. Vaccinia virus gene B7R encodes an 18-kDa protein that is resident in the endoplasmic reticulum and affects virus virulence Virology 267 1 2000 65 79 10648184
18 Jones D.T. Protein secondary structure prediction based on position-specific scoring matrices J Mol Biol 292 2 1999 195 202 10493868
19 Ferrè F. Clote P. DiANNA: a web server for disulfide connectivity prediction Nucleic Acids Res 33 2005 W230 W232 15980459
20 Doytchinova I.A. Flower D.R. VaxiJen: a server for prediction of protective antigens, tumour antigens and subunit vaccines BMC Bioinforma 8 2007 4
21 Krogh A. Larsson B. von Heijne G. Sonnhammer E.L. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes J Mol Biol 305 3 2001 567 580 11152613
22 Dimitrov I. Bangov I. Flower D.R. Doytchinova I. AllerTOP v.2--a server for in silico prediction of allergens J Mol Model 20 6 2014 2278 24878803
23 Nielsen M. Lundegaard C. Lund O. Prediction of MHC class II binding affinity using SMM-align, a novel stabilization matrix alignment method BMC Bioinforma 8 2007 238
24 Bui H.H. Sidney J. Peters B. Sathiamurthy M. Sinichi A. Purton K.A. Mothé B.R. Chisari F.V. Watkins D.I. Sette A. Automated generation and evaluation of specific MHC binding predictive tools: ARB matrix applications Immunogenetics 57 5 2005 304 314 15868141
25 Larsen J.E. Lund O. Nielsen M. Improved method for predicting linear B-cell epitopes Immunome Res 2 2006 2 Apr 24 16635264
26 Bui H.H. Sidney J. Dinh K. Southwood S. Newman M.J. Sette A. Predicting population coverage of T-cell epitope-based diagnostics and vaccines BMC Bioinforma 7 2006 153
27 Cheng J. Randall A.Z. Sweredoski M.J. Baldi P. SCRATCH: a protein structure and structural feature prediction server Nucleic Acids Res 33 2005 W72 W76 15980571
28 Yang J. Yan R. Roy A. Xu D. Poisson J. Zhang Y. The I-TASSER Suite: protein structure and function prediction Nat Methods 12 1 2015 7 8
29 Heo L. Park H. Seok C. GalaxyRefine: Protein structure refinement driven by side-chain repacking Nucleic Acids Res 41 2013 W384 W388 23737448
30 Lee G.R. Heo L. Seok C. Effective protein model structure refinement by loop modeling and overall relaxation Proteins 84 Suppl 1 2016 293 301 26172288
31 Laskowski R.A. MacArthur M.W. Thornton J.M. PROCHECK: validation of protein-structure coordinates. International Tables for Crystallography Vol F Chapter 21 4 2012 684 687
32 Kozakov D. Hall D.R. Xia B. Porter K.A. Padhorny D. Yueh C. Beglov D. Vajda S. The ClusPro web server for protein-protein docking Nat Protoc 12 2 2017 255 278 28079879
33 Kovacs J.A. Chacón P. Abagyan R. Predictions of protein flexibility: first-order measures Proteins 56 4 2004 661 668 15281119
34 Villalobos A. Ness J.E. Gustafsson C. Minshull J. Govindarajan S. Gene Designer: a synthetic biology tool for constructing artificial DNA segments BMC Bioinforma 7 2006 285
35 Sarker A. Rathore A.S. Gupta R.D. Evaluation of scFv protein recovery from E. coli by in vitro refolding and mild solubilization process Micro Cell Fact 18 1 2019 5
36 Sanchez-Trincado J.L. Gomez-Perosanz M. Reche P.A. Fundamentals and methods for T-and B-cell epitope prediction J Immunol Res 2017 2017
37 Galanis K.A. Nastou K.C. Papandreou N.C. Petichakis G.N. Pigis D.G. Iconomidou V.A. Linear B-cell epitope prediction for in silico vaccine design: A performance review of methods available via command-line interface Int J Mol Sci 22 6 2021 3210 33809918
38 Aiman S. Alhamhoomq Y. Ali F. Rahman N. Rastrelli L. Ahmed A. Khan A. Li C. Multi-epitope chimeric vaccine design against emerging Monkeypox virus via reverse vaccinology Techniques-A Bioinformatics and Immunoinformatics approach Front Immunol 2022 4645 10.3389/fimmu.2022.985450
39 Shantier S., Mustafa M., Abdelmoneim A., Fadl H., Elbager S., Makhawi A., Novel Multi Epitope-based Vaccine against Monkeypox Virus: Vaccinomic approach, 2022. https://doi.org/10.20944/preprints202206.0392.v1.
| 36508944 | PMC9724569 | NO-CC CODE | 2022-12-14 23:45:38 | no | J Infect Public Health. 2023 Jan 6; 16(1):107-116 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.11.033 | oa_other |
==== Front
Rev Francoph Lab
Rev Francoph Lab
Revue Francophone Des Laboratoires
1773-035X
1773-0368
Elsevier Masson SAS.
S1773-035X(22)00357-4
10.1016/S1773-035X(22)00357-4
Article
Tel niveau d’auto-immunité, telle sévérité de la Covid
Manus Jean-Marie
6 12 2022
12 2022
6 12 2022
2022 547 1010
Copyright © 2022 Elsevier Masson SAS. All rights reserved.
2022
Elsevier Masson SAS
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmc Une équipe de chercheurs de l’Université de New York annonce avoir identifié un lien entre le niveau d’auto-anticorps et la sévérité d’une infection Covid-19.
© DC Studio/stock.adobe.com
Dans une étude publiée par la revue Life Science Alliance [1], les chercheurs suggèrent que le taux d’auto-anticorps d’un patient peut être aussi considéré par l’équipe clinique comme un marqueur permettant d’identifier à un stade précoce de l’admission pour Covid-19 les patients à plus haut risque de forme sévère de l’infection à Sars-CoV-2.
Les auto-anticorps en question – anti-ADN et anti-phosphatidylsérine – prédisent donc le développement d’une forme de Covid-19 sévère.
Cette observation, que d’autres équipes dans le monde, notamment en France, ont signalée est une particularité de la Covid-19. Les chercheurs new-yorkais expliquent que des niveaux élevés d’auto-anticorps sont observés chez les patients Covid-19 mais leur contribution spécifique à la sévérité de la maladie et aux manifestations cliniques demeurent « pauvrement comprise ».
Ils ont réalisé une étude rétrospective chez 115 patients hospitalisés pour des signes différents de sévérité, afin d’analyser la génération d’anticorps auto-immuns envers différents antigènes : érythrocytes, phosphatidylsérine lipidique (PS) et ADN.
Des taux élevés d’auto-anticorps IgG contre les érythrocytes ont été observés dans une large proportion de patients (jusqu’à 36 %). Les anticorps anti-ADN et anti-PS, déterminés lors de l’admission à l’hôpital, étaient fortement corrélés avec le développement ultérieur d’une maladie sévère, correspondant alors à une valeur prédictive de 85,7 % et de 92,8 % respectivement.
Les patients affichant des valeurs positives pour au moins l’un des deux auto-anticorps représentaient 24 % de la totalité des cas sévères. Une analyse statistique a permis d’identifier de fortes corrélations entre des anticorps anti-ADN et des marqueurs de lésions cellulaires, de coagulation, des taux de neutrophiles et la taille des érythrocytes.
Ainsi, les auto-anticorps anti-ADN et anti-PS pourraient jouer un rôle important dans la pathogénie de la Covid-19 et pourraient être développés en tant que marqueurs prédictifs de la sévérité de l’infection et de ses manifestations cliniques spécifiques.
==== Refs
Références
1 Gomes C. Zuniga M. Crotty K.A. Autoimmune anti-DNA and anti-phosphatidylserine antibodies predict development of severe Covid-19 Life Sci Alliance 4 11 2021 e202101180 34504035
| 36505966 | PMC9724732 | NO-CC CODE | 2022-12-07 23:21:51 | no | Rev Francoph Lab. 2022 Dec 6; 2022(547):10 | utf-8 | Rev Francoph Lab | 2,022 | 10.1016/S1773-035X(22)00357-4 | oa_other |
==== Front
Rev Francoph Lab
Rev Francoph Lab
Revue Francophone Des Laboratoires
1773-035X
1773-0368
Elsevier Masson SAS.
S1773-035X(22)00366-5
10.1016/S1773-035X(22)00366-5
Article
Combien de temps un Covid-positif reste-t-il infectieux?
Manus Jean-Marie
6 12 2022
12 2022
6 12 2022
2022 547 1617
Copyright © 2022 Elsevier Masson SAS. All rights reserved.
2022
Elsevier Masson SAS
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmc Des preuves s’accumulent, selon lesquelles certains sujets peuvent continuer à transmettre le Sars-CoV-2 pendant beaucoup plus longtemps que les « quelques jours suggérés par des agences de santé publique ». Un article publié dans la revue Nature présente des aspects particuliers de cette pandémie qui livrerait une huitième vague cet automne, selon le ministère de la Santé et de la Prévention.
Des études [1,2] montrent que nombre de sujets Covid-positifs demeurent infectieux bien après la seconde semaine au cours de laquelle leurs symptômes sont apparus. De ce fait, alors qu’il est souvent conseillé aux sujets Sars-CoV-2- positifs de s'isoler pendant quelques jours seulement, les preuves s'accumulent que certains peuvent continuer à transmettre le virus beaucoup plus longtemps (plus de huit jours) qu’on le croit.
Nature [3] rappelle un point-clé : lorsque les centres for disease control d’Atlanta aux États-Unis ont réduit de moitié le temps d'isolement recommandé à cinq jours pour les patients Covid-19, ils ont déclaré que ce changement était motivé par la science, estimant que la plupart des transmissions du Sars-CoV-2 se produisent au début de la maladie, 24 à 48 heures avant l'apparition des symptômes et pendant deux ou trois jours après.
© PeopleImages/Istock
Nombre de scientifiques ont contesté et contestent encore cette décision, une série d'études confirmant que des Covid-19-positifs restent contagieux jusqu'à la deuxième semaine après avoir exprimé les premiers symptômes. La réduction de la durée recommandée du confinement – désormais courante dans le monde entier – serait plutôt motivée par la politique, disent-ils, plutôt que par de nouvelles données rassurantes !
« La durée pendant laquelle on reste infectieux n'a pas vraiment changé », selon Amy Barczak, infectiologue, du Massachusetts General Hospital (MGH) de Boston [2]. À ce stade – fin juillet – Il n'y a pas de données pour soutenir une durée de confinement réduite. Les données d’Amy Barczak, publiées sur medRxiv, suggérant qu'un quart des sujets contaminés par le variant Omicron pourraient encore être contagieux plus de huit jours après l’avoir rencontré.
Quand le virus éliminé laisse des traces derrière lui
Bien que la question soit simple « Pendant combien de temps une personne atteinte de Covid-19 est-elle contagieuse ? » les experts préviennent que la réponse est compliquée. « C’est une question de chiffres et de probabilité » remarque le virologue Benjamin Meyer de l'Université de Genève qui explique dans Nature que les variants émergents, la vaccination et les différents niveaux d'immunité naturelle renforcés par une infection antérieure peuvent tous influencer la rapidité avec laquelle un patient peut éliminer le virus de son organisme, et cela définit finalement quand il cesse d'être infectieux.
Les tests PCR pourraient donner un résultat positif même après qu’un sujet ne soit plus infectieux
« Dernier point étonnant, explique-il, les facteurs comportementaux comptent également. Les personnes qui ne se sentent pas bien ont tendance à moins se mêler aux autres, de sorte que la gravité des symptômes d'une personne peut influer sur la probabilité qu'elle en infecte d’autres ! »
Par ailleurs, la plupart des spécialistes seraient convaincus que les tests PCR peuvent donner un résultat positif même après qu’un sujet ne soit plus infectieux : c’est ce qui se produirait notamment lorsque les tests qui détectent l'ARN viral détectent aussi des restes non infectieux laissés derrière lui après l'élimination de la majeure partie du virus vivant. Étonnant, non ? En revanche, les tests à flux latéral ou « antigène rapide » – selon ce que rapporte Nature – offriraient une meilleure compréhension de l'infectiosité, car ils détectent les protéines produites lors de la réplication active du virus.
Alors ? Réponse dans un message très concis, selon ses termes, du Dr Emily Bruce, microbiologiste et généticienne moléculaire à l'Université du Vermont à Burlington : « Si vous êtes positif à l'antigène, vous ne devriez pas sortir et interagir étroitement avec des gens que vous ne voulez pas infecter ». Des messages oubliés mais souvent réentendus en ce moment, car le risque persiste. Ainsi, qu'en est-il de quelqu'un qui a été testé négatif sur un test de flux latéral pendant quelques jours mais qui reste fébrile avec une toux sèche ? Pour le Dr Bruce il faut se rappeler que même si les symptômes persistants peuvent sembler graves, ils n'indiquent pas toujours le risque d’une contagiosité continue, et que de nombreux symptômes sont causés par le système immunitaire et non directement par le virus lui-même ». Étonnant, non?
==== Refs
Références
1 Boucau J, Marino C, Regan J et al. Duration of viable virus shedding in SARS-CoV-2 omicron variant infection. medRxiv [Preprint]. 2022 Mar 2:2022.03.01.22271582. doi: 10.1101/2022.03.01.22271582. Update in: N Engl J Med. 2022 Jun 29.
2 Townsley H, Carr EJ, Russell TW et al. Non-hospitalised, vaccinated adults with COVID-19 caused by Omicron BA.1 and BA.2 present with changing symptom profiles compared to those with Delta despite similar viral kinetics.medRxiv 2022.07.07.22277367; doi: https://doi.org/10.1101/2022.07.07.22277367.
3 Adam D. How long is COVID infectious? What scientists know so far Nature 608 7921 2022 16 17 35883010
| 36505967 | PMC9724733 | NO-CC CODE | 2022-12-07 23:21:51 | no | Rev Francoph Lab. 2022 Dec 6; 2022(547):16-17 | utf-8 | Rev Francoph Lab | 2,022 | 10.1016/S1773-035X(22)00366-5 | oa_other |
==== Front
Rev Francoph Lab
Rev Francoph Lab
Revue Francophone Des Laboratoires
1773-035X
1773-0368
Elsevier Masson SAS.
S1773-035X(22)00352-5
10.1016/S1773-035X(22)00352-5
Article
L’Europe sanitaire et le booster additionnel d’un vaccin anti-Covid
Manus Jean-Marie
6 12 2022
12 2022
6 12 2022
2022 547 77
Copyright © 2022 Elsevier Masson SAS. All rights reserved.
2022
Elsevier Masson SAS
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmc Cet été, deux agences sanitaires de l’Union européenne : l’European centre for disease prevention and control et l’Agence européenne du médicament ont recommandé que la seconde dose-booster d’un vaccin mRNA Covid-19 soit envisagée pour les personnes entre 60 et 79 ans et celles avec problèmes médicaux les exposant à un haut risque de maladie sévère.
© Jon Anders Wiken/Stock.adobe.com
Les deux instances [1] ont rappelé, au printemps, que les plus de 80 ans et toute personne médicalement vulnérable devaient être candidats « si il y avait une résurgence des infections » à ce second booster, la « quatrième dose » du vaccin à ARNm : un rappel après le rappel de la vaccination initiale.
Alors qu’on évoquait une huitième vague automnale européenne – avec rappel des gestes-barrières – à partir d’une augmentation ici et là des hospitalisations l’European centre for disease prevention and control (ECDC) et l’Agence européenne du médicament (EMA) estimaient « critique » que les autorités de santé publique s’adressent aux 60 ans et plus et aux personnes médicalement vulnérables de tout âge en vue du second booster, celui-ci devant être souhaitablement administré entre quatre et six mois après le précédent à des patients qui pensaient en avoir fini avec leur vulnérabilité à Sars-CoV-2.
À ce sujet, ECDC et EMA rappelaient que « les vaccins autorisés actuelle-ment continuent d’être hautement efficaces à réduire les hospitalisations pour Covid-19, la maladie sévère et les décès dans le contexte de variants Sars-CoV-2 émergents ».
À la Commission européenne, la confiance dans les vaccins contre la Covid-19 est positive : « nos vaccins agissent et offrent un haut niveau de protection contre une maladie sévère et une hospitalisation. Avec des cas et des admissions qui augmentent de nouveau il est urgent que chacun soit vacciné et boosté aussi vite que possible, il n’y a pas de temps à perdre, a déclaré Stella Kyriakides [2], Commissaire européenne pour la santé et la sécurité alimentaire de l’UE, appelant les États-membres à proposer le rappel à tous les plus de 60 ans et aux plus vulnérables médicalement, c’est urgent pour se protéger, protéger ceux qui nous sont chers et tous les sujets vulnérables de nos populations ».
L’automne s’annonçant, il n’était pas encore évident que la « quatrième dose » puisse être recommandée à toute une population et non pas seulement aux plus de 60 ans. Car les plus jeunes, selon ECDC et EMA, ne sont pas à haut risque mais l’arrivée de la grippe saisonnière, qui peut elle aussi évoluer vers des complications létales, devait faire envisager la double vaccination grippe et Covid-19.
==== Refs
Références
1 www.ecdc.europa.eu/en/news-events/ecdc-and-ema-update-recommendations-additional-booster-doses-covid-19-vaccines1
2 www.who.int/europe/news/item/12-10-2022-joint-statement-by-commissioner-stella-kyriakides--who-regional-director-for-europe-dr-hans-henri-p.-kluge-and-director-of-the-ecdc-dr-andrea-ammon--working-together-towards-covid-19-and-seasonal-influenza-vaccinations-for-this-winter
| 36505968 | PMC9724734 | NO-CC CODE | 2022-12-07 23:21:51 | no | Rev Francoph Lab. 2022 Dec 6; 2022(547):7 | utf-8 | Rev Francoph Lab | 2,022 | 10.1016/S1773-035X(22)00352-5 | oa_other |
==== Front
Rev Francoph Lab
Rev Francoph Lab
Revue Francophone Des Laboratoires
1773-035X
1773-0368
Elsevier Masson SAS.
S1773-035X(22)00358-6
10.1016/S1773-035X(22)00358-6
Article
Comirnaty va enfin affronter le variant Omicron
Manus Jean-Marie
6 12 2022
12 2022
6 12 2022
2022 547 1111
Copyright © 2022 Elsevier Masson SAS. All rights reserved.
2022
Elsevier Masson SAS
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmc Début septembre, Pfizer et BioNTech ont reçu una avis positif de la Commission de l’EMA pour les médicaments humains (CHMP) pour mettre au service de la communauté des patients de l’Union européenne le vaccin de rappel bivalent Covid-19, adapté à Omicron BA.1.
Cette commission [1] a entériné les données d’innocuité, de tolérance et d’immunogénicité issues d’un essai de phase 2/3 de ce vaccin bivalent adapté (Sars-CoV-2 sauvage + Omicron BA.1). La campagne mondiale de la « quatrième dose » peut recommencer.
Pfizer et BioNTech [2] ont expliqué que leur dose de rappel de 30 µg de leur vaccin (Comirnaty Original/Omicron BA.1 15/15 µg) a été recommandée pour une autorisation de mise sur le marché conditionnelle par la CHMP de l’Agence européenne des médicaments (EMA) pour les patients à partir de 12 ans, la Commission européenne devant pour la bonne règle examiner la recommandation CHMP avant de donner le feu vert aux États-membres de l’Union Européenne (UE) et de l’Espace économique européen (EEE) qui suivent les directives de l’EMA.
La campagne mondiale de la « quatrième dose » peut recommencer
Le vaccin bivalent adapté Omicron BA.1 contient 15 µg d’ARNm codant pour la protéine de pointe de type sauvage du Sars-CoV-2, qui est présente dans le vaccin original Pfizer-BioNTech Covid-19, et 15 µg d’ARNm codant pour la pointe protéine du sous-variant Omicron BA.1.
Hormis l’ajout de la séquence d’ARNm correspondant à la protéine de pointe de BA.1, les autres composants du vaccin restent inchangés. Comme l’explique Albert Bourla, président-directeur général de Pfizer, « Omicron BA.1 offre aux habitants de l’UE, aux professionnels de santé et aux autorités de santé publique un moyen immédiat de commencer à renforcer l’immunité humaine contre le variant Omicron. Il a été cliniquement démontré que ce vaccin bivalent adapté au BA.1 avait un profil d’innocuité favorable avec une immunogénicité contre les souches de type sauvage et Omicron et pourrait servir d’élément-clé des stratégies de vaccination pour les mois à venir ».
© PX Media/stock.adobe.com
L’avis favorable de la CHMP confirme pour BioNTech/Pfizer que les objectifs d’immunogénicité et de sécurité des vaccins à ARNm, quelles que soient les biotechs les produisant depuis le début de la pandémie, adaptés aux variants de Sars-CoV-2 peuvent être atteints.
C’est à dire ? « Par rapport à une dose de rappel de notre vaccin Covid-19, dont l’utilisation est actuellement approuvée dans l’UE, le vaccin bivalent avec ARNm codant pour les protéines de type sauvage et la protéine de pointe BA.1 fournit des titres d’anticorps neutralisants plus élevés contre l’Omicron BA.1 et les sous-lignées BA.4/BA.5, précise le Pr Ugur Sahin, PDG et cofondateur de BioNTech. De plus une dose de rappel de notre vaccin bivalent adapté à Omicron BA.1 devrait préserver et élever l’étendue des réponses des lymphocytes B et T dans le but de fournir une immunité plus large contre la Covid-19 causée par Sars-CoV-2, y compris les sous-lignées d’Omicron ».
La recommandation d’aujourd’hui fait suite aux directives de l’EMA, de l’OMS et de la Coalition internationale des autorités de réglementation des médicaments pour faire progresser un candidat bivalent, et disposer dès que possible d’un vaccin adapté à Omicron pour les États-membres de l’UE et de l’EEE.
BioNTech et Pfizer [3] ont également déposé une demande à l’EMA pour une dose de rappel d’un vaccin bivalent Omicron BA.4/BA.5 afin de permettre des stratégies de vaccination flexibles. Cette application a été mise en examen début octobre. Un vaccin adapté à Omicron, basé sur le sous-variant BA.4/BA.5 a été autorisé par la Food and drug administration comme rappel pour les 12 ans et plus, le 31 août 2022.
==== Refs
Références
1 www.ema.europa.eu/en/news/first-adapted-covid-19-booster-vaccines-recommended-approval-eu
2 www.pfizer.com/news/press-release/press-release-detail/pfizer-and-biontech-announce-omicron-adapted-covid-19
3 https://www.ema.europa.eu/en/news/adapted-vaccine-targeting-ba4-ba5-omicron-variants-original-sars-cov-2-recommended-approval
| 36505965 | PMC9724735 | NO-CC CODE | 2022-12-07 23:21:51 | no | Rev Francoph Lab. 2022 Dec 6; 2022(547):11 | utf-8 | Rev Francoph Lab | 2,022 | 10.1016/S1773-035X(22)00358-6 | oa_other |
==== Front
J Evid Based Dent Pract
J Evid Based Dent Pract
The Journal of Evidence-Based Dental Practice
1532-3382
1532-3390
Mosby, Inc
S1532-3382(22)00147-6
10.1016/S1532-3382(22)00147-6
101819
Article
Table of Contents
6 12 2022
12 2022
6 12 2022
22 4 101819101819
2019
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmc
| 0 | PMC9724737 | NO-CC CODE | 2022-12-07 23:21:51 | no | J Evid Based Dent Pract. 2022 Dec 6; 22(4):101819 | utf-8 | J Evid Based Dent Pract | 2,022 | 10.1016/S1532-3382(22)00147-6 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01056-6
10.1016/j.revmed.2022.10.368
Ca083
Vascularite cutanée lupique secondaire à une vaccination par le vaccin Pfizer-BioNTech COVID-19 : à propos d’un cas
Chawad W. ⁎
Harmouche H.
Gougas A.
Daouaji H.
Ait Zine I.
Safae F.F.
Belkhettab S.
Ibourk Idrissi F.
Taouch A.
Mouatassim N.
Khibri H.
Ammouri W.
Maamar M.
Tazi Mezalek Z.
Adnaoui M.
Médecine interne et hématologie clinique, CHU Ibn Sina, Rabat, Maroc
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A461A462
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
L’outil le plus puissant pour lutter contre la pandémie COVID-19 reste à ce jour la vaccination. Bien que les différents vaccins commercialisés soient généralement bien tolérés, des évènements indésirables ont été rapportés dont l’apparition ou la réactivation d’une vascularite cutanée et le développement d’un lupus érythémateux cutané. Des peptides en commun entre la protéine spike et les protéines humaines suggèrent le développement d’une auto-immunité par le biais du mimétisme moléculaire. Ce mécanisme serait similaire à celui de l’auto-immunité induite par une infection au SARS-CoV-2. Le lupus érythémateux systémique est une maladie auto-immune de présentation protéiforme. De multiples facteurs (génétiques, environnementaux et immunologiques) interviennent dans sa pathogénie. Nous rapportons le cas d’une patiente ayant développée une vascularite cutanée lupique après avoir reçu la deuxième dose du vaccin Pfizer-BioNTech.
Observation
Il s’agit d’une patiente de 26 ans, qui a présenté 14 jours après la deuxième dose de vaccin Pfizer-BioNtech des lésions érythémateuses maculopapuleuses non prurigineuses augmentant de taille de façon centrifuge intéressant les membres et le visage en épargnant le front et les sillons nasogéniens. La biopsie cutanée a montré une lésion évoluée de vascularite leucocytoclasique des vaisseaux de petits calibres. La patiente a reçu de la prednisone a dose de 60 mg/j pendant deux semaines avec dégression progressive, associée à de la colchicine pendant 1 mois avec disparition complète des lésions. L’évolution a été marquée 4 mois plus tard, suite à une exposition solaire, par une récidive de la symptomatologie de base. Le bilan immunologique réalisé a retrouvé : des anticorps antinucléaires positifs à 1/160 de fluorescente mouchetée, des anticorps anti-SSA, anti-Ro52 et anti-ribosomes positifs. Les anti-ADN natifs et les ANCA était négatifs. La recherche d’une cryoglobulinémie est revenu négative. Les fractions C3 et C4 du complément étaient non consommées. La biologie standard a retrouvé une lymphopénie à 800 éléments/mm3 et une thrombopénie à 100 000 éléments/mm3. Le diagnostic de lupus systémique a été retenu sur les critères de classification EULAR/ACR 2019 (10 points). La patiente a été mise sous antipaludéens de synthèse (plaquenil 200 mg × 2/jour) et sous faibles doses de prednisone (5 mg) avec bonne évolution.
Discussion
Dans les études de phase 3, le vaccin Pfizer-BioNTech COVID-19 a démontré une efficacité de 95 % dans la prévention contre l’infection à SARS-CoV-2. La formation d’auto-anticorps et l’apparition de maladies auto-immunes ont été observées dans les suites d’une infection COVID. Les mécanismes physiopathologiques décrits sont : une réponse immunitaire excessive, un mimétisme moléculaire entre les protéines virales et celles de l’hôte et l’activation de plusieurs voies de l’inflammation.
Conclusion
Il n’existe pas de données définitives indiquant que les vaccins à ARNm seraient responsables du développement d’une auto-immunité, néanmoins la chronologie de la survenue des symptômes fait suggérer une possible association éthiopathogénique entre l’administration du vaccin et la survenue d’un lupus chez des sujets génétiquement prédisposés. Par conséquent nous espérons que ce cas clinique ne décourage nullement contre la poursuite de la vaccination, qui reste à ce jour le seul outil efficace dans la lutte contre la pandémie, mais qu’il sensibilise sur la possible survenue d’un lupus suite à une vaccination au sein d’une population prédisposée.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
==== Refs
Pour en savoir plus
Lemoine C, Padilla C, Krampe N, Doerfler S, Morgenlander A, Thiel B, et al. Systemic lupus erythematous after Pfizer COVID-19 vaccine: a case report. Clin Rheumatol 2022;41:1597–601.
| 0 | PMC9724751 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A461-A462 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.368 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)00710-X
10.1016/j.revmed.2022.10.022
Co010
COVID-19 sévère et maladies auto-immunes ou inflammatoires incidentes : une étude épidémiologique nationale à partir des données du PMSI en 2020
Helary A. 1
Mageau A. 1
Ruckly S. 2
Strukov A. 3
Papo T. 1
Timsit J.F. 4
Sacre K. 1⁎
1 Médecine interne, hôpital Bichat – Claude-Bernard, Paris
2 Iame Inserm Umr 1137, hôpital Bichat – Claude-Bernard, Paris
3 Msi, hôpital Bichat – Claude-Bernard, Paris
4 Réanimation médicale et infectieuse, hôpital Bichat, AP–HP, Paris
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A331A332
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Début août 2022, l’Organisation mondiale de la santé (OMS) recense plus de 580 millions de cas mondiaux cumulés d’infections à SARS-CoV-2 (COVID-19). Plusieurs arguments expérimentaux étayent le lien entre COVID-19 et maladies auto-immunes ou inflammatoires (MAII) : la production excessive de cytokines pro-inflammatoires a été associée aux formes graves de COVID-19 ; la présence d’anticorps sériques neutralisant l’IFN-α, décrite dans certaines MAII comme le lupus, est associée à la sévérité du COVID-19 ; des auto-anticorps spécifiques de certaines MAII sont fréquemment détectés chez les patients atteints de COVID-19 grave. La littérature médicale fait état, sous la forme de séries de cas ou d’observations isolées de MAII apparues au décours d’un COVID-19. Il n’existe pas d’étude épidémiologique comparative de grande ampleur analysant la survenue de cas incidents de MAII après un COVID-19. Nous avons analysé l’incidence de maladies auto-immunes ou inflammatoires après un sepsis lié au COVID-19 en comparaison à des sepsis non liés au COVID-19 chez des patients sans antécédent de MAII en France en 2020.
Patients et méthodes
Nous avons construit une cohorte à partir de la base de données médicales et administratives nationales d’hospitalisations (programme de médicalisation des systèmes d’information, PMSI) entre janvier 2011 et novembre 2020. Chaque patient ayant présenté un premier épisode de sepsis-COVID-19 était apparié de manière aléatoire à 2 patients contrôles ayant présenté un premier épisode de sepsis non-COVID-19 diagnostiqué dans un hôpital français en 2020. Le sepsis était défini par un code diagnostique d’infection et au moins un code se référant à une défaillance d’organe (diagnostic ou acte). Nous avons réalisé un appariement exact en utilisant l’âge ± 2 ans, le genre, l’existence d’un cancer actif, d’une hémopathie maligne, d’une infection VIH ou d’un antécédent de transplantation d’organe. Les patients ayant un antécédent de MAII était exclus de l’analyse.
Résultats
En 2020, nous avons apparié 68 175 patients ayant présenté un sepsis-COVID-19 à 136 350 patients avec un sepsis-non-COVID-19. L’incidence des MAI dans ces 2 populations était importante, atteignant, selon la MAII considérée (thrombopénie immunologique, anémie hémolytique auto-immune, lupus systémique, syndrome de Gougerot-Sjögren, Sclérodermie systémique, polyarthrite rhumatoïde, artérite à cellules géantes, pseudopolyarthrite rhizomélique, ou vascularites associées aux ANCA) 1,3 à 165 fois celle rapportée dans la population générale. L’analyse globale de survie à 9 mois sans MAII ne montrait par de surrisque associé spécifiquement au sepsis-COVID-19 comparativement au sepsis non-COVID-19 à l’exception de la polyarthrite rhumatoïde (HR : 1,76 ; IC95 % [1,28–2,43]) et de la pseudopolyarthrite rhizomélique (HR : 1,64 ; IC95 % [1,23–2,19]). Le sepsis-COVID-19, comparativement au sepsis-non-COVID-19 était associé à un moindre risque de vascularite à ANCA (HR : 0,22 ; IC95 % [0,12–0,42]).
Conclusion
Un premier épisode de sepsis en 2020 était associé à une incidence nettement plus élevée de MAII qu’en population générale avec un surrisque de polyarthrite rhumatoïde et de pseudopolyarthrite rhizomélique spécifiquement associé au sepsis-COVID-19 sepsis.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724752 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A331-A332 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.022 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01064-5
10.1016/j.revmed.2022.10.376
Ca091
Thrombose des veines hépatiques après la vaccination anti-COVID-19
Abdelmoula A. ⁎
Damak C.
Bouattour Y.
Frikha F.
Chabchoub I.
Ghribi M.
Ben Hamad M.
Ben Salah R.
Mona S.
Marzouk S.
Bahloul Z.
Médecine interne, CHU Hedi Chaker de Sfax, Sfax, Tunisie
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A466A466
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Différents types d’effets indésirables des vaccins ont été décrits. La thrombose est l’un des effets indésirables les plus graves et les plus atypiques du vaccin anti-COVID-19. Nous rapportons un cas de thrombose des veines hépatiques après le vaccin anti-COVID-19 avec insuffisance hépatique.
Observation
Nous rapportons le cas d’un patient de 26 ans ayant des antécédents médicaux de colite ulcéreuse (actuellement en rémission clinique et endoscopique). Il a été hospitalisé pour des vomissements et un ictère cutanéomuqueux durant 15 jours. L’interrogatoire du patient a révélé que cette symptomatologie est survenue 4 jours après le vaccin anti-COVID (Moderna). Nous avons constaté une altération de l’état général (anorexie, asthénie et perte de poids), un ictère cutanéomuqueux et un œdème généralisé. La biologie a montré une numération de la formule sanguine normale, une protéine C-réactive élevée à 112 mg/L et une vitesse de sédimentation à 35 mm 1re h. Il y avait des cytolyses avec une insuffisance hépatique (AST : 1880 UI/L [47 N] ; ALT : 620 UI/L [11 N] ; Bil T : 15 ; GGT : 247 ; PAL : 112 taux de prothrombine : 19 %; albuminémie : 19 g/L). Un scanner abdominal a montré une thrombose proximale complète des veines sus-hépatiques, une thrombose partielle de la branche portale gauche et une thrombose partielle de la veine splénique. L’enquête étiologique a éliminé une thrombophilie constitutionnelle et acquise (protéine C, protéine S ; anti-thrombine III ; résistance à la protéine C activée, anticorps antiphospholipides) ; une cause néoplasique (aucun signe clinique ni radiologique [scanner TAP]) ; et des causes inflammatoires (la colite ulcéreuse était en rémission). Le diagnostic de thrombose post-vaccinale a été retenu et confirmé par une enquête de pharmacovigilance. Le patient a été traité par un traitement anticoagulant avec une amélioration clinique complète. La normalisation du bilan hépatique a été obtenue après 2 mois de traitement. Le suivi global est de 12 mois.
Conclusion
Le site de thrombose le plus touché était le système veineux cérébral, ce qui est confirmé par les observations de la littérature. L’atteinte de la circulation portale est également fréquente. L’atteinte concomitante des veines sus-hépatiques et des veines portales avec insuffisance hépatique comme c’est le cas de notre patient n’a pas été rapportée.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724753 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A466 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.376 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01051-7
10.1016/j.revmed.2022.10.363
Ca078
Purpura extensif et infection SARS-CoV-2
Houache A. 1
Bouziani N. 1
Kella A. 1
Belabbas A. 1
Hakem D. 2⁎
1 Médecine interne, CHU Dr Boumediéne Bensmain, Kharouba, Mostaganem, Algérie
2 Médecine interne, hôpital Dr Mohammad-Lamine Debaghine, CHU Bab-El-Oued, Alger, Algérie
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A458A459
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
L’infection par le SARS-CoV-2 se manifeste principalement par des symptômes respiratoires cependant de multiples atteintes d’organes ont été rapporté et les téguments ne font pas exception. Nous en rapportons à ce propos une observation.
Observation
M. B.A., 54 ans sans antécédent particuliers présentait une asthénie, une fièvre, et une rhinite traitée symptomatiquement en ambulatoire. Quinze plus tard le patient consultait à nouveau pour un purpura fébrile rapidement extensif justifiant son admission en milieux hospitalier. L’anamnèse ne retrouvait pas de tares antérieures ni de prises médicamenteuses ou toxique. L’examen physique retrouvait un patient conscient coopérant score 15/15 sur l’échelle de Glasgow. Son IMC était estimé à 23,22. Il était apyrétique et stable sur le plan hémodynamique, eupnéique, sa TA était à 110/70 mmHg et son rythme cardiovasculaire réguliers avec une fréquence cardiaque à 70 bat/min. Sa saturation en O2 était à 99 %. Sur le plan cutané on mettait en évidence un purpura pétéchial diffus au dos, des membres supérieurs prenait un aspect ecchymotique aux membres inférieurs et infiltré au niveau du tronc. Ce purpura rapidement extensif n’était pas accompagné d’hémorragie des muqueuses ni de syndrome de Raynaud. L’examen cardiaque ne notait pas de souffle. L’examen neurologique ne retrouvait pas de signes méningés (pas de raideur de la nuque, signe de Kernig et de Brudzinski négatifs) ni de signes d’appel en faveur d’une neuropathie périphériques. Les réflexes cutanéoplantaires étaient normaux. L’examen abdominal retrouvait une splénomégalie type 2. Le reste de l’examen clinique était normal. La chimie urinaire ne retrouvait pas d’anomalies (leucocytes, protéine, sang négatifs). La NFS mettait en évidence une lymphopénie à 600/mL et un taux de plaquette à 293/L. Le TP était à 100 %, le TCK à 30,2 s sec, le taux de fibrinogène à 4,10 g/L et des D dimères à 8410 ng/mL. La CRP était à 135 mg/L. Le test antigénique covid était négatif et la sérologies COVID était positive à prédominance IgG. Les sérologies VIH, HVB et HVC, TPHA-VDRL, Wright étaient négatives de même que l’IDR à la tuberculine. Les tests hépatiques, la fonction rénale et les CPK étaient normaux et l’électrophorèse des protides ne montrait de pic monoclonal. La TDM thoracique et l’échocardiographie Doppler étaient sans anomalies. L’échographie abdominopelvienne confirmait la présence d’une splénomégalie stade 2. Au terme de ce bilan le purpura était rattaché à son infection COVID après élimination des autres causes infectieuses (endocardite d’Osler, hépatite C, CIVD, brucellose, tuberculeuse…). Sur le plan thérapeutique on préconisait un repos strict, un traitement symptomatique. L’évolution était spontanément favorable avec apyrexie, régression du purpura de la splénomégalie, normalisation des lymphocytes, diminution des Dimères à j8 avec un taux à 1462,26 ng/mL Le bilan immunologique AAN, anti-SSA/SSB, ANCA, cryoglobulinémie et APL était négatif.
Discussion
Tout purpura impose d’éliminer une cause infectieuse, une CIVD, un purpura d’Hénoch, une endocardite d’Osler, une septicémie à BGN notamment devant la notion de fièvre et de splénomégalie. Une vascularite type angéite de Zeek (médicamenteuse ou toxique, vaccination), une périartérite noueuse (PAN), une cryoglobulinémie, une vascularite à ANCA ne doivent pas être également occultées et doivent rechercher une atteinte rénale (hématurie, protéinurie, dégradation de la fonction rénale…), neurologiques périphériques à type notamment de multinévrites, une myocardite, une atteinte pulmonaire entrant dans le cadre d’un syndrome pneumo-rénal (hémorragies alvéolaires). Après avoir éliminé les autres causes infectieuses et devant une sérologie COVID positive et la régression spectaculaire des symptômes avec un recul suffisant le diagnostic d’un purpura post-COVID a été retenu. La biopsie cutanée n’a pas été justifiée et aurait pu apporter des informations supplémentaires (dépôts d’immuns complexes ? IgA ? microangiopathie ?) au niveau des lésions infiltrées du tronc.
Conclusion
Les dernières études montrent que les manifestations systémiques extrapulmonaires liées au syndrome post-COVID-19 existent chez environ 50 % à 80 % des patients symptomatiques. Les manifestations cutanées sont de plus en plus décrites et sont d’expressions plurielles exprimant le plus souvent une vascularite à immuns complexes comme celles décrites avec les infections virales HVC, VIH et les CIVD. Elles doivent être connues et reconnues par les praticiens impliquant la recherche d’une infection SARS-CoV-2 notamment en période de pandémie. Le pronostic restant lié aux atteintes viscérales associées.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
==== Refs
Pour en savoir plus
Asiri A, Alzahrani F, Alshehri S, AbdelQadir YH. New-Onset Henoch–Schonlein purpura after COVID-19 infection: a case report and review of the literature. Case Report | Open Access Volume 2022 | Article ID 1712651. https://doi.org/10.1155/2022/1712651. https://www.hindawi.com/journals/cripe/2022/1712651/.
| 0 | PMC9724754 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A458-A459 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.363 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)00751-2
10.1016/j.revmed.2022.10.063
Co051
Risque de poussée de lupus érythémateux systémique après un épisode sévère de COVID-19 : analyse de l’Entrepôt de Données de Santé de l’AP–HP
Mageau A. 1
Timsit J.F. 2
Papo T. 1
Sacre K. 1⁎
1 Médecine interne, hôpital Bichat – Claude-Bernard, Paris
2 Réanimation médicale et infectieuse, hôpital Bichat AP–HP, Paris
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A360A361
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Les interférons de type 1 étant impliqués dans la physiopathologie du lupus érythémateux systémique (LES) et dans la réponse immunitaire anti-COVID-19, un risque de poussée lupique COVID induite pourrait être suspecté. Nos objectifs étaient d’évaluer ce risque et de rechercher les facteurs associés à une poussée de LES après un épisode sévère de COVID-19.
Patients et méthodes
Nous avons mené une étude de cohorte rétrospective à l’aide de l’Entrepôt de Données de Santé (EDS) de l’Assistance publique–Hôpitaux de Paris (AP–HP) qui collecte l’ensemble des données médicales produites dans les 39 établissements de l’AP–HP en région parisienne. Un épisode sévère de COVID-19 était défini comme une pneumopathie associée au SARS-CoV-2 ayant nécessité une hospitalisation. Une poussée de LES était définie comme telle si les cliniciens en charge du patient avaient retenu ce diagnostic. Nous avons inclus chaque patient adulte ayant des antécédents de LES (défini par un code de diagnostic « M32 » CIM-10) et un séjour à l’hôpital avec un premier épisode de diagnostic de COVID-19 (défini par un code « U07.1 » CIM-10) entre mars 2020 et février 2022. Les cas de SLE incidents étaient exclus. Tous les dossiers médicaux ont été examinés individuellement pour récupérer les données démographiques, les caractéristiques du LES, les caractéristiques des épisodes de COVID-19 et le statut vaccinal. Les caractéristiques associées aux poussées de SLE post-COVID-19 ont été analysées à l’aide de procédures de régression logistique univariées.
Résultats
Parmi les 4 533 patients atteints de LES suivis à l’AP–HP et présents dans l’EDS, 128 (2,8 %) ont eu un séjour hospitalier avec un diagnostic de COVID-19 au cours de la période d’intérêt. Après examen de tous les dossiers médicaux individuels, nous avons exclus 38 patients ne répondant pas aux critères d’inclusion. En conséquence, 90 patients ont été inclus dans l’analyse ; 84 (93,3 %) étaient des femmes avec un âge médian [Q1–Q3] de 54,6 [40,8–68,3] ans au moment de l’hospitalisation pour COVID. Le délai médian entre le diagnostic de LES et l’épisode de COVID-19 était de 13,5 [5,6–22,8] ans. Soixante-treize (81,1 %) patients n’avaient reçu aucune dose de vaccin anti-SARS-CoV-2 avant l’épisode de COVID-19 et 9 (10 %) sont décédés au cours de leur hospitalisation initiale. Parmi les 90 survivants que nous avons pu suivre, nous avons observé 18 (20,0 %) poussées de LES post-COVID-19, dont 6 (30,0 %) sont survenues au cours du même séjour à l’hôpital. Le délai médian entre le début de l’épisode de COVID-19 et la poussée de LES était de 88 [22,5–137,5] jours. Nous avons observé 7 (38,9 %) poussées rénales dont 3 (16,7 %) correspondaient à des glomérulonéphrites lupiques de classe III ou IV. La seule caractéristique significativement différente entre les patients ayant présenté une poussée post-COVID et les autres était l’ancienneté du diagnostic de lupus : 6 [2,1–17,5] années dans le groupe poussée versus 14,1 [7,7–22,6] dans le groupe sans poussée. En revanche, nous n’avons pas observé de différence dans la proportion de patients vaccinés entre les deux groupes ou dans les marqueurs de sévérité de la pneumopathie COVID-19.
Conclusion
Les poussées de LES semblent être fréquentes après un épisode sévère de COVID-19. Une durée d’évolution de la maladie courte avant COVID était associée à un plus grand risque de poussée.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724755 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A360-A361 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.063 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01050-5
10.1016/j.revmed.2022.10.362
Ca077
Ischémies aiguës et infection COVID-19
Kella A.
Belabbas A.
Bouziani N.
Khedim A.
Relimi Y.
Habbi M.
Hakem D. ⁎
Médecine interne, CHU Dr Boumediéne Bensmain, Kharouba, Mostaganem, Algérie
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A458A458
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
L’infection au SARS-CoV-2 semble être une pathologie vasculaire dont le mécanisme incluent une hypercoagulabilité sanguine directe et une vasculaire suite à une réponse inflammatoire systémique [1]. Objet : nous rapportons à ce propos 3 observations de pneumopathie COVID-19 compliquée d’une ischémie aiguë des membres inférieurs et du membre supérieur.
Observation
-Observation 1 : patient âgé de 64 ans aux ATCD d’un DT2 depuis 10 ans ; sous insuline type prémix. La symptomatologie remonte à deux jours avant son hospitalisation faite d’une fièvre avec altération de l’état général et une dyspnée. La saturation en O2 était estimée à 87° en air ambiant. Le test PCR pour l’infection COVID-19 était positive. La TDM thoracique mettait en évidence une atteinte de 75 % du parenchyme pulmonaire. Le patient était mis sous traitement suivant le protocole international avec une thromboprophylaxie. L’évolution était marquée par l’apparition d’une thrombopénie à 75 000 avec une ischémie aiguë des deux membres inférieurs à j9 d’hospitalisation. Un Doppler artériel mettait en évidence une thrombose poplitée bilatérale. Il n’y avait pas de stigmates biologiques en faveur d’une thrombopénie immunologique induite à l’héparine « TIH ». Une chirurgie de sauvetage vasculaire était récusée du fait du risque opératoire majeur. Le patient était mis sous traitement médical optimalisé. L’évolution se faisait vers la nécrose du pied droit. -Observation 2 : patient âgé de 63 ans, sans ATCD particulier consultait en unité COVID pour une asthénie et une dyspnée. La TDM thoracique retrouvai une atteinte à 50 % du parenchyme pulmonaire et le test PCR positif confirmait l’infection COVID-19. Un traitement conventionnel selon les directives nationales associant une thromboprophylaxie était introduit. L’évolution était marquée par une ischémie du membre inférieur à j4 avec au Doppler la mise en évidence d’une thrombose artérielle jambière bilatérale. L’évolution vers la nécrose du gros orteil était observée en dépit du traitement initié. -Observation 3 : patiente âgée 77 ans au DT2 évoluant depuis 25 ans environ sous prémix et d’une HTA récente équilibrée sous monothérapie était hospitalisée pour une nécrose digitale du P1 du 5e rayon droit avec un Doppler des membres supérieurs sans particularité. Une enquête étiologique revenait négative en dehors d’une sérologie SARS-CoV-2 IgG positif–IgM négatif. La reprise de l’anamnèse notait la notion d’un syndrome grippal avec fièvre, toux et des arthromyalgies négligés non traités un mois avant son hospitalisation d’où l’hypothèse très probable retenue d’ischémie liée post-COVID-19.
Discussion
La survenue d’une ischémie aiguë des membres peut compliquer l’évolution de l’infection SARS-CoV-2 ; son diagnostic est clinique basé sur l’apparition des signes classiques sans spécificité liée à l’étiologie – SARS-CoV-2 ; son délai d’apparition est variable peut atteindre quelques semaines [2]. Aucun examen complémentaire ne doit retarder la prise en charge thérapeutique. Il ne permet pas de confirmer le mécanisme physiopathologique responsable dans la majorité des cas en dehors de l’hypothèse d’une TIH ou d’un accident embolique à partir d’un troubler du rythme (ACFA transitoire ou préexistant à l’infection COVID-19), à la suite d’un infarctus du myocarde (anévrisme) ou d’une plaque d’athérome rendue instable par l’inflammation au cours de l’infection COVID-19 notamment en cas de comorbidités (diabète, HTA, tabac, âge avancé…. Et d'sur AOMI méconnue, ou à partir d’un anévrisme de l’aorte abdominale…). Un traitement bien conduit au cours de la phase aiguë n’écarte pas le risque de survenue des évènements vasculaires thromboembolique [3]. Il n’existe pas de données spécifiques déterminant la durée du traitement en particulier thomboprophylactique qui reste à individualiser en fonction du terrain [4]. Le traitement chirurgical de revascularisation occupe une place primordiale dans la prise en charge afin d’éviter la constitution d’une ischémie dépassée mais au prix d’un collapsus et arrêt circulatoire (par insuffisance rénale et hyperkaliémie) à la lever d’obstacle suite à l’ischémie des masses musculaires [3]. La correction des facteurs aggravant l’ischémie (trouble du rythme ; bas débit ; la désaturation en O2 ….), la lutte contre l’œdème en particuliers après la chirurgie font partie de la prise en charge thérapeutique.
Conclusion
L’expression clinique de l’atteinte vasculaire au cours de l’infection SARS-CoV-2 est polymorphe va de l’atteinte veineuse à l’atteinte artérielle. Cette dernière apparaît être un facteur de risque pour les maladies vasculaires [1]. Une meilleure compréhension des mécanismes physiopathologiques débouchera sur de meilleures stratégies thérapeutiques tant préventives que curatives.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
==== Refs
Références
1 Gendon N, Smadja D. Service d’hématologie, université de Paris 2020.
2 COVID-19 and thrombotic or thromboembolic disease: implications for prevention. Am Coll Cardiol 2020.
3 High risk of thrombosis in patients in severe SARS-CoV-2 infection. Intensive Care Med 2020.
4 Traitement anticoagulant pour la prévention du risqué thrombotique chez les patients hospitalisés. La SRAR 2020.
Pour en savoir plus
Annales marocaine de médecine d’urgence et de réanimation. Juillet 2021.
| 0 | PMC9724756 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A458 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.362 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01063-3
10.1016/j.revmed.2022.10.375
Ca090
Thrombose veineuse extensive du membre supérieur après la vaccination anti-COVID-19
El-Ouakhoumi A. 1⁎
Zahlane M. 1
Benjilali L. 2
Essaadouni L. 1
1 Médecine interne, CHU Mohammed VI Marrakech, Marrakech, Maroc
2 Médecine interne, hôpital Arrazi, Marrakech, Maroc
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A465A465
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Les événements indésirables après la vaccination COVID-19 ont été une grande préoccupation au monde entier, faits majoritairement de réactions locales, syndrome pseudo-grippal, avec quelques cas de syndrome de Guillain-Barré, de myocardite et/ou de péricardite. En outre, de rares cas de thrombocytopénie thrombotique immunitaire induite par le vaccin (VITT) et de thromboses cérébrales ont été signalés surtout avec les vaccins à vecteur adénoviral. Nous rapportons un cas d’une thrombose du membre supérieur étendue à la veine axillaire, sous-clavière et jugulaire interne après vaccination COVID-19.
Observation
Patiente de 69 ans, sans antécédents pathologiques particuliers, sans facteurs de risque thrombogènes évidents, ayant présenté 3 semaines après sa première dose de vaccin AstraZeneca COVID-19 (ChAdOx1) une tuméfaction du membre supérieur gauche, au site de la vaccination, dure, chaude, douloureuse, sans signes infectieux. L’écho-Doppler veineux a objectivé une thrombophlébite superficielle et profonde du membre supérieur gauche étendue intéressant les veines radiales, cubitales, brachiales, axillaire, sous-clavière et jugulaire interne. Le bilan étiologique notamment infectieux, néoplasique, immunologique et de thrombophilie était négatif. Il n’y avait pas de thrombopénie. Sur le plan thérapeutique, la patiente a été mise sous héparinothérapie à dose curative avec relais par anti-vitamine K. L’évolution était favorable avec disparition de la tuméfaction du membre supérieur gauche.
Conclusion
La TVP aiguë après le vaccin AstraZeneca COVID-19 (ChAdOx1) est liée à l’activation de la cascade inflammatoire et à la coagulopathie. Il s’agit plutôt de thromboses cérébrales, abdominales ou d’embolies pulmonaires. La localisation au membre supérieur est exceptionnelle. Nous en discutons les mécanismes possibles.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724757 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A465 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.375 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)00787-1
10.1016/j.revmed.2022.10.099
Co087
Impact du vaccin ARNm BNT162b2 anti-SARS-CoV-2 sur la production d’interféron alpha et l’auto-immunité cellulaire au cours du lupus érythémateux systémique : le projet COVALUS
Mageau A. 1
Goulenok T. 1
Tchen J. 2
Roland Nicaise P. 3
Ferre V.M. 4
Delory N. 1
Francois C. 1
Papo T. 1
Charles N. 2
Sacre K. 1⁎
1 Médecine interne, hôpital Bichat – Claude-Bernard, Paris
2 Inserm U1149, hôpital Bichat – Claude-Bernard, Paris
3 Laboratoire d’immunologie, hôpital Bichat – Claude-Bernard, Paris
4 Virologie, hôpital Bichat – Claude-Bernard, Paris
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A385A386
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Alors que la vaccination anti-SARS-CoV-2 est devenue un enjeu important de la prise en charge des patients atteints de maladies auto-immunes systémiques telles que le lupus érythémateux systémique (LES), nous manquons de données sur l’impact de la vaccination par ARN messager sur l’auto-immunité. Notre objectif était de décrire l’impact de la vaccination par ARNm sur la production d’interféron alpha et l’auto-immunité cellulaire au cours du lupus.
Patients et méthodes
Nous avons réalisé une étude prospective, observationnelle, monocentrique au cours de laquelle nous avons inclus des patients atteints de LES, éligibles à la vaccination anti-SARS-CoV-2 par le vaccin BNT162b2 selon les recommandations françaises alors en vigueur. Nous avons secondairement exclu les patients non-naïfs vis-à-vis de l’infection par le SARS-CoV-2 à T0 et les patients sous mycophénolate, azathioprine ou rituximab. Les patients étaient évalués juste avant la première dose, puis à 1 mois (M1), à 3 mois (M3) et à 6 mois (M6) de leur première dose. En plus du suivi clinicobiologique habituel de la maladie, nous avons évalué la production d’IFN-alpha par les cellules dendritiques plasmacytoïdes (pDCs) en cytométrie de flux. Nous avons également quantifié l’auto-immunité T en mesurant la proportion de cellules T activées (CD154+ CD69+) après stimulation des cellules sanguines mononucléées périphériques (PBMCs) par des antigènes nucléaires (U1-RNP, histones, SS-A, SS-B).
Résultats
Trente-six patients lupiques et 11 contrôles sains ont été inclus. Les patients lupiques étaient majoritairement des femmes (31/36, 81 %) d’un âge médian de 44 [36–50] ans. Nous avons observé une augmentation significative de la proportion de pDCs produisant spontanément de l’IFN-alpha à M1 (1,27 % [0,6–2,6]) et M3 (1,25 % [0,87–1,83]) comparativement à T0 (0,64 [0,27–1,09] %, respectivement p < 0,001 et p < 0,01) chez les lupiques. Nous avons observé une augmentation similaire dans le groupe contrôle, mais uniquement à M1 (1,46 [0,95–2,12] %) comparativement à T0 (0,25 [0,12–0,47] %, p < 0,02). Par ailleurs, les pDCs des patients lupiques exprimaient à leur surface plus de marqueurs d’activation CD86 et HLA-DR à M3 que à T0. Concernant la proportion de cellules T auto-réactives des patients lupiques, nous avons observé, qu’après une hausse non significative entre T0 et M1, les cellules auto-réactives diminuaient significativement dans le temps (coefficient β d l’effet fixe associé au temps dans le modèle linéaire à effet mixte = −0,00067, p = 0,015). Nous n’avons pas observé une telle tendance chez les contrôles. Au cours du suivi, nous avons observé 2 poussées cliniques de la maladie. À l’exception de ces patients, nous n’avons pas constaté de modification significative du SLEDAI, du taux d’anti-dsDNA, de C3 ou de C4 dans la cohorte.
Conclusion
Chez les patients lupiques, la vaccination anti-SARS-CoV-2 par ARNm entraîne (1) une augmentation modeste infra-clinique de la production spontanée d’IFN-alpha par les pDCs et (2) une diminution des cellules T auto-réactives spécifiques des antigènes nucléaires.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724758 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A385-A386 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.099 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)00777-9
10.1016/j.revmed.2022.10.089
Co077
Stratégies innovantes de vaccination anti-pneumococcique par rapport au schéma standard chez les patients atteints de vascularites associées aux ANCA recevant du rituximab : essai contrôlé randomisé multicentrique (PNEUMOVAS)
Terrier B. 1⁎
Richert L. 2
Pugnet G. 3
Aumaître O. 4
Moranne O. 5
Diot E. 6
Karras A. 7
Bonnet F. 8
De Moreuil C. 9
Hachulla E. 10
Le Gallou T. 11
Lebas C. 12
Maurier F. 13
Rafat C. 14
Samson M. 15
Augusto J.F. 16
Janssen C. 17
Quéméneur T. 18
Batteux F. 19
Launay O. 20
Groupe français d’étude des vascularites
1 Médecine interne, hôpital Cochin, rue du Faubourg-Saint-Jacques, Paris
2 Inserm u897, Campus Carreire, Bordeaux
3 Service de médecine interne, CHU Toulouse Purpan, Toulouse
4 Médecine interne, CHU Gabriel-Montpied, Clermont-Ferrand
5 Néphrologie, hôpital de jour de Nîmes, Nîmes
6 Médecine interne, CHU de Tours, Tours
7 Néphrologie, HEGP, 15, rue Louis-Blanc, 75015 Paris
8 Médecine interne, hôpital Saint-André, Bordeaux
9 Service de médecine interne, CHU Brest Centre de Formation, Brest
10 Médecine interne, CHU, Lille
11 Médecine interne, centre hospitalier universitaire de Rennes, Rennes
12 Néphrologie, CHU de Lille, Lille
13 Service de médecins interne, hôpital Belle-Isle, Metz
14 Urgences néphrologiques et transplantation rénale, hôpital Tenon, AP–HP, Paris
15 Médecine interne et immunologie clinique, centre hospitalier universitaire F.-Mitterrand Dijon-Bourgogne, Dijon
16 Néphrologie, CHU Angers, Angers
17 Maladies infectieuses, CH Annecy Genevois, Épagny Metz-Tessy
18 Néphrologie-médecine interne, centre hospitalier de Valenciennes, Valenciennes
19 Laboratoire d’immunologie, hôpital Cochin, Paris
20 Fédération d’infectiologie, hôpital Cochin, Paris
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A378A379
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Les patients recevant des glucocorticoïdes et du rituximab (RTX) présentent un risque accru d’infections, en particulier d’infections invasives à pneumocoque. Les réponses vaccinales au virus de la grippe, au Streptococcus pneumoniae et au SARS-CoV-2 sous traitement par RTX sont fortement altérées. Chez les patients atteints de maladies auto-immunes recevant de tels traitements, en particulier ceux atteints de vascularites associées aux ANCA (AAV), il est donc nécessaire de développer des stratégies vaccinales anti-pneumococciques améliorées pour augmenter la réponse immunitaire et la protection vaccinale.
Patients et méthodes
Cet essai multicentrique de phase 2, randomisé, ouvert, a comparé deux stratégies innovantes de vaccin anti-pneumococcique « renforcées » au schéma de vaccination standard chez des patients atteints de VAA recevant un traitement par RTX. Des patients adultes atteints de VAA nouvellement diagnostiquée ou en rechute, présentant une maladie active (BVAS ≥ 3) et devant recevoir du RTX comme traitement d’induction (375 mg/m2/semaine pendant 4 semaines consécutives), ont été randomisés avec un rapport 1:1:1 dans trois bras parallèles : schéma standard associant une dose de vaccin pneumococcique conjugué 13-valent (PCV13) au jour 0 suivie d’une dose de vaccin non conjugué 23-valent (PPV23) au mois 5 (M5) (bras 1) ; double dose de PCV13 au jour 0 et au jour 7 suivie d’une dose de PPV23 à M5 (bras 2) ; ou 4 doses de PCV13 au jour 0 suivies d’une dose de PPV23 à M5 (bras 3). Le critère d’évaluation principal était la réponse immunitaire à M6 contre les 12 sérotypes de pneumocoque communs aux vaccins PCV13 et PPV23, classée selon quatre catégories ordonnées de réponse : réponse positive en anticorps contre 0–3, 4–6, 7–9 ou 10–12 sérotypes. Une réponse positive par sérotype était définie par un titre ELISA d’IgG spécifiques ≥ 1 μg/mL et une augmentation de deux fois par rapport au jour 0. Le critère d’évaluation primaire a été analysé dans un modèle de régression logistique à chances proportionnelles avec une correction de Bonferonni pour les 2 bras innovants. Les critères d’évaluation secondaires étaient les réactions locales et systémiques sollicitées 7 jours après chaque vaccination et tout événement indésirable lié ou pouvant être lié à l’immunisation vaccinale.
Résultats
Quatre-vingt-quinze participants ont été analysés dans la population modifiée en intention de traiter (âge moyen 60 ± 16,6 ans, 50 % d’hommes, 74 personnes atteintes d’une maladie nouvellement diagnostiquée, 66 d’une granulomatose avec polyangéite et 29 d’une polyangéite microscopique, BVAS moyen 15,3 ± 6,9), dont 30 affectés au bras 1, 32 au bras 2 et 33 au bras 3.
À M6, une réponse immunitaire contre 0–3, 4–6, 7–9 ou 10–12 sérotypes était observée chez 83,3 %, 13,3 %, 3,3 % et 0 % dans le bras 1 ; 56,3 %, 28,1 %, 15,6 % et 0 % dans le bras 2 ; et 60,6 %, 33,3 %, 6,1 % et 0 % dans le bras 3. Les patients du bras 2 étaient significativement plus susceptibles de se trouver dans une catégorie de réponse supérieures par rapport au régime standard après ajustement sur l’âge, avec un odds ratio proportionnel (pOR) de 4,1 (IC97,5 % : 1,1–15,9, p = 0,018), tandis que le bras 3 montrait une tendance non significative à améliorer les réponses vaccinales (pOR : 3,1, IC97,5 % : 0,8–11,9, p = 0,062). Une analyse de sensibilité sur une population per-protocole excluant les patients ayant subi des vaccinations ou des prises de sang hors des délais donnait des estimations concordantes.
Les réactions locales et/ou systémiques dans les 7 jours après chaque vaccination, et tout événement indésirable lié ou possiblement lié à la vaccination au cours des 6 premiers mois, sont survenus en plus grand nombre avec les schémas renforcés mais étaient principalement des réactions locales de grade 1 ou 2. Aucun événement indésirable grave lié à la vaccination n’a été observé.
Au cours du suivi, 8 poussées de vascularite sont survenues chez 6 patients, en médiane 87 jours après la dernière vaccination : un patient dans le bras 1, 2 dans le bras 2, et 3 dans le bras 3.
Conclusion
Chez les patients atteints de VAA recevant un traitement par RTX, une stratégie innovante de vaccination anti-pneumococcique renforcée, basée sur une double dose de PCV13 au jour 0 et au jour 7 suivie d’une dose unique de PPV23 à M5, améliore significativement les réponses en anticorps contre Streptococcus pneumoniae par rapport au schéma standard.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724759 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A378-A379 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.089 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)00881-5
10.1016/j.revmed.2022.10.193
Ca052
Maladie de Kikuchi-Fujimoto post-vaccin COVID-19 : à propos de 1 cas
Mekki S. ⁎
Ghribi M.
Bouattour Y.
Mouna S.
Frikha F.
Marzouk S.
Bahloul Z.
Service de médecine interne, faculté de médecine de Sfax, Sfax, Tunisie
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A441A442
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Avec le développement rapide de divers vaccins contre la maladie à coronavirus 2019 (COVID-19), des effets secondaires inhabituels ont été de plus en plus souvent signalés dans la littérature. La maladie de Kikuchi-Fujimoto (KF) ou la lymphadénite histiocytaire nécrosante est rarement observée après la vaccination. Trois cas de KF après la vaccination contre le COVID-19 ont été rapportés à ce jour. Nous rapportons un cas de KF post-vaccin COVID-19 en précisant les données cliniques, radiologiques et histologiques.
Observation
Il s’agit d’une patiente âgée de 60 ans, sans antécédents médicaux particuliers, qui a présenté à j15 de vaccin COVID-19 (Pfizer-BioNTech) une symptomatologie faite de signes généraux avec une asthénie, une anorexie et un amaigrissement chiffré à 4 kg sans fièvre concomitante à l’apparition d’adénopathies cervicales et axillaires gauches. Une écho-mammographie ainsi qu’une échographie cervicale étaient pratiqués qui ont montré la présence de mutliples adénopathies axillaires gauches dont la plus volumineuse mesurant 17 × 19 mm, des ganglions cervicaux bilatéraux sus et sous thyroïdiens, sus claviculaire et latérocervicale inférieure gauche. Une biopsie exérèse d’une adénopathie axillaire a conclu à une lymphadénite nécrosante de Kikuchi, sans signes de malignité ni granulome tuberculoïde. L’enquête infectieuse et immunologique était négative. Les anticorps antiprotéines S du SARS-COV2 étaient positifs à un titre > 250 UI/mL. La patiente était traitée par une corticothérapie à forte dose pendant 1 mois puis une dégression avec une bonne évolution clinique, biologique et radiologique.
Discussion
La maladie KF constitue une cause rare et bénigne d’adénopathies. C’est une entité anatomoclinique d’étiologie inconnue. La confirmation du diagnostic est toujours apportée par l’étude histologique ganglionnaire. Il est très rare que le KF survienne après une vaccination. La littérature décrit son apparition après la vaccination contre la grippe, le papillomavirus humain et l’encéphalite japonaise. Trois cas ont été décrits suite à la vaccination COVID-19 Pfizer, mais de nombreux articles ont décrit des aspects histologiques d’hyperplasie folliculaire réactionnelle. La KF a tendance à se manifester tardivement avec un intervalle allant de 10 à 35 jours. Il n’existe pas de traitement spécifique pour la KF, un traitement symptomatique avec des analgésiques est généralement suffisant, les stéroïdes sont parfois utilisés dans les cas les plus graves ou récurrents. Les 3 autres patients ont été traités par anti-inflammatoires non stéroïdiens.
Conclusion
La KF peut rarement survenir après la vaccination par COVID-19, les patients se présentent généralement avec une lymphadénopathie, de la fièvre et d’autres symptômes systémiques plus d’une semaine après l’application de la dose de vaccin. La présentation clinique initiale du KF peut être alarmante et imiter des infections, des maladies auto-immunes ou même des pathologies lymphprolifératives.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9724760 | NO-CC CODE | 2022-12-08 23:18:15 | no | Rev Med Interne. 2022 Dec 6; 43:A441-A442 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.193 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)00711-1
10.1016/j.revmed.2022.10.023
Co011
Impact de l’initiation d’un traitement par rituximab sur la réponse humoral à la vaccination anti-SARS-CoV-2 chez des patients ayant une maladie auto-immune et préalablement vaccinés
Oliosi E. 1
Flahault A. 2
Charre C. 3
Veyer D. 4
Combier A. 5
Lafont E. 6
Mouthon L. 7
Karras A. 8
Avouac J. 5
Terrier B. 9
Hadjadj J. 1⁎
1 Médecine interne, Cochin Port-Royal, Paris
2 Néphrologie, hôpital européen Georges-Pompidou, AP–HP, Paris
3 Virologie, hôpital Cochin, Paris
4 Virologie, hôpital européen Georges-Pompidou, AP–HP, Paris
5 Rhumatologie, hôpital Cochin, Paris
6 Médecine interne, hôpital européen Georges-Pompidou, AP–HP, Paris
7 Centre de référence des maladies systémiques auto-immunes rares, hôpital Cochin, Paris
8 Néphrologie, HEGP, 15, rue Louis-Blanc, 75015 Paris
9 Médecine interne, hôpital Cochin, rue du Faubourg-Saint-Jacques, Paris
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A332A332
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
L’infection à SARS-CoV-2 est plus sévère chez les patients ayant une maladie auto-immune traitée par rituximab. La réponse humorale à la vaccination anti-SARS-CoV-2 est fortement altérée chez les patients sous rituximab avec une production d’anticorps neutralisants corrélée au taux de lymphocytes B circulants. Notre objectif était d’évaluer l’impact de l’initiation d’un traitement par rituximab sur la réponse humorale au vaccin anti-SARS-CoV-2 chez des patients ayant une maladie auto-immune et préalablement vaccinés.
Patients et méthodes
Nous avons réalisé une étude rétrospective, descriptive, bicentrique, incluant des patients ayant été préalablement vaccinés par au moins 2 doses de vaccin anti-SARS-CoV-2 avec un taux d’anticorps anti-Spike protecteur (supérieur à 264 BAU/mL), et chez qui un traitement par rituximab était initié pour une maladie auto-immune, ou repris après plus d’un an d’interruption. Les critères de jugement principaux étaient l’évolution du taux d’anticorps anti-Spike à 3 mois et 6 mois après initiation du rituximab et la survenue d’une COVID-19. Un taux d’anticorps protecteur était défini par des IgG anti-Spike > 264 BAU/mL et une sérologie positive par un taux > 30 BAU/mL.
Résultats
Nous avons inclus 24 patients traités de novo par rituximab (17 femmes, âge médian 55 ans). Les pathologies les plus fréquentes étaient la polyarthrite rhumatoïde (29,2 %) et les vascularites associées aux ANCA (25 %). Le traitement le plus souvent associé était la corticothérapie orale (71 % des cas) avec une dose médiane de 17,5 (IQR : 6–40) mg/jour. Au moment de l’initiation du traitement par rituximab, 13 patients avaient reçu 2 doses de vaccin (54,2 %), 9 patients avaient reçu 3 doses (37,5 %) et 2 patients avaient reçu 4 doses (8,3 %). Le vaccin majoritaire était le BNT162B2 (83 %). Le délai médian entre la dernière dose de vaccin et l’administration du rituximab était de 35 jours (IQR : 15–54). Le taux d’anticorps était divisé par 3 à 3 mois de l’initiation du rituximab, et par presque 5 à 6 mois. À 3 mois, 13/14 (92,9 %) patients avaient un taux détectable d’anticorps anti-spike et 11/14 patients (78,6 %) avaient un taux protecteur. À 6 mois, 19/19 (100 %) des patients avaient un taux détectable et 13/19 (68,4 %) avaient un taux protecteur. Neuf patients ont reçu une dose supplémentaire de vaccin entre la première perfusion de rituximab et l’analyse à 6e mois sans impact significatif sur l’évolution du taux d’anticorps. Aucune variable n’était significativement associée à un taux protecteur d’anticorps à 6 mois, notamment les pathologies, les traitements associés et le taux d’IgG total. Cependant, une tendance vers des taux d’anticorps plus élevés à 6 mois était observée chez les patients ayant reçu 3 doses de vaccins avant le rituximab en comparaison avec ceux ayant reçu 2 doses. Trois patients ont développé une COVID-19 après initiation du traitement par rituximab, de forme modérée et sans nécessité d’hospitalisation dans tous les cas.
Conclusion
La baisse du taux d’anticorps anti-spike après initiation d’un traitement par rituximab semble rapide, mais comparable à la baisse constatée dans la population générale [1], avec néanmoins une majorité des patients gardant des taux protecteurs d’anticorps. Une fois le traitement par rituximab débuté, une dose de vaccin supplémentaire ne semblait pas permettre l’augmentation des taux d’anticorps.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
==== Refs
Référence
1 Zhang Z. Mateus J. Coelho C.H. Humoral and cellular immune memory to four COVID-19 vaccines Cell 185 14 2022 [2434–2451.e17]
| 0 | PMC9724761 | NO-CC CODE | 2022-12-08 23:18:16 | no | Rev Med Interne. 2022 Dec 6; 43:A332 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.023 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01065-7
10.1016/j.revmed.2022.10.377
Ca092
Troubles mnésiques persistants post-COVID-19 : intérêt de la recherche des anti-SARS-CoV-2 dans le LCR couplée à la spectro-IRM cérébrale ?
Zaouzaou L. 1⁎
Constans J.M. 2
Brochot E. 3
Bourgeois A.M. 4
Dernoncourt A. 1
Morain M. 1
Boulu X. 1
Karam J.D. 1
Duhaut P. 1
Schmidt J. 1
Salle V. 1
1 Médecine interne et recif, CHU Amiens, Amiens cedex
2 Radiologie, CHU Amiens, Amiens cedex
3 Virologie, CHU Amiens-Picardie Sud, Amiens
4 Biochimie, CHU Amiens, Amiens cedex
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A466A466
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
La maladie post-COVID-19 se développe quelle que soit la gravité de l’infection à SARS-CoV-2 initiale et comporte un large éventail de manifestations cliniques parmi lesquelles figurent de nombreux troubles neurologiques tels que les céphalées, l’anosmie, des troubles de la mémoire….
Observation
Nous rapportons le cas d’une patiente âgée de 68 ans adressée en consultation de médecine interne pour asthénie, troubles de la concentration et de la mémoire dans les suites d’une infection à SARS-CoV-2. Ses antécédents sont marqués par une HTA, un asthme, une thrombose veineuse profonde, une cholécystectomie, un œdème de Quincke avec choc anaphylactique aux AINS et une hypogammaglobulinémie modérée. On note une absence d’intoxication alcoolo-tabagique. La patiente a présenté en janvier 2021 une infection à SARS-CoV-2 avec fièvre et céphalées. La recherche de virus par PCR était négative lors de cet épisode avec secondairement une sérologie positive confirmant l’infection. Progressivement sont apparus des troubles de la mémoire et de la concentration. Son MMS était à 27/30.
L’IRM cérébrale en juin 2021 montrait de multiples hypersignaux de la substance blanche.
Un bilan a été réalisé en hôpital de jour avec réalisation d’une ponction lombaire.
Le LCR retrouvait l’absence de leucocytes (< 1 élément/mm3), des hématies à 24 éléments/mm3 et une protéinorachie à 0,45 g/L.
Les anticorps anti-rNMDA, anti-CASPR2, anti-rGABAb, anti-DPPX, anti-LGI1 et anti-AMPAr dans le LCR étaient négatifs.
Absence de bande surnuméraire à l’isofocalisation.
Les anticorps anti-SARS-CoV-2 dans le LCR étaient positifs à 108,7 UA/mL, la recherche de virus par PCR étant négative.
Le rapport de Delpech était normal à 0,52 mais n’excluant pas une synthèse intrathécale d’IgG. Par ailleurs, l’index albumine LCR/albumine sérum était normal.
Une polysomnographie ne retrouvait pas de syndrome d’apnées du sommeil.
Une spectroscopie par IRM cérébrale en mai 2022 a mis en évidence une réaction gliale ainsi qu’une dysfonction neuronale au niveau de l’hippocampe et de la protubérance.
Discussion
Les séquelles neuropsychologiques post-COVID-19 témoignent d’une neuro-inflammation liée à l’activation de la microglie et à une réaction auto-immune [1]. La persistance de troubles cognitifs associée à la présence d’anticorps anti-SARS-CoV-2 dans le LCR 6 mois après COVID-19 a été rapportée chez une patiente de 57 ans [2]. Dans notre observation, la présence de ces anticorps peut témoigner d’une neuro-inflammation constatée sur la spectroscopie par IRM cérébrale. Une exploration biologique plus approfondie du LCR ainsi qu’une exploration par spectroscopie par IRM cérébrale pourraient être suggérées chez les patients présentant des troubles cognitifs plusieurs mois après une infection à SARS-CoV-2.
Conclusion
D’autres études seront nécessaires afin de mieux définir la place de la recherche des anticorps anti-SARS-CoV-2 dans le LCR chez les patients ayant des séquelles neurologiques majeures liées au COVID-19.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
==== Refs
Références
1 Castanares-Zapatero D. Chalon P. Kohn L. Dauvrin M. Detollenaere J. Maertens de Noordhout C. Pathophysiology and mechanism of long COVID: a comprehensive review Ann Med 54 2022 1473 1487 35594336
2 Borsche M. Reichel D. Fellbrich A. Lixenfeld A.S. Rahmöller J. Vollstedt E.J. Persistent cognitive impairment associated with cerebrospinal fluid anti-SARS-CoV-2 antibodies six months after mild COVID-19 Neurol Res Pract 3 2021 34 34148546
| 0 | PMC9724766 | NO-CC CODE | 2022-12-08 23:18:16 | no | Rev Med Interne. 2022 Dec 6; 43:A466 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.377 | oa_other |
==== Front
Rev Med Interne
Rev Med Interne
La Revue De Medecine Interne
0248-8663
1768-3122
Published by Elsevier Masson SAS
S0248-8663(22)01062-1
10.1016/j.revmed.2022.10.374
Ca089
La réponse humorale aux vaccins anti-SARS-CoV-2 chez les patients suivis pour maladies immuno-inflammatoires
Bourguiba R. ⁎
Boudokhane M.
Myriam A.
Imen A.
Jomni T.
Syrine B.
Douggui M.H.
Médecine interne, hôpital des forces de sécurité intérieure, Tunis, Tunisie
⁎ Auteur correspondant.
6 12 2022
12 2022
6 12 2022
43 A465A465
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
La pandémie au SARS-CoV-2 a été responsable de milliers de décès de par le monde. La vaccination anti-SARS-CoV-2 a montré son efficacité contre les formes graves de l’infection par le SARS-CoV-2. Plusieurs études se sont intéressées à la réponse humorale et cellulaire aux vaccins anti-SARS-CoV-2 chez les patients suivis pour maladies auto-immunes et inflammatoires sous traitements immunosuppresseurs ou immunomodulateurs. Il a été démontré que les patients sous traitements anti-CD20 avaient une mauvaise réponse humorale au vaccin. Le but de notre était d’étudier la réponse humorale chez les patients sous immunosuppresseurs conventionnels et biothérapies.
Patients et méthodes
Les patients suivis pour maladies immuno-inflammatoires sous immunosuppresseurs ou immunomodulateurs ayant reçu au moins une dose de vaccin anti-SARS-CoV-2 ont été inclus. Une sérologie anti-SARS-CoV-2 (dosage IgM et IgM) VIDAS® a été prélevée. L’étude a été approuvée par le comité d’éthique local de l’hôpital des forces des sécurité de l’intérieur, La Marsa, Tunisie.
Résultats
Soixante patients ont été colligés, l’âge moyen était de 50,82 ± 12,39 ans. Les maladies immuno-inflammatoires étaient : la maladie Crohn (n = 28), polyarthrite rhumatoïde (n = 9), rectocolite hémorragique (n = 5), maladie de Behçet (n = 5), lupus érythémateux systémique (n = 3), syndrome de Sjögren (n = 3), sarcoïdose (n = 2), maladie de Takayasu (n = 1). Aucun traitement de fond n’a été arrêté après la vaccination. Dix-neuf patients étaient sous biothérapies : infliximab (n = 12), adalimumab (n = 3), etanercept (n = 2), ustekinumab (n = 1), tocilizimab (n = 1). Dix patients étaient sous combothérapie : anti-TNF et azathioprine (n = 10). Quarante-trois patients étaient sous immunosuppresseurs conventionnels : azathioprine (n = 18), méthotrexate (n = 16), corticoïdes > 10 mg/j (n = 12). Tous les patients avaient au moins reçu une dose de vaccin : le nombre de dose a été réparti comme suit : 1 dose (n = 7), deux doses (n = 33), 3 doses (n = 17), 4 doses (n = 3). Une infection par le SARS-CoV-2 étaient notées chez 21 patients. Les vaccins reçus étaient : BNT162b2 (n = 38), Sinovac/Sinopharme (n = 12), chAdOx1-S (n = 10), mRNA-1273 (n = 1) et Ad26.Cov2.S (n = 1).
Le délai moyen entre la réalisation de la sérologie et la dernière dose de la vaccination était de 6,38 ± 2,73 mois. Le taux moyen des IgG était de 40,41 ± 12,46 UI.
Conclusion
Notre étude a montré une réponse humorale significative après l’administration de la deuxième dose du vaccin anti-SARS-CoV-2, ce résultat est en accord avec les études rapportées dans la littérature. Braun et al. [1] ont montré sur une série de 264 patients suivis pour maladies inflammatoires ayant reçus 2 doses de vaccins BNT162b2, la réponse humorale étaient corrélée à la durée d’évolution de la maladie inflammatoire et le type du traitement reçu : le méthotrexate et les anti-CD20 entraînent une diminution de la réponse humorale, le traitement par les anti-JAK, anti-TNF et anti-IL ne semblent pas modifier le taux d’anticorps neutralisants [2].
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
==== Refs
Références
1 Haberman R.H. Herati R.S. Simon D. Samanovic M. Blank R.B. Tuen M. Methotrexate hampers immunogenicity to BNT162b2 mRNA COVID-19 vaccine in immune-mediated inflammatory disease [Internet] Rheumatology 2021 [cited 2022 Jun 30. Available from: http://medrxiv.org/lookup/doi/10.1101/2021.05.11.21256917]
2 Braun-Moscovici Y. Kaplan M. Braun M. Markovits D. Giryes S. Toledano K. Disease activity and humoral response in patients with inflammatory rheumatic diseases after two doses of the Pfizer mRNA vaccine against SARS-CoV-2 Ann Rheum Dis [Internet] 80 2021 1317 1321 [cited 2022 Jun 30. Available from: https://ard.bmj.com/lookup/doi/10.1136/annrheumdis-2021-220503] 34144967
| 0 | PMC9724770 | NO-CC CODE | 2022-12-08 23:18:16 | no | Rev Med Interne. 2022 Dec 6; 43:A465 | utf-8 | Rev Med Interne | 2,022 | 10.1016/j.revmed.2022.10.374 | oa_other |
==== Front
Virology
Virology
Virology
0042-6822
1096-0341
Elsevier Inc.
S0042-6822(22)00214-8
10.1016/j.virol.2022.12.003
Brief Communication
Long-term persistence of IgG antibodies in recovered COVID-19 individuals at 18 months post-infection and the impact of two-dose BNT162b2 (Pfizer-BioNTech) mRNA vaccination on the antibody response: Analysis using fixed-effects linear regression model
Dehgani-Mobaraki Puya a
Wang Chao b
Floridi Alessandro c
Floridi Emanuela c
Dawoodi Sunny d
Zaidi Asiya K. e∗
a Associazione Naso Sano, Umbria Regional Registry of Volunteer Activities, Via Luca Benincasa 2, San Mariano, 06073, Perugia, Italy
b Health & Social Care Statistic, Faculty of Health, Social Care and Education, Kingston University and St George's, University of London, London, SW17 0RE, UK
c Centro Ricerche Analisi Biochimico Specialistiche, Perugia, Italy
d RIPAS Hospital, Brunei
e Associazione Naso Sano, Italy
∗ Corresponding author.
6 12 2022
6 12 2022
2 9 2022
21 11 2022
2 12 2022
© 2022 Elsevier Inc. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
This era of emerging variants needs a thorough evaluation of data on the long-term efficacy of immune responses in vaccinated as well as recovered individuals, to understand the overall evolution of the pandemic. In this study, we aimed to assess the dynamics of IgG response over 18 months in n = 36 patients from the Umbria region in Italy, who had a documented history of COVID-19 infection in March 2020, and then compared the impact of two-dose BNT162b2 (Pfizer-BioNTech) vaccination on the antibody responses of these patients with the ones who did not receive any dose of vaccine. This is the longest observation (March 2020–September 2021) for the presence of antibodies against SARS-CoV-2 in recovered individuals along with the impact of 2 dose-BNT162b2 vaccination on these responses. Fixed-effect regression models were used for statistical analysis which could be also used to predict future titer trends. At 18 months, 97% participants tested positive for anti-NCP hinting towards the persistence of infection-induced immunity even for the vaccinated individuals. Our study findings demonstrate that while double dose vaccination boosted the IgG levels in recovered individuals 161 times, this “boost” was relatively short-lived. The unvaccinated recovered individuals, in contrast, continued to show a steady decline but detectable antibody levels. Further studies are required to re-evaluate the timing and dose regimen of vaccines for an adequate immune response in recovered individuals.
==== Body
pmc1 Introduction
This era of emerging variants needs a thorough evaluation of data on the long-term efficacy of immune responses in vaccinated as well as recovered individuals, to understand the overall pandemic evolution. Vaccine equity is an evolving global challenge that needs prompt alternate strategies for equal allocation across all countries in line with WHO's campaign for Vaccine Equity (Who.int, 2021). Recent data has suggested post-infection immunity to be durable, long-acting, and protective. (Gaebler et al., 2021; Abu-Raddad et al., 2021; Anand et al., 2021; Egbert et al., 2021; Turner et al., 2021; Gallais et al., 2021; Dehgani-Mobaraki et al., 2021). However, recovered individuals have also been advised to receive complete vaccination similar to naïve individuals with no previous history of COVID-19 infection. Whether two doses are required in recovered individuals or a single dose could offer adequate protection, is still a matter of debate. There are countries that have given three/four doses of vaccines to all individuals (recovered/naïve).
Real-time data indicates that recovered individuals have protective immunity that could be sufficiently enhanced by a single dose of the vaccine rather than a double dose. Unfortunately, neither recall responses nor ideal vaccine dosing regimens have been studied in them (Saadat et al., 2021; Ebinger et al., 2021). When compared to natural infection, vaccination elicits a more specific response which is greater in magnitude and is largely focused on the spike-receptor binding domain (S-RBD) rather than Nucleocapsid (NCP). Recent data suggests impaired protection, predominantly viral neutralization, and vaccine effectiveness against the variants of concern (Gregory Hames et al., 2021).
In this study, we aimed to assess the dynamics of IgG response over 18 months in patients who had a documented history of COVID-19 infection in March 2020, and then compared the impact of two-dose BNT162b2 (Pfizer-BioNTech) vaccination on the antibody responses of these patients with the ones who did not receive any dose of vaccine (NCT05038475). (Clinical and Immunological Responses After).
2 Materials and methods
2.1 Patient cohort
A monocentric longitudinal observational study was conducted in patients based in the Umbria region of Italy who had tested positive for SARS-CoV-2 in March 2020 by RT-qPCR, aimed to analyze the presence of antibodies against SARS-CoV-2. These RT-PCR tests were performed by the Local health regulatory authorities according to the national guidelines and standard operating protocols. The patients were managed as per the set protocols by the treating doctor, based on the severity of the disease. On recovery, all subjects were informed about this study for seroprevalence and were invited for voluntary participation. After detailed written informed consent, serological samples were collected at timed intervals and antibody levels were analyzed using the MAGLUMI® 2019-nCoV lgM/lgG chemiluminescent analytical system (CLIA) assay and the MAGLUMI® SARS-CoV-2 S-RBD IgG CLIA. (New Industries Biomedical Engineering Co., Ltd [Snibe], Shenzhen, China). Both these immunoassays were granted Emergency Use Authorization by the US Food and Drug Administration, anti-NCP during the initial months of the pandemic followed by anti-S-RBD later in 2020 (MAGLUMI. nCoV IgM, 2019).
Using a standardized questionnaire, the participants were asked to provide information about their COVID-19 clinical history along with symptoms and treatment undertaken. Although reminders were sent for follow-up periodically, participation in the study was completely voluntary. The study participants did not receive any compensation or any other benefit but were informed individually about their antibody status.
The blood samples were collected after informed consent by the patients and with the approval of the ethics committee of the Associazione Naso Sano (Document number ANS-2020/001) at an accredited lab (Laboratory of Nuclear Lipid BioPathology, CRABION, Perugia, Italy). Data collection and analysis were masked from the main principal investigator, who was also a part of the study sample to avoid observer bias (World Health Organization, 2020; Xydakis et al., 2020). The study was conducted in accordance with the Declaration of Helsinki and national and institutional standards.
3 Patient selection
Study timeline and assays used: The antibody levels were analyzed in two phases, based on the availability of immunoassays at the time, i.e from May 2020 to January 2021, antibodies against Nucleocapsid (NCP) were analyzed using the MAGLUMI® 2019-nCoV lgM/lgG CLIA assay through sequential serum samples. Time was treated as a factor and six different time points were defined; (T0-T5). The first blood sample was collected in May 2020, 2 months after infection (march), and was defined as T0. Consecutive serological samples were analyzed at three months (T1), five months (T2), seven months (T3), eight months (T4) ten months (T5) post-infection in June, August, October, November of 2020, and January 2021 respectively.
The second phase comprised of the introduction of a different immunoassay; MAGLUMI® SARS-CoV-2 S-RBD IgG CLIA to detect antibodies against Spike-Receptor Binding Domain from Feb 2021 to September 2021.
Analysis using both the immunoassays (anti-NCP and anti-S-RBD) for each patient of the study group could not be done throughout 18 months due to practical challenges associated with it. Italy was one of the first countries to be affected by COVID-19 in early 2020 and the anti-S-RBD immunoassay received emergency approval only later in 2020.
From the original cohort, in n = 24 individuals, anti-NCP was analyzed from May 2020 to January 2021 and then anti-S-RBD was analyzed from February 2021 to September 2021. For n = 12 (8 female and 4 male) participants, who met the eligibility criteria for participation (documented history of COVID-19 in March 2020) that were enrolled in this study in Feb 2021, only anti-S-RBD could be analyzed from February 2021 to September 2021 making the final sample size of n = 36 participants.
4 Introduction of the vaccinated subgroup
Since the legal provisions adopted by the Italian Ministry of Health advised vaccination, irrespective of previous disease status, n = 21 participants were gradually vaccinated from mid-March 2021. Out of these, n = 19 participants continued to follow up voluntarily for antibody titer assessment at monthly intervals.
Therefore at 18 months, the study participants were divided into two groups:
Group A: Recovered and received two doses of BNT162b2 vaccine (n = 21). From this recovered vaccinated group, n = 2 (P23 and P30 in the figure) failed to follow up post-vaccination making the final sample size of n = 19.
Group B: Recovered individuals who were unvaccinated (n = 15).
For the month of September 2021 (18 months post-infection), the immunoassay to detect anti-NCP was reintroduced so that even vaccinated individuals could be analyzed for the persistence of infection-induced immunity not generated by mRNA spike-based vaccination. Only qualitative assessment could be done for the presence of anti-NCP antibodies at this point.
Although the study sample is small, the presented data can be useful in providing information on the longer-term dynamics of relevant antibodies since the beginning of the pandemic and the impact of vaccination on antibody responses.
5 Analytical system used
For analysis at 18 months, the study participants were divided into two groups based on their vaccination status.
Group A: recovered and received two doses of BNT162b2 vaccine (n = 19) and Group B: recovered individuals who were unvaccinated (n = 15).
As per the Specifications, the anti-SARS-CoV-2 S-RBD IgG assay had a sensitivity of 100% with CI [99.9%–100.0%] at ≥15 days post symptom onset, and specificity of 99.6%; CI [98.7%–100.0%]. High-concentration samples were diluted automatically by analyzers and the recommended dilution was 1:9 with the diluent in the kit. The sample, buffer, and magnetic microbeads coated with S-RBD recombinant antigen were mixed thoroughly and incubated, forming immune complexes. After precipitation, decanting of supernatant, and performing a wash cycle, ABEI labeled with anti-human IgG antibody was added, and incubated to form complexes. Again after precipitation in a magnetic field, decanting of supernatant, and performing another wash cycle, the Starter 1 + 2 were added to initiate a chemiluminescent reaction. The light signal was measured by a photomultiplier as relative light units (RLUs), which is proportional to the concentration of SARS-CoV-2 S-RBD IgG presented in the sample. The measurements and interpretation of results were made according to the manufacturer's instructions. The analyzer automatically calculates the concentration in each sample using a calibration curve which is generated by a 2-point calibration master curve procedure. The results were expressed in AU/mL. A result less than 1.00 AU/mL (<1.00 AU/mL) was considered to be non-reactive while a result greater than or equal to 1.00 AU/mL (≥1.00 AU/mL) was considered to be reactive (https://www.fda.gov/medical-devices/coronavirus-disease-2019-covid-19-emergency-use-authorizations-medical-devices/vitro-diagnostics-euas., 2019; https://www.snibe.com/zh_en/en_newsView.aspx?id = 647).
6 Statistical analysis
To model how antibody response evolves over time, we used a fixed-effects linear regression while adjusting for the testing method/immunoassay used (NCP or S-RBD), vaccination status, and temporal effects. To model the non-linear effects of time, a restricted cubic spline function for time (month) was applied in the regression model using the method by Orsini and Greenland (2011) to visualize and interpret the results (Orsini and Greenland, 2011). To account for serial correlation, the error term is modeled as a first-order autoregression (AR) process.
7 Results
Fig. 1 shows how the IgM and IgG levels have evolved over time for the recovered subjects and the impact of two-dose BNT162b2 (Pfizer-BioNTech) vaccination on them.Fig. 1 Fig. 1 shows how the IgM and IgG levels have evolved over 18 months for n = 21 of Group A: recovered subjects and the impact of two-dose BNT162b2 (Pfizer-BioNTech) vaccination on the antibody responses, where Fig. 1(a) shows the IgM titer trend (red) and Fig. 1(b) shows the IgG titer trend (red line) over 18 months in each subject as per the patient number that was assigned randomly. The blue vertical line represents the time of vaccination. From this recovered vaccinated group, n = 2 (P23 and P30 in the figure) failed to follow up post-vaccination.
Fig. 1
Fig. 1(a) shows the IgM response (red) over 18 months in each subject as per the patient number that was assigned randomly. The blue vertical line represents the time of vaccination. Fig. 1(b) shows the IgG response (log scale) (red) over 18 months in each subject as per the patient number that was assigned randomly. The blue vertical line represents the time of vaccination. The graphs demonstrate that vaccination has an enormous impact on antibody responses. It was interesting to note that, although there was a rapid and steep rise in the antibody responses immediately post-vaccination, this rise was short-lived. This explains the role of boosters that might, perhaps “boost” the titer levels for a brief duration. Whether the titer trend after vaccination drops back to pre-vaccination levels in a few weeks or in fact, lowers below the baseline needs further analysis on a larger population.
Fig. 2(a) visualizes the modeling results for the four scenarios, based on immunoassay employed (anti-NCP or anti-S-RBD) and completion of two doses of BNT162b2 (Pfizer–BioNTech) mRNA vaccine: 1) unvaccinated and tested against NCP; 2) vaccinated and tested against NCP; 3) unvaccinated and tested against S-RBD; and 4) vaccinated and tested against S-RBD. IgG values were log-transformed to avoid negative values.Fig. 2 Fig. 2(a) visualizes the modeling results for the four scenarios, based on immunoassay employed (anti-NCP or anti-S-RBD) and completion of two doses of BNT162b2 (Pfizer–BioNTech) mRNA vaccine: 1) unvaccinated and tested against NCP; 2) vaccinated and tested against NCP; 3) unvaccinated and tested against S-RBD; and 4) vaccinated and tested against S-RBD. IgG levels were log-transformed to avoid negative values. Fig. 2(b) describes the differences in antibody responses between a given month and the baseline month (month 2), holding all the other variables (vaccination status and testing method) constant and Fig. 2(c) describes the modeling results to determine the effect of vaccination on antibody levels (IgM and IgG) in recovered subjects. The IgM levels against NCP are lower, by an amount of 0.324, relative to spike-RBD IgM levels. Similarly, IgG levels against NCP are lower than spike-RBD IgG levels by a factor of 0.119. The vaccines lead to a large increase in antibody responses: vaccination increases IgM by amount of 0.773, and IgG by 161 times.
Fig. 2
Fig. 2(b) describes the differences in antibody responses between a given month and the baseline month (month 2), holding all the other variables (vaccination status and testing method) constant. This allows visualization of antibody response dynamics regardless of other factors since this model assumes an independent temporal effect. Statistical models (especially Fixed-effect regression models) are flexible to diminish the impact of confounding by (unmeasured) fixed/time-invariant factors.
Fig. 2(c) describes the modeling results to determine the effect of vaccination on antibody responses (IgM and IgG) in recovered subjects. IgM responses against NCP are lower by an amount of 0.324, relative to the response against spike-RBD. Similarly, IgG response against NCP is lower than that against spike-RBD by a factor of 0.119.
A double-dose vaccination with BNT162b2 in recovered individuals induced an enormous but short-lived increase in antibody levels. While the IgM response increased by an amount of 0.773, the IgG response increased 161 times.
However, we do agree that a caveat should be noted that the model assumed that anti-S-RBD responses only affect the level but not the trend/shape of the curve relative to anti-NCP. Further analysis of the interaction between time and type of antibody is not statistically significant which supports our original model.
It was interesting to note that all participants from Group B (n = 15) tested positive for anti-S-RBD for 18 months post-infection, with no case reported of reinfection despite multiple waves of mutant strains encountered in the region since March 2020. Also, 33 out of the 34 participants (97%) tested positive for anti-NCP at 18 months. We hypothesize that anti-NCP immunoassays could be a valuable tool for the assessment of persisting infection-induced immunity and should be re-adopted for analysis even for vaccinated individuals. In terms of strain, whether the infection with the wild-type conferred a more robust protective immunity against re-infection as compared to infection with other mutant strains needs further large-scale analysis.
8 Discussion
In this study, we longitudinally analyzed the antibody responses in patients that recovered from COVID -19, 18 months after SARS-CoV-2 infection (wild type), and divided them into two groups for comparison based on vaccination with two doses of BNT162b2. Group A: Recovered and received two doses of the BNT162b2 vaccine (n = 21). From this recovered vaccinated group, n = 2 (P23 and P30 in the figure) failed to follow up post-vaccination making the final sample size of n = 19, and Group B: Recovered individuals who were unvaccinated (n = 15).
To our knowledge, at the time of writing this manuscript, this is one of the longest observational studies that report the persistency of antibodies against SARS-CoV-2 at 18 months and the impact of 2 doses of BNT162b2 vaccine in seropositive convalescent patients.
Our observations have been in line with previously published studies suggesting the presence and persistence of long-term protective immunity after SARS-CoV-2 infection. A study by Gaebler et al. concluded that “memory B cell response to SARS-CoV-2 evolves between 1.3 and 6.2 months after infection in a manner that is consistent with antigen persistence” (Gaebler et al., 2021). The famous Qatar study by Abu-Raddad LJ et al., observed the estimated efficacy of immunity against reinfection to be 95% over 7 months (Abu-Raddad et al., 2021). This finding was similar to the findings by Anand et al. that anti-S-RBD IgM in plasma decayed rapidly, whereas the reduction of IgG was less prominent up to 8 months (Anand et al., 2021). Thereafter, a study was conducted on 3015 healthcare workers by Egbert ER. et al. demonstrating spike antibodies to SARS-CoV-2 to be durable for up to 10 months after natural infection (Egbert et al., 2021). Later, Turner et al. (n = 77, mild infection) observed that anti-S-IgG was detectable at 11 months. This could be attributed to the fact that S-binding bone marrow plasma cells (BMPCs) are quiescent, indicating that they are part of a long-lived compartment (Turner et al., 2021). When it comes to reduction in risk of infection, these antibodies might offer protection for up to 13 months (Gallais et al., 2021). Our previous study for infection-acquired immunity against SARS-CoV-2 at 14 months has been published (Dehgani-Mobaraki et al., 2021).
Vaccination-induced immunity in individuals with no history of previous infection has been relatively short-lived (half-life of 69–173 days) with requirements of boosters to augment antibody responses against subsequent variants of concern, especially the Omicron variant (Levin et al., 2021). This variant harbors more mutations compared to prior variants, and therefore “efficiently” escapes humoral immunity induced by primary vaccination. It was observed by Garcia-Beltran et al. that additional mRNA vaccine doses (boosters) tend to demonstrate cross-neutralizing responses against Omicron as well. This could be due to “affinity maturation of the existing antibodies” or new shared epitopes of variants serving as targets (Garcia-Beltran et al., 2022a). The authors also observed a 9-fold increase in the geometric mean neutralization titer (GMNT) for Delta pseudoviruses in individuals that received three-doses of BNT162b as compared to two doses, and an increase in cross-neutralization of the Omicron variant by 27-fold for individuals who received three doses of BNT162b vaccine. This neutralization signal could be correlated with antibody levels against SARS-CoV-2 spike-RBD measured by anti-spike-RBD immunoassays that are widely available (Garcia-Beltran et al., 2021). Therefore, our focus on antibody levels against spike-RBD is valuable.
Our study findings demonstrate that while double-dose vaccination boosted the IgG levels in recovered individuals 161 times, this “boost” was relatively short-lived. The unvaccinated recovered individuals, in contrast, continued to show a steady decline but detectable antibody levels. Whether or not the present vaccines induce sterilizing immunity remains unclear. They trigger serological IgA response but do not generate mucosal IgA at the virus site of entrance. Vaccine breakthrough infections may be because of deficient local preventive immunity due to absence of mucosal IgA (Piano Mortari et al., 2021).
Upon mRNA BNT162b2 vaccination, neutralizing antibodies are boosted significantly in the serum, but not at the mucosal sites and saliva. This indicates poor activation of oral mucosal immunity that fails to limit the viral attachment to target cells at the entry site (Azzi et al., 2021).
In contrast, convalescent individuals that recovered from COVID-19 infection (also known as hybrid immunity) demonstrated the presence of specific IgA indicating that natural infection conferred mucosal immunity.[22,23 24] (Piano Mortari et al., 2021; Azzi et al., 2021; Sterlin et al., 2021).
Vaccination-induced immunity in individuals with a history of prior infection was associated with high levels of wild-type neutralization titers even when vaccinated distantly. (>6 months) (Garcia-Beltran et al., 2022b).
In line with a study by Hall et al., our study results demonstrate that a previous COVID-19 infection (infection-acquired immunity) appeared to elicit robust, long-term, and sustained levels of SARS-CoV-2 antibodies in individuals who received two doses of BNT162b2. It was also observed that individuals with hybrid immunity developed the “highest and most durable protection”. (Hall. et al., 2021).
Our study had some limitations. First, a small sample size. Second, ideally, simultaneous antibody titer detection of each patient at each time point using both NCP and S-RBD assay would have given the best results for comparative analysis but the S-RBD assay received emergency approval only later in 2020.
The nucleocapsid protein (NCP) of the SARS-CoV-2 is considered a biomarker associated with natural exposure. Since this NCP is not present in spike-based vaccines, there is no vaccine-induced response against it in vaccinated individuals. The type of antibodies produced and the type of cell-mediated immune response triggered depends on the mechanism of virus/vaccine product exposure (infection or vaccination). For example, mRNA and adenovirus vector vaccines induce antibodies against the spike protein of SARS-CoV-2; therefore a positive test for spike protein specific IgM or IgG in vaccinated individuals could indicate prior infection or vaccination. To differentiate between these two situations, a positive Nucleocapsid-protein assay could be used to indicate prior infection. Therefore, testing for anti-NCP could be an important tool to analyze the type of existing immunity in an individual (vaccine-induced/natural/hybrid). Moreover, Vaccinated individuals who were previously exposed usually test positive for antibodies against NCP (hybrid immunity) while the vaccinated individuals who have never been exposed lack anti-NCP antibodies (vaccine-induced immunity) (Assis et al., 2021; Krutikov et al., 2022). The most recent study in 2022, by Di Chiara et al. has also suggested that anti–S-RBD antibodies persist up to 18 months after infection, even in children bringing hope for the future of antibody dynamics and protection against reinfection.
The major strengths of our study include the longest observation (March 2020–September 2021) for the presence of antibodies against SARS-CoV-2 in recovered individuals along with the impact of 2 dose-BNT162b2 vaccination on these levels. Secondly, mathematical models: fixed-effect regression models were used for statistical analysis that diminishes the impact of confounding by (unmeasured) fixed/time-invariant factors. These modeling results could also be used to predict future titer trends. Thirdly, since the immunoassay to detect anti-NCP was reintroduced at 18 months, even the vaccinated individuals could be analyzed for the persistence of infection-induced immunity not generated by mRNA spike-based vaccination. Finally, combining the data for both was necessary because this makes it possible to explore the differences between anti-NCP and anti-S-RBD (as also shown by the regression results in Fig. 2(c)). This is similar to an interrupted time series (ITS) analysis which typically looks at how an ‘intervention’ affects the temporal pattern between before and after periods using whole datasets. Our data is rich in that it is both cross-sectional (multiple patients) and time-series which allows us to adopt complex longitudinal models.
9 Conclusion
This work provides further evidence of long-term persistence of immune response in individuals recovered from SARS-CoV-2 wild-type infection and the impact of two-dose BNT162b2 (Pfizer-BioNTech) mRNA vaccination on these antibody responses which could help in vaccination strategies and prevention policies.
CRediT authorship contribution statement
Puya Dehgani-Mobaraki: Conceptualization, Writing – review & editing. Chao Wang: Software, Methodology, Formal analysis, Visualization. Alessandro Floridi: Methodology, Investigation, Resources. Emanuela Floridi: Methodology, Investigation, Resources. Sunny Dawoodi: Writing – review & editing. Asiya K. Zaidi: Data curation, Writing – original draft, preparation, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
Data will be made available on request.
Acknowledgment
The authors would like to acknowledge the assistance provided by the Rotary Club Fortebraccio Montone for the successful conduction of this research study.
==== Refs
References
Abu-Raddad L.J. Chemaitelly H. Coyle P. Malek J.A. Ahmed A.A. Mohamoud Y.A. Younuskunju S. Ayoub H.H. Al Kanaani Z. Al Kuwari E. Butt A.A. Jeremijenko A. Kaleeckal A.H. Latif A.N. Shaik R.M. Abdul Rahim H.F. Nasrallah G.K. Yassine H.M. Al Kuwari M.G. Al Romaihi H.E. Al-Thani M.H. Al Khal A. Bertollini R. SARS-CoV-2 antibody-positivity protects against reinfection for at least seven months with 95% efficacy EClinicalMedicine 35 2021 100861 10.1016/j.eclinm.2021.100861 May Epub 2021 Apr 28. PMID: 33937733; PMCID: PMC8079668
Anand S.P. Prévost J. Nayrac M. Longitudinal analysis of humoral immunity against SARS-CoV-2 Spike in convalescent individuals up to 8 months post-symptom onset Cell Rep Med 2 6 2021 100290 10.1016/j.xcrm.2021.100290
Assis R. Jain A. Nakajima R. Distinct SARS-CoV-2 antibody reactivity patterns elicited by natural infection and mRNA vaccination npj Vaccines 6 2021 132 10.1038/s41541-021-00396-3 34737318
Azzi L. Dalla Gasperina D. Veronesi G. Shallak M. Ietto G. Iovino D. Baj A. Gianfagna F. Maurino V. Focosi D. Maggi F. Ferrario M.M. Dentali F. Carcano G. Tagliabue A. Maffioli L.S. Accolla R.S. Forlani G. Mucosal immune response in BNT162b2 COVID-19 vaccine recipients EBioMedicine 75 2021 103788 10.1016/j.ebiom.2021.103788 Dec 23 Epub ahead of print. PMID: 34954658; PMCID: PMC8718969
Clinical and Immunological Responses after SARS-CoV-2 Infection Causing COVID-19 - Full Text View 2020 ClinicalTrials.gov
Dehgani-Mobaraki P. Zaidi A.K. Yadav N. Floridi A. Floridi E. Longitudinal observation of antibody responses for 14 months after SARS-CoV-2 infection Clin. Immunol. 230 2021 108814 10.1016/j.clim.2021.108814 Sep Epub 2021 Jul 31. PMID: 34343708; PMCID: PMC8325385
Di Chiara C. Cantarutti A. Costenaro P. Long-term immune response to SARS-CoV-2 infection among children and adults after mild infection JAMA Netw. Open 5 7 2022 e2221616 10.1001/jamanetworkopen.2022.21616
Ebinger J.E. Fert-Bober J. Printsev I. Antibody responses to the BNT162b2 mRNA vaccine in individuals previously infected with SARS-CoV-2 Nat. Med. 27 2021 981 984 10.1038/s41591-021-01325-6 33795870
Egbert E.R. Xiao S. Colantuoni E. Durability of spike immunoglobin G antibodies to SARS-CoV-2 among health care workers with prior infection JAMA Netw. Open 4 8 2021 e2123256 10.1001/jamanetworkopen.2021.23256
Gaebler C. Wang Z. Lorenzi J.C.C. Muecksch F. Finkin S. Evolution of antibody immunity to SARS-CoV-2 Nature 591 7851 2021 639 644 10.1038/s41586-021-03207-w Mar Epub 2021 Jan 18. PMID: 33461210; PMCID: PMC8221082 33461210
Gallais F. Gantner P. Bruel T. Velay A. Planas D. Wendling M.J. Bayer S. Solis M. Laugel E. Reix N. Schneider A. Glady L. Panaget B. Collongues N. Partisani M. Lessinger J.M. Fontanet A. Rey D. Hansmann Y. Kling-Pillitteri L. Schwartz O. De Sèze J. Meyer N. Gonzalez M. Schmidt-Mutter C. Fafi-Kremer S. Evolution of antibody responses up to 13 months after SARS-CoV-2 infection and risk of reinfection EBioMedicine 71 2021 103561 10.1016/j.ebiom.2021.103561 Sep Epub 2021 Aug 27. PMID: 34455390; PMCID: PMC8390300
Garcia-Beltran W.F. Lam E.C. Astudillo M.G. Yang D. Miller T.E. Feldman J. Hauser B.M. Caradonna T.M. Clayton K.L. Nitido A.D. Murali M.R. Alter G. Charles R.C. Dighe A. Branda J.A. Lennerz J.K. Lingwood D. Schmidt A.G. Iafrate A.J. Balazs A.B. COVID-19-neutralizing antibodies predict disease severity and survival Cell 184 2 2021 476 488 Jan 21 33412089
Garcia-Beltran W.F. St Denis K.J. Hoelzemer A. Lam E.C. Nitido A.D. Sheehan M.L. Berrios C. Ofoman O. Chang C.C. Hauser B.M. Feldman J. Roederer A.L. Gregory D.J. Poznansky M.C. Schmidt A.G. Iafrate A.J. Naranbhai V. Balazs A.B. mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant Cell 2022 10.1016/j.cell.2021.12.033 Jan 6:S0092-8674(21)01496-3 Epub ahead of print. PMID: 34995482; PMCID: PMC8733787
Garcia-Beltran W.F. St Denis K.J. Hoelzemer A. Lam E.C. Nitido A.D. Sheehan M.L. Berrios C. Ofoman O. Chang C.C. Hauser B.M. Feldman J. Roederer A.L. Gregory D.J. Poznansky M.C. Schmidt A.G. Iafrate A.J. Naranbhai V. Balazs A.B. mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant Cell 2022 10.1016/j.cell.2021.12.033 Jan 6:S0092-8674(21)01496-3 Epub ahead of print. PMID: 34995482; PMCID: PMC8733787
Gregory M. Hames Thomas Does infection with or vaccination against SARS-CoV-2 lead to lasting immunity? Lancet Respir. Med. 2021 10.1016/S2213-2600(21)00407-0 Published Online October 21, 2021
Hall V. Foulkes S. Insalata F. Effectiveness and durability of protection against future SARS-CoV-2 infection conferred by COVID-19 vaccination and previous infection; findings from the UK SIREN prospective cohort study of healthcare workers March 2020 to September 2021 medRxiv 11 29 2021 21267006 10.1101/2021.11.29.21267006
https://www.fda.gov/medical-devices/coronavirus-disease-2019-covid-19-emergency-use-authorizations-medical-devices/vitro-diagnostics-euas
https://www.snibe.com/zh_en/en_newsView.aspx?id=647
Krutikov M. Palmer T. Tut G. Fuller C. Azmi B. Giddings R. Shrotri M. Kaur N. Sylla P. Lancaster T. Irwin-Singer A. Hayward A. Moss P. Copas A. Shallcross L. Prevalence and duration of detectable SARS-CoV-2 nucleocapsid antibodies in staff and residents of long-term care facilities over the first year of the pandemic (VIVALDI study): prospective cohort study in England Lancet Healthy Longev 3 1 2022 e13 e21 10.1016/S2666-7568(21)00282-8 Jan Epub 2021 Dec 16. PMID: 34935001; PMCID: PMC8676418 34935001
Levin E.G. Lustig Y. Cohen C. Fluss R. Indenbaum V. Amit S. Doolman R. Asraf K. Mendelson E. Ziv A. Rubin C. Freedman L. Kreiss Y. Regev-Yochay G. Waning immune humoral response to BNT162b2 covid-19 vaccine over 6 months e84 N. Engl. J. Med. 385 24 2021 10.1056/NEJMoa2114583 Epub 2021 Oct 6 PMID: 34614326; PMCID: PMC8522797
MAGLUMI. nCoV IgM/IgG - Letter of Authorization 2019 fda.gov (Google Scholar)
Orsini N. Greenland S. A procedure to tabulate and plot results after flexible modeling of a quantitative covariate STATA J. 11 1 2011 1 29
Piano Mortari E. Russo C. Vinci M.R. Terreri S. Highly specific memory B cells generation after the 2nd dose of BNT162b2 vaccine compensate for the decline of serum antibodies and absence of mucosal IgA Cells 10 10 2021 2541 10.3390/cells10102541 Sep 26 PMID: 34685521; PMCID: PMC8533837 34685521
Saadat S. Rikhtegaran Tehrani Z. Logue J. Binding and neutralization antibody titers after a single vaccine dose in health care workers previously infected with SARS-CoV-2 JAMA 325 14 2021 1467 1469 10.1001/jama.2021.3341 33646292
Sterlin D. Mathian A. Miyara M. IgA dominates the early neutralizing antibody response to SARS-CoV-2 Sci. Transl. Med. 13 577 2021 eabd2223 10.1126/scitranslmed.abd2223
Turner J.S. Kim W. Kalaidina E. Goss C.W. Rauseo A.M. Schmitz A.J. Hansen L. Haile A. Klebert M.K. Pusic I. O'Halloran J.A. Presti R.M. Ellebedy A.H. SARS-CoV-2 infection induces long-lived bone marrow plasma cells in humans Nature 595 7867 2021 421 425 10.1038/s41586-021-03647-4 Jul Epub 2021 May 24. PMID: 34030176 34030176
Who.int Vaccine equity declaration [online] Available at: https://www.who.int/campaigns/vaccine-equity/vaccine-equity-declaration> 2021
World Health Organization International guidelines for certification and classification (coding) of COVID-19 as cause of death Available from: https://www.who.int/classifications/icd/Guidelines_Cause_of_Death_COVID-19-20200420-EN.pdf?ua=1 Document Number: WHO/HQ/DDI/DNA/CAT 2020 June 1
Xydakis M.S. Dehgani-Mobaraki P. Holbrook E.H. Smell and taste dysfunction in patients with COVID-19 Lancet Infect. Dis. 20 9 2020 1015 1016 10.1016/S1473-3099(20)30293-0 Sep Epub 2020 Apr 15. PMID: 32304629; PMCID: PMC7159875 32304629
| 36516688 | PMC9725186 | NO-CC CODE | 2022-12-13 23:16:45 | no | Virology. 2023 Jan 6; 578:111-116 | utf-8 | Virology | 2,022 | 10.1016/j.virol.2022.12.003 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00025
10.1097/INF.0000000000003733
3
Letters to the Editor
Frequency of SARS-CoV-2 Positivity Among Children Presenting With Gastroenteritis in Emergency Department
Corso Chiara Maria MD 1
Pediatric Emergcy Department Fondazione IRCCS Ca’ Granda, Ospedale Maggiore Policlinico, Department of Clinical Sciences and Community Health, Università degli Studi di Milano, Milan, Italy
Marchisio Paola MD [email protected]
Pediatric Highly Intensive Care Unit, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Milan, Italy
Agostoni Carlo MD 3
Department of Clinical Sciences and Community Health, Università degli Studi di Milano, Pediatric Intermediate Care Unit, Fondazione IRCCS Ca’ Granda, Ospedale Maggiore Policlinico, Milan, Italy
Casazza Giovanni MD [email protected]
Department of Clinical Sciences and Community Health, Università degli Studi di Milano, Milan, Italy
Costantino Giorgio MD [email protected]
Department of Clinical Sciences and Community Health, Università degli Studi di Milano, Emergency Department Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy
Milani Gregorio P. MD 6
Pediatric Emergcy Department Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Department of Clinical Sciences and Community Health, Università degli Studi di Milano, Milan, Italy
Address for Correspondence: Carlo Agostoni, Fondazione IRCCS Ca’ Granda, Ospedale Maggiore Policlinico, via della Commenda 9, 20122, Milan, Italy. E-mail: [email protected].
11 10 2022
1 2023
11 10 2022
42 1 e38e39
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
STATUSONLINE-ONLY
==== Body
pmcTo the Editors:
SARS-CoV-2 infection typically presents with respiratory symptoms, such as rhinitis, cough and dyspnea.1 Although children affected by SARS-CoV-2 might also present uniquely with vomiting or diarrhea,2 the frequency of SARS-CoV-2 positivity among children presenting with an acute gastroenteritis is currently unknown.3,4 The primary aim of this study was to investigate the frequency of SARS-CoV-2 positivity among children with gastroenteritis. The secondary aims were to compare the frequency of SARS-CoV-2 positivity among children with gastroenteritis, with an acute respiratory disease or without any symptom of infection. At the emergency department (ED) of the Fondazione Ca’ Granda Policlinico, Milan, Italy, children presenting with gastroenteritis or a respiratory disease undergo a nasopharyngeal swab for the detection of SARS-CoV-2 by a molecular test. Children requiring an immediate hospitalization are also tested, regardless of their symptoms.5 For this study, we included all children <18 years of age visiting the ED from December 21, 2020, to March 20, 2022, with gastroenteritis (ie, >3 episodes of vomiting or diarrhea in 24 hours). For each child with gastroenteritis, one child with a respiratory disease (rhinitis, cough or dyspnea) and one child requiring hospitalization for conditions without any symptom of infection (eg, trauma) during the same period were included. These children were matched for age (±6 months) and encounter date (±10 days) with those presenting with gastroenteritis.
Subjects with both gastrointestinal and respiratory symptoms or without a SARS-CoV-2 swab test were excluded. Children reporting a close contact with a subject with SARS-CoV-2 in the previous 7 days were also excluded. Information on age, sex, date of encounter and the presence of fever (≥37.5 °C) was retrospectively extracted.
Data are given as median and interquartile range or frequency, percentage and 95% confidence interval (CI). The χ2-test was used to compare the frequency of SARS-CoV-2 positivity among the 3 study groups.
In the study period, 423 children presented to the ED with acute gastroenteritis. The molecular test for SARS-CoV-2 was not available in 13 children. Therefore, 410 children [3.7 (1.4–8.2) years] with gastroenteritis, 410 with a respiratory disease and 410 hospitalized without any infectious symptom were included (Table 1). Fever was more frequent (P < 0.001) in children with respiratory disease than in those with gastroenteritis. The prevalence of SARS-CoV-2 positivity was similar in children with gastroenteritis and with a respiratory disease [8.0% (5.6%–11%) vs. 6.6% (4.4%–9.4%), respectively; P = 0.5], and lower in those without any infectious symptom [3.2% (1.7%–5.4%); P = 0.004]. Among children with fever, those with gastroenteritis and those with a respiratory disease had a similar frequency of SARS-CoV-2 positivity [8.8% (5.3%–14%) vs. 6.0% (3.5%–9.6); P = 0.3].
TABLE 1. Characteristics of Children With Acute Gastroenteritis, Respiratory Disease or Without Any Symptom of Infection Admitted to the ED Between December 21, 2020, and March 20, 2022
Gastroenteritis Respiratory Disease No Symptom
N 410 410 410
Age, y 3.9 [1.5–8.3] 3.4 [1.4–6.8] 3.9 [1.2–8.7]
Male 239 (58) 231 (56) 243 (59)
Fever 205 (50) 265 (65)* 0
SARS-CoV-2 positivity 33 (8.0)§ 27 (6.6)‡ 13 (3.2)†
Children of the 3 groups are matched for age and encounter date. Data are presented as median and interquartile range or frequency and percentage (%).
* P < 0.001 gastroenteritis vs. respiratory disease.
‡ P = 0.5 gastroenteritis vs. respiratory disease.
§ P = 0.004 gastroenteritis vs. no symptom.
† P = 0.034 respiratory disease vs. no symptom.
These data suggest that the frequency of SARS-CoV-2 is similar in cases with acute gastroenteritis and with a respiratory disease. Furthermore, both children with respiratory disease and gastroenteritis are more frequently SARS-CoV-2 positive than children without any symptom of infection. Although data were retrospectively collected in a single center, this study has potential implications for clinicians and health policymakers and helps assessing the pretest probability of SARS-CoV-2 positivity among children in ED.
The study was partially supported by a grant of the Italian ministry of Health (Ricerca Corrente 2020).
The authors have no conflicts of interest to disclose.
The funding agencies had no role in study design, data collection, analysis or interpretation, or writing of the report
All authors contributed to the study design. C.M.C. collected data. G.C. performed data analysis, P.M., G.C., C.A. and G.P.M. contributed to data interpretation. G.P.M. wrote the first draft of the manuscript. All authors reviewed the first draft of the manuscript. All authors approved the manuscript as submitted.
==== Refs
REFERENCES
1. Yonker LM Neilan AM Bartsch Y . Pediatric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2): clinical presentation, infectivity, and immune responses. J Pediatr. 2020;227 :45–52.e5. doi:10.1016/j.jpeds.2020.08.037.32827525
2. Sayed IA Bhalala U Strom L . Gastrointestinal manifestations in hospitalized children with acute SARS-CoV-2 infection and multisystem inflammatory condition: an analysis of the VIRUS COVID-19 registry. Pediatr Infect Dis J. 2022;41 :751–758. doi:10.1097/INF.0000000000003589.35622434
3. Dipasquale V Passanisi S Cucinotta U . Implications of SARS-COV-2 infection in the diagnosis and management of the pediatric gastrointestinal disease. Ital J Pediatr. 2021;47 :71. doi:10.1186/s13052-021-01020-9.33761992
4. Galanopoulos M Gkeros F Doukatas A . COVID-19 pandemic: pathophysiology and manifestations from the gastrointestinal tract. World J Gastroenterol. 2020;26 :4579–4588. doi:10.3748/wjg.v26.i31.4579.32884218
5. Milani GP Bottino I Rocchi A . Frequency of children vs adults carrying severe acute respiratory syndrome coronavirus 2 asymptomatically. JAMA Pediatr. 2021;175 :193–194. doi:10.1001/jamapediatrics.2020.3595.32926119
| 36476536 | PMC9725712 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 11; 42(1):e38-e39 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003733 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00007
10.1097/INF.0000000000003758
3
COVID Reports
Clinical Characteristics, Transmission Rate and Outcome of Neonates Born to COVID-19-Positive Mothers: A Prospective Case Series From a Resource-Limited Setting
https://orcid.org/0000-0002-1405-4822
Rood Marloes MD *†
ten Kate Lisa MD [email protected]
†
https://orcid.org/0000-0001-9487-1218
Boeddha Navin P. MD, PhD [email protected]
†‡
van ‘t Kruys Kevin MD [email protected]
†
From the * Department of Pediatrics, Rijnstate Hospital Arnhem, Wagnerlaan, Arnhem, The Netherlands
† Department of Pediatrics, Academic Pediatric Hospital Suriname, Academic Hospital Suriname, Flustraat, Paramaribo, Suriname
‡ Department of Pediatrics, Erasmus MC-Sophia Children’s Hospital, University Medical Center Rotterdam, Wytemaweg, Rotterdam, The Netherlands.
Address for correspondence: Marloes Rood, MD, Department of Pediatrics, Rijnstate Hospital Arnhem, Wagnerlaan 55, 6815 AD, Arnhem, The Netherlands. E-mail: [email protected]. or Department of Pediatrics, Academic Pediatric Hospital Suriname, Academic Hospital Suriname, Flustraat, Paramaribo, Suriname.
01 11 2022
1 2023
01 11 2022
42 1 3542
29 9 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
Background:
Coronavirus disease (COVID-19) infection during pregnancy could damage the placenta, but data on neonates born to COVID-19-positive mothers is scarce. In this case series, we aim to describe clinical characteristics, transmission rate and outcomes at 3 months of age among neonates born to mothers with COVID-19 diagnosed near the time of delivery.
Methods:
Prospective, multicenter case series from Suriname. We collected clinical data of neonates born to mothers with COVID-19 infection between June and August 2021. COVID-19 swabs were taken within 5 days and 2 weeks after birth. Follow-up took place at 3 months.
Results:
We enrolled 18 neonates. However, 18/18 (100%) mothers were infected in the third trimester and 10/18 (55.6%) had severe COVID-19 infection requiring ICU admission and 2/10 (20%) died. In total 16/18 (77.8%) neonates were born after cesarean section and 13/18 (72.2%) were born preterm (median 35 weeks, Interquartile range 32 + 4–38 + 0). Neonatal intensive care unit admission was needed in 7/18 (38.9%) neonates. Respiratory symptoms occurred in 12/18 (66.7%), 5/18 (27.8%) were suspected of early-onset sepsis and 1/18(5.6%) of late-onset sepsis. One preterm neonate developed necrotizing enterocolitis. A nasopharyngeal swab was positive in 1/18 (5.5%) neonates within 5 days of life and in 0/11 (0%) neonates after 2 weeks. Follow-up showed mild neurodevelopmental delay in 2/14 (14.3%) patients.
Conclusion:
We describe a high proportion of severely ill mothers due to COVID-19 infection with subsequent cesarean delivery and prematurity. Accounting for gestational age at birth, the neonatal clinical course and findings at follow-up appeared similar to neonates born to COVID-19-negative mothers. Maternal vaccination is recommended to prevent neonatal risks associated with prematurity and cesarean delivery.
COVID-19
neonates
follow
up
==== Body
pmcIn March 2020, the World Health Organization (WHO) declared the coronavirus infection disease (later termed as COVID-19) as a global pandemic.1 Emerging strains of COVID-19 continue to cause severe mortality and morbidity.2,3
COVID-19 infects all age groups including newborns and elderly, with pregnant women being a specific vulnerable group.3 Placental tissue of pregnant women with COVID-19 infection is associated with histological changes, possibly compromising the fetal environment subsequently leading to adverse neonatal outcome.4 Furthermore, a vertical and horizontal transmission rate of COVID-19 has been described in up to 10%, indicating the need for adequate preventive interventions including mode of delivery and isolation measures.5–7
In neonates born to mothers with COVID-19, most of the COVID-19-positive neonates were symptomatic and a significant proportion of them required intensive care admission.6,8 In addition, COVID-19-negative neonates were also more likely to be admitted to the neonatal intensive care unit (NICU).7,9,10 The increase in NICU admission of neonates born to mothers with COVID-19 could possibly be related to more premature deliveries.11 Follow-up data of neonates born to COVID-19-positive mothers is scarce and has been limited to 30 days after hospital discharge.12
The aim of this case series is to describe the clinical characteristics and transmission rate of neonates born to COVID-19-positive mothers. Moreover, we report the outcome at the age of 3 months of neonates born to COVID-19-positive mothers.
METHODS
Setting and Study Patients
This is a country-wide prospective case series from all 5 hospitals in Suriname, South America. From June 1 to July 31 2021, we enrolled all neonates born to mothers with a COVID-19 infection during the present pregnancy. We excluded stillbirths. The diagnosis of a COVID-19 infection was based on laboratory confirmation of a polymerase chain reaction (PCR) of a nasopharyngeal swab. Only symptomatic women were tested.
Clinical Data and Nasopharyngeal Swab Collection
Data on clinical presentation, illness severity, management and outcome were collected prospectively. All included neonates underwent a nasopharyngeal swab for PCR diagnostics within 5 days after birth to test vertical transmission.13,14 All PCR swabs within 5 days were taken before the neonate had any physical contact with their mother. A second nasopharyngeal swab 2 weeks after birth was performed to study the possibility of horizontal transmission in a COVID-19-positive environment. At the age of 3 months, all included children were invited to visit the pediatrician at the Academic Hospital Paramaribo for follow-up. The following data was collected: specific complaints reported by the caregivers, total amount of emergency room or outpatient clinic visits, weight, head circumference, general physical examination and the neurodevelopment by the Van Wiechen scheme.15 P-values were determined for weight and head circumference at birth and at 3-month follow-up. For term children and preterm children at 3-month follow-up we used the WHO growth standards 2006 and 2007 and the WHO Antro analyzer.16–18 For premature children at birth, we used the Fenton preterm growth standards 2013 and the Fenton 2013 Growth Calculator.19,20 For 1 child the follow-up was before the corrected gestational age of 40 weeks and the Fenton preterm growth standards 2013 was used.
COVID-19 Protocols
We followed local and hospital-specific protocols concerning medical care. Neonates with the indication for admission to the neonatal intensive care, high care or medium care received standard medical care regardless of their COVID-19 status. Parents of the infants were allowed to visit their child once they were declared COVID-19 negative by the national public health institute, “Bureau voor Openbare Gezondheidszorg”. Neonates without indication for admission were, according to international guidelines of the WHO, allowed to room in with their mothers.1 Breastfeeding and skin-to-skin contact were permitted.
Ethical Aspects
The study protocol was approved by the Medical Ethical Committees “Commissie Mens gebonden Wetenschappelijk Onderzoek”, by the medical director of the Academic Hospital Paramaribo and by all the pediatric departments of the participating hospitals. Informed written consent was obtained from at least one of the parents or caregivers.
RESULTS
We enrolled 18 neonates born to COVID-19-positive mothers. Demographics and health profile of pregnant women are shown in Table 1. All pregnant women had singleton pregnancies and were diagnosed with COVID-19 infection during the third trimester of their pregnancy. Comorbidities were found in only 2 women; 1 woman suffered from pre-eclampsia and 1 with diabetes mellitus. All women had COVID-19 symptoms; 9/18 (50.0%) had mild to moderate severity of the disease and 8/18 (44.4%) had severe COVID-19 infection requiring ICU admission of which 2 (11.1%) died. The disease severity of 1 woman was unknown. Only 4/18 (22.2%) women gave birth through vaginal delivery. The remaining 14/18 (77.8%) neonates were born after cesarean section of which 12 were performed on the maternal indication. In 10/12 (83.3%) cases, the maternal indication was impending respiratory failure. The other 2 maternal indications were cesarean section in the medical history and placenta previa.
TABLE 1. Demographic and Health Profile of Pregnant Woman with COVID-19
All Patients (n = 18) Preterm <37 Weeks (n = 13) Term ≥37 Weeks (n = 5)
Maternal age (year) Median 33.0 34.5 33.0
IQR 25.5–37.0 25.3–37.0 28.0–41.0
Severity of disease* Mild 3/18 (16.7%) 2/13 (15.4%) 1/5 (20.0%)
Moderate 6/18 (33.3%) 3/13 (23.1%) 3/5 (60.0%)
Severe 8/18 (44.4%) 7/13 (53.8%) 1/5 (20.0%)
Unknown 1/18 (5.6%) 1/13 (7.7%) 0/5 (0.0%)
Comorbidity None 16/18 (88.9%) 11/13 (84.6%) 13/13 (100.0%)
Pre-eclampsia 1/18 (5.6%) 1/13 (7.7%) 0/13 (0.0%)
Diabetes mellitus 1/18 (5.6%) 1/13 (7.7%) 0/13 (0.0%)
Dexamethason Yes 13/18 (72.2%) 11/13 (84.6%) 2/5 (40.0%)
No 4/18 (22.2%) 1/13 (7.7%) 3/5 (60.0%)
Unknown 1/18 (5.6%) 1/13 (7.7%) 0/5 (0.0%)
Gestational age (weeks) Median 35 + 0 33 + 0 39 + 0
IQR 32 + 4–38 + 0 32 + 1–35 + 2 38 + 1–39 + 5
Mode of delivery Cesarean section 14/18 (77.7%) 11/13 (84.6%) 3/5 (60.0%)
Maternal 12/18 (66.7%) 10/13(76.9%) 2/5 (40.0%)
Neonatal 2/18 (11.1%) 1/13 (7.7%) 1/5 (20.0%)
Vaginal 4/18 (22.2%)* 2/13 (15.4%) 2/5 (40.0%)
*Severe = ICU-admission; moderate = oxygen requirement with nasal cannula, venturi mask and/or non-rebreather mask; mild = no treatment needed and/or asymptomatic.
Clinical Characteristics of Neonates Born to COVID-19-Positive Mothers
Clinical characteristics, management and outcome of neonates born to COVID-19-positive mothers are summarised in Table 2. Of 18 neonates, 9 (50%) were male and 9 (50%) were female. In total 13/18 (72.2%) neonates were born preterm and only 5/18 (27.8%) were born at full-term gestation. The median gestational age was 35 weeks (Interquartile range 32 + 4–38 + 0). Among them, 17 neonates were born appropriate for their gestation and only 1 (5.6%) was born small for gestational age. However, 7 of the 18 (38.9%) neonates included had birth weights above 2.5 kg, 9/18 (50%) had 1.5–2.5 kg and 2 (11.1%) of the neonates weighed less than 1.5 kg. Appearance, Pulse, Grimace, Activity, Respiration scores were >5 at 5 minutes for all neonates. No neonatal asphyxia occurred in the study cohort.
TABLE 2. Clinical Characteristics, Management, SARS CoV-2 PCR Testing, and Outcome of Neonates Born to COVID-19-Positive Mothers–Summary Data
All Patients (n = 18) Preterm <37 Weeks (n = 13) Term ≥37 Weeks (n = 5)
Clinical characteristics
Birthweight (g) Median 2261 1870 2920
IQR 1765–2903 1740–2290 2850–3250
Head circumference (cm) Median 31.5 31.0 34.0
IQR 31.0–33.5 30.0– 31.9 32.0–34.0
APGAR at 5 min <5 0/18 (0%) 0/13 (0.0%) 0/5 (0.0%)
>5 17/18 (94.4%) 12/13 (92.3%) 5/5 (100.0%)
Unknown 1/18 (5.6%) 1/13 (7.7%) 0/5 (0.0%)
Respiratory disorder Yes 12/18 (66.7%) 10/13 (76.9%) 2/5 (40.0%)
No 6/18 (33.3%) 3/13 (23.1%) 3/5 (60.0%)
Circulatory disorder Yes 1/18 (5.6%) 1/13 (7.7%) 0/5 (0.0%)
No 17/18 (94.4%) 12/13 (92.3%) 5/5 (100.0%)
Gastrointestinal disorder Yes 2/18 (11.1%) 2/13 (15.4%) 0/5 (0.0%)
No 16/18 (88.9%) 11/13 (84.6%) 5/5 (100.0%)
Metabolic disorder Yes 5/18 (27.8%) 5/13(38.5%) 0/5 (0.0%)
No 13/18 (72.2%) 8/13 (61.5%) 5/5 (100.0%)
Hematologic disorder Yes 2/18 (11.1%) 2/13 (15.4%) 0/5 (0.0%)
No 16/18 (88.9%) 11/13 (84.6%) 5/5 (100.0%)
Infectious disorder Yes 7/18 (38.9%) 5/13 (38.5%) 2/5 (40.0%)
No 11/18 (61.1%) 8/13 (61.5%) 3/5 (60.0%)
Neurologic disorder Yes 0/18 (0.0%) 0/13 (0.0%) 0/5 (0.0%)
No 18/18 (100.0%) 13/13 (100.0%) 5/5 (100.0%)
Management
Admission NICU 7/18(38.9%) 7/13 (53.8%) 0/5(0.0%)
High-care 6/18 (33.3%) 4/13 (30.8%) 2/5 (40.0%)
No hospital admission 5/18 (27.8%) 2/13 (15.4%) 3/5 (60.0%)
Hospital lenght of stay (days) Median 8 9 4
IR 5–9 7–10 4–5
Respiratory support Invasive ventilation 1/18(5.6%) 1/13(7.7%) 0/5(0.0%)
CPAP 8/18 (44.4%) 8/13 (61.5%) 0/5 (0.0%)
Low-flow 3/18 (16.7%) 1/13 (7.7%) 2/5 (40.0%)
None 6/18 (33.3%) 3/13 (23.1%) 3/5 (60.0%)
Caffeine Yes 5/18(27.8%) 5/13(38.5%) 0/5(0.0%)
No 13/18 72.2%) 8/13 (61.5%) 5/5 (100.0%)
Antibiotics Yes 7/18(38.9%) 5/13(38.5%) 2/5(40.0%)
No 11/18 (61.1%) 8/13 (61.5%) 3/5 (60.0%)
Blood transfusion Yes 2/18(11.1%) 2/13(15.4%) 0/5(0.0%)
No 16/18 (88.9%) 11/13 (84.6%) 5/5 (100.0%)
Phototherapy Yes 4/18 (22.2%) 4/13(30.8%) 0/5(0.0%)
No 14/18 (77.8%) 9/13 (69.2%) 5/5 (100.0%)
Cranial ultrasound Indication 2/18(11.1%) 2/13(15.4%) 0/5(0.0%)
Abnormalities 0/2 (0.0%) 0/2 (0.0%)
Outcome at 3 months
Symptoms Yes 3/18 (16.7%) 3/13(23.1%) 0/5(0.0%)
No 11/18 (61.1%) 6/13 (46.2%) 5/5 (100.0%)
Unknown 4/18 (22.2%) 4/13 (30.8%) 0/5 (0.0%)
Weight (g) Median 4255 4230 4940
IQR 4083–4618 3940–4400 4560–5250
Unknown 6/18 (33.3%) 4/13 (30.8%) 2/5 (40.0%)
Head circumference (cm) Median 37.3 37.0 38.8
IQR 36.6–37.6 36.5–37.5 37.7–38.9
Unknown 6/18 (33.3%) 4/13 (30.8%) 2/5 (40.0%)
General physical examination Abnormalities 1/18(5.6%) 1/13(7.7%) 0/5(0.0%)
No abnormalities 12/18 (66.7%) 8/13 (61.5%) 4/5 (80.0%)
Unknown 5/18 (27.8%) 4/13 (30.8%) 1/5 (20.0%)
Neurodevelopment Abnormalities 2/18(11.1%) 2/13(15.4%) 0/5(0.0%)
No abnormalities 11/18 (61.1%) 7/13 (53.8%) 4/5 (80.0%)
Unknown 5/18 (27.8%) 4/13 (30.8%) 1/5 (20.0%)
SARS CoV-2 PCR
≤5 days positive 1/18(5.6%) 1/13(7.7%) 0/5 (0.0%)
2–3 weeks positive 0/11(0.0%) 0/9(0.0%) 0/2(0.0%)
CPAP, continuous positive airway pressure; NICU, NICU, neonatal intensive care unit; PCR, polymerase chain reaction.
Seven of 18 neonates were admitted to the NICU. Six of 18 neonates were admitted to high-care units and 5/18 were sent home directly after birth. Respiratory symptoms occurred in 12/18 (66.7%) neonates. Although 4/18 (22.2%) neonates suffered from transient tachypnea of the newborn, of which one was supported with continuous positive airway pressure (CPAP) and the other 3 received low flow. In total 8/18 (44.4%) neonates developed neonatal respiratory distress syndrome, most of them received respiratory support with CPAP. Only for 1 neonate, with grade 3 Neonatal Respiratory Distress Syndrome, invasive ventilation was required. Also, 6/18 (33.3%) neonates did not develop any respiratory symptoms.
Infectious problems were seen in 7/18 (38.9%) neonates. Five were suspected of early-onset neonatal sepsis, for which empirical antibiotic therapy was given in different regimens because of different hospital guidelines among the participating centers. One neonate received antibiotics during 72 hours for suspected late onset sepsis. All blood cultures were negative. In one neonate, a urinary tract infection was suspected. Treatment with antibiotics was given for 72 hours. Negative urinary cultures were found. Four of 18 children had hyperbilirubinemia, for which phototherapy was administered.
One preterm neonate admitted to the NICU developed Necrotizing Enterocolitis, which was treated sufficiently with antibiotics and discontinued enteral feeding.
Two children were admitted with hematologic problems. One of them developed anemia after maternal blood loss during a cesarean section, this was corrected with one blood transfusion. The other preterm neonate developed anemia several weeks after birth, for which one transfusion was given.
None of the neonates died during the hospital stay.
The details of the 18 neonates and their mothers are shown in Table 3.
TABLE 3. Clinical Characteristics and Outcomes of Neonates Born to COVID-19-Positive Mothers (n = 18), Specified by Neonate
Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7 Case 8 Case 9
Maternal characteristics
Age (years) 37 23 43 36 46 Unknown 32 36 37
Trimester of COVID 3rd 3rd 3rd 3rd 3rd 3rd 3rd 3rd 3rd
Severity disease* Severe Moderate Severe Moderate Mild Moderate Moderate Mild Severe
Comorbidity† No No No No No No No No No
Dexamethasone Yes Yes Yes Yes No Yes Yes No Yes
Gestational age (weeks) 33 + 0 32 + 4 28 + 3 38 + 2 40 + 3 32 + 4 33 + 0 35 + 3 31 + 0
Mode of delivery ‡ CS, maternal indication CS, maternal indication CS, maternal indication CS, maternal indication CS, neonatal indication VD CS, maternal indication CS, neonatal indication CS, maternal indication
Clinical characteristics neonate
Sex Male Female Female Female Male Female Male Female Male
Birthweight (gram) with P-value 2014 (p50) 1745 (p44) 1290 (p82) 2850 (p19.3) 3780 (p80.3) 1825 (p52) 2231 (p71) 3265 (p96) 1740 (p69)
Head circumference (cm) with P-value 31.5 (p79) 31.0 (p86) 28.0 (p97) 31.5 (p2.2) 35.0 (p66.4) 30.0 (p67) 33.5 (p98) 34.0 (p93) 31.0 (p96)
APGAR score 8–9–10 8–9–10 10–7–10 10–10–10 9–10–10 9–10–10 6–8–9 7–8–8 2–9–9
Respiratory disorder NRDS, grade unknown NRDS, grade 2
Apnea NRDS, grade 3 No No NRDS, grade 1 Transient Tachypnea of the Newborn No NRDS, grade unknown
Apnea
Circulatory disorder No No No No No No No No No
Gastrointestinal disorder No No Necrotizing Enterocolitis, grade 1A No No Vomiting No No No
Metabolic disorder Un-conjugated hyperbili-rubinemia, physiological
Acute Kidney Injury, prerenal Un-conjugated hyperbili-rubinemia, physiological No No No No No No Un-conjugated hyperbili-rubinemia, physiological
Hematologic disorder No Anaemia No No No No No No No
Infectious disorder Urinary Tract Infection, suspected (UC unknown) No Late Onset Sepsis, suspected (BC negative) No No Early onset neonatal sepsis, suspected (BC unknown) No No No
Neurologic disorder No No No No No No No No No
Management
ICU admission Yes, NICU Yes, NICU Yes, NICU No No Yes, NICU Yes, NICU No Yes, NICU
Hospital length of stay (days) Unknown Unknown Unknown No admission No admission Unknown 10 No admission 8
Maximum respiratory support CPAP CPAP Invasive ventilation No No CPAP CPAP No CPAP
Caffeine No Yes Yes No No No No No Yes
Antibiotics§ Augmentin; stopped after 72 u with negative UC No Augmentin/Amikacin; stopped after 72 u with negative BC No No Augmentin/
Gentamycin; Stopped after 48 u with negative BC No No No
Transfusion blood products No Yes No No No No No No No
Phototherapy Yes Yes No No No No No No Yes
Cranial ultrasound No indication No abnormalities No abnormalities No indication No indication No indication No indication No indication No indication
SARS-2-COVID PCR
≤5 days Negative Negative Negative Negative Negative Negative Negative Negative Negative
2–3 weeks Not executed Negative Negative Not executed Not executed Negative Negative Not executed Negative
Outcome 3 months
Reported symptoms Unknown No No No No No No Unknown No
Weight (gram) with P-value Unknown 4230 (p0.9) 2370 (p7) Unknown Unknown 4280 (p6.6) 4550 (p8.6) Unknown 4820 (p0.9)
Head circumference (cm) with P-value Unknown 37.5 (p7.0) 32.8 (p31) Unknown Unknown 36.6 (p6.9) 38.0 (p22.8) Unknown 37.5 (p0.5)
General physical examination Unknown No abnormalities No abnormalities No abnormalities Unknown No abnormalities No abnormalities Unknown Systolic souffle at 2 weeks
Neurodevelopment ¶ Unknown No abnormalities Delay (does not follow) No abnormalities Unknown Delay (does not smile back) No abnormalities Unknown No abnormalities
Case 10 Case 11 Case 12 Case 13 Case 14 Case 15 Case 16 Case 17 Case 18
Maternal characteristics
Age (years) 23 26 38 33 25 22 36 33 33
Trimester of COVID positive test 3rd 3rd 3rd 3rd 3rd 3rd 3rd 3rd 3rd
Severity disease* Moderate Unknown Severe, died. Moderate Severe Severe Severe Severe, died Mild
Comorbidity† No No No No No No Pre-eclampsia No Diabetes mellitus
Dexamethasone No Unknown Yes No Yes Yes Yes Yes Yes
Gestational age (weeks) 39+0 35+3 31+5 38+0 33+5 35+0 (Ballard-score) 35+0 39+0 36+3
Mode of delivery ‡ VD CS, maternal indication CS, maternal indication VD CS, maternal indication CS, maternal indication CS, maternal indication CS, maternal indication vaginal delivery, induction because of microcephaly
Clinical characteristics neonate
Sex M M F M F F M F M
Birthweight (gram) with P-value 2850 (p14.2) 2455 (p37) 1700 (p95) 3250 (p42.1) 1870 (p32) 1402 (p1) 2290 (p32) 2920 (p24.1) 3027 (p68)
Head circumference (cm) with P-value 34.0 (p35.8) 31.0 (p20) Unknown 32.0 (p2.6) 30.0 (p40) 29.5 (p9) 31.0 (p27) 34.0 (p54.1) 33.0 (p51)
APGAR score 9–9–9 7–9–10 7–8 9–9 9–9–9 Unknown 9–9 10–10 10–10
Respiratory disorder Transient Tachypnea of the Newborn No NRDS, grade unknown
Apnoe Transient Tachypnea of the Newborn NRDS, grade unknown
Apnoe NRDS, grade 2 Transient Tachypnea of the Newborn No No
Circulatory disorder No No Tachycardy, No No No No No No
Gastrointestinal disorder No No No No No No Feeding difficulties No No
Metabolic disorder No No No No Un-conjugated hyperbili-rubinemia, physiological No No No No
Hematologic disorder No 1. Anemia No No No No No No No
Infectious disorder Infection, suspected (BC unknown) No No Early onset neonatal sepsis, suspected (BC negative) No Early onset neonatal sepsis, suspected (BC negative) Early onset neonatal sepsis, suspected (BC unknown) No No
Neurologic disorder No No No No No No No No No
Management
ICU admission Head Circumference Head Circumference Head Circumference Head Circumference Head Circumference Yes, NICU Head Circumference No No
Hospital length of stay (days) 3 4 Unknown 5 9 7 14 No admission No admission
Maximum respiratory support Low-flow No CPAP Low-flow CPAP CPAP Low-flow No No
Caffeine No No Yes No Yes No No No No
Antibiotics Augmentin, duration unknown No No Augmentin, stopped after 5 days. No Benzyl-penicillin/
Gentamycin; stopped after 7 days Augmentin, duration unknown No No
Transfusion blood products No Yes No No No No No No No
Phototherapy No No No No Yes No No No No
Cranial ultrasound findings No indication No indication No indication No indication No No indication No indication No indication No indication
SARS-2-COVID PCR
≤5 days Negative Negative Negative Negative Positive Negative Negative Negative Negative
2–3 weeks Negative Negative Negative Not executed Negative Not executed Negative Negative Not executed
Outcome 3 months
Reported symptoms No Unknown No No Reflux symptoms respiratory Syncytial virus bronchiolitis
Obstipation Reflux symptoms No Unknown
Weight (gram) with P-value 4940 (p18.0) Unknown 3940 (p0.6) 5560 (p51.1) 4400 (p5.9) 3410 (p0.0) 4130 (p0.6) 4180 (p11.2) Unknown
Head circumference (cm) with P-value 39.0 (p47.2) Unknown 37.5 (p13.7) 36.6 (p1.7) 36.5 (p3.5) 34.0 (p0.0) 37.0 (p2.3) 38.8 (p78.1) Unknown
General physical examination No abnormalities Unknown No abnormalities No abnormalities No abnormalities No abnormalities No abnormalities No abnormalities Unknown
Neurodevelopment No abnormalities Unknown No abnormalities No abnormalities No abnormalities No abnormalities No abnormalities No abnormalities Unknown
CPAP, continuous positive airway pressure; CS, cesarean section; NICU, neonatal intensive care unit; PCR, polymerase chain reaction; VD, vaginal delivery.
*Severe = ICU-admission; moderate = oxygen requirement with nasal cannula, venturi mask and/or non-rebreather mask; mild = no treatment needed and/or asymptomatic.
†Diabetic mellitus/gravidarum, hypertension/pregnancy induced hypertension/pre-eclampsy, epilepsy, sickle cell; serology (VDRL, HbsAg, HIV).
‡VD = vaginal delivery; CS = cesarean section.
§BC = blood culture; UC = urine culture.
¶Neurodevelopment according to “Van Wiechen” at 3 months (follows with eyes and head, smiles back, raises head in prone position at 45 degree.
SARS-CoV-2 PCR Testing in Neonates
A nasopharyngeal swab to detect severe acute respiratory syndrome coronavirus-2 PCR was collected in all 18 newborns within the first 5 days of life before they had physical contact with their mother. One of the 18 neonates (5.5%) was positive. This neonate tested positive within 24 hours after delivery and was negative when retested at 2 weeks of age.
At 2 weeks postpartum, repeat PCR testing was performed in 11/18 (61.1%) included neonates. Swabs could not be collected in 7 neonates because of loss to follow-up or because no permission of parents was obtained to repeat the testing. None of the 11 (0/11) (0.0%) PCR tests were positive, indicating no horizontal transmission in this study.
Three Months Follow-Up
A total of 13 of 18 infants were seen by a pediatrician for follow-up at the outpatient clinic. Of the 5 infants who were not seen by a pediatrician, 1 infant could not come because of travel distance and costs, therefore a consult by phone was performed; 4 infants were lost to follow-up.
At the age of 3 months, 1 infant was diagnosed with bronchiolitis and obstipation; 2 infants reported complaints of reflux. From 12 of the 13 infants with follow-up at 3 months, weight and head circumference were noted. For 11 of 12 infants, weight and head circumference curves were according to the expected growth curve or did increase. Only 1 of 12 children, case 3, had a decrease in the growth curve for both weight and head circumference. With general physical examination, no infants had any abnormalities. One infant had a cardiac murmur at the age of 2 weeks, at follow-up the murmur was not found anymore with no cardiac anomalies on cardiac ultrasound.
Twelve of 14 infants had normal neurodevelopment according to Van Wiechen; 2 infants had a neurodevelopmental delay. One of them, an ex-preterm infant born at 28 weeks, also had deviating growth of the skull.
The neonate with a positive nasopharyngeal swab, who was born preterm and received CPAP and phototherapy in the neonatal period, suffered from gastro-esophageal problems as reflux at 2–3-month follow-up. No other problems were observed or reported at follow-up.
DISCUSSION
In this case series, we describe the clinical characteristics, transmission rate and middle-term outcome of infants born to infected women during the severe acute respiratory syndrome coronavirus-2 pandemic.
We observed that more than 2/5 of the mothers with COVID-19 infection were severely ill, of which 2 women died. In our study period, COVID-19 vaccination in pregnant women was not actively recommended yet. The severity of the COVID-19-disease resulted in our study population of almost 80% in cesarean section and in more than 70% in preterm births. These findings of a higher incidence of cesarean section and preterm births are in line with previous systematic reviews.8,21–23
Both, cesarean delivery and prematurity are important observations because of the possibility of adverse events in neonates.24–26 Moreover, our data underscore the importance of vaccination in pregnant women, to prevent severe illness and to prevent potential complications of cesarean section and/or prematurity.
Despite increased risk of adverse events, none of the neonates suffered from asphyxia. Respiratory symptoms occurred in almost 70% of neonates, which was higher than reported in a systematic review of Dhir, et al.8 However, this review involved less premature babies, which could partly explain less frequent respiratory symptoms.
We report a vertical transmission rate of 5.5% and a horizontal transmission rate of 0%. However, we did not confirm vertical transmission via other means, such as placental tissue, amniotic fluid or serologic evidence. Previous studies reported COVID-19-positive rates of 0%–10.8% in swabs taken within 72 hours after birth. We hypothesized that horizontal transmission could occur because of COVID-19-positive members in the household, potentially leading to delayed infection and complications of COVID-19. However, none of the neonates in our study were COVID-19-positive at the age of 2 weeks. Our findings support the current WHO guidelines of breastfeeding and skin-to-skin contact for neonates of COVID-19 positive mothers. Separating neonates after birth from COVID-19-positive mothers brings more harm than the risk of horizontal transmission.
Viral infections are known to cause long-term morbidity.27 Although, the long-term outcome of COVID-19 infection in adults has been described, little is known about the outcome of neonates. At follow-up, 3 months after birth, we did not observe any COVID-19-related problems. Neither we observed more/other non-COVID related problems compared to neonates born to COVID-19-negative mothers. No other literature was found on follow-up at 3 months or longer. The longest time to follow-up of neonates born to COVID-19-positive mothers was 4 weeks, showing adequate growth and outcome at 1 month of life.28 Our study shows no COVID-19-related problems even at the age of 3 months.
The 1 neonate who tested positive on a nasopharyngeal swab, did not reveal any important clinical findings during the first 3 months of life. This is consistent with the current literature. A systematic review of Trevisanuto et al.29 with a median follow-up of 10 days, shows that most neonates with COVID-19 infection are asymptomatic or presented with mild symptoms.
We hypothesize that longer follow-up of neonates born to COVID-19-positive mothers will not reveal new findings. However, in some cases of congenital infections, symptoms may appear after years. Therefore, future research on COVID-19-positive neonates should monitor neurodevelopment.
This study is limited by the relatively low number of included cases. Based on the number of COVID-19-positive PCR tests in the country, we estimate that approximately 24 neonates were born to mothers with a COVID-19 infection during the study period. We may have missed pregnant women who did not need hospital admission and COVID-19-positive pregnant women who were asymptomatic (pregnant women admitted to the hospital were only tested in case of symptoms).30 Also, all maternal COVID-19 infections occurred in the third trimester, whereas it is possible that an infection earlier in pregnancy could result in more severe outcome. We also did not take into account different COVID-19 strains which could cause different phenotypes in the neonate. Nevertheless, this prospective, country-wide case series from a resource-limited setting is the first to study potential horizontal transmission and outcome at 3 months, thereby contributing to current knowledge of neonates from COVID-19-infected mothers.
In conclusion, this case series shows a high rate of severely ill pregnant women due to COVID-19 infection with direct consequences on neonates because of preterm birth and delivery through cesarean section. More attention is needed on maternal vaccination, to prevent neonatal risk due to preterm birth and cesarean section. The neonatal clinical course and findings at follow-up at 3 months did not seem to differ from neonates born to COVID-19-negative mothers. Furthermore, we show vertical transmission rates of 5% and no horizontal transmission at the age of 2 weeks. Future studies should investigate neonatal outcomes after maternal COVID-19 infection in the 1st or 2nd trimester, preferably with subtyping of different COVID-19 strains.
ACKNOWLEDGMENTS
We acknowledge the co-investigators from the participating centers: Dr. W. Zijlmans (Diakonessenhuis hospital), Drs. N. Braafheid (Sint Vincentius hospital). and Drs. R. Morpurgo (‘s Lands hospital).
The authors have no conflicts of interest or funding to disclose.
Marloes Rood and Lisa ten Kate contributed equally.
==== Refs
REFERENCES
1. World Health Organisation. New research highlights risks of separating newborns from mothers during COVID-19 pandemic. Available at: https://www.who.int/news/item/16-03-2021-new-research-highlights-risks-of-separating-newborns-from-mothers-during-covid-19-pandemic. Published March 16, 2021. Accessed February 11, 2022.
2. Tareq AM bin ET Dhama K . Impact of SARS-CoV-2 delta variant (B.1.617.2) in surging second wave of COVID-19 and efficacy of vaccines in tackling the ongoing pandemic. Human Vaccin Immunother. 2021;17 :4126–4127.
3. Lassi ZS Ana A Das JK . A systematic review and meta-analysis of data on pregnant women with confirmed COVID-19: clinical presentation, and pregnancy and perinatal outcomes based on COVID-19 severity. J Global Health. 2021;11 :05018.
4. Husen MF van der Meeren LE Verdijk RM . Unique severe covid-19 placental signature independent of severity of clinical maternal symptoms. Viruses. 2021;13 :1670.34452534
5. Tolu LB Ezeh A Feyissa GT . Vertical transmission of severe acute respiratory syndrome coronavirus 2: a scoping review. PLoS One. 2021;16 :e0250196.33886645
6. Malik S Surve S Wade P . Clinical characteristics, management, and short-term outcome of neonates born to mothers with COVID-19 in a tertiary care hospital in India. J Trop Pediatr. 2021;67 :fmab054.34114628
7. Papapanou M Papaioannou M Petta A . Maternal and neonatal characteristics and outcomes of covid-19 in pregnancy: an overview of systematic reviews. Int J Environ Res Public Health. 2021;18 :596.33445657
8. Dhir SK Kumar J Meena J . Clinical features and outcome of SARS-CoV-2 infection in neonates: a systematic review. J Trop Pediatr. 2021;67 :fmaa059.32856065
9. Han Y Ma H Suo M . Clinical manifestation, outcomes in pregnant women with COVID-19 and the possibility of vertical transmission: a systematic review of the current data. J Perinat Med. 2020;48 :912–924.33068387
10. Alipour Z Samadi P Eskandari N . Relationship between coronavirus disease 2019 in pregnancy and maternal and fetal outcomes: retrospective analytical cohort study. Midwifery. 2021;102 :103128.34474247
11. Metz TD Clifton RG Hughes BL . Association of SARS-CoV-2 infection with serious maternal morbidity and mortality from obstetric complications. JAMA. 2022;327 :748–759.35129581
12. Angelidou A Sullivan K Melvin PR . Association of maternal perinatal SARS-CoV-2 infection with neonatal outcomes during the COVID-19 pandemic in massachusetts. JAMA Network Open. 2021;4 :e217523.33890989
13. RIVM. Coronavirus disease COVID-19. Available at: https://www.rivm.nl/en/coronavirus-covid-19/coronavirus-disease-covid-19. Published 2022. Accessed May 30, 2022.
14. Debrabandere ML Farabaugh DC Giordano C . A review on mode of delivery during COVID-19 between December 2019 and April 2020. Am J Perinatol. 2021;38 :332–341.33285608
15. Verkerk PH Reerink JD Herngreen WP . Het Van Wiechenschema in de Praktijk. 1993. https://repository.tno.nl/islandora/object/uuid:ccb9707a-7bf4-406f-aa41-efbd539bec95. Accessed February 28, 2022.
16. de Onis M ; WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr. 2006;95 :76–85.
17. de Onis M . WHO Child Growth Standards: Head Circumference-for-Age, Arm Circumference-for-Age, Triceps Skinfold-for-Age and Subscapular Skinfold-for-Age: Methods and Development. American Journal of Clinical Nutrition. Available at: https://www.who.int/publications/i/item/9789241547185. Published November 12, 2007. Accessed February 26, 2022.
18. WHO. WHO Anthro Survey Analyser. Available at: https://www.who.int/tools/child-growth-standards/software. Accessed February 26, 2022.
19. Chou JH Roumiantsev S Singh R . PediTools electronic growth chart calculators: applications in clinical care, research, and quality improvement. J Med Internet Res. 2020;22 :e16204.32012066
20. Fenton TR Kim JH . A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr. 2013;13 :59.23601190
21. Zaigham M Andersson O . Maternal and perinatal outcomes with COVID-19: a systematic review of 108 pregnancies. Acta Obstet Gynecol Scand. 2020;99 :823–829.32259279
22. de Medeiros KS Sarmento ACA Costa APF . Consequences and implications of the coronavirus disease (COVID-19) on pregnancy and newborns: a comprehensive systematic review and meta-analysis. Int J Gynecol Obstet. 2021;156 :394–405.
23. Khan DSA Hamid LR Ali A . Differences in pregnancy and perinatal outcomes among symptomatic versus asymptomatic COVID-19-infected pregnant women: a systematic review and meta-analysis. BMC Pregnancy and Childbirth. 2021;21:801.
24. National Institutes of Health. What is a C-section? https://www.nichd.nih.gov/health/topics/labor-delivery/topicinfo/c-section. Published January 9, 2017. Accessed February 11, 2022.
25. Furdon SA Clark DA . Prematurity. MedScape. https://emedicine.medscape.com/article/975909-overview#a1. Published October 13, 2017. Accessed February 11, 2022.
26. Gregory KD Jackson S Korst L . Cesarean versus vaginal delivery: whose risks? whose benefits? Am J Perinatol. 2012;29 :7–18.21833896
27. Silasi M Cardenas I Kwon JY . Viral infections during pregnancy. Am J Reprod Immunol. 2015;73 :199–213.25582523
28. Falsaperla R Giacchi V Lombardo G . Neonates born to COVID-19 mother and risk in management within 4 weeks of life: a single-center experience, systematic review, and meta-analysis. Am J Perinatol. 2021;38 :1010–1022.34082444
29. Trevisanuto D Cavallin F Cavicchiolo ME . Coronavirus infection in neonates: a systematic review. Arch Dis Child Fetal Neonatal Ed. 2021;106 :330–335.32943533
30. Het Bureau Openbare Gezondheidszorg (BOG) Suriname/National Registry Suriname. Available at: https://bogsuriname.com/wp-content/uploads/2021/07/21JUL2021.pdf. Published 2021. Accessed July 12, 2022.
| 36476523 | PMC9725735 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 1; 42(1):35-42 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003758 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00006
10.1097/INF.0000000000003740
3
COVID Reports
Monoclonal Antibody and Antiviral Therapy for Mild-to-Moderate COVID-19 in Pediatric Patients
https://orcid.org/0000-0002-9344-5185
Vora Surabhi B. MD, MPH *
Englund Janet A. MD [email protected]
*
Trehan Indi MD, MPH, DTM&H [email protected]
*
Waghmare Alpana MD [email protected]
*
Kong Ada PharmD [email protected]
†
Adler Amanda BA [email protected]
‡
Zerr Danielle M. MD, MPH [email protected]
*
From the * Department of Pediatrics, University of Washington, Seattle, Washington
† Department of Pharmacy, Seattle Children’s Hospital, Seattle, Washington
‡ Division of Infectious Diseases, Seattle Children’s Hospital, Seattle, Washington.
Address for correspondence: Surabhi B. Vora, MD, MPH, Seattle Children’s Hospital, 4800 Sand Point Way NE, Seattle, WA 98105. E-mail: [email protected].
21 10 2022
1 2023
21 10 2022
42 1 3234
14 9 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
Multiple antiviral and monoclonal antibody therapies are now available for mild-moderate COVID-19 in high-risk patients ≥12 years of age. However, data for the use of these agents in children is limited. We reviewed 94 pediatric patients for whom early therapy was requested since the emergence of the Omicron variant and describe patient characteristics, treatment logistics and associated short-term events.
SARS-CoV-2
pediatrics
monoclonal antibodies
antivirals
SDCT
==== Body
pmcThe surge of the SARS-CoV-2 Omicron variant (B.1.1.529) coincided with new treatment options for mild and moderate COVID-19 in high-risk adolescents and adults. The oral antivirals nirmatrelvir/ritonavir and monoclonal antibody sotrovimab1 were given Emergency Use Authorization (EUA) in the US for children ≥12 years of age, although studies leading to their approval did not include pediatric patients.2,3 Furthermore, a 3-day course of remdesivir reduced hospitalizations in high-risk outpatients but the study included only 8 patients <18 years of age.4 Molnupiravir was also only studied in adult patients and not authorized for children.5 In sum, data describing the use of any of these agents in pediatric patients is limited.
Based on US Food and Drug Administration (FDA) EUA criteria, pediatric-specific guidance,6 and national prioritization schema,7 our institution prioritized monoclonal antibodies and early antivirals for individuals who were severely immunocompromised or incompletely vaccinated with risk factors including severe obesity, medical complexity with respiratory technology dependence, or multiple other risk factors. Overall, 95% of SARS-CoV-2 strains were Omicron variant in our region by late December 2021.8 Here, we describe characteristics of patients approved and not approved for therapy, treatment-related process measures and short-term outcomes associated with early COVID-19 therapy in high-risk pediatric patients.
MATERIALS AND METHODS
In early 2021, our tertiary care children’s hospital established a COVID-19 Therapeutics Committee to review requests for early treatment of SARS-CoV-2 infection in high-risk patients. This Committee consisted of experts in Infectious Diseases, Emergency Medicine, and Pharmacy. The Committee added new antiviral therapies and monoclonal antibodies for the treatment of mild and moderate infection as these agents became available. Requests came primarily from pediatric subspecialty providers at our institution; however, community providers also applied for therapy for their patients. We encouraged requests for symptomatic severely immunocompromised patients including transplant and oncology patients, those with primary immunodeficiencies, severe obesity, or significant underlying cardiac or pulmonary conditions or other medical complexities (see Table, Supplemental Digital Content 1, http://links.lww.com/INF/E847, for further definition of high-risk conditions). However, providers could submit an intake for any potentially eligible patient, including those who were asymptomatic. Mild COVID-19 was defined as symptoms of viral illness or upper respiratory tract infection and moderate infection included patients with signs or symptoms of pneumonia without sustained hypoxia. The intake process required submission of information about patient age, weight, underlying condition(s), SARS-CoV-2 test positivity (type of test and date), symptom onset, and vaccination status.
To determine which treatment(s) to offer, we considered each individual’s age and weight, geographic location, vaccination status and ability to respond to vaccination, potential drug interactions, underlying renal and liver function and current infusion center or inpatient bed availability. We prioritized oral therapy (nirmatrelvir/ritonavir) when a patient was not hospitalized and eligible based on timing of infection, age and drug interactions. For those for whom nirmatrelvir/ritonavir was not an option, a monoclonal antibody (sotrovimab) was considered next during periods when it was available. Administration in our outpatient infusion center was preferred; when not possible, the Emergency Department (ED) was used for administration. A 3-day course of remdesivir was favored for eligible patients <12 years of age or <40 kg, patients who were already hospitalized, and patients who were not eligible for other therapies. As with the monoclonal antibody therapy, infusion of remdesivir in the outpatient infusion center was arranged when possible; if necessary, a patient could receive dosing in the ED or as an inpatient. Assessment of baseline renal and hepatic function was recommended before remdesivir administration. See Table, Supplemental Digital Content 2, http://links.lww.com/INF/E848) for treatment prioritization schema.
After IRB approval, medical records for patients for whom early COVID-19 treatment was requested between December 22, 2021, and January 30, 2022, were reviewed for patient demographics, underlying conditions, SARS-CoV-2 vaccination status, days since first symptom or positive test (PCR or rapid antigen), medication adherence, treatment associated adverse events, and ED visits and hospitalizations within 7 days after request intake. A Fisher-Exact test was used to calculate whether there was a significant difference between approved and not approved requests in terms of ethnicity.
RESULTS
COVID-19 treatment requests for 94 patients were reviewed; 66 (70%) received approval for sotrovimab, nirmatrelvir/ritonavir, or remdesivir (Table 1). Molnupiravir was not recommended for any patients. Immunocompromised patients comprised most requests (66%), with malignancy being the most frequent immunocompromising condition. The most common reasons for denial of therapy were fully vaccinated status (46%) and not being considered in the highest risk categories (61%). Supplies of the various agents varied throughout the study time period; nirmatrelvir/ritonavir and sotrovimab were especially limited in the early weeks. Only 1 patient who was deemed eligible did not receive any treatment caused by lack of availability. Self-identified race and ethnicity categorizations were similar between those for whom therapy was approved versus not approved.
TABLE 1. Characteristics of Patients for Whom COVID-19 Therapy Was Requested and Approved
Patient Characteristics Therapy Requested (N, %), N = 94 Therapy Approved (N, %), N = 66
Median age in y (range) 13 (0.6-24) 12 (0.6-24)
<5 16 (17) 12 (18)
5–11 20 (21) 13 (20)
12–17 49 (52) 33 (50)
18+ 9 (10) 8 (12)
Gender
Female 43 (46) 30 (45)
Ethnicity
Hispanic 33 (35) 24 (36)
Non-Hispanic 61 (65) 42 (64)
Race
White or Caucasian 41 (44) 27 (41)
Black or African-American 5 (5) 4 (6)
Asian 5 (5) 2 (3)
Hawaiian/Pacific-Islander 4 (4) 4 (6)
Alaska Native/American-Indian 3 (3) 3 (5)
Multiple 7 (7) 5 (7)
Other* 23 (24) 19 (29)
Unknown/Refused 6 (6) 2 (3)
Underlying condition(s)†
Immunocompromised 62 (66) 45 (68)
Malignancy 29 (31) 23 (35)
Solid organ transplant 9 (10) 7 (11)
Hematopoietic cell transplant 4 (4) 3 (5)
Primary immunodeficiency 6 (6) 5 (8)
Rheumatologic condition 5 (5) 3 (5)
Other immunocompromise 9 (10) 4 (6)
Obesity 13 (14) 12 (18)
Chronic lung disease 17 (18) 11 (17)
Chronic kidney disease 4 (4) 2 (3)
Congenital heart disease 6 (6) 3 (5)
Diabetes mellitus 3 (3) 1 (2)
Sickle cell disease 1 (1) 1 (2)
Vaccine status
Ineligible (<5 y of age) 16 (17) 12 (18)
Unknown 1 (1) 1 (2)
Eligible (5+ y of age) 77 (82) 53 (80)
0 doses 35 (37) 28 (42)
1 dose 6 (6) 6 (9)
2 doses 25 (27) 15 (23)
3 doses 11 (12) 4 (6)
Not approved (N = 28)‡
Fully vaccinatedc 13 (46) N/A
Not in highest risk categories 17 (61)
Too late in infection§ 4 (14)
No treatment available 1 (4)
Tixagevimab/cilgavimab recipient 1 (4)
* The majority of those who chose “Other” race also chose Hispanic ethnicity.
† Patients may have more than 1 indication.
‡ Defined as ≥2 doses of SARS-CoV-2 vaccination per Centers for Disease Control guidance.
§ Defined as ≥7 d for patients <12 y of age and ≥10 d for patients ≥12 y of age.
Median age of children who received therapy was 16, 14.5 and 8 years for sotrovimab, nirmatrelvir/ritonavir and remdesivir, respectively (see Table, Supplemental Digital Content 3, http://links.lww.com/INF/E849). Median days from start of infection to first dose of therapy was 3 for sotrovimab (range: 0–6) and nirmatrelvir/ritonavir (range: 1–6) and 2 (range: 0–9) for remdesivir. Therapy was not given to 19 patients, despite approval, most often because of family refusal or improving symptoms. Two patients did not receive remdesivir because baseline ALT was too high. Remdesivir treatment occurred most frequently in the inpatient setting (59% of the time), while sotrovimab was administered more often in the outpatient infusion center (55% of the time). Adverse events were rare following all agents; 1 patient had chest pain during infusion of sotrovimab and for 2 days after, leading to an ED visit. Another patient was seen in the ED for an episode of shortness of breath during a course of nirmatrelvir/ritonavir. Both were discharged in good condition. No patients had a rise in alanine aminotransferase or creatinine after remdesivir. Three patients were hospitalized during the 7-day follow-up period for potentially COVID-19 related disease: one who was not approved for treatment, one who was approved for sotrovimab but did not receive it, and one after treatment with remdesivir. See Table, Supplemental Digital Content 3, http://links.lww.com/INF/E849, for further details.
DISCUSSION
We provide one of the first descriptions of early COVID-19 treatment in high-risk pediatric patients infected with the Omicron variant of SARS-CoV-2. We found that these therapies were generally well-tolerated and subsequent COVID-19-related ED visits or hospitalizations were uncommon. Although logistical challenges because of both medication supply and administration were present during the study period, they did not significantly impact our ability to provide therapy to those deemed eligible by our internal criteria.
While most SARS-CoV-2 infected children do well without therapy, treatment of mild or moderate infection should be considered in those at highest risk of progression to severe disease. Recent data demonstrates that hospitalization rates in children 0–4 years increased significantly during times of Omicron-variant predominance; information regarding treatment in pediatric patients has thus become even more important.9 However, despite FDA EUA for sotrovimab and nirmatrelvir/ritonavir, and now bebtelovimab,10 in patients ≥12 years of age and more recent FDA approval of remdesivir for children <12 years, safety and efficacy data for these agents in pediatric patients remain scarce. In this study, we begin to demonstrate the feasibility of allocation and administration of these agents in pediatric patients, especially high-risk outpatients.
Several patients in our study received therapy as inpatients. These were generally patients hospitalized for reasons other than COVID-19 who were found to be positive for SARS-CoV-2 during admission screening or patients for whom outpatient administration of IV therapeutics was not feasible because of limited capacity in our infusion center or ED. Remdesivir most required inpatient admission due to the need for multiple intravenous doses over several days. As availability of nirmatrelvir/ritonavir has increased in the ensuing months since our study, admissions solely for the administration of remdesivir decreased.
Many patients who were approved for therapy did not ultimately receive it. Reasons for this varied by therapy but family refusal was most common in those approved for remdesivir. Other reasons included symptoms improving without therapy and preremdesivir baseline ALT being too high. Only one patient did not receive therapy because of limited infusion center capacity.
Although we saw no indicators of disparities in our approval process, we continue work to explore potential equity issues as documented in the distribution of COVID-19 therapies in adults.11,12 Disparities before the actual referral submission would not be visible in our analysis as we did not include all patients who were potentially eligible for COVID-19 treatment.
Limitations of this study include its observational and retrospective nature, lack of controls, small sample size at a single center, short follow-up period, and the possibility that not all events and outcomes were reported in the medical record. Given this, the efficacy of these treatments could not be assessed. In addition, systematic monitoring of adverse events was not performed. As symptomatic patients with severe underlying conditions were more likely to have therapies requested and approved, assessment of treatment impact on broader populations is difficult. Larger studies examining the safety and efficacy of early treatment of COVID-19 specifically in high-risk pediatric patients are required.
Supplementary Material
J.A.E. receives research support from AstraZeneca and Pfizer and is a consultant for AstraZeneca, Moderna and Meissa Vaccines. A.W. reports grant support from Ansun Biopharma, Allovir, GSK/Vir, and Pfizer and is an Advisory Board Member for Kyorin Pharmaceutical. All other authors have no conflicts of interest relevant to this work to disclose.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.pidj.com).
==== Refs
REFERENCES
1. Gandhi RT Malani PN del Rio C . COVID-19 Therapeutics for nonhospitalized patients. JAMA. 2022;327 :617–618.35029659
2. Hammond J Leister-Tebbe H Gardner A . Oral nirmatrelvir for high-risk nonhospitalized adults with Covid-19. N Engl J Med. 2022;386 :1397–1408.35172054
3. Gupta A Gonzalez-Rojas Y Juarez E . Early treatment for Covid-19 with SARS-CoV-2 neutralizing antibody sotrovimab. N Engl J Med. 2021;385 :21.
4. Gottlieb RL Vaca CE Paredes R . Early remdesivir to prevent progression to severe Covid-19 in outpatients. N Engl J Med. 2022;386 :305–315.34937145
5. Fact Sheet for Healthcare Providers: Emergency Use Authorization for Lagevrio (molnupiravir) Capsules. August 2022. Available at: https://www.fda.gov/media/155054/download. Accessed September 7, 2022.
6. Wolf J Abzug MJ Anosike BI . Updated guidance on use and prioritization of monoclonal antibody therapy for treatment of COVID-19 in adolescents. J Pediatric Infect Dis Soc. 2022;11 :177–185.35107571
7. The COVID-19 Treatment Guidelines Panel’s Interim Statement on Patient Prioritization for Outpatient Anti-SARS-CoV-2 Therapies or Preventive Strategies When There Are Logistical or Supply Constraints. National Institutes of Health COVID 19 Treatment Guidelines, December 2021.
8. UW Virology COVID-19 Dashboard (washington.edu). Accessed September 7, 2022.
9. Marks KJ Whitaker M Agathis NT . Hospitalization of infants and children aged 0-4 years with laboratory-confirmed COVID-19-COVID-NET, 14 states, March 2020–February 2022. MMWR. 2022;71 :429–436.35298458
10. Fact Sheet for Healthcare Providers: Emergency Use Authorization for Bebtelovimab. August 2022. Available at: https://pi.lilly.com/eua/bebtelovimab-eua-factsheet-hcp.pdf. Accessed September 7, 2022.
11. Wiltz JL Feehan AK Molinari NM . Racial and ethnic disparities in receipt of medications for treatment of COVID-19-United States, March 2020-August 2021. MMWR. 2022;71 :96–102.35051133
12. Gold JAW Kelleher J Magid J . Dispensing of oral antiviral drugs for treatment of COVID-19 by zip code-level social vulnerability- United States, December 23, 2021-May 21, 2022. MMWR. 2022;71 :825-829.35737571
| 36476522 | PMC9725736 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 21; 42(1):32-34 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003740 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00010
10.1097/INF.0000000000003751
3
COVID Reports
Positive Predictive Value of Rapid Antigen Tests in School During SARS-CoV-2 Omicron Variant Surge
Jang Eun Jung MPH [email protected]
*
https://orcid.org/0000-0003-2733-0715
Choe Young June MD [email protected]
†
Yun Go-Woon MS [email protected]
*
Kim Ryu Kyung MS [email protected]
*
Jeong Heegwon BS ‡
Lee Sangwon PhD [email protected]
*
Park Young-Joon MD, MPH *
From the * Director for Epidemiological Investigation Analysis, Korea Disease Control and Prevention Agency, Cheongju, Republic of Korea
† Department of Pediatrics, Korea University Anam Hospital, Seoul, Republic of Korea
‡ Student Health Policy Division, Ministry of Education, Sejong, Republic of Korea.
Address for correspondence: Young-Joon Park, MD, MPH, (28159) Korea Disease Control and Prevention Agency, Osong Health Technology Administration Complex, 187, Osongsaengmyeong 2-ro, Osong-eup, Heungdeok-gu, Cheongju-si, Chungcheongbuk-do, Korea. E-mail: [email protected].
18 10 2022
1 2023
18 10 2022
42 1 e6e8
21 9 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
In Korea, we conducted a national observational study to calculate the positive predictive value of SARS-CoV-2 rapid antigen tests in K-12 schools during the Omicron variant surge in March 2022. The weekly positive predictive value ranged from 86.4% to 93.2%. The positive predictive value was the highest among elementary school students with symptoms (95.7%) and lowest among teachers/staff without symptoms (70.9%).
rapid antigen test
RAT
COVID-19
SARS-CoV-2
school
screening
STATUSONLINE-ONLY
SDCT
==== Body
pmcThe coronavirus disease 2019 (COVID-19) has led to the closure of schools worldwide, resulting in significant disruption in children’s learning.1 In South Korea, schools K-12 and universities opened in the spring of 2022 during the Omicron variant surge.2 The Ministry of Education introduced guidelines outlining the use of random screening in the rapid antigen test (RAT) for symptomatic and asymptomatic students and teachers/staff.3 RAT aims to provide timely identification of infection, but there are concerns regarding the positive predictive value of testing.4 This study investigated the true and false positives of the RAT in a large sample of students and teachers/staff serially screened throughout South Korea during the SARS-CoV-2 Omicron variant surge.
METHODS
Since March 2022, nationwide RAT screening for SARS-CoV-2 has been implemented in kindergartens (4–6 years of age), elementary schools (grades 1–6), middle schools (grades 7–9) and high schools (grades 10–12). Students with and without symptoms were tested twice weekly, and teacher/staff were screened once weekly. Those who had a close contact history with patients with SARS-CoV-2 were screened 3 times per week voluntarily. Each student was asked to report about symptoms (multiple choice), including fever, chills, cough, headache, myalgia, sore throat, runny nose, loss of taste/smell, fatigue, nausea, vomiting or diarrhea. The RAT results were reported through the Student Health Self-Diagnosis App developed by the Ministry of Education. If the RAT result was positive, the patient would be referred for a confirmatory polymerase chain reaction (PCR) test within 72 hours. We merged the symptoms and RAT results data with the PCR data collected between March 2 and April 3, 2022. Individuals with incomplete information (ie, input errors, blank dates) and a history of SARS-CoV-2 infection were excluded from the analysis. The RAT results were compared with the PCR results, and the positive predictive value and binomial 95% confidence intervals were calculated. The weekly positivity rate of COVID-19 in the RAT and PCR was estimated by dividing the number of enrolled students and teachers/staff with COVID-19 by the total number of tested individuals.
The study was reviewed and approved by the Korea Disease Control and Prevention Agency Institutional Review Board (2021-12-03-PE-A).
RESULTS
Between March 2 and April 3, 2022, of the 2,822,544 RAT results collected, 136,902 (4.9%) were positive, of which 126,042 were confirmed by PCR, resulting in a positive predictive value of 92.1% (Table 1). The RAT positivity rate ranged from 68.7% to 83.8% among individuals with symptoms, while the rate ranged from 0.3% to 0.7% in individuals without symptoms (Table 1). The weekly SARS-CoV-2 RAT positivity rate in South Korea ranged from 4.9% to 5.9%, whereas the positive predictive value of RAT ranged from 86.4% to 93.2% (Figure 1, Supplemental Digital Content 1, http://links.lww.com/INF/E851). The positive predictive value was the highest among elementary school students with symptoms (95.7%) and lowest among teachers/staff without symptoms (70.9%) (Table 1). In all groups, the RAT positive predictive value was higher in patients with symptoms than that in those without symptoms (Fig. 1).
TABLE 1. Result of Nationwide School COVID-19 RAT Screening Reflexed to PCR Confirmation During SARS-CoV-2 Omicron Variant Surge in South Korea, March 2 to April 3, 2022
Symptom* Status Tested Persons RAT Screening Reflex PCR Confirmation†
Positive Negative Positive Negative
n % n % n % n %
Symptomatic Students Kindergarten 8974 6872 76.6 2102 23.4 6499 94.6 373 5.4
Elementary school 79,435 56,246 70.8 23,189 29.2 53,820 95.7 2426 4.3
Middle school 36,551 25,119 68.7 11,432 31.3 23,770 94.6 1349 5.4
High school 28,679 20,083 70.0 8596 30.0 18,228 90.8 1855 9.2
Teachers/staffs 13,060 10,943 83.8 2117 16.2 9088 83.1 1855 17.0
Asymptomatic Students Kindergarten 172,746 1133 0.7 171,613 99.3 969 85.5 164 14.5
Elementary school 1,087,744 7534 0.7 1,080,210 99.3 6659 88.4 875 11.6
Middle school 614,195 4568 0.7 609,627 99.3 3773 82.6 795 17.4
High school 474,160 3392 0.7 470,768 99.3 2519 74.3 873 25.7
Teachers/staffs 307,000 1012 0.3 307,000 99.7 717 70.9 295 29.2
Total 2,822,544 136,902 4.9 2,686,654 95.1 126,042 92.1 10,860 7.9
* Symptom refers to fever, chills, cough, headache, myalgia, sore throat, runny nose, loss of taste/smell, fatigue, nausea, vomiting, and diarrhea.
† All positive results from RAT screening are reflexed to undergo PCR testing for confirmation.
FIGURE 1. The positive predictive value of rapid antigen test screening among students and teachers/staff by presence or absence of symptoms (fever, chill, cough, headache, myalgia, sore throat, runny nose, loss of taste/smell, fatigue, nausea, vomiting, and diarrhea).
DISCUSSION
The overall rate of true positive results among the RAT tests for COVID-19 was high, especially in students with symptoms, which is consistent with other studies.5 Establishing an early diagnosis of COVID-19 requires interpreting test results appropriately. Currently, RAT is widely available and its appropriate use may improve case detection and therefore could reduce transmission and absenteeism.6 As this study was conducted parallel with the policy of returning to school if the RAT result was negative, the results may inform the policymakers to minimize school closure due to the COVID-19 outbreak.
The probability of infection depends on the SARS-CoV-2 prevalence in the community, which was high in South Korea with the Omicron variant surge during the observation period, which increased the performance of RAT in schools.7,8 During the peak of the COVID-19 community outbreak, a positive RAT result is highly likely to represent a true SARS-CoV-2 infection. Furthermore, a negative test result suggests a low possibility of COVID-19. Here, most students presenting with respiratory symptoms and fever with a positive RAT result would have COVID-19, and conversely, if the prevalence is low, positive RAT results are more likely to be false-positive. The study limitations include the convenience sample based on passive surveillance using volunteered symptoms and RAT positivity, therefore may lack clear generalizability. Moreover, reported symptoms were aggregated as symptomatic versus asymptomatic, therefore, we were unable to assess the most significant symptom suggestive of RAT or PCR positivity.
Interpreting SARS-CoV-2 RAT appropriately can provide useful information for decision-making in the educational setting. RAT screening during the community outbreak of SARS-CoV-2 provides a high positive predictive value; therefore, provides helpful information to policymakers to optimize the school health policy.
ACKNOWLEDGMENTS
We thank the Ministry of the Interior and Safety, Si/Do and Si/Gun/Gu, medical health centers staff, and medical facilities for their efforts in responding to the COVID-19 outbreak.
Supplementary Material
The authors have no funding or conflicts of interest to disclose.
Eun Jung Jang and Young June Choe contributed equally to the manuscript.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.pidj.com).
==== Refs
REFERENCES
1. Esposito S Principi N . School closure during the coronavirus disease 2019 (COVID-19) pandemic: an effective intervention at the global level? JAMA Pediatr. 2020;174 :921–922.32401277
2. Kim EY Choe YJ Park H . Community Transmission of SARS-CoV-2 omicron variant, South Korea, 2021. Emerg Infect Dis. 2022;28 :898–900.35171760
3. Ministry of Education. COVID-19 related guidelines in education setting [Korean]. Available at: https://www.moe.go.kr/boardCnts/listRenew.do?boardID=72756&renew=72756&m=031304&s=moe. Accessed April 7, 2022.
4. Peeling RW Olliaro PL Boeras DI . Scaling up COVID-19 rapid antigen tests: promises and challenges. Lancet Infect Dis. 2021;21 :e290–e295.33636148
5. Dinnes J Deeks JJ Berhane S . Rapid, point-of-care antigen and molecular-based tests for diagnosis of SARS-CoV-2 infection. Cochrane Database Syst Rev. 2021;3 :Cd013705.33760236
6. Schecter-Perkins EM Doron S Johnston R . A test-to-stay modified quarantine program for COVID-19 in schools. Pediatrics. 2022;149 :e2021055727.35132435
7. Good CB Hernandez I Smith K . Interpreting COVID-19 test results: a Bayesian approach. J Gen Intern Med. 2020;35 :2490–2491. doi:10.1007/s11606-020-05918-8.32495086
8. Lee JJ Choe YJ Jeong H . Importation and transmission of SARS-CoV-2 B.1.1.529 (Omicron) variant of concern in Korea, November 2021. J Korean Med Sci. 2021;36 :e346.34962117
| 36476526 | PMC9725737 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 18; 42(1):e6-e8 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003751 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00008
10.1097/INF.0000000000003743
3
COVID Reports
Serious Infusion Reactions in Two Adolescents Receiving Bebtelovimab
Peters Megan E. MD [email protected]
*
Strayer Jill RPh [email protected]
†
Wald Ellen R. MD *
From the * Department of Pediatrics, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin
† Department of Pharmacy, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin.
Address for correspondence: Ellen R. Wald, MD, Department of Pediatrics, University of Wisconsin, Madison, 600 Highland Ave, H4/458 CSC, Madison, WI 53792. E-mail: [email protected].
22 11 2022
1 2023
22 11 2022
42 1 e1e3
15 9 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
There is scant literature available for pediatric prescribers regarding safety and efficacy of monoclonal antibody formulations against coronavirus disease 2019 (COVID-19). Here, we present 2 cases of serious infusion reactions in adolescent patients receiving the monoclonal antibody bebtelovimab and a succinct review of available antiviral medications for pediatric patients with mild or moderate COVID-19.
adolescents
coronavirus disease 2019 (COVID-19)
monoclonal antibody
bebtelovimab
adverse reactions
STATUSONLINE-ONLY
==== Body
pmcThe coronavirus disease 2019 (COVID-19) pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has now led to the infection of over 13.5 million children in the United States.1 The morbidity and mortality from COVID-19 is higher in pediatric patients with comorbid conditions including obesity and immunodeficiency.2 Therapeutic options to treat mild to moderate infections in adult and adolescent outpatients at increased risk for progression to severe COVID-19 include antivirals and monoclonal antibodies. Safety and efficacy data are minimal in children and adolescents who have received monoclonal antibodies for COVID-19, and there is less experience with their use in pediatric patients. Here, we present serious adverse reactions associated with the monoclonal antibody bebtelovimab in 2 adolescents.
Case 1 is a 16-year-old male who had congenital posterior urethral valves, which led to end-stage kidney disease resulting in a renal transplant in August 2021. His current immunosuppressive regimen included mycophenolate and tacrolimus. He was fully vaccinated and received a booster against COVID-19 1 year and again 5 months before presentation, respectively. He tested positive for SARS-CoV-2 on a home antigen test 4 days before coming to the infusion center. His symptoms included sore throat, cough, rhinorrhea, and fatigue. His creatinine was 1.14 mg/dL. He was eligible to receive bebtelovimab owing to his age >12 years, weight >40 kg, and immunocompromised status.
Bebtelovimab 175 mg was administered over 1 minute per manufacturer’s instructions through a peripheral IV and flushed slowly through the IV according to standard practice in our unit. Within 2 minutes, the patient began complaining of difficulty breathing. On examination, he was flushed in appearance with notable agitation and respiratory distress. He had marked facial swelling. No urticaria were noted, but the skin over his face and trunk was flushed. No wheezing was heard on chest auscultation. Oxygen, intramuscular epinephrine, and intravenous diphenhydramine were administered. The patient’s respiratory symptoms resolved within 5 minutes; facial swelling improved. He was admitted overnight to observe for possible rebound symptoms of anaphylaxis. He was discharged home the following morning after an uneventful night.
Case 2 is a 15-year-old female with moderate persistent asthma managed with inhaled steroids and albuterol. She was fully vaccinated against COVID-19 vaccination and received a booster dose 1 year and again 5 months before presentation, respectively. She tested positive for SARS-CoV-2 4 days before coming to the infusion center on day 6 of her illness. She had a sporadic cough but no other symptoms. She was eligible for bebtelovimab owing to her age >12 years, weight >40 kg, and her chronic respiratory condition of asthma.
Informed by the previous experience with case 1, and with the current patient’s atopic tendencies, the care team opted to administer intravenous diphenhydramine before administering bebtelovimab 175 mg IV over 1 minute. The medication was flushed slowly through the IV tubing over 1 minute. Two minutes following administration, the patient reported abdominal discomfort and difficulty breathing. She developed diffuse flushing. On examination, she was dyspneic. There was no swelling of her lips or tongue and no urticarial lesions. Lungs were clear to auscultation. Oxygen was administered via facemask. She was given a dose of intramuscular epinephrine, followed by intravenous diphenhydramine, and her symptoms improved within minutes. She was discharged from the infusion center 3 hours later with a prescription for an epinephrine injector in the event of rebound symptoms.
DISCUSSION
Preventing the Progression of Mild or Moderate COVID-19
There are several options for management of outpatients with mild or moderate COVID-19, which aim to prevent disease progression resulting in hospitalization or death. These therapeutic options are specifically aimed toward adolescents and adults who are at risk for hospitalization or death from COVID-19 and are not intended for general use. Choosing an optimal therapy should be informed by multiple considerations, including an individual patient’s risk factors for severe COVID-19, potential drug-drug interactions, and availability of treatments in a specific geographic area. Our patients’ risk factors for progression to severe COVID-19 were immunosuppression from solid organ transplantation and moderate persistent asthma. Both conditions are associated with “moderate risk of severe COVID-19.”3 Current recommendations from the National Institutes of Health favor the use of Paxlovid (Pfizer, NYC, NY) or Remdesivir (Gilead Sciences, Foster City, CA) in the setting of mild to moderate COVID in adults; if these options are not available or clinically cannot be used, bebtelovimab (Eli Lilly Company, Indianapolis, IN) can be considered.4 However, the paucity of pediatric-specific data has been a limitation in the development of similar recommendations from the American Academy of Pediatrics.5
Potential Therapies in Children
Paxlovid is an attractive first choice as oral treatment in adolescent outpatients at high risk for progression to severe disease, in the absence of potential significant drug-drug interactions. Paxlovid, effective against all variants of SARS-CoV-2, is a combination of 2 protease inhibitors, nirmatrelvir, and ritonavir. Nirmatrelvir is the component with activity against SARS-CoV-2; ritonavir interferes with its degradation and allows higher levels of nirmatrelvir for a longer time. If started within 5 days of onset of symptoms, Paxlovid has shown 89% effectiveness in reducing hospitalizations and death compared with placebo in an unvaccinated population of adult patients at high risk for progression to severe disease.6 While oral medications for home treatment are extremely attractive, the complexities of using Paxlovid are considerable and relate primarily to the large number of potential drug interactions encountered by the infected individuals most likely to benefit from use of the drug. An excellent resource available to guide clinicians is the US Food and Drug Administration Paxlovid Patient Eligibility Screening Checklist, which offers a comprehensive list of drugs and their potential for serious drug interactions with Paxlovid.7 The guidance of a pharmacist is much appreciated in the decision-making process. For example, to administer Paxlovid safely to the patient who had received a renal transplant requires careful management of his tacrolimus dosing, owing to the potential for either (1) toxicity from supratherapeutic drug levels or (2) graft rejection in the setting of subtherapeutic drug levels. For the patient with asthma, no adjustments in dosage would be required, although she was not eligible to receive Paxlovid as she presented on day 6 of illness.
Remdesivir is the next choice for treatment. Remdesivir, an antiviral agent, targets the highly conserved viral RNA-dependent RNA polymerase and is active against all variants of SARS-CoV-2. It is currently approved for use in all pediatric patients >28 days and >3 kg and should be administered within 7 days of the onset of symptoms. Remdesivir requires 3 successive days of intravenous infusions of the drug; it has been shown to result in an 87% lower risk of hospitalization or death than placebo in an unvaccinated population of adult patients at high risk for progression to severe disease.8 Although the safety profile is very acceptable, in our institution a 3-day requirement for the infusion center presents logistic problems for administration over the weekend. Both of our patients would have been eligible for this treatment and presumed to benefit, although neither received remdesivir owing to these logistical constraints.
The last choice is monoclonal antibody. Treatment of COVID-19 adult outpatients at high risk for progression to severe disease with some monoclonal antibodies has reduced the risk of hospitalization and death by up to 85%.9 The administration of monoclonal antibodies requires an intravenous infusion of a one-time dose within 7 days of onset of symptoms. The major impediment to use has been loss of activity against new variants that have appeared and circulated. The single monoclonal antibody retaining activity against currently circulating subvariants of Omicron is bebtelovimab. However, its newness precludes the availability of published or unpublished data on effectiveness in high-risk populations. Before the circulation of the omicron variant, an unpublished trial among low-risk outpatients showed a reduction in time to sustained symptom resolution from 8 to 6 days compared with placebo.10,11 Consequent to their characterization as proteins monoclonal antibodies are known to confer elevated risk for immunogenicity or hypersensitivity reactions.12 However, only 0.3% of patients were reported to have had infusion reactions in the information concerning bebtelovimab,10 which is comparable to rates of infusion reactions for other types of anti-SARS-CoV-2 monoclonal antibodies.3 The unexpected intensity and frequency of both of our patients developing immediate, severe infusion reactions (anaphylaxis) following administration of bebtelovimab prompted us to bring this to attention.
Our Experience With Anti-SARS-CoV-2 Monoclonal Antibodies
Since 2021 and before the availability of bebtelovimab, we treated 27 adolescents with monoclonal antibody infusions against SARS-CoV-2, including bamlanivimab, casirivimab+imdevimab, bamlanivimab+etesevimab, tixagevimab+cilgavimab, and sotrovimab. Of those patients, 3 had symptoms consistent with a mild infusion reaction, including skin flushing or shortness of breath. Since the availability of bebtelovimab, both patients receiving this drug developed severe reactions consistent with anaphylaxis, and 1 required hospitalization overnight. As there are limited data available for health-care providers concerning the risks versus benefits of monoclonal antibody infusions, we present these cases so that outpatient prescribers may better inform their decision-making for the care of patients with COVID-19.
CONCLUSION
Pediatric outcomes of those who receive monoclonal antibody therapies against COVID-19 are rarely described. Here, we report severe infusion reactions for 2 patients receiving bebtelovimab, one of whom required escalation in care, and describe the treatment options to prevent progression of disease in patients with mild or moderate COVID-19. Prescribers of anti-COVID-19 monoclonal antibody therapies should be aware of the potential for severe infusion reactions or anaphylaxis from these medications.
The authors have no funding or conflicts of interest to disclose.
==== Refs
REFERENCES
1. American Academy of Pediatrics. Children and COVID-19: State-Level Data Report. Summary of Findings. Available at: https://www.aap.org/en/pages/2019-novel-coronavirus-covid-19-infections/children-and-covid-19-state-level-data-report/. Accessed June 14, 2022.
2. Woodruff RC Campbell AP Taylor CA . Risk factors for severe COVID-19 in children. Pediatrics. 2022;149 :e2021053418.34935038
3. Wolf J Abzug M Anosike B . Updated guidance on use and prioritization of monoclonal antibody therapy for treatment of COVID-19 in adolescents. J Pediatric Infect Dis Soc. 2022;11 :177–185.35107571
4. National Institutes of Health. Therapeutic Management of Nonhospitalized Adults with COVID-19. Available at: https://www.covid19treatmentguidelines.nih.gov/management/clinical-management/nonhospitalized-adults--therapeutic-management/?utm_source=site&utm_medium=home&utm_campaign=highlights. Accessed June 14, 2022.
5. American Academy of Pediatrics. Management Strategies in Children and Adolescents with Mild to Moderate COVID-19. Available at: https://www.aap.org/en/pages/2019-novel-coronavirus-covid-19-infections/clinical-guidance/outpatient-covid-19-management-strategies-in-children-and-adolescents/. Accessed June 14, 2022.
6. United States Food and Drug Administration. Fact Sheet for Healthcare Providers: Emergency Use Authorization for Paxlovid. Available at: https://www.fda.gov/media/155050/download. Accessed June 14, 2022.
7. United States Food and Drug Administration. PAXLOVID Patient Eligibility Screening Checklist Tool for Prescribers. Available at: https://www.fda.gov/media/158165/download. Accessed June 14, 2022.
8. Gottlieb RL Vaca CE Paredes R . Early remdesivir to prevent progression to severe Covid-19 in outpatients. N Engl J Med. 2022;386 :305–315.34937145
9. Gupta A Gonzalez-Rojas Y Juarez E . Early treatment for COVID-19 with SARS-CoV-2 neutralizing antibody sotrovimab. N Engl J Med. 2021;285 :1941–1950.
10. United States Food and Drug Administration. Fact Sheet for Healthcare Providers: Emergency Use Authorization for Bebtelovimab. Available at: https://www.fda.gov/media/156152/download. Accessed June 14, 2022.
11. Dougan M Azizad M Chen P . Bebtelovimab, Alone or Together with Bamlanivimab and Etesevimab, as a Broadly Neutralizing Monoclonal Antibody Treatment for Mild to Moderate, Ambulatory COVID-19. Available at: https://www.medrxiv.org/content/10.1101/2022.03.10.22272100v1. Accessed August 29, 2022.
12. Picard M Galvão VR . Current knowledge and management of hypersensitivity reactions to monoclonal antibodies. J Allergy Clin Immunol Pract. 2017;5 :600–609.28110056
| 36476524 | PMC9725739 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 22; 42(1):e1-e3 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003743 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00009
10.1097/INF.0000000000003750
3
COVID Reports
Bulging Anterior Fontanelle Caused by Severe Acute Respiratory Syndrome Coronavirus-2
Sethuraman Chidambaram MBBS, MD, MRCPCH *
Holland Jonathon MRCPCH, MB BChir, BA [email protected]
†
Priego Gema MD, FRCP [email protected]
‡
Khan Faizullah MBBS, BSc [email protected]
*
Johnson Richard MB BChir [email protected]
*
Keane Morgan MB BCh, MRCP, MRCPCH [email protected]
*
From the * Department of Paediatrics, Queens Hospital, Barking, Havering, and Redbridge University Hospital Trust, Rom Valley Way, Romford RM7 0AG, United Kingdom
† Department of Paediatric Neurology, Great Ormond Street Hospital for Children NHS Trust, Great Ormond Street, London WC1N 3JH, United Kingdom
‡ Department of Radiology, Queens Hospital, Barking, Havering, and Redbridge University Hospital Trust, Rom Valley Way, Romford RM7 0AG, United Kingdom.
Address for correspondence: Chidambaram Sethuraman, MBBS, MD Paediatrics, MRCPCH, Department of Paediatrics, Queens Hospital, Barking, Havering and Redbridge University Hospital Trust, Rom Valley Way, Romford RM7 0AG, United Kingdom. E-mail: [email protected]
14 10 2022
1 2023
14 10 2022
42 1 e4e5
16 9 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
Neurologic manifestations of the 2019 novel coronavirus disease in children are varied. We present the case of a 9-month-old child with bulging anterior fontanelle caused by severe acute respiratory syndrome coronavirus-2.
bulging anterior fontanelle
severe acute respiratory syndrome coronavirus-2
coronavirus disease 2019
children
STATUSONLINE-ONLY
==== Body
pmcCASE PRESENTATION
A 9-month-old Afro-Caribbean boy presented to the emergency department in April 2022 with a 1-day history of fever (maximum axillary temperature 39.6°C) and an episode of vacant staring lasting for a few seconds associated with labored breathing. He returned to his normal self within a few minutes. He had symptoms of rhinorrhea and cough for 1 day before fever. There was no diarrhea, vomiting, or rash.
The patient was born at term via normal vaginal delivery, and there were no concerns at birth. His mother was unvaccinated against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). Past medical history was noted of eczema and sickle cell trait. He has achieved age-appropriate milestones and has been immunized as per the UK schedule to date. There is no family history of febrile seizures.
On arrival at the emergency department, his axillary temperature was 39.4°C. He was tachycardic with a heart rate of 174/min. His other observations were respiratory rate 40/min, blood pressure 90/53 mm Hg, capillary refill time <2 seconds and oxygen saturation 97 % in air.
He appeared tired but was responsive to commands. His cardiac, pulmonary, and abdominal examinations were normal. On neurologic examination, he was found to have a bulging and pulsatile anterior fontanelle (AF) while sleeping and in both supine and upright positions. There were no focal neurologic deficits.
The initial laboratory investigations included blood and urine samples. His total leukocyte count was 15 (reference range for age 6–17 × 109 L) with 50% neutrophils and C-reactive protein of less than 1 mg/L (normal range 0–5 mg/L). His urea, creatinine, bilirubin, alanine aminotransferase, albumin and electrolyte levels were within normal limits. His urine dip urinalysis was negative.
In view of having a bulging fontanelle together with fever and an episode of vacant staring, the possibility of raised intracranial pressure secondary to meningitis/encephalitis was considered. Computed tomography of the brain performed before lumbar puncture showed bulging of the AF and widening of the major calvarial sutures, consistent with raised intracranial pressure, with no evidence of hydrocephalus. Lumbar puncture was unremarkable with clear cerebrospinal fluid (CSF), <1 WBC, glucose 3.5 mmol/L and protein 0.25 g/L (laboratory normal range 0.05–0.29 g/L). CSF culture for herpes simplex virus-1 and 2, varicella zoster virus and human herpes virus-6 were negative. His blood culture was sterile. Noncontrast magnetic resonance imaging (MRI) brain done after 3 days of admission showed no structural abnormality and reduction of bulging through AF; however, there was nonspecific finding of 3 tiny foci of increased signal intensity in the deep white matter of the right superior parietal region, right superior frontal lobe and left supramarginalis subcortical white matter. His nasopharyngeal aspirate was negative for influenza virus (A and B), rhinovirus, parainfluenza virus (1, 2, and 3), adenovirus, respiratory syncytial virus and enterovirus and positive for SARS-CoV-2 RNA (respiratory multiplex polymerase chain reaction) which was confirmed again with a nasopharyngeal swab (dedicated polymerase chain reaction). Further genome analysis showed omicron variant of SARS-CoV-2.
The child was treated with ceftriaxone, clarithromycin and acyclovir, which were discontinued after 4 days with negative CSF and blood cultures. He did not receive steroids, intravenous immunoglobulin, remdesivir or monoclonal antibodies. The patient improved clinically in the ward, and his AF was normal at discharge on day 5. He was reviewed after 3 weeks in clinic, at which time his AF remained normal and no other concerns had arisen.
DISCUSSION
In contrast to other respiratory viruses, children typically have less severe symptoms with SARS-CoV-2 infection, and the commonly proposed hypothesis behind this include age-related increased endothelial damage in adults, higher density and increased affinity of angiotensin converting enzyme 2 receptors in adults and high prevalence of pre-existing comorbidities in adults.1
Neurologic manifestations in children with COVID-19 are varied. SARS CoV-2 virus has neurotropic potential like other respiratory viruses, such as influenza and respiratory syncytial virus.2 Evidence of neurotropism with resultant astrocytic and neuronal injury is provided by studies demonstrating elevated serum levels of biomarkers such as glial fibrillary acidic protein and serum neurofilament light chain.3 Retrograde transmission through the olfactory neurons, disinhibition of renin angiotensin pathway and cytokine-induced neuroinflammation are some of the possible mechanisms behind the neurologic effects of COVID-19.2 The presentations commonly described in the literature include headaches overlapping with idiopathic intracranial hypertension,4 demyelinating disorders, stroke, encephalopathy and cerebral edema.5
Bulging of the AF in an infant is often a sign of raised intracranial pressure, and the differential diagnosis is wide, including infections, intracranial bleeds, space occupying lesions and trauma. With regard to infection, in a study done on 153 febrile children with bulging AF, 1 child had bacterial meningitis and the other etiologies were aseptic meningitis (26.7%), upper respiratory tract infection (18.3%), viral not otherwise specified (15.6%), roseola infantum (8.5%), acute otitis media (6.5%) and pneumonia (4.5%).6 In our case with sterile CSF, blood, urine, negative human herpes virus-6 and negative nasopharyngeal aspirate for all respiratory viruses except for SARS-CoV-2, COVID-19 was the likely etiology. This is further supported by the finding of foci of increased signal intensity in the white matter in MRI which has been described as the most common radiological abnormality in children with COVID-19. In a multicenter and multinational study done looking at the neuroimaging manifestations of children with COVID-19, patchy or confluent areas of T2 hyperintensity in the gray and white matter with or without reduced diffusion or enhancement was the commonly reported one and found in 28 (74%) of 38 children.7 A similar presentation with fever and raised fontanelle because of SARS-CoV-2 in a 4-month-old has been recently published in the literature.8
Two years into the pandemic, SARS-CoV-2 has been associated with many different pathological presentations, and we suggest considering SARS-CoV-2 as a potential differential, and diagnosis of exclusion, in a child with a bulging fontanelle.
FIGURE 1. A: Coronal reconstruction shows marked bulging of the brain parenchyma through the AF (solid white arrows). The remaining brain was unremarkable on the computed tomography. B: MRI fluid-attenuated inversion recovery axial sequence showing focal areas of subcortical/deep white matter signal abnormality (dashed white arrows), which were nonspecific, but has been extensively described in SARS-CoV-2 infection.
The authors have no funding or conflicts of interest to disclose.
==== Refs
REFERENCES
1. Zimmermann P Curtis N . Why is COVID-19 less severe in children? A review of the proposed mechanisms underlying the age-related difference in severity of SARS-CoV-2 infections. Arch Dis Child. 2021;106 :429–439.
2. Govil-Dalela T Sivaswamy L . Neurological effects of COVID-19 in children. Pediatr Clin North Am. 2021;68 :1081–1091.34538300
3. Kanberg N Ashton NJ Andersson LM . Neurochemical evidence of astrocytic and neuronal injury commonly found in COVID-19. Neurology. 2020;95 :e1754–e1759.32546655
4. Verkuil LD Liu GT Brahma VL . Pseudotumor cerebri syndrome associated with MIS-C: a case report. Lancet (London, England). 2020;396 :532.
5. LaRovere KL Riggs BJ Poussaint TY . Neurologic involvement in children and adolescents hospitalized in the United States for COVID-19 or multisystem inflammatory syndrome. JAMA Neurol. 2021;78 :536–547.33666649
6. Shacham S Kozer E Bahat H . Fever and bulging fontanelle in infants. Arch Dis Child. 2008;93 :pw183.
7. Lindan CE Mankad K Ram D ; ASPNR PECOBIG Collaborator Group. Neuroimaging manifestations in children with SARS-CoV-2 infection: a multinational, multicentre collaborative study. Lancet Child Adolesc Health. 2021;5 :167–177.33338439
8. Schiff J Brennan C . Covid-19 presenting as a bulging fontanelle. Am J Emerg Med. 2021;43 :81–82.33548683
| 36476525 | PMC9725740 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 14; 42(1):e4-e5 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003750 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
36102739
00022
10.1097/INF.0000000000003707
3
ESPID Reports and Reviews
Real Life Evidence From the Use of COVID-19 mRNA Vaccines in Pediatric Populations
Lagousi Theano MD, PhD [email protected]
*†
Tsagkli Panagiota MD [email protected]
*
Spoulou Vana MD, MPhil, PhD *†
* From the Immunobiology and Vaccinology Research Laboratory, National and Kapodistrian University of Athens, Athens, Greece
† 1st Department of Pediatrics, “Aghia Sophia” Children’s Hospital, Athens, Greece.
Correspondence: Vana Spoulou, MD, MPhil, PhD, 1st Department of Pediatrics, “Aghia Sophia” Children’s Hospital, Athens 11527, Greece. E-mail: [email protected].
14 9 2022
1 2023
14 9 2022
42 1 e32e34
18 7 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
STATUSONLINE-ONLY
==== Body
pmcSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) hit pediatric populations globally, accounting for approximately 13.5 million coronavirus disease 2019 (COVID-19) cases, although with lower morbidity rates compared with adults and mortality rate less than 0.02%.1 Nevertheless, children paid their own toll to SARS-CoV-2 pandemic due to a novel clinical entity called multisystem inflammatory syndrome (MIS-C), described solely in children and adolescents.1 Moreover, indirect effects of the pandemic attributed to interruption of education and social isolation had more profound effect on younger ages.
Several vaccines against COVID-19 have been approved for emergency use by the Food and Drug Administration (FDA) and the European Medicines Agency, aiming to protect not only high-risk individuals but also the general population from severe disease. Besides, the widespread implementation of the vaccines aimed to change the dynamics of the pandemic by decreasing transmission rates and building herd immunity.
Considering the potential benefits of universal vaccination, many countries have recommended the vaccination of all pediatric populations over the age of 5 years old, with 2 doses of mRNA vaccines; BNT162b2 (Pfizer/New York/USA-BioNTech/Mainz/Germany) and mRNA-1273 (Moderna/Massachusetts/USA). As of June 2022, the FDA revised the emergency use authorization for BNT162b2 as a 3-dose series to include children 6 months to 4 years old, with the first 2 doses administered 21 days apart and the third dose administered at least 60 days after dose 2 with a dosage of 3 µg. This vaccine was already approved for individuals 5 years of age and older (10 µg for 5–11 years old and 30 µg for ages above 12). The FDA also approved mRNA-1273 for those 6 months to 17 years old. The dose is 100 µg for ages 12 years and up, 50 µg for ages 6–11 years and 25 µg for ages 6 months to 5 years. The European Medicines Agency authorized BNT162b2 for children 5–11 years old (10 µg) and mRNA-1273 for those 6–11 years of age (50 µg). Both vaccines are also approved for children 12 years old and above (30 µg of BNT162b2 and 100 µg of mRNA-1273). All European Union countries are offering vaccination for all children 5−17 years old except Sweden where vaccination for children 5−11-year-old, is recommended only for those with risk factors. Fifteen European Union countries recommend boosters for adolescents, whereas the United States extended this recommendation to children 5 years of age. For immunocompromised adolescents, there is recommendation for 2 boosters with an interval of 4 months between the 2 doses.
Despite significant knowledge gaps on the safety against vaccine-associated rare adverse events, duration of protection offered, and most importantly, the contribution of vaccines to the control of viral transmission, approximately 20 million doses have been administered to those younger than 18 years old.2
Here, we review the prelicensure and real-life evidence of mRNA vaccines regarding safety and effectiveness in the prevention of infection and disease including MIS-C, and their effect on viral spread in the light of continuous emergence of novel variants of concern and discuss to what extent the implementation of mRNA vaccines has met initial expectations.
PRELISENCURE PEDIATRIC CLINICAL TRIALS
The safety and immunogenicity of the 2 mRNA vaccines, BNT162b2 and mRNA-1273, were evaluated by placebo-controlled, observer-blinded studies among adolescents who received a dual dose schedule and clinical effectiveness was assessed by bridging immunogenicity data with those obtained by adult studies.
Safety, immunogenicity and efficacy of the BNT162b2 vaccine was evaluated in 2260 adolescents 12–15 years of age who received a 2-dose schedule of 30 µg with a 21-day interval and a 2-month follow-up.3 BNT162b2 had a favorable safety profile, with mild to moderate reactogenicity. Any adverse events resolved in 1–2 days; pain at the site of injection was the most frequently reported local reaction. Headache and fatigue were the main systemic reactions, mainly following the second dose. Immunogenicity of BNT162b2 among 12-to-15-year-old adolescents was noninferior to that observed in the 16-to-25-year-old population and a 100% vaccine efficacy against COVID-19 was recorded during 1-month follow-up period.
A similar study evaluated the safety, immunogenicity and efficacy of mRNA-1273 in 3732 healthy adolescents who received either a dual dose of 100 µg of the vaccine or placebo, 28 days apart.4 The person-years of follow-up for efficacy were 513–522 in the mRNA-1273 group and 238–248 in the placebo group and 83 days for safety. The most common adverse reactions included injection-site pain, headache and fatigue. Vaccine immunogenicity met the noninferiority criterion compared with adult studies, while its efficacy against COVID-19 was >90%. Nevertheless, the study had some limitations including a lower incidence of mild COVID-19 among adolescents compared with adults that may have affected the efficacy analysis results. Besides, this study demonstrated that mRNA-1273 vaccine was approximately 40% effective in preventing asymptomatic SARS-CoV-2 infection in adolescents post second dose.
Immunogenicity of BNT162b2 was also assessed among 2268 children 5-to-11-years-old who received 2 doses of 10 µg of vaccine or placebo.5 Immune responses were comparable to those recorded among vaccine recipients 16-to-25-year-olds who had received dual immunization with 30-µg vaccines. The median follow-up for vaccine efficacy against COVID-19 was 2.3 months. The BNT162b2 vaccine was safe, immunogenic and efficacious. This schedule showed a similar safety profile with that observed in adolescents and adults but with a higher incidence of injection-site irritation and lower incidence of systemic events, that is, fever and chills. There were no reported events of MIS-C, myocarditis or pericarditis. A robust virus-neutralizing response was observed which was comparable to that among recipients 16-to-25-years-old with a 90.7% vaccine efficacy against CΟVID-19. Notably, among children who developed COVID-19, symptoms were milder in vaccine recipients.
A clinical trial was also conducted to assess the efficacy of mRNA-1273 vaccine schedule (2 doses of 50 µg with a 28-day interval) in 4016 children 6–11 years of age. The study took place during the era of B1.617.2 (Delta) variant and vaccine efficacy against COVID-19, assessed 14 days or more following the second dose, was 88.0%.6
In conclusion, phase 3 studies despite the caveats of the small numbers of young participants and the very short follow-up period showed a favorable safety profile and met the noninferiority criterion, paving the way toward the authorization for emergency use of mRNA vaccines in pediatric populations ≥5 years old.
REAL-LIFE EVIDENCE FOR mRNA VACCINE SAFETY AND EFFECTIVENESS AGAINST SARS-COV-2 INFECTION AND DISEASE
Soon after the introduction of massive vaccination campaigns in adolescents, early safety signals for vaccine-associated myocarditis were detected. The accumulation of cases among adolescents and young adults was up to 5 times higher than the pre-pandemic incidence of all-cause acute myocarditis in the general population.7 Following these early findings, active surveillance programs confirmed an elevated risk for myocarditis in mRNA COVID-19 vaccine recipients, particularly among males 12–29 years of age, with an incidence range 3.9–4.7 per 100,000 second vaccine doses.7 Fortuitously, similar safety signals have not been detected in children receiving the pediatric dose of BNT162b2.
Pediatric use of mRNA vaccines was generated shortly before the emergence of the Delta variant. Several studies conducted in the second half of 2021 showed an effectiveness above 90% 1 month post dual vaccination that decreased up to 67% 4–5 months thereafter although effectiveness against hospital admissions remained high up to 6 months.8 Remarkably, vaccine effectiveness (VE) against ICU admission was 98% while all deaths occurred in unvaccinated adolescents.9 Importantly, during the dominance of Delta variant, 2 doses of BNT162b2 reduced the likelihood of MIS-C by 91% in adolescents.10
Considering that neutralizing antibodies correlate with protection against mucosal infection, the observed waning of antibody titers soon after primary immunization could explain the progressively higher rates of breakthrough disease among fully vaccinated subjects. In contrast, protection against hospitalization and severe COVID-19 is mediated by B and T cellular immunity which has been shown to have longer duration than humoral response.11 The important role of cellular immunity for protection against severe COVID-19 is further supported by the fact that the interval between onset of the disease and clinical deterioration is long enough to allow B and T memory cells pass from a dormant state to activation and ultimately lead to the resolution of infection.
mRNA VE AGAINST BA2.12.1 (OMICRON) VARIANT
Omicron variant displaying a highly mutated S protein spread globally in the last trimester of 2021 and its dominance was associated with increased incidence of breakthrough cases raising concerns about VE against the new variant.
Serum neutralizing activity against Omicron evaluated in children and adolescents was significantly reduced compared with the ancestral strains, implying further reduction of VE against the novel variant.12 This hypothesis was confirmed by epidemiological studies from the United States during the Omicron wave, where VE against symptomatic infection for children 5–11 years of age was estimated at 60.1% 2–4 weeks and 28.9% 2 months after the second dose. Among adolescents 12–15 years old, the estimated VE was 59.5% and 16.6%, respectively.13
Based on adult studies showing that booster doses improve protection against Omicron by augmenting neutralizing antibody concentrations, Advisory Committee on Immunization Practices expanded the eligibility for a third dose to everyone over 5 years old. However, the merit of a booster in healthy children requires monitoring of the duration of antibodies stimulated by the additional dose. In contrast, enhancement of protection from severe COVID induced by a booster could be very important since Omicron is very contagious and the number of children who will become seriously ill will increase. If there is sufficient evidence that additional doses enhance protection against severe COVID through enrichment of B and T cellular memory this will be in favor for recommending boosters in children at risk. From the public health point of view, the effect of a booster could promote a barrier of immunity against viral spread since there is evidence for reduced transmission by booster-vaccinated individuals compared with dual-vaccinated individuals.14 However, this effect could be transient if there is a rapid immunity waning following booster.
CONCLUSIONS
Real-life evidence confirmed that mRNA vaccines protect pediatric populations from severe COVID-19 and possibly MIS-C but failed to prevent infection and viral transmission due to the rapid waning of humoral immunity against continuously emerging new variants. The elucidation of the immunological characteristics of recall responses is required to access the potential benefits of boosters before universal recommendations are issued. Most importantly, if continued monitoring of disease severity is reassuring that novel variants cause mild disease, it will be essential to move towards the perception that COVID-19 vaccines should be used for protection of children at risk against an endemic virus rather than utilized to interrupt the pandemic.
The authors have no funding or conflicts of interest to disclose.
==== Refs
REFERENCES
1. CDC. Demographic Trends of COVID-19 cases and deaths in the US reported to CDC. Available at: https://covid.cdc.gov/covid-data-tracker. Accessed May 23, 2022.
2. ECDC. Data on COVID-19 vaccination in the EU/EEA. Available at: https://www.ecdc.europa.eu/en/publications-data/data-covid-19-vaccination-eu-eea. Accessed May 19, 2022.
3. Frenck RW Klein NP Kitchin N . Safety, immunogenicity, and efficacy of the BNT162b2 COVID-19 vaccine in adolescents. N Engl J Med. 2021;385 :239–250.34043894
4. Ali K Berman G Zhou H . Evaluation of mRNA-1273 SARS-CoV-2 Vaccine in Adolescents. N Engl J Med. 2021;385 :2241–2251.34379915
5. Walter EB Talaat KR Sabharwal C . Evaluation of the BNT162b2 COVID-19 vaccine in children 5 to 11 years of age. N Engl J Med. 2022;386 :35–46.34752019
6. Creech CB Anderson E Berthaud V . Evaluation of mRNA-1273 COVID-19 vaccine in children 6 to 11 years of age. N Engl J Med. 2022;386 :2011–2023.35544369
7. CDC. Pfizer-BioNTech COVID-19 Vaccine Reactions & Adverse Events. Available at: https://www.cdc.gov/vaccines/covid-19/info-by- product/pfizer/reactogenicity.html. Accessed April 27, 2022.
8. Tartof SY Slezak JM Fischer H . Effectiveness of mRNA BNT162b2 COVID-19 vaccine up to 6 months in a large integrated health system in the USA: a retrospective cohort study. Lancet. 2021;398 :1407–1416.34619098
9. Olson SM Newhams MM Halasa NB . Effectiveness of BNT162b2 Vaccine against critical COVID-19 in adolescents. N Engl J Med. 2022;386 :713–723.35021004
10. Zambrano LN Olson S . Effectiveness of BNT162b2 (Pfizer-BioNTech) mRNA vaccination against multisystem inflammatory syndrome in children among persons aged 12–18 years. MMWR. 2022;71 :52–58.35025852
11. Goel RR Painter MM Apostolidis SA . mRNA vaccines induce durable immune memory to SARS-CoV-2 and variants of concern. Science. 2021;374 :abm0829.34648302
12. Girard B Tomassini JE Deng W ea . mRNA-1273 Vaccine-elicited neutralization of SARS-CoV-2 Omicron in adolescents and children. medRxiv [Preprint]. 2022.
13. Fleming-Dutra KE Britton A Shang N . Association of Prior BNT162b2 COVID-19 vaccination with symptomatic SARS-CoV-2 infection in children and adolescents during omicron predominance. J Am Med Association. 2022;327 :2210–2219.
14. Wald A . Booster vaccination to reduce SARS-CoV-2 transmission and infection. J Am Med Association. 2022;327 :327–328.
| 36102739 | PMC9725741 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 14; 42(1):e32-e34 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003707 | oa_other |
==== Front
Pediatr Infect Dis J
Pediatr Infect Dis J
INF
The Pediatric Infectious Disease Journal
0891-3668
1532-0987
Lippincott Williams & Wilkins Hagerstown, MD
00011
10.1097/INF.0000000000003759
3
COVID Reports
Multisystem Inflammatory Syndrome in Children Associated With SARS-CoV-2 Infection in KwaZulu-Natal, South Africa
https://orcid.org/0000-0002-1517-0358
Chinniah Kogielambal MBChB *†
https://orcid.org/0000-0003-3092-7790
Bhimma Rajendra PhD [email protected]
*‡
https://orcid.org/0000-0003-4940-0534
Naidoo Kimesh Loganathan MBChB, PhD [email protected]
*§
https://orcid.org/0000-0003-3082-9234
Archary Moherndran PhD [email protected]
*§
https://orcid.org/0000-0001-9244-6285
Jeena Prakash MBChB, PhD [email protected]
*§
https://orcid.org/0000-0001-5363-0602
Hoosen Ebrahim MBChB [email protected]
*‡
Singh Shivani MBChB, DCH SA [email protected]
*‡
Lawler Melissa MBChB [email protected]
*‖
Naby Fathima MBChB [email protected]
***
https://orcid.org/0000-0001-9665-2035
Masekela Refiloe FCCP, PhD [email protected]
*§
From the * Department of Paediatrics and Child Health, University of KwaZulu-Natal, Durban, South Africa
† Department of Paediatrics, Mahatma Gandhi Memorial Hospital, Durban, South Africa
‡ Department of Paediatrics, Inkosi Albert Luthuli Central Hospital, Durban, South Africa
§ Department of Paediatrics, King Edward VIII Hospital, Durban, South Africa
‖ Department of Paediatrics, Prince Mshyeni Memorial Hospital, Durban, South Africa
** Department of Paediatrics, Pietermaritzburg Metropolitan Hospitals Complex, Durban, South Africa.
Address for correspondence: Kogielambal Chinniah, MBChB, FC Paeds, Cert Paediatric Rheumatology, Department of Paediatrics and Child Health, School of Clinical Medicine, University of KwaZulu-Natal, 719 Umbilo Road, Congella 4013, South Africa. E-mail: [email protected].
06 12 2022
1 2023
06 12 2022
42 1 e9e14
30 9 2022
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
Background:
Multisystem inflammatory syndrome in children (MIS-C) following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been infrequently described in Africa.
Objective:
To describe the clinical characteristics, outcomes and associations of severe disease in children hospitalized with MIS-C in KwaZulu-Natal.
Methods:
Retrospective multicenter study of children (0–13 years) who met the Centers for Disease Control and Prevention criteria for MIS-C. Children with shock were compared with children without shock to determine the characteristics of severe MIS-C.
Results:
Twenty-nine children with MIS-C were identified, the mean age was 55 (SD ±45) months, 25 (86%) were Black-African, and 8 (28%) had pre-existing comorbidities. The predominant presenting symptoms included fever 29 (100%), gastrointestinal symptoms 25 (83%), skin rash 19 (65%), and shock 17 (59%). Children with shock had significantly increased CRP (P = 0.01), ferritin (P < 0.001), troponin-T (P = 0.02), B-type natriuretic peptide (BNP) (P = 0.01), and lower platelets (P = 0.01). Acute kidney injury (P = 0.01), cardiac involvement (P = 0.02), and altered levels of consciousness (P = 0.03) were more common in children with shock. The median length of hospital stay was 11 (IQR 7–19) days, with a mortality of 20.6%. Children who did not survive had significantly higher ferritin levels 1593 (IQR 1069–1650) ng/mL versus 540 (IQR 181–1156) ng/mL; P = 0.03) and significantly more required mechanical ventilation (OR 18; confidence interval 1.7–191.5; P = 0.005).
Conclusions:
Hospitalized children with MIS-C in KwaZulu-Natal had more aggressive disease and higher mortality than children in better-resourced settings. Markedly elevated biomarkers and critical organ involvement were associated with severe disease. Risk factors for poor outcomes include higher ferritin levels and the need for mechanical ventilation.
paediatrics
multisystem inflammatory syndrome in children
SARS-CoV-2
STATUSONLINE-ONLY
SDCT
==== Body
pmcMultisystem Inflammatory Syndrome in children (MIS-C) is a novel systemic hyperinflammatory, multiorgan disorder temporally associated with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection described in children and adolescents younger than 21 years of age. The Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO) have developed criteria for MIS-C.1,2 The CDC criteria for MIS-C include individuals <21 years of age, with laboratory evidence or history of close contact with SARS-CoV-2 infection, fever >24 hours, severe disease requiring hospitalization with at least 2 organ systems involved, laboratory evidence of inflammation, and all other plausible diagnoses to be excluded.1 The exact pathophysiology of MIS-C remains unknown, although it is considered an exaggerated innate and adaptive response of the immune system following SARS-CoV-2 infection in susceptible individuals.3
Since April 2020, reports from Europe, United Kingdom, United States of America and the Western Cape in South Africa reported cases of MIS-C in children with a median age of 9–11 years and mortality rates between 0% and 4%.4–8 However, India, Pakistan, and Colombia have reported MIS-C in younger children (median age 2–7 years) with mortality rates ranging from 9% to 20%.9–12 There are few studies from Africa; these include a study by Butters et al,4 a letter by Webb et al8 from Cape Town, a case series from Sokunbi et al13 from Nigeria and 2 case reports from Moodley P et al14 and Onyeaghala et al.15 In the analysis of 469 children hospitalized in 6 sub-Saharan African countries with SARS-CoV-2 infection, 6.1% were noted to have suspected or confirmed MIS-C.16 While MIS-C was reported in this large multicentre study, the study was limited because of a lack of laboratory and infrastructural capacity to confirm the diagnosis in most countries where the study was performed. Outcomes of children in Sub-Saharan Africa may differ because of the higher burden of childhood diseases like malnutrition, human immunodeficiency virus infection (HIV), tuberculosis (TB) and poorer overall SARS-CoV-2 vaccine coverage.17 MIS-C has also been reported to be more severe in children of Black-African ethnicity living in Europe, the United Kingdom and the Americas and children from a lower socioeconomic background.18,19 Therefore, there is a need to describe the characteristics of MIS-C and the possible associations of severe disease in African children. KwaZulu-Natal is the second-most populous province in South Africa, with the second-highest documented COVID infections, where 86% of the population is Black-African, and 52% of the population live below the poverty line.20,21 Thus, we undertook this study in KwaZulu-Natal.
MATERIALS AND METHODS
This retrospective study was conducted at 5 referral hospitals in KwaZulu-Natal, and records of children hospitalized with MIS-C between June 1, 2020, and April 30, 2021, were analyzed. Cases were identified from the admission register at each hospital by the attending pediatricians, who were co-investigators in this study. Records that met the CDC criteria for MIS-C were included, while records that did not meet the CDC criteria or had an alternative etiologic diagnosis, including those with other microbial causes for the inflammation, were excluded.1 Demographics, clinical data, laboratory tests, imaging, treatment and outcomes were abstracted from the patient’s medical records; this was captured on a case report form and entered into a Microsoft Excel database for Windows [version 16.60 (22041000)]. Medical records which could not be retrieved were excluded. The patient’s race was abstracted from the records as reported by the caregivers. The 5 hospitals enrolled included 2 regional hospitals, 2 tertiary hospitals and 1 central hospital.
These sites were selected as these hospitals have access to laboratory facilities and the capacity to provide critical care to children.
Nutritional disorders were defined using the WHO child growth standards and expressed as z-scores. Weight/length growth charts were used for children under 5 years, and the body mass index (BMI) for children 5 years and older.22 Laboratory tests were analyzed by the National Health Laboratory Service; test results were interpreted according to the National Health Laboratory reference ranges. SARS-CoV-2 RT-PCR was performed on nasopharyngeal swabs of all children. Children who had negative COVID-19 tests but were suspected of having MIS-C were tested for the IgG antibody using the Abbott Architect SARS-CoV-2-IgG (Abbott IL, United States) test for the qualitative detection of SARS-CoV-2 nucleocapsid protein antibodies, to determine exposure to COVID-19. Cardiac involvement was defined as any myocardial, pericardial, endocardial or coronary artery involvement on echocardiography or elevated cardiac biomarkers (troponin-T ≥100 or B-type natriuretic peptide (BNP) ≥1000). Myocardial involvement was defined as fractional shortening ≤28% or ejection fraction less than 50%, or elevated cardiac biomarkers. Coronary artery dilatation was reported if the artery diameter was greater than 2.5 z-scores on echocardiography.23 Shock was identified from the clinical records, supported by evidence of hypotension (systolic arterial pressure less than the fifth percentile for age), capillary refill time >2 seconds or prescription of inotropes. Acute kidney injury (AKI) was defined according to the kidney disease: Initiating Global Outcome (KDIGO) classification.24 Whittaker et al identified 3 patterns of MIS-C in children, a group with severe disease who developed shock and organ involvement, a group with persistent fever and elevated inflammatory markers without features of organ involvement and a third group with features of Kawasaki Disease.5 In this study, children who developed shock, were defined as having “severe disease” and were compared with children who did not develop shock to identify possible associations of severe MIS-C. Morbidity was reported if a child had clinical or biochemical evidence of organ dysfunction on discharge.
Statistical Analysis
Baseline demographic and clinical characteristics were summarized using frequencies, proportions, medians and interquartile ranges. These were applied to categorical and continuous variables. Comparisons were performed using the Pearson Chi-squared test and Fisher exact test for categorical variables, student t-test for normally distributed variables and Mann-Whitney test for data that were not normally distributed. A P value of <0.05 was considered statistically significant. Risk factors associated with death were identified using Chi-square tests or Fisher’s exact test for categorical variables and Mann-Witney tests for numeric variables. Logistic regression was then used to examine the strength of the association and odds ratios (OR) with 95% confidence limits reported. Risk factors potentially associated with shock or mortality were examined. Only the bivariate P values and OR are reported. Because of the small sample size and missing data, a multivariable model was not used to adjust for possible confounders. All analyses were performed using Stata software version 17 (StataCorpLLC). Approval for the study was obtained from the University of KwaZulu-Natal Biomedical Research Ethics Committee (BREC 00002121/2020).
RESULTS
Thirty-five children were identified with a suspected diagnosis of MIS-C during the 11-month study period, which included the first and second SARS-CoV-2 waves in South Africa. Two children were excluded as a temporal association with SARS-CoV-2 infection could not be established, and 4 had alternative causes for the inflammation. Twenty-nine children were included in the study, 27 (93%) had laboratory evidence of SARS-CoV-2 infection, and 2 had close SARS-CoV-2 contacts. SARS-CoV-2 RT-PCR was positive on 16/29 (55%), and the SARS-CoV-2 antibody was positive on 12/16 (75%). One child tested positive for both the SARS-CoV-2 PCR and the antibody test.
Table 1 describes the demographics and clinical characteristics of the participants. The mean age was 55 (SD ±45) months with 17 (59%) ≤5 years, 13 (45%) males and 25 (86%) Black-African. Pre-existing comorbidities were documented in 8 (28%) children, including congenital heart disease 1 (3%), system onset juvenile idiopathic arthritis 1 (3%), HIV 1 (3%), moderate acute malnutrition 3 (10%) and obesity 2 (7%). The most common presenting symptoms were fever 29 (100%), gastrointestinal symptoms 25 (86%) and rash 19 (65%). The medium duration of fever was 5 IQR (4–7) days. The mucocutaneous lesions included discreet erythematous maculopapular rash over the trunk and limbs, diffuse erythroderma, urticarial lesions with pruritis, peripheral edema of hands and feet, desquamation, perioral erythema, cracking of the lips and bilateral bulbar conjunctival injection. Acute abdominal pain was documented in 17 (59%) children, with 4 (14%) requiring laparotomy.
TABLE 1. Demographics, Clinical Features and Treatment of Children With Shock Compared With Those Without Shock, With MIS-C in KwaZulu-Natal
Total (n = 29) Patients With Shock (n = 17) Patients Without Shock (n = 12) P Odds Ratio (OR) 95% Confidence interval (CI)
Age (months), mean (SD) 55 (±45) 66 (±50) 50 (±32) 0.26 1.16 0.96–1.41
Sex
Male 13 (45%) 4 (30%) 9 (70%) ref ref ref
Female 16 (55%) 13 (81%) 3 (19%) 0.01 9.75 1.74–54.85
Ethnicity
African 25 (86%) 15 (60%) 10 (40%) 0.70 1.5 0.18–12.46
Indian 4 (14%) 2 (50%) 2 (50%) ref ref ref
Comorbidity 8 (28%) 7 (88%) 1 (12%) 0.05 7.7 0.8–74
Laboratory tests
COVID PCR-positive 16 (55%) 10 (63%) 6 (37%) 0.64 1.43 0.32–6.32
SARS-CoV-2 IgG (n = 16) 12 (75%) 8 (67% 4 (33%) 0.15 6.0 0.46–77.7
Clinical features
Fever 29 (100%) 17 (59%) 12 (41%)
Vomiting 19 (65%) 13 (68%) 6 (32%) 0.15 3.25 0.66–15.98
Abdominal pain 17 (59%) 11 (65%) 6 (35%) 0.43 1.83 0.40–8.27
Diarrhea 15 (52%) 9 (60%) 6 (40%) 0.88 1.13 0.26–4.94
Skin Rash 19 (65%) 12 (63%) 7 (37%) 0.50 1.71 0.36–8.08
Conjunctival Injection 14 (48%) 9 (64%) 5 (36%) 0.55 1.58 0.35–7.00
Mucositis 16 (55%) 9 (56%) 7 (43%) 0.77 0.80 0.18–3.57
Cardiac involvement 15 (52%) 12 (80%) 3 (20%) 0.02 7.20 1.35–38.33
Respiratory distress 12 (41%) 9 (75%) 3 (25%) 0.14 3.38 0.67–17.00
Acute kidney injury 12 (41%) 11 (92%) 1 (8%) 0.01 20.17 2.07–196.4
ALOC 10 (34%) 9 (90%) 1 (10%) 0.03 11.37 1.17–590.4
Treatment
Corticosteroids* 21 (72%) 15 (71%) 6 (29%) 0.04 7.5 1.2–48.1
IVIG† 21 (72%) 14 (66%) 7 (33%) 0.22 3.33 0.61–18.1
Inotropic drugs 15 (52%) 15 (100%) 0 (0%) <0.001 76.83 9.41–inf
ICU Admission 11 (38%) 10 (91%) 1 (9%) 0.01 14.25 1.47–742.2
Mechanical Ventilation 10 (34%) 9 (90%) 1 (10%) 0.03 11.37 1.17–590.4
*Corticosteroids: IV methylprednisone 10 mg/kg/day × 3 days (taper) or IV dexamethasone: 0.5 mg/kg/d × 5–10 days.
†IVIG-IV: immumonoglobulin 2 g/kg over 48 hours.
The laboratory findings are described in Table 2. Children who presented in shock had significantly higher CRP (211 vs. 106 mg/L; P = 0.01), ferritin (1447 vs. 204 ng/mL; P < 0.001), D-dimers (8.5 vs. 1.8 mg/L; P = 0,01), BNP (2500 vs. 163 ng/L; P = 0,01), troponin-T (110 vs. 40 ng/mL; P = 0.02) and lower mean platelet levels (89 vs. 277 × 109/L; P = 0.02) and albumin (23 vs. 33 g/L; P = 0.001) compared with those without shock. Significantly more cardiac involvement (OR 7.2; confidence interval [CI] 1.35–38.33); P = 0.02; acute kidney injury (AKI; OR 20.17; CI 2.07–196.4); P = 0.01 and altered level of consciousness (OR 11:37; CI 1.17–590.4); P = 0.03 was observed in children with shock. Echocardiography was performed after the first week of admission in 20 of 29 (69%) of children by a Pediatric Cardiologist. Myocardial involvement was documented in 13 of 26 (50%), pericardial effusion 2 of 20 (10%) and no coronary artery dilatations or aneurysms were detected. The extent of multisystem involvement is reflected in 24 (83%) children having ≥4 organ-systems involved and 18 (62%) having ≥6 organ-systems involved.
TABLE 2. Laboratory Features of Children With Shock Compared With Those Without Shock, With MIS-C in KwaZulu-Natal
Total (n = 29) Patients With Shock (n = 17) Patients Without Shock (n=12) P NHLS (Normal Range)
Leucocyte count (×109/L)*† 12 (9.1–21.3) 14.5 (9.2–2) 11.4 (8.1–17.7) 0.39 3.9–10.2
Lymphocyte (×109/L)*† 1.4 (1.1–2.1) 1.4 (1–1.9) 1.5 (1.3–2.5) 0.28 1.90–4.30
Platelet (×109/L)*† 133 (78–312) 89 (76–159) 277 (151–459) 0.01 150–440
C-reactive protein (mg/L)*† 181 (106–233) 211 (167–246) 106.5 (74–164.5) 0.01 <10
Ferritin (ng/mL)*† (n = 27) 631 (226–1593) 1447 (860–1650 204 (94.5–470 <0.001 7–84
D-dimer (mg/L)*† (n = 28) 3.9 (1.2–12.2) 8.6 (2.4–14.1) 1.8 (0.2–3.5) 0.01 0.00–0.25
BNP (ng/L)*† (n = 26) 935 (152–4579) 2500 (746–4944) 163.5 (78–590) 0.01 <100
Troponin (ng/mL)*† (n = 25) 74 (40–124) 110 (60–435) 40 (20–84) 0.02 ≤50
Albumin (g/L)*† 24 (22–33) 23 (20–24) 33.5 (31–43.5) <0.001 32–47
ALT(U/L)*† 62 (30–109) 64 (48–122) 40 (11.5–103) 0.18 5–30
Creatinine (umol/L)*† 59 (25–123) 87 (59–149) 24.5 (20.5–38.5) <0.001 40–72
Sodium mmol/L*† 135 (131–138) 134 (128–138) 138 (133–138) 0.29 136–145
*IQR.
#Median.
ALT indicates alanine transaminase; BNP, B-type natriuretic peptide; IQR, interquartile range; NHLS, National Health Laboratory Service.
The treatment is described in Table 1. Nine children (34%) required invasive ventilation, 1 child did not have access to invasive ventilation and inotropic drugs were prescribed in 15 (52%). Twenty-three children (79%) received specific treatment for MIS-C; 21 (72%) received intravenous immunoglobulin (IVIG) 2 g/kg over 48 hours, 6 (21%) received a second dose of IVIG, 21 (72%) received corticosteroids and 19 (66%) received both IVIG and steroids. Of the children who received corticosteroids, 14 (67%) received intravenous methylprednisolone at 10 mg/kg for 3–5 days with tapering and 7 (33%) received intravenous dexamethasone at 0.5 mg/kg/day × 7–10 days. Aspirin was prescribed in 9 (31%), while 4 (14%) received low molecular weight heparin. All children received antibiotics, none of the admission microbial cultures were positive, and 2 children developed ICU-acquired nosocomial sepsis. Children who developed shock required significantly more corticosteroids (OR 7.5; CI 1.2–48.1; P < 0.04), inotropic drugs (OR 76.83; CI 9.41–inf; P < 0.001), intensive care unit (ICU) admission (OR 14.25; CI 1.47–742.2; P < 0.01) and invasive ventilation (OR 11.37; CI 1.17–590.4; P < 0.03) compared with those without shock. According to the American Heart Association criteria, 7 children were classified as having incomplete Kawasaki disease (KD).25 The mean age was 84 (SD ±40) months, and all had fever and mucocutaneous signs. Shock was documented in 5 (71%), myocardial injury 4 (57%), AKI 4 (57%), thrombocytopenia 5 (71%) and 6 (86%) required corticosteroids. None had evidence of coronary artery aneurysms on late echocardiography.
The median length of hospital stay was 11 (IQR 7–19) days. A full recovery was documented in 14 (48%) of children at discharge, and 4 (14%) recovered within 8 weeks of discharge; these children had discharge diagnoses of intermittent fevers 2 (7%), mild myocardial dysfunction 1 (3.5%), and a small pericardial effusion 1 (3.5%). Residual morbidity was documented in 5 (17%) children; spastic quadriplegia 1 (3.5%), speech and cognitive deficits 1 (3.5%), amputation of the foot 1 (3.5%), cardiomyopathy 1 (3.5%) and relapse of the systemic onset of juvenile idiopathic arthritis 1 (3.5%). The CT angiography of the child with spastic quadriplegia demonstrated multiple areas of stenosis in the anterior cerebral artery and the angiogram of the child with gangrene demonstrated arterial occlusion of the vessels of the foot. The mortality was 20.6%, and the mean age of children who did not survive was 48.3 (SD:65,3). In the mortality analysis, we found that children who did not survive had significantly higher ferritin levels 1593 (IQR 1069–1650) ng/mL versus 540 (IQR 181–1156) ng/mL; P = 0.03) and significantly more of them required mechanical ventilation (P < 0.005; OR = 18; CI:1.7–191.5) compared with those who survived. There were no significant differences between the survivors and the nonsurvivors with regard to age, sex, inflammatory markers, cardiac biomarkers or organ involvement. Four of the 6 (66%) children who did not survive met the 2016 Pediatric Rheumatology International Trials Organization criteria for macrophage activating syndrome,26 compared with 4 of 23 (17%) of those who survived. Because of resource constraints, biologics and extracorporeal membrane oxygenation therapy were not available.
DISCUSSION
This study reports on 29 children hospitalized with MIS-C in KwaZulu-Natal, South Africa. It is the first study that highlights the significant mortality and morbidity associated with MIS-C in a poorly resourced African setting with high rates of HIV, Tuberculosis and malnutrition. The 5 hospitals selected are referral hospitals; thus, there may be an overrepresentation of children with severe MIS-C. In contrast to studies from high-income countries,5–7 we found that these children tended to be younger, had more aggressive disease and had higher mortality. We found that 59% of children were ≤5 years of age compared with 42% in the group of children under 13 years reported by Dufort et al from the United States.7 A younger age for MIS-C was also reported in other low-middle-income countries, including India, Pakistan and Colombia.10–12 In addition, we found that previously unidentified comorbidities like HIV and malnutrition, although limited by small numbers, may contribute to the higher morbidity and mortality in this setting.27,28 This study adds to the evidence that significantly elevated biomarkers and critical organ involvement may be associated with severe MIS-C, and it suggests that higher ferritin levels and the need for mechanical ventilation confers a higher mortality risk in this context.
The predominant presentation of MIS-C in the current study was that of a febrile illness with gastrointestinal and mucocutaneous symptoms associated with shock and multisystem disease. We found that significantly increased CRP, ferritin, BNP, troponin, creatinine, low platelets and albumin were associated with severe MIS-C. These biomarkers have crucial clinical utility in identifying, managing, and monitoring children with high risk for severe MIS-C, especially in Sub-Saharan Africa, where access to higher levels of care is very limited.
Critical organ involvement was more common in our series, with altered levels of consciousness (34% vs. 9%) and AKI (41% vs. 22%) compared with the United Kingdom.5 A third of the children presented with an altered level of consciousness, highlighting the delay in presentation of MIS-C in our setting. Lipton et al observed significantly elevated inflammatory markers in 26 children with MIS-C who had AKI, suggesting that AKI is also a marker of severe disease.29 Coronary artery dilatations and aneurysms have been described on early echocardiography in children with MIS-C, which resolved on subsequent imaging.5,6,30 The absence of coronary artery dilatations and aneurysms may be explained by the delayed echocardiography performed in our setting because of resource constraints. A quarter of our participants presented with the KD phenotype; however, they were older, had severe circulatory, cardiac, renal and coagulation abnormalities and required more corticosteroids, which differs from the classic KD. This supports the evidence that MIS-C is distinct from KD and may represent a new systemic inflammatory disease.25
Consensus has not been reached regarding the treatment of MIS-C, and the optimal use of IVIG and corticosteroids remains controversial.31,32 The treatment of MIS-C was not standardized however most children received IVIG as the first line of therapy; corticosteroids were added if there was a poor response to IVIG or if the child had shock or organ involvement. Most children received both IVIG and corticosteroids. The cost of IVIG may be prohibitive in resource-limited countries, and the large volumes of fluids required to reconstitute IVIG may exacerbate myocardial dysfunction. Licciardi et al documented a 67.7% response in 31 children treated with IV methylprednisolone alone as the first line of treatment, suggesting that steroids could be used as first-line therapy.33 However, a larger randomized study is required to verify this. When compared with the series by Butters et al from Cape Town, we observed lower platelet count (133 vs. 189 × 109/L), higher troponin levels (74 vs. 38 ng/mL), greater use of inotropes (52% vs. 38%) and more invasive ventilation (34% vs. 11.8%), suggesting a more aggressive inflammatory process in our cohort. This may possibly be because of a delay in accessing healthcare and the limited access to higher levels of care and advanced therapies in our study. The lower rates of ICU admissions (38% vs. 80%) compared with the cohort from the United States reflect the limited availability of ICU facilities. There is a critical shortage of intensive care beds for children in KwaZulu-Natal, with ICUs only available at the central hospital and one of the tertiary hospitals enrolled. A limited number of patients are offered “interim ventilation” at regional hospitals until an ICU bed is available at the central hospital. The central hospital is the only public hospital in the province to provide renal replacement therapy and perform echocardiography by Pediatric Cardiologists to a population of 10 million, reflecting resource constraints in this population.
The 20% mortality in this series is higher than that reported in the literature, which ranges from 0% to 4% in high-income countries to 9% to 20% in LMIC. We found that the higher ferritin levels conferred a higher risk for mortality which is consistent with the report from Colombia. Two-thirds of children who did not survive had evidence of macrophage activating syndrome,26 reflecting the aggressive inflammation associated with MIS-C in our patients. The increased mortality in this study may be attributed to severe inflammation associated with delayed presentation, delay in accessing higher levels of care and poor access to advance therapies because of resource constraints or a more aggressive inflammatory process in the African patient. KwaZulu-Natal is a province with high poverty levels (52.4% vs. 21.3%) compared with the Western Cape. According to a Statistics South Africa 2018 general household survey, 62% of the child population of KwaZulu-Natal is classified as rural, while 94% of the children in the Western Cape were classified as urban dwellers.18 Rural areas in South Africa have limited access to health care facilities, primarily regional and tertiary levels of health care. Javalkar et al also reported that lower socioeconomic status and Black-African ethnicity were independent risk factors for MIS-C.34 About half of the population in KwaZulu-Natal live below the poverty line in poor socioeconomic circumstances.18 In a retrospective surveillance study in the United States, Abrams et al found that non-Hispanic Black children were more likely to require mechanical ventilation and have more severe outcomes than White children.15 Broad et al reported similar findings in the United Kingdom.16 A larger prospective study with genetic and immunopathological investigations is required to explore these factors in our setting.
CONCLUSION
MIS-C, in a resource-constrained setting, is an aggressive disease with substantial mortality. Significantly elevated biomarkers and critical organ involvement are associated with severe MIS-C. This study highlights the need for rapid recognition and treatment of MIS-C. It underscores the premise that social determinants of health possibly contribute to poorer outcomes in more impoverished communities.
Limitations of the Study
A major limitation is that this is a small study and is underpowered to detect statistically significant differences between groups. Children under the age of 13 years were enrolled from referral facilities, therefore, there is a bias toward younger children and severe diseases. The study also reports on children from areas with limited health resources and poor socioeconomic backgrounds, with both these factors impacting clinical presentation and outcomes.
ACKNOWLEDGMENTS
We express gratitude to Ms Cathy Connolly (Biostatistician), School of Public Health, University of KwaZulu-Natal) and Mrs Leora Sewnarain for assistance with formatting and language review.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.
The authors have no funding or conflicts of interest to disclose.
The data supporting this study’s findings are available from the corresponding author, KC, upon reasonable request.
KC was responsible for study design, KC, KLN, EH, SS, ML, and FB for data collection, KC, RB, KLN, MA, PJ, and RM for data analysis and KC, RB, KLN, MA, PJ, and RM for drafting of the manuscript. All authors were responsible for the manuscript review. All authors read and approved the final article.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.pidj.com).
==== Refs
REFERENCES
1. Centers for Disease Control and Prevention (CDC). Multi-system inflammatory syndrome in children (MIS-C) associated with coronavirus disease 2019 (COVID-19). 2020. Available at: https://emergency.cdc.gov/han/2020/han00432.asp. Accessed May 10, 2022.
2. World Health Organization (WHO). Multi-system inflammatory syndrome in children and adolescents temporally related to COVID-19. 2020. Available at: https://www.who.int/publications-detail-redirect/multisystem-inflammatory-syndrome-in-children-and-adolescents-with-covid-19. Accessed May 10, 2022.
3. Henderson LA Yeung RSM . MIS-C: early lessons from immune profiling. Nat Rev Rheumatol. 2021;17 :75–76.33349661
4. Butters C Abraham DR Stander R . The clinical features and estimated incidence of MIS-C in Cape Town, South Africa. BMC Pediatr. 2022;22 :241.35501710
5. Whittaker E Bamford A Kenny J ; PIMS-TS Study Group and EUCLIDS and PERFORM Consortia. Clinical characteristics of 58 children with a pediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2. JAMA. 2020;324 :259–269.32511692
6. Feldstein LR Rose EB Horwitz SM . Overcoming COVID-19 investigators; CDC COVID-19 response team. multisystem inflammatory syndrome in U.S. children and adolescents. N Engl J Med. 2020;383 :334–346.32598831
7. Dufort EM Koumans EH Chow EJ . Multisystem inflammatory syndrome in children in New York State. N Engl J Med. 2020;383 :347–358.32598830
8. Webb K Abraham DR Faleye A ; Cape Town MISC-Team. Multi-system inflammatory syndrome in children in South Africa. Lancet Child Adolesc Health. 2020;4 :e38.32835654
9. Nayak S Panda PC Biswal B ; EICOMISC Study Group. Eastern India Collaboration on Multisystem Inflammatory Syndrome in Children (EICOMISC): a multicenter observational study of 134 cases. Front Pediatr. 2022;10 :834039.35377583
10. Dhanalakshmi K Venkataraman A Balasubramanian S . Epidemiological and clinical profile of pediatric inflammatory multisystem syndrome - temporally associated with SARS-CoV-2 (PIMS-TS) in Indian children. Indian Pediatr. 2020;57 :1010–1014.32769230
11. Mohsin SS Abbas Q Chowdhary D . Multisystem inflammatory syndrome (MIS-C) in Pakistani children: a description of the phenotypes and comparison with historical cohorts of children with Kawasaki disease and myocarditis. PLoS One. 2021;16 :e0253625.34153080
12. Acevedo L Piñeres-Olave BE Niño-Serna LF . Mortality and clinical characteristics of multi-system inflammatory syndrome in children (MIS-C) associated with covid-19 in critically ill patients: an observational multicenter study (MISCO study). BMC Pediatr. 2021;21 :516.34794410
13. Sokunbi O Akinbolagbe Y Akintan P . Clinical presentation and short-term outcomes of multisystemic inflammatory syndrome in children in Lagos, Nigeria during the COVID-19 pandemic: a case series. EClinicalMedicine. 2022;49 :101475.35747195
14. Moodley P Tsitsi JML Reddy DL . A case of multi-system inflammatory syndrome in an African adolescent male: case report. Pan Afr Med J. 2021;38 :174.33995781
15. Onyeaghala C Alasia D Eyaru O . Multisystem inflammatory syndrome (MIS-C) in an adolescent Nigerian girl with COVID-19: a call for vigilance in Africa. Int J Infect Dis. 2021;105 :124–129.33582372
16. Nachega JB Sam-Agudu NA Machekano RN ; African Forum for Research and Education in Health (AFREhealth) COVID-19 Research Collaboration on Children and Adolescents. Assessment of clinical outcomes among children and adolescents hospitalised with COVID-19 in 6 sub-Saharan African countries. JAMA Pediatr. 2022;176 :e216436.35044430
17. Modjadji P Madiba S . The double burden of malnutrition in a rural health and demographic surveillance system site in South Africa: a study of primary schoolchildren and their mothers. BMC Public Health. 2019;19 :1087.31399048
18. Abrams J Oster M Godfred-Cato S . Factors linked to severe outcome in multi-system inflammatory syndrome in children (MIS-C) in the USA: a retrospective surveillance study. Lancet Child Adolesc Health. 2021;5 :323–331.33711293
19. Broad J Forman J Brighouse J . Post-COVID-19 paediatric inflammatory multi-system syndrome: association of ethnicity, key worker and socio-economic status with risk and severity. Arch Dis Child. 2021;106 :1218–1225.33727312
20. Republic of South Africa, Department of Statistics South Africa. Five facts about poverty in South Africa. Available at: https://www.statssa.gov.za/?p=12075. Accessed May 10, 2022.
21. Statistics South Africa (STATS SA). Poverty in perspective. Contextualising Poverty and Inequality in the KZN, with further emphasis on the generational and endemic nature of poverty as it affects women and the girl child in the Limpopo. Available at: https://www.parliament.gov.za/storage/app/media/1_Stock/Events_Institutional/2020/womens_charter_2020/docs/19-02-2021/20210212_Womens_Charter_Review_KZN_19th_of_Feb_afternoon_Session_Final.pdf. Accessed May 24, 2022.
22. World Health Organization (WHO)/United Nations Children’s Fund (UNICEF). WHO Growth Standards and the Identification of Severe Acute Malnutrition in Infants and Children. Available at: https://apps.who.int/iris/bitstream/ handle/10665/44129/9789241598163_eng.pdf. Accessed July 20, 2022.
23. Matsubara D Kauffman HL Wang Y . Echocardiographic findings in pediatric multisystem inflammatory syndrome associated with COVID-19 in the United States. J Am Coll Cardiol. 2020;76 :1947–1961.32890666
24. Kidney Disease Improving Global Outcomes clinical practice guideline for acute kidney injury. Kidney Int Suppl. 2012;2 :1–138. Available at: https://kdigo.org/wp-content/uploads/2016/10/KDIGO-2012-AKI-Guideline-English.pdf. Accessed May 25, 2022.
25. Council on Cardiovascular Disease in the Y, Committee on Rheumatic Fever E, Kawasaki D, American Heart A. Diagnostic guidelines for Kawasaki disease. Circulation. 2001;103 :335–336.11208699
26. Ravelli A Minoia F Davì S ; Paediatric Rheumatology International Trials Organisation. 2016 Classification criteria for macrophage activation syndrome complicating systemic juvenile idiopathic arthritis: a European League Against Rheumatism/American College of Rheumatology/Paediatric Rheumatology International Trials Organisation Collaborative Initiative. Arthritis Rheumatol. 2016;68 :566–576.26314788
27. Van Heerden J Nel J Moodley P . Multisystem inflammatory syndrome (MIS): a multicentre retrospective review of adults and adolescents in South Africa. Int J Infect Dis. 2021;111 :227–232.34428544
28. Republic of South Africa, Department of Statistics (STATS SA). Mid-year population estimates, Statistical Release P0302. 2021. Available at: http://www.statssa.gov.za/publications/P0302/P03022021.pdf. Accessed May 19 2022.
29. Lipton M Mahajan R Kavanagh C . AKI in COVID-19-associated multisystem inflammatory syndrome in children (MIS-C). Kidney360. 2021;2 :611–618.35373052
30. Capone CA Misra N Ganigara M . Six month follow-up of patients with multi-system inflammatory syndrome in children. Pediatrics. 2021;148 :e2021050973.34326176
31. Crosby L Balasubramanian S Ramanan AV . Steroids or intravenous immunoglobulin as first line in MIS-C in LMICs. Lancet Rheumatol. 2021;3 :e615–e616.34337437
32. Villacis-Nunez DS Jones K Jabbar A . Short-term outcomes of corticosteroid monotherapy in multisystem inflammatory syndrome in children. JAMA Pediatr. 2022:176:576–584.34779842
33. Licciardi F Baldini L Dellepiane M . MIS-C treatment: is IVIG always necessary? Front Pediatr. 2021;9 :753123.34805048
34. Javalkar K Robson VK Gaffney L . Socio-economic and racial and/or ethnic disparities in multisystem inflammatory syndrome. Pediatrics. 2021;147 :e2020039933.33602802
| 36476527 | PMC9725742 | NO-CC CODE | 2022-12-08 23:18:16 | no | Pediatr Infect Dis J. 2023 Jan 6; 42(1):e9-e14 | utf-8 | Pediatr Infect Dis J | 2,022 | 10.1097/INF.0000000000003759 | oa_other |
==== Front
Anesth Analg
Anesth Analg
ANE
Anesthesia and Analgesia
0003-2999
1526-7598
Lippincott Williams & Wilkin Hagerstown, MD
36219579
00012
10.1213/ANE.0000000000006214
3
Original Research Articles
Original Clinical Research Report
10
Thrombosis-Related Loss of Arterial Lines in the First Wave of COVID-19 and Non–COVID-19 Intensive Care Unit Patients
Zon Rebecca L. MD *†
Merz Lauren E. MD, MSc ‡
Fields Kara G. MS †§
Grandoni Jessica PharmD ∥
Stuart Jessica C. MD ‡
Occhiogrosso Rachel H. MD *
Li Linda MD ¶
Baron Rebecca M. MD †#
Fredenburgh Laura E. MD †#
Woolley Ann E. MD, MPH †**
Connors Jean M. MD †††
Frendl Gyorgy MD, PhD †§
From the * Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
† Harvard Medical School, Boston, Massachusetts
Departments of ‡ Medicine
§ Anesthesiology, Perioperative and Pain Medicine,
∥ Pharmacy Services, Brigham and Women’s Hospital, Boston, Massachusetts
¶ Department of Anesthesiology, Cedars-Sinai Medical Center, Los Angeles, California
# Division of Pulmonary and Critical Care Medicine
** Department of Infectious Diseases, Brigham and Women’s Hospital, Boston, Massachusetts
†† Department of Medicine, Hematology Division, Brigham and Women’s Hospital, Boston, Massachusetts.
Address correspondence to Rebecca L. Zon, MD, Department of Medical Oncology, Dana-Farber Cancer Institute, 450 Brookline Ave, Boston, MA 02215. Address e-mail to [email protected].
11 10 2022
1 2023
11 10 2022
136 1 7078
25 7 2022
Copyright © 2022 International Anesthesia Research Society
2022
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
BACKGROUND:
Patients with coronavirus disease 2019 (COVID-19) can present with severe respiratory distress requiring intensive care unit (ICU)–level care. Such care often requires placement of an arterial line for monitoring of pulmonary disease progression, hemodynamics, and laboratory tests. During the first wave of the COVID-19 pandemic in March 2020, experienced physicians anecdotally reported multiple attempts, decreased insertion durations, and greater need for replacement of arterial lines in patients with COVID-19 due to persistent thrombosis. Because invasive procedures in patients with COVID-19 may increase the risk for caregiver infection, better defining difficulties in maintaining arterial lines in COVID-19 patients is important. We sought to explore the association between COVID-19 infection and arterial line thrombosis in critically ill patients.
METHODS:
In this primary exploratory analysis, a multivariable Fine-Gray subdistribution hazard model was used to retrospectively estimate the association between critically ill COVID-19 (versus sepsis/acute respiratory distress syndrome [ARDS]) patients and the risk of arterial line removal for thrombosis (with arterial line removal for any other reason treated as a competing risk). As a sensitivity analysis, we compared the number of arterial line clots per 1000 arterial line days between critically ill COVID-19 and sepsis/ARDS patients using multivariable negative binomial regression.
RESULTS:
We retrospectively identified 119 patients and 200 arterial line insertions in patients with COVID-19 and 54 patients and 68 arterial line insertions with non-COVID ARDS. Using a Fine-Gray subdistribution hazard model, we found the adjusted subdistribution hazard ratio (95% confidence interval [CI]) for arterial line clot to be 2.18 (1.06–4.46) for arterial lines placed in COVID-19 patients versus non–COVID-19 sepsis/ARDS patients (P = .034). Patients with COVID-19 had 36.3 arterial line clots per 1000 arterial line days compared to 19.1 arterial line clots per 1000 arterial line days in patients without COVID-19 (adjusted incidence rate ratio [IRR] [95% CI], 1.78 [0.94–3.39]; P = .078).
CONCLUSIONS:
Our study suggests that arterial line complications due to thrombosis are more likely in COVID-19 patients and supports the need for further research on the association between COVID-19 and arterial line dysfunction requiring replacement.
SDCT
==== Body
pmcKEY POINTS
Question: Is there an increased subdistribution hazard of arterial line-associated thrombosis and subsequent arterial line dysfunction in intensive care unit (ICU) patients with COVID-19 compared to ICU patients with non–COVID-19–related sepsis/ARDS?
Findings: An increased subdistribution hazard of arterial line dysfunction requiring replacement due to thrombosis was detected in COVID-19 patients in our study compared to non–COVID-19 sepsis/ARDS patients in our primary analysis, but this finding was not robust to a sensitivity analysis estimating the incidence rate ratio of arterial line-associated thrombosis.
Meaning: Larger prospective studies are warranted regarding whether ICU patients with COVID-19 have a greater subdistribution hazard of arterial line dysfunction due to thrombosis.
Arterial lines are commonly used in the intensive care unit (ICU) and may require replacement in the setting of thrombosis or malfunction. A 2014 Cochrane review found that most arterial lines fail after 36 to 158 hours with failure defined as pressure waveform dampening or catheter clotting.1 During the March–June 2020 wave of the COVID-19 pandemic, anecdotal observations suggested that arterial line replacements due to failure were more frequent in patients with COVID-19 than in non–COVID-19 ICU patients. In addition, senior clinicians in our hospital noted that arterial lines were more difficult to place and required a greater number of attempts than a typical ICU patient despite initial blood return and ultrasound confirmation of needle position. These repeated attempts led to clinicians spending long periods of time in patient rooms placing arterial lines and/or troubleshooting dysfunctional lines increasing their exposure to the SARS-COV-2 virus.
One possibility explaining the observed increased failure rate of arterial lines in COVID-19 ICU patients is a hypercoagulable state that drives increased rates of arterial thrombosis. Existing evidence finds an increased incidence of thromboembolic events in patients with COVID-19, and in particular those who require ICU care.2–4 Although the mechanisms underlying this hypercoagulable state are incompletely understood, the inflammatory and hypercoagulable state caused by COVID-19,5 characterized by elevated coagulation parameters, may play a role.6 Although the increased incidence of venous and arterial thromboembolic events has been well documented in patients with COVID-19, data on thrombosis of arterial lines are mostly confined to case reports and case series.7
The primary aim of this exploratory study was to test the hypothesis that arterial line dysfunction and thrombosis are associated with COVID-19 infection in critically ill patients. Our primary analysis targeted the subdistribution hazard (primary analysis) and incidence rate (sensitivity analysis) of arterial line dysfunction requiring replacement due to thrombosis in patients with COVID-19 infection versus those with sepsis/ARDS. Our exploratory secondary aims were to examine potential associations between different anticoagulation strategies, the subdistribution hazard of arterial line-associated thrombosis, and the incidence rate of arterial lines placed, time to arterial line removal for any reason, odds of removal for clot, and the number of arterial line insertion attempts in COVID-19 patients.
METHODS
We performed a retrospective cohort study of all patients with COVID-19 in the ICU at Brigham and Women’s Hospital (BWH) or its affiliates (Brigham and Women’s Faulkner Hospital and North Shore Medical Center) who were monitored with an arterial line at any time during the first wave of the COVID-19 pandemic (March 2020 and May 2020). Eligible patients were identified by searching the electronic health record (EHR) for an arterial line billing code from March 2020 to May 2020. The study was approved by the Institutional review board (IRB) and the requirement for written informed consent was waived by the IRB. SARS-COV-2 infection was determined by positive reverse-transcription polymerase chain reaction on nasopharyngeal or endotracheal samples. Comparison patients without COVID-19 were identified through the registry of critical illness (RoCI), a collated database and biorepository of ICU patients at BWH.8 Patients from the RoCI database were included in reverse chronological order starting March 2018 and included if their ICU diagnosis was sepsis or ARDS and if they had an arterial line placed during their ICU course. RoCI patients were included from March 2018 to May 2020. Patients included in the control group from the first COVID-19 wave tested negative for SARS-CoV-2 by PCR. Arterial lines were placed primarily using 20-gauge Becton-Dickinson Incite Autoguard Winged Catheters and occasionally the Arrow Radial Artery Catheterization Set. A guidewire was not routinely used. All lines were placed using Brigham and Women’s standardized procedures. The radial artery was the preferred and most commonly cannulated site. See Supplemental Digital Content 1, Supplemental Material 1, http://links.lww.com/AA/E42 for specific details on arterial line insertion and management.
Data extracted from the EHR included vital status, age on ICU admission, date of birth, and body mass index (BMI). Data collated by manual extraction included date of ICU admission and SARS-COV-2 infection, history of clotting disorder, history of hypercoagulability, the presence of deep vein thrombosis (DVT) or pulmonary embolism (PE), and renal replacement therapy (RRT) during admission. For each patient, information about every unique arterial line was recorded, including anatomic location, the number of insertion attempts, D-Dimer at time of placement, duration of line (in days) inclusive of placement date, reason for line removal, and D-dimer at removal. The reason for removal of each arterial line was reviewed. The reason for removal was defined as thrombosis if thrombosis was seen on ultrasound, dampening of the arterial line waveform occurred, or resistance to manual flushing or blood withdrawal difficulties was experienced.1 We also recorded data on anticoagulant use before admission and indication, agent and intensity of anticoagulation while admitted, and any thrombotic events. For the analysis, the individual arterial lines were stratified based on the highest dose of anticoagulation.
Anticoagulant regimens were defined as standard prophylaxis, intermediate prophylaxis, and treatment dosing. Standard prophylactic dosing was defined as enoxaparin 40 mg daily (or 30 mg adjusted for renal impairment) or heparin 5000 units (U) every (Q) 8 hours (h) (for average weight patients). For obese patients defined as ≥120 kg or BMI ≥35, enoxaparin 40 mg twice daily (BID), enoxaparin 0.5 mg/kg daily (max dose 100 mg daily), or heparin 7500 U Q 8 hours was recommended. For low body weight <50 kg patients, enoxaparin 30 mg daily or heparin 5000 U BID to three times daily (TID) was recommended. Intermediate prophylactic anticoagulation included enoxaparin 40 mg BID and heparin 7500 U TID (average weight patients). For obese patients, enoxaparin 0.5 mg/kg BID (max dose 100 mg BID) or heparin 10,000 U TID was recommended. For low body weight patients, enoxaparin 30 mg BID or heparin 7500 U TID was recommended. Therapeutic anticoagulation was defined as enoxaparin 1 mg/kg BID, heparin infusion (targeting aPTT 1.5–2× baseline), apixaban 5 mg BID or 10 mg BID, rivaroxaban 15 mg or 20 mg daily, and warfarin (international normalized ratio [INR] goal >2.0). Apixaban 2.5 mg BID was considered prophylactic, unless age, weight, and renal function called for reduction of dosing in which case it was considered therapeutic. Notably, the institutions included in this study implemented intermediate-dose thrombosis prophylaxis for COVID-19 ICU patients (1.5–2× the standard anticoagulant dose) midway through the study period on April 24, 2020.
Statistical Analysis
The magnitude and direction of the differences in demographics, comorbidities, length of ICU stay, and death between COVID-19 ICU and sepsis/ARDS ICU patients were quantified as standardized differences.9 Standardized differences with magnitude ≥0.1 were taken to indicate a notable difference between groups.10 Demographics, comorbidities, length of ICU stay, and death were also compared between groups using 2-sample t-tests or Wilcoxon rank-sum tests for continuous variables, and χ2 or Fisher exact tests for nominal categorical variables. For the primary analysis, an extension of the multivariable Fine-Gray subdistribution hazard model for clustered data (here, multiple arterial lines per patient) was used to estimate the association between arterial lines placed in COVID-19 patients versus those with non–COVID-19 sepsis/ARDS and the subdistribution hazard of arterial line removal for clot while accounting for arterial line removal for any other reason (eg, ICU discharge and patient death) as a competing risk.11
As a sensitivity analysis for the primary analysis, the number of arterial line clots observed per 1000 arterial line days was compared between COVID-19 ICU and non–COVID-19 sepsis/ARDS ICU patients using multivariable negative binomial regression. A multivariable Fine-Gray subdistribution hazard model for clustered data was used to estimate the association between arterial lines placed in COVID-19 ICU patients before the anticoagulation protocol change versus COVID-19 ICU patients after the anticoagulation protocol change versus non–COVID-19 sepsis/ARDS ICU patients and the subdistribution hazard of arterial line removal for clot.
A multivariable Fine-Gray subdistribution hazard model for clustered data was also used to estimate the association of treatment versus intermediate versus prophylactic anticoagulation (defined as the maximum level while each line was in place) with the subdistribution hazard of arterial line removal for clot among lines placed in COVID-19 ICU patients during which anticoagulation did not change beyond one level (n = 164 arterial lines). The time to arterial line removal for any reason was compared between arterial lines placed in COVID-19 ICU versus non–COVID-19 sepsis/ARDS ICU patients using multivariable Cox proportional hazards regression with cluster robust standard errors. Multivariable logistic regression with cluster robust standard errors was used to estimate the association between arterial lines placed in COVID-19 ICU versus sepsis/ARDS ICU patients and the odds of clot as reason for removal. The number of arterial line insertion attempts was compared between arterial lines placed in COVID-19 ICU versus sepsis/ARDS ICU patients using multivariable zero-truncated Poisson regression (where >5 attempts were analyzed as 5 attempts) with cluster robust standard errors. The association between COVID-19 ICU versus sepsis/ARDS ICU patients and the incidence rate of arterial line placement was estimated using multivariable zero-truncated Poisson regression. All regression models included covariates for patient age, sex, BMI, and the presence of hypercoagulable diseases excluding obesity. The proportional subdistribution hazards assumption of each Fine-Gray model was assessed using a score test based on modified weighted Schoenfeld residuals.12 The proportional hazards assumption of the Cox model was evaluated using a score test based on scaled Schoenfeld residuals. For the count outcome models, overdispersion was assessed via estimation of the negative binomial dispersion parameter. Count outcomes were modeled using Poisson regression when there was no evidence of overdispersion, and with negative binomial regression when overdispersion was detected. A χ2 test was used to assess whether D-dimer ≥500 ng/mL (considered clinically abnormal) at insertion of the patient’s first arterial line was associated with the development of least one arterial line removal for clot among COVID-19 ICU patients. All statistical hypothesis tests were evaluated at a 2-sided alpha level of 0.05 with no correction for multiple testing given the exploratory nature of the study. Statistical analyses were performed using SAS software version 9.4 (SAS Institute) and the crrSC package implemented in R software version 4.2.0 (R Foundation for Statistical Computing).
No a priori power calculation was performed. However, per recommendation by Althouse,13 we calculated our power to detect a hazard ratio of 1.3 for arterial line removal due to clot for COVID-19 ICU versus sepsis/ARDS ICU observations with the study sample size and observed outcome incidence. With 200 arterial lines in COVID-19 ICU patients, 68 sepsis/ARDS ICU arterial lines, and an overall 29.5% incidence of arterial line removal for clot, we had 17% power to detect a hazard ratio of 1.3 at a 2-sided alpha level of 0.05 using a Cox proportional hazards model.
RESULTS
Patient selection and inclusion for the study are outlined in Figure 1. One hundred and nineteen patients (receiving 200 arterial lines) with COVID-19 and 54 patients with sepsis or ARDS (receiving 68 arterial lines) were included in the final analysis. Patient demographics, comorbidities, length of ICU stay, and death within 45 days of ICU admission are presented in Table 1. A significant proportion of the COVID-19 cohort met the definition for clinical ARDS.3
Table 1. Comparison of Demographics, Comorbidities, Length of Stay, and Death Between COVID-19 ICU and Non–COVID-19 Sepsis/ARDS ICU Patients
COVID-19 ICU Non–COVID-19 Sepsis/ARDS ICU Standardized difference P value
Characteristic (n = 119) (n = 54)
Age at ICU admission (y), mean ± SD 64.1 ± 15.4 62.7 ± 16.2 0.085 .601
Female, n (%) 51 (42.9) 22 (40.7) 0.043 .794
BMI, kg/m2, mean ± SD 28.9 ± 6.8 28.8 ± 8.5 0.020 .909
History of known clotting disorder, n (%) 1 (0.8) 0 0.130 .999
Hypercoagulable diseases including obesity, n (%) 61 (51.3) 35 (64.8) −0.277 .097
Hypercoagulable diseases excluding obesity, n (%) 22 (18.5) 18 (33.3) −0.344 .032
History of deep vein thrombosis 5 (4.2) 6 (11.1) −0.262 .100
History of pulmonary embolism 8 (6.7) 3 (5.6) 0.049 .999
History of stroke 12 (10.1) 7 (13.0) −0.090 .575
On renal replacement therapy, n (%) 25 (21.0) 7 (13.0) 0.216 .207
ICU length of stay (d), median (Q1–Q3) 16 (8–25) 12 (7–18) 0.325 .054
Death within 45 d of ICU admission, n (%) 51 (42.9) 24 (44.4) −0.032 .845
Standardized differences with magnitude ≥0.1 were taken to indicate a notable difference between groups.10 Variables were compared between groups using 2-sample t tests or Wilcoxon rank-sum tests for continuous variables, and χ2 or Fisher exact tests for nominal categorical variables.
Abbreviations: ARDS, acute respiratory distress syndrome; BMI, body mass index; COVID-19‚ coronavirus disease 2019; ICU, intensive care unit; Q‚ quartile; SD, standard deviation.
Figure 1. Flow diagram of COVID-19 and non–COVID-19 sepsis/ARDS ICU patient selection. ARDS indicates acute respiratory distress syndrome; COVID-19‚ coronavirus disease 2019; ICU, intensive care unit; RoCI, registry of critical illness database.
Standardized differences between COVID and non-COVID cohorts with absolute value ≥0.1 include: history of known clotting disorder, hypercoagulable diseases (defined by thrombophilia, cancer, and history of DVT/PE/stroke) in the absence or presence of obesity, history of DVT, and RRT use. ICU length of stay was longer in the COVID-19 ICU versus sepsis/ARDS ICU group (16 [8–25] vs 12 [7–18] days; standardized difference, 0.325). The incidence of death within 45 days of ICU admission did not differ between groups (42.9% for COVID vs 44.4% for non–COVID-19 cohort).
The cause-specific cumulative incidence of arterial line removal due to clotting in COVID-19 ICU patients and sepsis/ARDS ICU patients is shown in Figure 2. A multivariable Fine-Gray model estimated an adjusted subdistribution hazard ratio (SHR) (95% confidence interval [CI]) for arterial line clot of 2.18 (1.06, 4.46) for arterial lines placed in COVID-19 versus sepsis/ARDS patients (P = .034) (Table 2). In addition, we found an 87% increase in the hazard of an arterial line clot occurring in women compared to men (adjusted SHR [95% CI], 1.87 [1.15–3.05]; P = .012) (Table 2). An increased hazard of arterial line removal for clot was also detected between lines placed in COVID-19 ICU patients before the anticoagulation protocol change (n = 143 arterial lines) versus non–COVID-19 sepsis/ARDS patients (adjusted SHR [95% CI], 2.44 [1.18–5.04]; P = .016), but not between lines placed in COVID-19 ICU patients after the anticoagulation protocol change (n = 57 arterial lines) versus non–COVID-19 sepsis/ARDS patients (adjusted SHR [95% CI], 1.64 [0.69–3.87]; P = .260) (Supplemental Digital Content 1, Table 1, http://links.lww.com/AA/E42).
Table 2. Multivariable Fine-Gray Subdistribution Hazard Models for Comparison of Time to Arterial Line Clot Between Arterial Lines Placed in COVID-19 ICU Versus Non–COVID-19 Sepsis/ARDS ICU Patients and Between Anticoagulation Levels Among Lines Placed in COVID-19 ICU Patients
Parameter Adjusted subdistributionhazard ratio (95% CI) P value
COVID-19 ICU versus non–COVID-19 sepsis/ARDS ICU 2.18 (1.06–4.46) .034
Age (per 1 y increase) 1.00 (0.98–1.02) .970
Female versus male 1.87 (1.15–3.05) .012
Body mass index (per 1 kg/m2 increase) 1.01 (0.97–1.04) .760
Hypercoagulable diseases excluding obesity 0.93 (0.54–1.59) .780
Intermediate versus prophylaxis 0.60 (0.30–1.21) .150
Treatment versus prophylaxis 0.30 (0.13–0.72) .007
Treatment versus intermediate 0.50 (0.19–1.31) .160
Age (per 1 y increase) 1.00 (0.98–1.03) .790
Female versus male 1.74 (0.85–3.56) .130
Body mass index (per 1 kg/m2 increase) 0.99 (0.94–1.05) .800
Hypercoagulable diseases excluding obesity 1.03 (0.48–2.23) .930
Multivariable Fine-Gray subdistribution hazard models for clustered data were used to compare the subdistribution hazard of arterial line removal for clot between (1) arterial lines placed in COVID-19 ICU versus non–COVID-19 sepsis/ARDS ICU patients and (2) maximum anticoagulation level for lines placed in COVID-19 ICU patients during which anticoagulation did not change beyond one level (n = 164 arterial lines). These Fine-Gray models accounted for arterial line removal for any reason besides clot as a competing risk, as well as the correlation between multiple arterial lines from the same patient.
Abbreviations: ARDS, acute respiratory distress syndrome; CI, confidence interval; COVID-19‚ coronavirus disease 2019; ICU, intensive care unit.
Figure 2. Plot of cause-specific cumulative incidence of removal of arterial lines due to clot formation for COVID-19 ICU and non–COVID-19 sepsis/ARDS ICU patients. Arterial line removal for reason besides clot was treated as a competing risk. ARDS indicates acute respiratory distress syndrome; COVID-19‚ coronavirus disease 2019; ICU, intensive care unit.
Arterial line characteristics such as the total number of arterial lines needed for a patient during ICU stay, the number of lines removed from presumed clot, and the number of insertion attempts are shown in Table 3. The number of arterial lines placed per 1000 ICU days did not differ between COVID and non-COVID patients (91.4 vs 86.5, respectively; adjusted incidence rate ratio [IRR] [95% CI], 1.05 [0.80–1.40]; P = .711). The median (first quartile [Q1] and third quartile [Q3]) longevity of each arterial line also did not differ between COVID and non-COVID patients (6 [3–13] days in COVID-19 ICU patients vs 6.5 [3–10] days in non-COVID patients) (adjusted hazard ratio [95% CI], 0.82 [0.62–1.09]; P = .170). However, an arterial line placed in a COVID-19 ICU patient had an adjusted 2.56 odds ratio for being removed for confirmed or clinically suspected clotting compared to an arterial line placed in a sepsis/ARDS patient (95% CI, 1.18–5.52; P = .017). There were 36.3 arterial clots per 1000 arterial line days in patients with COVID-19 and 19.1 arterial line clots per 1000 arterial line days in non–COVID-19 sepsis/ARDS patients (adjusted IRR [95% CI], 1.78 [0.94–3.39]; P = .078). We found no association between COVID-19 and the incidence rate of arterial line insertion attempts (adjusted IRR [95% CI], 0.92 [0.60–1.42]; P = .720).
Table 3. Comparison of Arterial Line Characteristics Between COVID-19 ICU and Non–COVID-19 Sepsis/ARDS ICU Patients
Characteristic COVID-19 ICU Non–COVID-19 sepsis/ARDS ICU Adjusted effect size (95% CI) P value
Arterial line level n = 200 arterial lines n = 68 arterial lines
Time to arterial line removal for any reason (d), median (Q1–Q3) 6 (3–13) 6.5 (3–10) 0.82 (0.62–1.09)a .170
Clot as reason for removal, n (%) 68 (34.0) 11 (16.2) 2.56 (1.18–5.52)b .017
Number of insertion attempts, n (%) 0.92 (0.60–1.42)c .720
1 126 (68.1) 42 (62.7)
2 33 (17.8) 15 (22.4)
3 15 (8.1) 6 (9)
4 6 (3.2) 4 (6)
≥5 5 (2.7) 0 (0)
Group level
n = 2187 ICU days n = 786 ICU days
Arterial lines per 1000 ICU days 91.4 86.5 1.05 (0.80–1.40)d .711
n = 1875 arterial line days n = 575 arterial line days
Clots per 1000 arterial line days 36.3 19.1 1.78 (0.94–3.39)e .078
All comparisons were adjusted for age, sex, BMI, and hypercoagulable diseases excluding obesity.
Abbreviations: ARDS, acute respiratory distress syndrome; CI, confidence interval; COVID-19‚ coronavirus disease 2019; ICU, intensive care unit; Q‚ quartile.
a Effect size is an adjusted hazard ratio estimated with a multivariable Cox proportional hazards model using cluster robust standard errors.
b Effect size is an adjusted odds ratio estimated with a multivariable logistic regression model using cluster robust standard errors.
c Effect size is an adjusted incidence rate ratio estimated with a multivariable zero-truncated Poisson regression model using cluster robust standard errors. Placements with >5 attempts were analyzed as 5 attempts.
d Effect size is an adjusted incidence rate ratio obtained from a multivariable zero-truncated Poisson regression model.
e Effect size is an adjusted incidence rate ratio obtained from a multivariable negative binomial regression model.
Figure 3. Plot of cause-specific cumulative incidence of removal of arterial lines for clot for COVID-19 ICU patients stratified by maximum anticoagulation strategy used. Arterial line removal for reason besides clot was treated as a competing risk. COVID-19 indicates coronavirus disease 2019; ICU‚ intensive care unit.
The cause-specific cumulative incidence of arterial line removal for clot by anticoagulation intensity (none, standard prophylactic, intermediate prophylactic, and treatment dose) among COVID-19 ICU patients is shown in Figure 3. Treatment dose was associated with an adjusted 70% decrease in the hazard of a clot occurring compared to standard prophylactic anticoagulation (adjusted SHR [95% CI], 0.30 [0.13–0.72]; P = .007). No difference was detected in the hazard of clot between intermediate dose and standard prophylactic anticoagulation (adjusted SHR [95% CI], 0.60 [0.30–1.21]; P = .150) or between treatment and intermediate dose anticoagulation (adjusted SHR [95% CI], 0.50 [0.19–1.31]; P = .160) (Table 2). The length of arterial line patency and percentage of arterial lines removed for presumed clot stratified by anticoagulation strategy and COVID-19 disease status are shown in the Supplemental Digital Content 1, Table 2, http://links.lww.com/AA/E42. No association was detected between D-dimer level ≥500 ng/mL at first arterial line insertion and the development of arterial line clot in patients with COVID-19 in the ICU (P = .185).
DISCUSSION
High rates of arterial line failure have been anecdotally observed among COVID-19 ICU patients since the beginning of the pandemic. In this retrospective study, we found that COVID-19 ICU patients had an increased hazard of arterial line thrombosis (adjusted SHR [95% CI], 2.18 [1.06–4.46]) when compared with non–COVID-19 patients (P = .034), with 2.56 times the odds of arterial line removal for confirmed or clinically suspected clotting in a COVID-19 patient compared to a non–COVID-19 patient (95% CI, 1.18–5.52; P = .017). The incidence rate of arterial line clots did not differ between COVID-19 and non–COVID-19 ICU patients, although the IRR point estimate was clinically relevant (36.3 arterial clots per 1000 arterial line days versus 19.1 arterial line clots per 1000 arterial line days; adjusted IRR [95% CI], 1.78 [0.94–3.39]; P = .078). We found no evidence for a difference in the number of arterial lines placed per 1000 ICU days between the aforementioned groups (adjusted IRR [95% CI], 1.05 [0.80–1.40]; P = .711).
Although exploratory only, our data suggest a relationship between COVID-19 infection and arterial line dysfunction. Because the number of insertion attempts between patients with COVID-19 versus non–COVID-19 did not differ, procedural vascular trauma during placement was unlikely to have contributed to the increased risk of line failure from thrombosis. However, 5 COVID-19 patients required 5 or more attempts for arterial line placement compared to no patients in our non–COVID-19 ICU cohort. As both patient discomfort and the risk of caregiver infection are increased with the multiple attempts, increased difficulty with arterial line insertion is clinically relevant and hopefully will drive further investigations to understand the underlying pathophysiology and develop countermeasures. Finally, while general population data on sex-based differences in venous thromboembolism risk are mixed,14 prior data indicate that male sex is associated with greater arterial line patency.15 In our model, female sex was an independent risk factor for arterial line failure. The 1.87 adjusted SHR of arterial line clot for women in our study was higher than previously reported and should also promote investigation as to an interaction between COVID-19 infection and female sex. Estrogen is unlikely to play a role in our finding, as the average age of COVID-19 and non–COVID-19 groups was 64.1 years and 62.7 years old, respectively.
The role of anticoagulation and its intensity in COVID-19 is still under investigation. The INSPIRATION Trial found no difference in venous or arterial events, extracorporeal membrane oxygenation (ECMO) use, or mortality at 30 days between COVID-19 patients in the ICU using an intermediate anticoagulation strategy compared to standard prophylactic anticoagulation.16 The National Institutes of Health (NIH) sponsored Accelerating COVID-19 Therapeutic Interventions and Vaccines 4 ACUTE (ACTIV-4a), in combination with a Randomised, Embedded, Multi-factorial, Adaptive Platform Trial for Community-Acquired Pneumonia (REMAP CAP) and Antithrombotic Therapy to Ameliorate Complications of COVID-19 (ATTACC), multiplatform trial, which assessed full dose anticoagulation versus standard prophylactic dose in COVID-19 ICU patients, found no benefit of therapeutic dose anticoagulation in this population.17 Current American Society of Hematology guidelines suggest using standard prophylactic anticoagulation over intermediate or treatment dosing for COVID-19 ICU patients.18 Other trials are currently investigating the optimal anticoagulation regimen for COVID-19 ICU patients.
Our data suggest that treatment dose anticoagulation compared with standard prophylaxis dose led to longer duration of functioning of arterial lines without clotting. We observed no significant difference between intermediate versus treatment dosing or intermediate versus prophylactic anticoagulation dosing on arterial line functioning. Although preserving arterial line function is not itself a compelling reason for full anticoagulation, it is a potentially relevant consideration given the risk of caregiver infection with invasive procedures.
Before the COVID-19 pandemic, data were conflicted on the benefit of heparin flushes compared to normal saline flushes in maintaining the functionality of arterial catheters.1 A 2020 observational cohort study found improved arterial line patency duration in COVID-19 patients when using low-dose heparinized saline, suggesting a potential benefit in this patient population.19 However, the effects of systemic or local anticoagulation on arterial lines have not been further tested in large, prospective studies.
Our study has several limitations. The small and retrospective nature limited our ability to detect small effect sizes and raises the possibility of confounders that may have affected the effect we observed. The limited sample size posed challenges to extensive, granular confounder adjustment given low prevalence comorbidities and concerns for overfitting. Rapidly evolving protocols during the first COVID wave, such as an institutional shift in April 2020 to intermediate dose prophylactic anticoagulation for all COVID-19 ICU patients, may also have confounded our results. The impact of other evolving COVID-19 therapeutics on our outcome was not assessed. Although we sought contemporaneous non–COVID-19 ICU patients, this comparator group was comprised primarily of patients hospitalized before the onset of the pandemic due to suspension of most non–COVID-19 studies at the beginning of the pandemic. Finally, our study focused only on the first COVID wave, and subsequent waves may have had different clinical characteristics. Given these limitations, our results should be interpreted as hypothesis generating.
In conclusion, patients with severe COVID-19 frequently require arterial line placement. Because of the risk of infection, minimizing provider exposure while placing, managing, and replacing arterial lines is important to reducing patient to caregiver infection. Our finding is exploratory only but suggests an increased subdistribution hazard of arterial line removal for clot in COVID-19 ICU versus non–COVID-19 sepsis/ARDS ICU patients. Treatment dose anticoagulation compared to prophylactic dose resulted in a significant decrease in the subdistribution hazard of arterial line removal for clot. Further studies should be undertaken to investigate the association of COVID-19 with arterial line thromboses and the role anticoagulation in minimizing arterial line dysfunction in patients with COVID-19.
DISCLOSURES
Name: Rebecca L. Zon, MD.
Contribution: This author helped with development of study concept, data collection, and writing contribution.
Conflicts of Interest: R. L. Zon is a consultant and stockholder for Amagma Therapeutics.
Name: Lauren E. Merz, MD, MSc.
Contribution: This author helped with data collection and writing contribution.
Conflicts of Interest: None.
Name: Kara G. Fields, MS.
Contribution: This author helped with statistical analysis.
Conflicts of Interest: None.
Name: Jessica Grandoni, PharmD.
Contribution: This author helped with data collection and writing contribution.
Conflicts of Interest: None.
Name: Jessica C. Stuart, MD.
Contribution: This author helped with data collection and writing contribution.
Conflicts of Interest: None.
Name: Rachel H. Occhiogrosso, MD.
Contribution: This author helped with data collection and writing contribution.
Conflicts of Interest: None.
Name: Linda Li, MD.
Contribution: This author helped with data collection.
Conflicts of Interest: None.
Name: Rebecca M. Baron, MD.
Contribution: This author helped with data collection.
Conflicts of Interest: R. M. Baron is on Advisory Boards for Genentech and Merck.
Name: Laura E. Fredenburgh, MD.
Contribution: This author helped with data collection.
Conflicts of Interest: L. E. Fredenburgh reports research funding to the institution from Bayer, outside the submitted work.
Name: Ann E. Woolley, MD, MPH.
Contribution: This author helped with data collection.
Conflicts of Interest: None.
Name: Jean M. Connors, MD.
Contribution: This author helped with the development of the study concept and writing contribution.
Conflicts of Interest: J. M. Connors reported receiving personal fees from Bristol Myers Squibb, Pfizer, Abbott, Alnylam, Takeda, Roche, and Sanofi and that her institution has received research funding from CSL Behring.
Name: Gyorgy Frendl, MD, PhD.
Contribution: This author helped with the development of the study concept and writing contribution.
Conflicts of Interest: G. Frendl reports funding from the National Heart, Lung, and Blood Institute (1UG3-HL140177-01A1; PI: Marcos Vidal-Melo; site-PI: Gyorgy Frendl), and from Sage Pharmaceuticals, Alexion Pharmaceuticals, and bioMerieux for clinical trials unrelated to this project.
This manuscript was handled by: Avery Tung, MD, FCCM.
Supplementary Material
GLOSSARY
ACTIV-4a Accelerating COVID-19 Therapeutic Interventions and Vaccines 4 ACUTE
ARDS acute respiratory distress syndrome
ATTACC Antithrombotic Therapy to Ameliorate Complications of COVID-19
BID twice daily
BMI body mass index
BWH Brigham and Women’s Hospital
CI confidence interval
COVID-19 coronavirus disease 2019
DVT deep vein thrombosis
ECMO extracorporeal membrane oxygenation
EHR electronic health record
ICU intensive care unit
INR international normalized ratio
IRB institutional review board
IRR incidence rate ratio
LDHS low-dose heparinized saline
NIH National Institutes of Health
PE pulmonary embolism
Q quartile
REMAP CAP A Randomised, Embedded, Multi-factorial, Adaptive Platform Trial for Community-Acquired Pneumonia
RoCI registry of critical illness
RRT renal replacement therapy
SHR subdistribution hazard ratio
TID three times daily
Funding: None.
Conflicts of Interest: See Disclosures at the end of the article.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.
G. Frendl and J. Connors contributed equally and share last authorship.
==== Refs
REFERENCES
1. Robertson-Malt S Malt GN Farquhar V Greer W . Heparin versus normal saline for patency of arterial lines. Cochrane Database Syst Rev. 2014;5:CD007364.24825673
2. Klok FA Kruip MJHA van der Meer NJM . Incidence of thrombotic complications in critically ill ICU patients with COVID-19. Thromb Res. 2020;191 :145–147.32291094
3. Moll M Zon RL Sylvester KW . VTE in ICU patients with COVID-19. Chest. 2020;158 :2130–2135.32710891
4. Llitjos JF Leclerc M Chochois C . High incidence of venous thromboembolic events in anticoagulated severe COVID-19 patients. J Thromb Haemost. 2020;18 :1743–1746.32320517
5. Spiezia L Boscolo A Poletto F . COVID-19-related severe hypercoagulability in patients admitted to intensive care unit for acute respiratory failure. Thromb Haemost. 2020;120 :998–1000.32316063
6. Han H Yang L Liu R . Prominent changes in blood coagulation of patients with SARS-CoV-2 infection. Clin Chem Lab Med. 2020;58 :1116–1120.32172226
7. Al-Samkari H Karp Leaf RS Dzik WH . COVID-19 and coagulation: bleeding and thrombotic manifestations of SARS-CoV-2 infection. Blood. 2020;136 :489–500.32492712
8. Dolinay T Kim YS Howrylak J . Inflammasome-regulated cytokines are critical mediators of acute lung injury. Am J Respir Crit Care Med. 2012;185 :1225–1234.22461369
9. Austin PC . An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46 :399–424.21818162
10. Austin PC . Using the standardized difference to compare the prevalence of a binary variable between two groups in observational research. Simul Computat. 2009;38 :1228–1234.
11. Zhou B Fine J Latouche A Labopin M . Competing risks regression for clustered data. Biostatistics. 2012;13 :371–383.22045910
12. Zhou B Fine J Laird G . Goodness-of-fit test for proportional subdistribution hazards model. Stat Med. 2013;32 :3804–3811.23625840
13. Althouse AD . Post hoc power: not empowering, just misleading. J Surg Res. 2021;259 :A3–A6.32814615
14. Montagnana M Favaloro EJ Franchini M Guidi GC Lippi G . The role of ethnicity, age and gender in venous thromboembolism. J Thromb Thrombolysis. 2010;29 :489–496.19536458
15. American Association of Critical-Care Nurses. Evaluation of the effects of heparinized and nonheparinized flush solutions on the patency of arterial pressure monitoring lines: the AACN thunder project. Am J Crit Care. 1993;2 :3–15.8353575
16. Sadeghipour P Talasaz AH Rashidi F ; INSPIRATION Investigators. Effect of intermediate-dose vs standard-dose prophylactic anticoagulation on thrombotic events, extracorporeal membrane oxygenation treatment, or mortality among patients with COVID-19 admitted to the intensive care unit: the INSPIRATION randomized clinical trial. JAMA. 2021;325 :1620–1630.33734299
17. Investigators R-C Investigators AC-a Investigators A . Therapeutic anticoagulation with heparin in critically ill patients with COVID-19. N Engl J Med. 2021;385 :777–789.34351722
18. Cuker A Tseng EK Nieuwlaat R . American Society of Hematology 2021 guidelines on the use of anticoagulation for thromboprophylaxis in patients with COVID-19. Blood Adv. 2021;5 :872–888.33560401
19. Maurer LR Luckhurst CM Hamidi A . A low dose heparinized saline protocol is associated with improved duration of arterial line patency in critically ill COVID-19 patients. J Crit Care. 2020;60 :253–259.32920504
| 36219579 | PMC9725743 | NO-CC CODE | 2022-12-08 23:18:16 | no | Anesth Analg. 2023 Jan 11; 136(1):70-78 | utf-8 | Anesth Analg | 2,022 | 10.1213/ANE.0000000000006214 | oa_other |
==== Front
Lancet Microbe
Lancet Microbe
The Lancet. Microbe
2666-5247
The Author(s). Published by Elsevier Ltd.
S2666-5247(22)00335-4
10.1016/S2666-5247(22)00335-4
Correspondence
Omicron sublineage recombinant XBB evades neutralising antibodies in recipients of BNT162b2 or CoronaVac vaccines
Zhang Xiaojuan a
Chen Lin-Lei a
Ip Jonathan Daniel a
Chan Wan-Mui a
Hung Ivan Fan-Ngai b
Yuen Kwok-Yung acde
Li Xin ac
To Kelvin Kai-Wang acde
a State Key Laboratory for Emerging Infectious Diseases, Department of Microbiology, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
b Department of Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
c Department of Microbiology, Queen Mary Hospital, Hong Kong Special Administrative Region, China
d Department of Clinical Microbiology and Infection Control, The University of Hong Kong-Shenzhen Hospital, Shenzhen, China
e Centre for Virology, Vaccinology and Therapeutics, Hong Kong Science and Technology Park, Hong Kong Special Administrative Region, China
6 12 2022
6 12 2022
© 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmcThe SARS-CoV-2 omicron variant XBB sublineage, a BA.2.10.1–BA.2.75 recombinant classified as variant under monitoring by WHO, has been found in 35 countries,1 and has become the dominant strain in Singapore. There is early evidence suggesting that XBB might be associated with a higher risk of reinfection.2 A previous study using a pseudovirus neutralisation test and sera from individuals who received CoronaVac (Sinovac) found that XBB is the most immunoevasive sublineage.3
We assessed the neutralisation of XBB.1 and XBB.3 compared with BA.5.2 (a widely circulating strain since July, 2022) and the ancestral strain, using a live virus neutralisation test.4, 5 XBB.1 differs from XBB.3 due to an extra spike mutation: Gly252Val. We included sera specimens from 30 individuals who received two to four doses of BNT16b2 (Pfizer-BioNtech) or CoronaVac with or without previous SARS-CoV-2 infection (seven [23%] individuals who received two vaccinations and had a previous BA.2 infection; seven [23%] who received three vaccinations and had a previous BA.2 infection; nine [30%] who received three vaccinations and had no previous SARS-CoV-2 infections; and seven [23%] who received four vaccinations and had no previous SARS-CoV-2 infections; appendix p 3). Overall, the geometric mean 50% neutralising antibody titre (NT50 GMT) was lower for XBB strains (XBB.1, 26.0; XBB.3, 19.4) than the ancestral strain (436.1; XBB.1 16·8-fold, p<0.0001; XBB.3: 22·5-fold, p<0·0001) or BA.5.2 strain (87.4; XBB.1 3·4-fold, p=0·0191; XBB.3 4·5-fold, p<0·0001), but the difference between XBB.1 and XBB.3 was not statistically significant (p=0·17; appendix p 1). All subgroups with different history of vaccination or infection had a statistically significantly lower GMT against XBB.1 or XBB.3 than those against the ancestral strain (appendix p 1).
Paired acute and convalescent serum specimens were available for one patient with BA.5.2 and two patients with XBB. The patient who had previously had a BA.5.2 infection, had a 16·2 times higher NT50 GMT against the ancestral strain and a 16·5 times higher GMT NT50 against BA.5.2 for the convalescent serum than the acute serum, but the acute and convalescent sera had similar NT50 against XBB.1 or XBB.3 (appendix p 2). The patient who had previously had an XBB.1 infection had an increase in their NT50 GMT against the ancestral strain by 7·9 times, BA.5.2 by 29·6 times, XBB.1 by 21·3 times, and XBB.3 by 28·1 times (appendix p 2). The patient who had previously had an XBB.3 infection had an increase in their NT50 GMT against XBB.1 by 10·8 times and XBB.3 by 6·9 times; this patient had a 2·1 times increase in their NT50 GMT against the ancestral strain and a 3·2 times increase against BA.5.2. In summary, our data showed that both XBB.1 and XBB.3 were much more immunoevasive than ancestral strain and BA.5.2. This immunoevasion is consistently seen in patients with different history of vaccination or infection. Since patients infected with BA.5.2 might not elicit neutralising antibody against XBB sublineage, patients who have been infected with BA.5 or those with bivalent vaccine might have a higher risk of reinfection or vaccine breakthrough infection from XBB sublineage than previous sublineages.
Supplementary Material
Supplementary appendix
IF-NH received payment from MSD for speaking at the COVID-19 Regional Expert Input Forum 2021; is on the advisory board of Pfizer on COVID-19 management in 2022; and was on the advisory board of Gilead on evolving treatment landscape in COVID-19 in 2021. K-YY and KK-WT report collaboration with SinoVac and Sinopharm. All other authors declare no competing interests. XZ and L-LC contributed equally. We would like to acknowledge Deborah Ho and Polly Pang for their help in patient recruitment; and Kwok-Hung Chan, Allen Chu, Brian Chan, and Ricky Zhang for their technical support. This work was supported by Health and Medical Research Fund, the Food and Health Bureau, The Government of the Hong Kong Special Administrative Region (HKSAR; COVID1903010 [Project 1]), Consultancy Service for Enhancing Laboratory Surveillance of Emerging Infectious Diseases and Research Capability on Antimicrobial Resistance for Department of Health of the HKSAR; the Theme-Based Research Scheme (T11/707/15) of the Research Grants Council, HKSAR; Emerging Collaborative Project of Guangzhou Laboratory (EKPG22-01); and Emergency COVID-19 Project (2021YFC0866100), Major Projects on Public Security, National Key Research and Development Program; and donations of Richard Yu and Carol Yu, Shaw Foundation Hong Kong, Michael Seak-Kan Tong, May Tam Mak Mei Yin, Lee Wan Keung Charity Foundation, Hong Kong Sanatorium and Hospital, Respiratory Viral Research Foundation Limited, Hui Ming, Hui Hoy and Chow Sin Lan Charity Fund, Chan Yin Chuen Memorial Charitable Foundation, Marina Man-Wai Lee, the Jessie and George Ho Charitable Foundation, Kai Chong Tong, Tse Kam Ming Laurence, Foo Oi Foundation, Betty Hing-Chu Lee, and Ping Cham So.
==== Refs
References
1 WHO COVID-19 weekly epidemiological update - 26 October 2022 https://www.who.int/publications/m/item/weekly-epidemiological-update-on-covid-19---26-october-2022 2022
2 WHO TAG-VE statement on omicron sublineages BQ.1 and XBB. 2022 https://www.who.int/news/item/27-10-2022-tag-ve-statement-on-omicron-sublineages-bq.1-and-xbb
3 Cao Y Jian F Wang J Imprinted SARS-CoV-2 humoral immunity induces convergent Omicron RBD evolution bioRxiv 2022 published online Oct 30. 10.1101/2022.09.15.507787
4 Chen LL Chu AW Zhang RR Hung IF To KK Serum neutralisation of the SARS-CoV-2 omicron sublineage BA.2 Lancet Microbe 3 2022 e404 35373159
5 Chen LL Abdullah SMU Chan WM Contribution of low population immunity to the severe Omicron BA.2 outbreak in Hong Kong Nat Commun 13 2022 3618 35750868
| 36493789 | PMC9725777 | NO-CC CODE | 2022-12-08 23:18:16 | no | Lancet Microbe. 2022 Dec 6; doi: 10.1016/S2666-5247(22)00335-4 | utf-8 | Lancet Microbe | 2,022 | 10.1016/S2666-5247(22)00335-4 | oa_other |
==== Front
Lancet Microbe
Lancet Microbe
The Lancet. Microbe
2666-5247
The Author(s). Published by Elsevier Ltd.
S2666-5247(22)00289-0
10.1016/S2666-5247(22)00289-0
Articles
Leveraging an established neighbourhood-level, open access wastewater monitoring network to address public health priorities: a population-based study
Bowes Devin A PhD abm†
Driver Erin M PhD ae†
Kraberger Simona PhD c
Fontenele Rafaela S PhD cd
Holland LaRinda A BS c
Wright Jillian BS am
Johnston Bridger BS a
Savic Sonja BS a
Engstrom Newell Melanie BS ab
Adhikari Sangeet PhD ae
Kumar Rahul PhD a
Goetz Hanah PhD b
Binsfeld Allison BS am
Nessi Kaxandra BS a
Watkins Payton BS a
Mahant Akhil BS a
Zevitz Jacob BS a
Deitrick Stephanie PhD j
Brown Philip MS k
Dalton Richard MS k
Garcia Chris BS k
Inchausti Rosa MA l
Holmes Wydale MPA l
Tian Xiao-Jun PhD h
Varsani Arvind PhD cdg
Lim Efrem S PhD cd
Scotch Matthew Prof PhD af
Halden Rolf U Prof PhD aeim*
a The Biodesign Institute Center for Environmental Health Engineering, Arizona State University, Tempe, AZ, USA
b School for Engineering of Matter, Transport, and Energy, Arizona State University, Tempe, AZ, USA
c The Biodesign Institute Center for Fundamental and Applied Microbiomics, Arizona State University, Tempe, AZ, USA
d School of Life Sciences, Arizona State University, Tempe, AZ, USA
e School for Sustainable Engineering and the Built Environment, Arizona State University, Tempe, AZ, USA
f College of Health Solutions, Arizona State University, Tempe, AZ, USA
g Center for Evolution and Medicine, Arizona State University, Tempe, AZ, USA
h School of Biological and Health Systems Engineering, Arizona State University, Tempe, AZ, USA
i Global Futures Laboratory, Arizona State University, Tempe, AZ, USA
j Enterprise GIS & Data Analytics, Information Technology, Tempe, AZ, USA
k Municipal Utilities, Tempe, AZ, USA
l Strategic Management and Diversity Office, Tempe, AZ, USA
m OneWaterOneHealth, The Arizona State University Foundation, The Biodesign Institute, Arizona State University, Tempe, AZ, USA
* Correspondence to: Prof Rolf U Halden, The Biodesign Institute Center for Environmental Health Engineering, Arizona State University, Tempe, AZ 85287, USA
† Contributed equally
6 12 2022
6 12 2022
© 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
Before the COVID-19 pandemic, the US opioid epidemic triggered a collaborative municipal and academic effort in Tempe, Arizona, which resulted in the world's first open access dashboard featuring neighbourhood-level trends informed by wastewater-based epidemiology (WBE). This study aimed to showcase how wastewater monitoring, once established and accepted by a community, could readily be adapted to respond to newly emerging public health priorities.
Methods
In this population-based study in Greater Tempe, Arizona, an existing opioid monitoring WBE network was modified to track SARS-CoV-2 transmission through the analysis of 11 contiguous wastewater catchments. Flow-weighted and time-weighted 24 h composite samples of untreated wastewater were collected at each sampling location within the wastewater collection system for 3 days each week (Tuesday, Thursday, and Saturday) from April 1, 2020, to March 31, 2021 (Area 7 and Tempe St Luke's Hospital were added in July, 2020). Reverse transcription quantitative PCR targeting the E gene of SARS-CoV-2 isolated from the wastewater samples was used to determine the number of genome copies in each catchment. Newly detected clinical cases of COVID-19 by zip code within the City of Tempe, Arizona were reported daily by the Arizona Department of Health Services from May 23, 2020. Maricopa County-level new positive cases, COVID-19-related hospitalisations, deaths, and long-term care facility deaths per day are publicly available and were collected from the Maricopa County Epidemic Curve Dashboard. Viral loads of SARS-CoV-2 (genome copies per day) measured in wastewater from each catchment were aggregated at the zip code level and city level and compared with the clinically reported data using root mean square error to investigate early warning capability of WBE.
Findings
Between April 1, 2020, and March 31, 2021, 1556 wastewater samples were analysed. Most locations showed two waves in viral levels peaking in June, 2020, and December, 2020–January, 2021. An additional wave of viral load was seen in catchments close to Arizona State University (Areas 6 and 7) at the beginning of the fall (autumn) semester in late August, 2020. Additionally, an early infection hotspot was detected in the Town of Guadalupe, Arizona, starting the week of May 4, 2020, that was successfully mitigated through targeted interventions. A shift in early warning potential of WBE was seen, from a leading (mean of 8·5 days [SD 2·1], June, 2020) to a lagging (−2·0 days [1·4], January, 2021) indicator compared with newly reported clinical cases.
Interpretation
Lessons learned from leveraging an existing neighbourhood-level WBE reporting dashboard include: (1) community buy-in is key, (2) public data sharing is effective, and (3) sub-ZIP-code (postal code) data can help to pinpoint populations at risk, track intervention success in real time, and reveal the effect of local clinical testing capacity on WBE's early warning capability. This successful demonstration of transitioning WBE efforts from opioids to COVID-19 encourages an expansion of WBE to tackle newly emerging and re-emerging threats (eg, mpox and polio).
Funding
National Institutes of Health's RADx-rad initiative, National Science Foundation, Virginia G Piper Charitable Trust, J M Kaplan Fund, and The Flinn Foundation.
==== Body
pmcIntroduction
Triggered by the SARS-CoV-2 pandemic, the use of wastewater-based epidemiology (WBE) as a potentially powerful, rapid, and inexpensive tool to inform public health decision making has seen a remarkable increase globally. For decades, WBE has been used to track chemical and biological threats, with numerous studies underscoring its efficacy and usefulness for understanding and managing community health.1, 2 At the onset of the SARS-CoV-2 pandemic, substantial delays in conventional and individualised clinical testing, due in part to an overwhelmed health-care system and resource limitations,3 positioned WBE as a promising supplemental tool for assessing SARS-CoV-2 spread at the population level.4, 5, 6, 7 Early data from the beginning of the pandemic suggested that SARS-CoV-2 levels in wastewater and sewage sludge offered early indication of clinical confirmed infections, disease, and mortality in a community.8, 9
Research in context
Evidence before this study
We searched PubMed using the terms “wastewater-based epidemiology” AND “SARS-CoV-2” AND “COVID-19”, for articles that reported SARS-CoV-2 detection in wastewater published from January, 2020, to April, 2020. No language restrictions were applied to the search. The field of wastewater-based epidemiology (WBE) has been active for decades, using analytical tools to monitor a wide range of chemical and biological agents indicative of various aspects of human health, behaviour, and exposure in support of public health strategies and interventions. In response to a demand for enhanced data resolution early in the COVID-19 pandemic, researchers realised the benefits of spatially discrete sampling campaigns within urban sewersheds that could target study populations at the neighbourhood, campus, and even building (dormitory) level. This new approach to population health monitoring by WBE has raised important questions regarding what constitutes ethical practices, and how to shield the emerging science from concerns of privacy, inappropriate use of data, and the potential stigmatisation of populations.
Added value of this study
This study highlights pioneering work in the use of wastewater-based epidemiology conducted at the neighbourhood level for monitoring both substance use (opioids) and the spread of COVID-19, showcasing the early adoption of community-forward WBE to manage health risks. A guiding principle in study design was maintenance of ethical standards in data collection and data use by engaging community members early on, obtaining buy-in before commencing monitoring efforts, and sharing data equitably with all stakeholders. This academic and municipal partnership showcases an early-established (2018) neighbourhood-level wastewater monitoring effort for opioid use accompanied by an online data dashboard that served to rapidly launch a similar model for SARS-CoV-2 and COVID-19 to effectively aid in the public health response using a non-invasive and inclusive approach to community health protection.
Implications of all the available evidence
The time lag observed between recognition of a new public health threat (eg, SARS-CoV-2) and the development and deployment of public health response (eg, development and scale-up of clinical testing and vaccination capacity) will not be limited to the COVID-19 pandemic, as already evidenced by mpox (formerly known as monkeypox). The lessons learned in this study serve to inform municipalities and other institutions on how to develop a wastewater-based surveillance system capable of adapting to future public health threats, including chemical exposures, infectious disease agents, and antibiotic resistance, while maintaining ethically sound data collection, dissemination, and management strategies.
The City of Tempe, Arizona, with a residential population of around 200 000, had been an early adopter of WBE for the purpose of tracking opioid use, which began in May, 2018, and led to the launch of a fully interactive, public-facing, open access WBE dashboard in February, 2019.10 In this municipal and academic partnership, Tempe and Arizona State University (ASU) first engaged with community stakeholders in a series of public town halls and workshops to achieve a high level of ethical practice and community acceptance of the use of WBE.11 Once established, monthly wastewater samples were shared, with subsequent analysis and joint reporting of opioid use-trends within the community (oxycodone, codeine, heroin, fentanyl, and metabolites; μg per day per 1000 people) in five urban sewersheds.10 From this effort, a routine was developed that served to enhance the public health response by integrating Tempe Fire Medical Rescue, crisis intervention services, and other stakeholders into a workgroup that relied on WBE data to guide resource deployment by community need (figure 1 ). With this existing framework in place, Tempe and ASU were in a unique position at the start of the COVID-19 pandemic to leverage this previously established collaboration to rapidly monitor SARS-CoV-2. Quantitative assessments of virus levels in wastewater soon followed, with objectives to identify hotspots of infection early, and implement targeted interventions where appropriate. The pre-existing, neighbourhood-level wastewater monitoring network offered an opportunity to test the potential of WBE to serve as an early warning system that might reveal virus presence and spread before clinical case data from testing of individuals.12, 13 Thus, the overarching goals of our study were (1) to compare WBE data to newly reported clinical cases, related hospitalisations, and associated deaths at a high temporal and spatial resolution (ie, county, city, zip code [postal code], and neighbourhood levels), and (2) to determine whether the concurrent pandemic monitoring by WBE produced data and information not available or obvious from clinical testing.Figure 1 Schematic of the Arizona State University and City of Tempe academic and municipal partnership
A WBE monitoring network established in 2018 for monitoring opioid use was leveraged to enable a rapid transition to monitoring SARS-CoV-2 during the COVID-19 global pandemic (starting in 2020) with work products including the world's first WBE-informed public-facing interactive online dashboards to combat the opioid and COVID-19 epidemics through a data-driven targeted public health response. ASU=Arizona State University. LC-MS/MS=liquid chromatography-tandem mass spectrometry. RT-qPCR=reverse transcription quantitative PCR. WBE=wastewater-based epidemiology.
Methods
Study locations
This study was conducted within the City of Tempe, Arizona, and the Town of Guadalupe, Arizona, (ie, Greater Tempe), with an estimated residential population of approximately 200 000, and home to Arizona State University. For opioid monitoring before the pandemic, the Greater Tempe area was divided into five sewer catchments (Areas 1–5; appendix p 8) as determined by recurring compliance monitoring. Two additional catchments from an adjacent municipality that contributed to Tempe wastewaters were collected to isolate the Tempe-associated sewage signal. To improve spatial resolution, additional sampling locations were identified based on ease of collection (Area 6). Three other permanent locations needed infrastructure modifications or approvals before onboarding, including Area 7, the Town of Guadalupe (appendix p 8), and Tempe St Luke's Hospital, which had an active COVID-19 ward at the time this study took place (appendix p 9). In total, 11 wastewater catchments were analysed in this study.
The Institutional Review Board of Arizona State University determined this study was not research involving human participants and was exempt from formal review (STUDY00006069).
Sample collection, processing, and analysis
The predefined sampling strategy for opioid monitoring consisted of 7 consecutive days of sample collection each month across variable weeks from permanent, subsurface sampling stations. Although this sampling strategy was sufficient for long-term trend assessment of opioid use, SARS-CoV-2 monitoring required increased temporal resolution based on infection rate dynamics. Accordingly, flow-weighted and time-weighted 24 h composite samples of untreated wastewater were collected at each sampling location within the wastewater collection system for 3 days each week (Tuesday, Thursday, and Saturday) beginning in April, 2020, (Area 7 and Tempe St Luke's Hospital were added in July, 2020). Samples were collected either with an automated refrigerated or portable sampler (Teledyne ISCO, Lincoln, NE, USA) using a mixture of wet and dry ice for cooling (appendix p 10). Wastewater flow was monitored by an ISCO LaserFlow meter (Teledyne ISCO, Lincoln, NE, USA), located within a nearby manhole, or estimated based on historic data (appendix p 11). Composite samples were transferred to high-density polyethylene bottles stored in coolers with ice for transport and processed on the same day to minimise degradation.
Samples were processed and analysed according to published studies.5, 14, 15 Briefly, raw wastewater samples were analysed for SARS-CoV-2 following sequential steps of filtration (0·45 μm polyether sulfone membrane), viral concentration by ultracentrifugation (10 kDa molecular weight cutoff centrifugal filters), nucleic acid extraction (50 μL resultant RNA extract), and TaqMan-based (Invitrogen, Waltham, MA, USA) reverse transcription quantitative PCR (RT-qPCR) targeting the SARS-CoV-2 E (envelope) gene. Spike-and-recovery experiments were performed using murine hepatitis virus (as a surrogate) based on previously published methods.5, 16, 17 Full method details for sample processing and analysis, including primer and probe sequences and RT-qPCR methods, are in the appendix (pp 2–3, 6–7).
Population estimates
Resident populations for each sewershed were estimated using 2010 census block group data. Employment estimates were obtained from the Maricopa Association of Governments (MAG) 2019 employment data and included the following classifications: employees living outside of Tempe (non-resident, employed) and Tempe residents (resident, employed). To correct for changes in employment numbers during lockdown events (commercial closures) and telecommuting activities, we used available MAG average weekday traffic volume (compared with non-lockdown events) in Maricopa County. This percentage was used to correct the non-resident (Tempe employed) employment numbers. Student population estimates were obtained from publicly available campus resident data18 and changes in wastewater flow volume.
Clinical data
Newly detected clinical cases of COVID-19 by zip code within the City of Tempe, Arizona, were reported daily by the Arizona Department of Health Services. The City of Tempe began extracting and archiving these data on May 23, 2020. Daily case data are not available before this date (data are in aggregate as total cases from the start of the pandemic). Data on Maricopa County-level new positive cases, COVID-19-related hospitalisations, deaths, and long-term care facility deaths per day are publicly available and were collected from the Maricopa County Epidemic Curve Dashboard.19 Data on clinical testing capacity at both the city level (Tempe) and state level (Arizona) were acquired through publicly available datasets.19, 20
Statistical analysis
Measured concentrations of SARS-CoV-2 (genome copies per L) in each sewer catchment (x) were transformed to viral load per day (genome copies per day) using the following equation:
Viralload(genomecopiesperday)=Cx×Qx
where Cx (genome copies per L) is the measured concentration in a given wastewater catchment and Qx is the total daily volumetric flow rate (L per day) for that wastewater catchment (appendix p 11). In cases where one sewer catchment flowed into another (appendix p 12), a mass balance was performed whereby the viral load (genome copies per day) from the contributing catchment was subtracted from the receiving catchment to isolate individual signals.
Statistical assessments were conducted in MATLAB R2021a (MathWorks, Natick, MA, USA). Root mean square error, a statistical test that measures the standard deviation of the residuals (ie, the predicted data), was used to calculate the offset between different compared data categories using the following equation:
Rootmeansquareerror=∑in(xi-yi)2
where n is the number of observations in each specific dataset, x i is the viral load (genome copies per day) of SARS-CoV-2 in wastewater, and y i refers to external datasets: either the newly detected clinical cases, COVID-19-related hospitalisations, or COVID-19-related deaths. Data were assessed between April 1, 2020, and March 31, 2021, using individual waves of infection corresponding to up to three events peaking in June–July, 2020, August 2020, and December, 2020–January, 2021. Data were shifted from 0 to 20 days in both directions (forwards and reverse) for each of the comparisons. The data resolution between clinical cases and wastewater testing were different (daily vs three times per week); thus, clinical results that did not have a corresponding wastewater data point were omitted from the assessment post-shift.
Role of the funding source
The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
Results
Between April 1, 2020, and March 31, 2021, a total of 1556 wastewater samples were analysed across the Greater Tempe area. The number of samples collected per catchment during the study varied from 155 samples in Area 1 to 103 samples in Area 7 and 101 samples in Tempe St Luke's Hospital, with observed differences in the number of samples resulting from occasional sampler malfunctioning and staggered onboarding of sampling locations. The mean total number of SARS-CoV-2 detections per catchment throughout the study was 66 (SD 36), with a minimum of four (Area 3) to a maximum of 116 (Area 6). When detected, the mean SARS-CoV-2 concentration was 617 000 (SD 2·075 million) gene copies per L (median 251 450 gene copies per L [IQR 3500–260300]), indicative of great fluctuations in virus levels over time. Detailed concentration information is provided in the appendix (p 9).5, 14
SARS-CoV-2 viral loads were calculated for each sample using wastewater flow data provided by Tempe (appendix p 11). Flow rates in catchment Areas 1–7 had data recorded at 2-min intervals in real time using permanent laser flow meters, and the Town of Guadalupe and Tempe St Luke's Hospital had only historical flow data available. Flow varied from a maximum in Area 1 of 54·5 million L per day (SD 6·6) to a minimum of 0·106 million L per day (historical estimate) for the hospital location; median 3·0 million L per day (IQR 0·4 million–16·8 million). Mass balances that corrected wastewater sample co-mingling in Areas 1–3 (appendix p 12) resulted in viral loads ranging from 6 × 1010 to 1 × 1013 genome copies per day (appendix p 13). The distributions in viral load varied between each location, with Areas 1, 2, and 6 having smoother distributions in viral load over time, and other locations such as Areas 3, 5, and 7 showing more isolated single-day spikes in activity. Most locations showed two waves in viral levels peaking in June, 2020, and December, 2020–January, 2021. However, in catchments close to ASU, an additional unique wave of viral load was seen (Areas 6 and 7) at the beginning of the fall (autumn) semester in late August, 2020 (appendix p 13).
Estimated Tempe (Areas 1–7) sewer catchment subpopulations ranged from a minimum of 6698 in Area 5 and a maximum of 142 520 in Area 1, the largest geographical catchment area (appendix p 5). Variability in Tempe data was a function of the total numbers of residents, employed individuals, and the number of students in the catchment areas. The population of the Town of Guadalupe (6500) was determined using US census data,21 and the hospital location was omitted from this population analysis as the number of hospital staff and patients was unknown.
The result of these efforts ultimately culminated in eight collection locations viewable online by the public (Areas 1–7 of Tempe and the Town of Guadalupe) on an interactive dashboard that went live the week of May 4, 2020 (appendix p 8), with the Town of Guadalupe displayed on a separate tab (appendix p 8). The dashboard displays each catchment area overlaid on a street-level city map so that users can geospatially identify contributing locations to each catchment. Data are shown as the logarithm of genome copies per L of wastewater and are presented as a weekly average consisting of the Tuesday, Thursday, and Saturday collected samples. Text and infographics accompany these data, such as: overview of WBE basics, how to properly interpret the data, and how data are used by the city. Additionally, the SARS-CoV-2 wastewater dashboard is nested in a Community COVID-19 Health Site that contains pertinent information provided by the Centers for Disease Control and Prevention, city demographic information, and positive clinical cases reported by zip code. During the initial lockdown from May to early June, 2020, the wastewater-informed data dashboard revealed remarkably elevated virus concentrations (genome copies per L) in the Town of Guadalupe beginning the week of May 11, 2020, to the week of June 8, 2020, that later decreased after prompt interventions such as face-mask mandates and community education in town halls beginning in June, 2020 (appendix p 8).
Measured viral loads per day of SARS-CoV-2 within each sewer catchment area in Tempe were aggregated and partitioned to their respective zip codes according to their geographical distribution (appendix p 14). The five-digit, Tempe, Arizona zip codes used in this study are 85281, 85282, 85283, and 85284, and will be referred to here as ZC-1, ZC-2, ZC-3, and ZC-4, respectively. From this analysis, wastewater-derived SARS-CoV-2 peaks in activity correlated with newly detected clinical cases per day in three distinct waves of activity: June 2020, late August, 2020 (fall [autumn] semester 2020), and December, 2020–January, 2021. Comparisons between spikes of SARS-CoV-2 viral loads in wastewater and clinical case data showed that peaks in wastewater preceded positive clinical cases by 7 days (ZC-1), 6 days (ZC-2), 11 days (ZC-3), and 10 days (ZC-4) mean of 8·5 days (SD 2·1), during the first wave of the pandemic (June, 2020), and again during the late-August, 2020, wave (6 days; figure 2 ). Tempe aggregated viral loads were compared to Maricopa County Public Health data20 (figure 3 ), which again showed peaks of SARS-CoV-2 viral loads in wastewater preceding new clinically reported cases by 2 days, COVID-19-related hospitalisations by 16 days, and deaths by 18 days during the first wave of the pandemic in June, 2020. During the December, 2020–January, 2021 wave, wastewater was no longer a leading indicator in any region in Tempe. Trends either directly aligned with newly reported clinical cases (ZC-1) or lagged behind clinical case data by 2 days (ZC-2 and ZC-3), 4 days (ZC-4; figure 2), and 2 days (aggregate); mean of −2·0 days (SD 1·4; figure 2). At the county level, wastewater lagged behind clinical results, peaking at 3 days behind Maricopa reported cases, 3 days behind SARS-CoV-2-related hospitalisations, and 1 day behind SARS-CoV-2-related deaths (figure 3).Figure 2 Comparison of wastewater-derived viral loads of SARS-CoV-2 (genome copies per day) and clinical cases of COVID-19 in Tempe, Arizona
(A) SARS-CoV-2 genome copies per day in the four zip codes (ZC-1 to ZC-4) and in aggregate, overlaid with newly reported clinical cases. Numbers are the number of days the wastewater signal leads (positive number) or lags (negative number) clinical cases, determined by root mean square error analysis. (B) Reported number of saliva-based tests processed per day for faculty, staff, and the general public (grey), students only (black), and a 7-day trailing average (green dotted line). Vertical grey dotted lines indicate noteworthy enhancements of testing efficacy: laboratory capacity increased to 12 000 tests per day in November, 2020, and January, 2021 was reported to have the highest volume of saliva-based tests received. These improvements are a probable cause of wastewater monitoring transitioning from a leading to lagging or real-time indicator of community viral presence.
Figure 3 Wastewater-derived SARS-CoV-2 viral load (genome copies per day) compared with county-level and state-level clinical datasets
(A) Peaks in SARS-CoV-2 viral load (genome copies per day) in Tempe, Arizona wastewater compared with Maricopa County, Arizona new clinically detected cases of COVID-19, and COVID-19-related hospitalisations and deaths. Numbers are the number of days the wastewater signal leads (positive number) or lags (negative number) clinical cases. (B) Total number of diagnostic tests for COVID-19 conducted per week (light pink) in the state of Arizona, demonstrating increased testing efficacy starting from 30 available sites state-wide in May, 2020, to more than 500 available testing sites state-wide in October, 2020. (C) Reported timeline of events during time of study serves to support that as clinical testing improved throughout the pandemic, wastewater monitoring transitioned from a leading indicator to a lagging or real-time indicator. *The Arizona State University Biodesign Clinical Testing Laboratory developed a saliva-based test in partnership with the Arizona Department of Health Services to offer free, accessible testing state-wide. †Three universities: Arizona State University, Northern Arizona University, and University of Arizona.
Discussion
We employed WBE to monitor SARS-CoV-2 in Greater Tempe, Arizona, USA, by implementing a unique, high-frequency (three samples per week), and high spatial resolution (ie, neighbourhood-level) sampling approach in conjunction with immediate, open access data sharing with the public via an online dashboard. The present work is unique in that it provides historical context to the use of WBE, and how it can serve to reshape the philosophy of wastewater-informed public data sharing, and displaying public health information pertaining to both the opioid epidemic in the USA and the global COVID-19 pandemic. This study further illustrates that sub-sewershed wastewater monitoring produces actionable data, and can be conducted ethically with support from the community and relevant stakeholders.
The measured values of SARS-CoV-2 in wastewater were in line with those reported from other wastewater monitoring studies,9, 22 with the maximum concentration of 37·6 million gene copies of SARS-CoV-2 per L. This measurement occurred at the hospital location, which had an active COVID-19 ward at the time of collection on Jan 11, 2021, during peak pandemic conditions. Higher relative standard deviations (RSDs) for SARS-CoV-2 concentrations recorded for a given week (three observations per week) occurred in locations with a higher proportion of commercial businesses, including Area 4 (RSD 83%) and Area 5 (RSD 93%), compared with areas with largely residential catchments (Area 1 [58%] and Area 2 [65%]). This finding might explain the relatively smoother trends over time in Areas 1 and 2 compared with areas with higher transient populations that showed isolated single-day spikes in viral presence. These results suggest that a high-frequency sample collection approach should be considered in catchments with a higher proportion of transient populations, which might be susceptible to greater variability in wastewater-derived viral measurements from day to day.
Estimating population size by study area was challenging due to collecting wastewater from within the sewer infrastructure rather than determining population served by a wastewater treatment plant as historically conducted.23 Because of net importation of people to the city for work, it was important not only to quantify the number of residents but also the number of non-residents employed and transient student populations, a task accomplished using MAG and on-campus student resident data. To date, no other WBE study has reported use of employment data to refine population estimates within collection systems. MAG data needed to be corrected for lockdown activities, for which we used Arizona Department of Transportation arterial traffic flow data (eg, 40% decrease in arterial traffic equated to a 40% decrease in employment populations). Campus resident data was appropriate to assess temporal changes in student populations due to online courses; however, numbers did not account for changes during holiday travel or off-campus housing; thus, percentage changes in wastewater flow were used to estimate population changes. For instance, wastewater flow from Area 6 (university-adjacent catchment) increased by 20% throughout the academic year; therefore, the population in that area was assumed to increase proportionally.
The northern-most zip code of Tempe that encompasses ASU (ZC-1) was the only zip code that showed viral increases in late August, 2020. Contributions to viral load by university students within a given community was not just isolated to Tempe; contributions also occurred in other communities that housed large universities.24, 25 These results align with preliminary assessments of wastewater and clinical case data that suggested monitoring wastewater provided an early warning capacity ranging between 2 and 21 days, demonstrating that wastewater can serve as an early indicator of future clinical case load, morbidity, and mortality.9, 22
Although wastewater monitoring has demonstrated changes as a leading indicator at the wastewater treatment plant level,26 to our knowledge, our study is still the first report of capturing changes in lead-time dynamics of viral presence at the sub-catchment, zip-code level. This consecutive decrease in lead time between wastewater measurements and clinical testing might best be explained by notable capacity improvement in availability and frequency of clinical testing largely driven by saliva-based tests (from 30 sites in May, 2020, to 500 sites in October, 2020) that took place in parallel over the course of this study (Figure 2, Figure 3).18, 27 This unique clinical testing environment differs from other reported studies where the drivers of lead and lag might be affected by other parameters,28, 29 suggesting that the greatest benefits of WBE might be early detection of disease outbreaks in situations where a substantial health-care response has not yet been mounted for many reasons (ie, clinical testing site scarcity, testing fatigue, vaccination campaigns, and widespread use of at-home rapid tests).
The importance of neighbourhood-level sampling was demonstrated in our study and allowed for the identification of isolated hotspots, such as the Town of Guadalupe, Arizona, where the municipalities' wastewater co-mingles with Tempe's Area 3. However, in Area 3 SARS-CoV-2 was only detected four times during the entire year-long sampling campaign (appendix p 13), implying that the elevated SARS-CoV-2 signal originating in the Town of Guadalupe was attenuated beyond detectability at the Area 3 sampling location, and was visible only through high spatial resolution monitoring. Clinical testing in theory might have rendered the Town's infection hotspot visible; however, testing was scarce in the area, thereby potentially obscuring what was happening within the community.
The limitations of our study include the use of a single-gene target (E gene) analysis to determine SARS-CoV-2 presence, which was among one of the first gene targets reported by WHO in 2020. Additional viral targets might have improved our detection frequencies; however, the efficiency of the assay did not change over the course of this study, suggesting it to be reliable for this purpose. Sewage temperature, travel time, and storage are known to influence the stability of labile wastewater-borne biomarkers such as viral RNA;30 however, similar to most WBE studies, data were not corrected for these variables.
This study illustrates that a major challenge to neighbourhood-level monitoring by WBE is not only assay development, but also creating partnerships with city personnel, gaining trust from community members, establishing the sampling network and methodology, understanding which establishments or buildings are contributing to a given collected sample, and how these populations change (eg, on weekdays, at weekends, and during closures or events). These factors lead to detailing and understanding the primary outcomes of this type of investigation—how data should be protected, shared, and used to inform public health decision making. Thus, this work can serve to inform municipalities interested in adopting and implementing WBE programmes to monitor already known and newly emerging public health threats, and further, afford the ability to rapidly transition between threats, chemical or biological, if and when necessary. Our study can therefore be transferable across disciplines for a wide variety of applications in support of further understanding and investigating public health datasets.
Data sharing
All data needed to evaluate the conclusions in the Article are present in the main Article text and in the appendix. The SARS-CoV-2 Wastewater Monitoring Dashboard and COVID-19 Community Health Site in the City of Tempe, Arizona are available at https://covid19.tempe.gov/.
Declaration of interests
EMD is a managing member of AquaVitas, a company working in the field of wastewater-based epidemiology. RUH is a managing member of AquaVitas and founder of the Arizona State University non-profit project OneWaterOneHealth operating in the same intellectual space. All other authors declare no competing interests.
Supplementary Material
Supplementary appendix
Acknowledgments
This study was funded by the National Institutes of Health's RADx-rad initiative (U01LM013129-02S2), National Science Foundation (2028564), Virginia G Piper Charitable Trust (LTR 05/01/12), J M Kaplan Fund (30009070), and The Flinn Foundation. We thank the City of Tempe for their diligent collection of wastewater samples for this project throughout the course of the COVID-19 pandemic. We also thank Arizona State University students Nivedita Biyani and Indrayudh Mondal for their help in sample pickups, as well as Bryce McFayden and Michaela Shope for their help with sample processing.
Contributors
DAB and EMD conceptualised the study, conducted the laboratory work, did the sample processing, data generation, data analysis, and wrote the original draft of the manuscript. SK, RSF, LAH, AV, and ESL did method development, sample processing and analysis, and reviewed the manuscript. HG did data analysis. JW, BJ, SS, MEN, SA, RK, AB, KN, PW, AM, and JZ did sample processing. SD, PB, RD, and CG are City of Tempe personnel or collaborators who provided external datasets required for data analysis. RI and WH managed the Tempe wastewater-based epidemiology programme. X-JT oversaw data analysis and edited the manuscript. AV, ESL, and MS oversaw method development, including sample processing and analysis, and edited the manuscript. RUH conceived the study, provided oversight on the ASU wastewater-based epidemiology programme in conjunction with Tempe, and did data reporting. DAB and EMD accessed and verified the data. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.
==== Refs
References
1 Gracia-Lor E Castiglioni S Bade R Measuring biomarkers in wastewater as a new source of epidemiological information: current state and future perspectives Environ Int 99 2017 131 150 28038971
2 Choi PM Tscharke BJ Donner E Wastewater-based epidemiology biomarkers: past, present and future Trends Analyt Chem 105 2018 453 469
3 Sanche S Lin YT Xu C Romero-Severson E Hengartner N Ke R High contagiousness and rapid spread of severe acute respiratory syndrome coronavirus 2 Emerg Infect Dis 26 2020 1470 1477 32255761
4 Sherchan SP Shahin S Ward LM First detection of SARS-CoV-2 RNA in wastewater in North America: a study in Louisiana, USA Sci Total Environ 743 2020 140621 32758821
5 Wright J Driver EM Bowes DA Johnston B Halden RU Comparison of high-frequency in-pipe SARS-CoV-2 wastewater-based surveillance to concurrent COVID-19 random clinical testing on a public US university campus Sci Total Environ 820 2022 152877 34998780
6 Gonzalez R Curtis K Bivins A COVID-19 surveillance in southeastern Virginia using wastewater-based epidemiology Water Res 186 2020 116296 32841929
7 Hamouda M Mustafa F Maraqa M Rizvi T Aly Hassan A Wastewater surveillance for SARS-CoV-2: lessons learnt from recent studies to define future applications Sci Total Environ 759 2021 143493 33190883
8 Peccia J Zulli A Brackney DE Measurement of SARS-CoV-2 RNA in wastewater tracks community infection dynamics Nat Biotechnol 38 2020 1164 1167 32948856
9 Wu F Zhang J Xiao A SARS-CoV-2 titers in wastewater are higher than expected from clinically confirmed cases mSystems 5 2020 e00614 e00620 32694130
10 Tempe County Gov Tempe opioid wastewater collection dashboard 2019 https://arcg.is/ey0Ha
11 Withycombe-Keeler L, Halden R, Selin C. The future of wastewater sensing workshop guide. Nov 2–3, 2015.
12 Ahmed W Tscharke B Bertsch PM SARS-CoV-2 RNA monitoring in wastewater as a potential early warning system for COVID-19 transmission in the community: a temporal case study Sci Total Environ 761 2021 144216 33360129
13 Medema G Heijnen L Elsinga G Italiaander R Brouwer A Presence of SARS-coronavirus2 RNA in sewage and correlation with reported COVID-19 prevalence in the early stage of the epidemic in the Netherlands Environ Sci Technol Lett 7 2020 511 516
14 Holland LA Kaelin EA Maqsood R An 81-nucleotide deletion in SARS-CoV-2 ORF7a identified from sentinel surveillance in Arizona (January to March 2020) J Virol 94 2020 e00711 e00720 32357959
15 Fontenele RS Kraberger S Hadfield J High-throughput sequencing of SARS-CoV-2 in wastewater provides insights into circulating variants Water Res 205 2021 117710 34607084
16 Besselsen DG Wagner AM Loganbill JK Detection of rodent coronaviruses by use of fluorogenic reverse transcriptase-polymerase chain reaction analysis Comp Med 52 2002 111 116 12022389
17 Ahmed W Bertsch PM Bivins A Comparison of virus concentration methods for the RT-qPCR-based recovery of murine hepatitis virus, a surrogate for SARS-CoV-2 from untreated wastewater Sci Total Environ 739 2020 139960 32758945
18 Arizona State University ASU's COVID-19 management strategy & data update https://eoss.asu.edu/health/announcements/coronavirus/management 2020
19 Maricopa County Gov Maricopa county COVID-19 data https://www.maricopa.gov/5786/COVID-19-Data 2020
20 Tempe County Gov COVID-19 dashboard https://covid19.tempe.gov 2020
21 United States Census Beurau QuickFacts Guadalupe town. Arizona https://www.census.gov/quickfacts/guadalupetownarizona 2021
22 Nemudryi A Nemudraia A Wiegand T Temporal detection and phylogenetic assessment of SARS-CoV-2 in municipal wastewater Cell Rep Med 1 2020 100098 32904687
23 Centazzo N Frederick BM Jacox A Cheng SY Concheiro-Guisan M Wastewater analysis for nicotine, cocaine, amphetamines, opioids and cannabis in New York City Forensic Sci Res 4 2019 152 167 31304444
24 Gibas C Lambirth K Mittal N Implementing building-level SARS-CoV-2 wastewater surveillance on a university campus Sci Total Environ 782 2021 146749 33838367
25 Fox MD Bailey DC Seamon MD Miranda ML Response to a COVID-19 outbreak on a university campus - Indiana, August 2020 MMWR Morb Mortal Wkly Rep 70 2021 118 122 33507894
26 Xiao A Wu F Bushman M Metrics to relate COVID-19 wastewater data to clinical testing dynamics Water Res 212 2022 118070 35101695
27 Office of the Governor Doug Ducey Primer: Arizona continues to ramp up testing https://azgovernor.gov/governor/news/2020/10/primer-arizona-continues-ramp-testing 2020
28 Olesen SW Imakaev M Duvallet C Making waves: defining the lead time of wastewater-based epidemiology for COVID-19 Water Res 202 2021 117433 34304074
29 Safford HR Shapiro K Bischel HN Wastewater analysis can be a powerful public health tool—if it's done sensibly PNAS 119 2022 1
30 Hart OE Halden RU Computational analysis of SARS-CoV-2/COVID-19 surveillance by wastewater-based epidemiology locally and globally: feasibility, economy, opportunities and challenges Sci Total Environ 730 2020 138875 32371231
| 36493788 | PMC9725778 | NO-CC CODE | 2022-12-08 23:18:16 | no | Lancet Microbe. 2022 Dec 6; doi: 10.1016/S2666-5247(22)00289-0 | utf-8 | Lancet Microbe | 2,022 | 10.1016/S2666-5247(22)00289-0 | oa_other |
==== Front
J Am Acad Dermatol
J Am Acad Dermatol
Journal of the American Academy of Dermatology
0190-9622
1097-6787
Published by Elsevier on behalf of the American Academy of Dermatology, Inc.
S0190-9622(21)02991-1
10.1016/j.jaad.2021.12.025
Article
Asymmetric patterns of patient access to in-person and teledermatologic healthcare during the COVID-19 pandemic
Vasavda Chirag BS ab
Tang Olive BA bc
Kwatra Shawn G. MD b
Ho Byron K. MD b†
Grossberg Anna L. MD b∗†
a The Solomon H. Snyder Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD, USA
b Department of Dermatology, Johns Hopkins University School of Medicine, Baltimore, MD, USA
c Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
∗ Primary correspondence to: Anna L. Grossberg, MD, 200 N Wolfe Street Rubenstein 2107 Department of Dermatology Johns Hopkins University School of Medicine Baltimore, MD 21287 USA phone: (410) 955-2049 eml:
† co-corresponding authors
18 12 2021
18 12 2021
16 6 2021
7 12 2021
9 12 2021
© 2021 Published by Elsevier on behalf of the American Academy of Dermatology, Inc.
2021
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
KEYWORDS
healthcare disparities
healthcare access
remote care
health policy
teledermatology
telemedicine
COVID-19
==== Body
pmcFUNDING/SUPPORT
This work was supported by The Maryland Dermatologic Society MDS127961 (to A.L.G.) and NIH T32 GM136577 (to C.V.) and NIH F30 DK120160 (to O.T.).
CONFLICTS OF INTEREST AND FINANCIAL DISCLOSURES
Dr. Kwatra is an advisory board member for Pfizer, Regeneron Pharmaceuticals, Galderma, and Menlo Therapeutics. The remaining authors state no conflict of interest.
IRB APPROVAL
This study was conducted in accordance with the guidelines and regulations as approved by the Johns Hopkins School of Medicine Institutional Review Board (IRB00257728) with a waiver of informed consent.
| 34933041 | PMC9726204 | NO-CC CODE | 2022-12-13 23:17:38 | no | J Am Acad Dermatol. 2023 Jan 18; 88(1):167-169 | utf-8 | J Am Acad Dermatol | 2,021 | 10.1016/j.jaad.2021.12.025 | oa_other |
==== Front
World Neurosurg
World Neurosurg
World Neurosurgery
1878-8750
1878-8769
Elsevier Inc.
S1878-8750(22)01684-9
10.1016/j.wneu.2022.11.132
Original Article
The Effect of COVID-19 Vaccines on Stroke Outcomes: A Single-Center Study
El Naamani Kareem 1
Amllay Abdelaziz 1
Chen Ching-Jen 2
Capone Stephen 3
Abbas Rawad 1
Sioutas Georgios S. 1
Munoz Alfredo 1
Yudkoff Clifford J. 1
Carreras Angeleah 1
Sambangi Abhijeet 1
Hunt Adam 1
Jain Paarth 1
Stine Emily A. 4
Sathe Anish 1
Smit Rupert 1
Yazbeck Fouad 5
Tjoumakaris Stavropoula I. 1
Gooch Michael R. 1
Herial Nabeel A. 1
Rosenwasser Robert H. 1
Zarzour Hekmat 1
Schmidt Richard F. 1
El-Ghanem Mohammad 6
Jabbour Pascal M. 1∗
1 Department of Neurological Surgery, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA
2 Department of Neurosurgery, The University of Texas Health Science Center, Houston, Texas, USA
3 Department of Neurology, Virginia Tech Carilion Clinic, Roanoke, Virginia, USA
4 Psychology Department, College of Arts and Sciences, Arcadia University, Glenside, Pennsylvania, USA
5 University of Washington, Seattle, Washington, USA
6 Department of Neurology, University of Arizona, Tuscon, Arizona, USA
∗ To whom correspondence should be addressed: Pascal M. Jabbour, M.D.
7 12 2022
7 12 2022
17 10 2022
29 11 2022
© 2022 Elsevier Inc. All rights reserved.
2022
Elsevier Inc.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
One of the defining narratives of the COVID-19 pandemic has been the acceptance and distribution of vaccine. To compare the outcomes of COVID-19 positive vaccinated and unvaccinated stroke patients.
Methods
This is a single-center retrospective study of COVID-19-vaccinated and unvaccinated stroke patients between April 2020 and March 2022. All patients presenting with stroke regardless of treatment modalities were included. National Institutes of Health Stroke Scale was used to assess stroke severity. The primary outcome was functional capacity of the patients at discharge.
Results
The study cohort comprised 203 COVID-19 positive stroke patients divided into 139 unvaccinated and 64 fully vaccinated patients. At discharge, the modified Rankin scale score was significantly lower in the vaccinated cohort (3[1–4] vs. 4[2–5], odds ratio = 0.508, P = 0.011). At 3 months of follow-up, the median modified Rankin scale score was comparable between both cohorts.
Conclusions
Although vaccination did not show any significant difference in stroke patient outcomes on follow-up, vaccines were associated with lower rates of morbidity and mortality at discharge among stroke patients during the pandemic.
Key words
COVID-19
Stroke
Vaccination
Abbreviations and Acronyms
ANCOVA, Analysis of Covariance
ASPECTS, Alberta Stroke Program Early Computed Tomography Score
COVID-19, Coronavirus Disease 2019
LVO, Large Vessel Occlusion
mRS, modified Rankin Scale
NIHSS, National Institute of Health Stroke scale
SARS-CoV-2, Severe Acute Respiratory Syndrome Coronavirus 2
tPA, tissue plasminogen activator
==== Body
pmcIntroduction
After COVID-19 was declared a public health emergency by the World Health Organization, it was not until 9 months later that the first American outside of a clinical trial received the first dose of the vaccine.1 , 2 In our prior study and associated literature it has been established that COVID-19 is an independent factor of unfavorable outcomes in stroke patients.3 , 4 Despite finding an association between COVID-19 and stroke severity, information on how COVID-19 vaccination could have any further impact on patients with strokes remains scarce. In this study we compare the outcomes of vaccinated and unvaccinated COVID-19 positive-stroke patients to provide further evidence about the efficacy of vaccines in decreasing stroke severity and improving stroke outcomes.
Materials and Methods
Patient Population
This was a single-center retrospective study of COVID-19-positive vaccinated and unvaccinated stroke patients between April 2020 and March 2022. The institutional review board of participating institutions reviewed and approved the study, and patient consent was waived.
Diagnosis of COVID-19 was established using reverse-transcriptase–polymerase-chain-reaction assays of nasopharyngeal samples for identification of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). COVID-19 vaccines included the following: 1) Comirnaty (BNT 162b2 mRNA vaccine) by Pfizer BionTech, 2) Ad26.COV2. S adenovirus vaccine by Johnson & Johnson/Janssen, and 3) Spikevax (mRNA-1273vaccine) by Moderna. The inclusion criteria was all COVID-19-positive patients presenting with stroke regardless of treatment modalities during the study timeline.
Charts were reviewed and collected data for the pandemic cohort included the following: 1) baseline characteristics including age, gender, and race; 2) past medical and social history; 3) Vaccination status including unvaccinated, and fully vaccinated defined as at least 2 doses of vaccination (2 doses, 3 doses, or 4 doses) (Moderna or Pfizer) or 1 dose for Janssen; 4) stroke characteristics included the Alberta stroke program early computed tomography score, National Institutes of Health Stroke Scale score (NIHSS), and number of tandem occlusions; 5) treatment characteristics included tissue plasminogen activator administration and the need for mechanical thrombectomy; and 6) Functional outcomes included modified Rankin scale (mRS) at discharge, and at 3 months of follow-up and mortality. For the prepandemic cohort, the NIHSS score on admission was collected.
Primary and Secondary Outcomes
The primary outcome was the functional capacity of the patients at discharge based on ordinal mRS shift analysis. Secondary outcomes included NIHSS at 24 hours postprocedure, NIHSS at discharge, mortality at discharge, mRS up to 3 months of follow-up, mortality up to 3 months, NIHSS up to 3 months, the need for decompressive hemicraniectomy, and the length of hospital stay. Good functional outcome was defined by mRS 0–2, and poor functional outcome was defined by mRS 3–6.
Statistical Analysis
All statistical analyses were performed using the Stata software (version 17.0, College Station, Texas, USA). The remaining methods section is in the supplementary material folder. Following the pandemic, the cohort was divided into unvaccinated and vaccinated groups. Between-group baseline characteristics were compared using χ2 or Fisher's exact tests for categorical variables, and Student's t or Mann-Whitney U tests for continuous variables, where appropriate. Associations between outcomes and vaccination groups were assessed using ordinal logistic, binary logistic, and linear regression, wherever appropriate. Odds ratios (ORs) or betas and corresponding 95% confidence intervals were reported. Fisher's exact test was performed for outcomes with zero frequency cells. These relationships were then adjusted in multivariable models including baseline variables with P values < 0.100 in univariable comparisons. All tests were two-tailed, and P value of < 0.050 was considered statistically significant. Missing data were not imputed.
Results
The study cohort consisted of 203 COVID-19 positive stroke patients divided into 139 unvaccinated patients and 64 fully vaccinated patients.
Comparison of Baseline Characteristics Between Fully Vaccinated and Unvaccinated COVID-19 Positive Patients Presenting with Stroke
With respect to patient demographics, the mean age, gender, and race distributions were comparable between both cohorts. With respect to past medical history, the number of patients with atrial fibrillation (34.4% vs.19.4%; P = 0.021), and peripheral vascular disease (17.2% vs.7.2%; P = 0.03) was significantly higher in the vaccinated cohort compared to the unvaccinated cohort. On admission, the number of patients on antiplatelet and anticoagulation medications was comparable between both cohorts. Compared to fully vaccinated patients, unvaccinated patients had a significantly higher admission NIHSS (median:10[3–19] vs. 4[1.5–9]; P < 0.001). Lastly, the number of patients requiring mechanical thrombectomy was significantly higher in the unvaccinated cohort (27.3% vs.12.5%; P = 0.019) (Table 1 ).Table 1 Comparison of Baseline Characteristics Between COVID-19-Positive Unvaccinated Versus Fully Vaccinated Patients Presenting with Stroke
Unvaccinated (n = 139) Fully Vaccinated (n = 64) P-Value
Age, mean (SD) years 65.8 (15.2) 67.7 (13.9) 0.391
Male, n (%) 75/139 (54.0) 33/64 (51.6) 0.751
Race, n (%) 0.805
Caucasian 80/138 (58.0) 40/64 (62.5)
African America 41/138 (29.7) 17/64 (26.6)
Asian 11/138 (8.0) 3/64 (4.7)
Hispanic 5/138 (3.6) 3/64 (4.7)
Other 1/138 (0.7) 1/64 (1.6)
Hypertension, n (%) 104/139 (74.8) 55/64 (85.9) 0.074
Diabetes mellitus, n (%) 60/139 (43.2) 24/64 (37.5) 0.446
CHF, n (%) 15/139 (10.8) 10/64 (15.6) 0.330
Cancer, n (%) 16/139 (11.5) 9/64 (14.1) 0.607
Atrial fibrillation, n (%) 27/139 (19.4) 22/64 (34.4) 0.021
Prior stroke, n (%) 33/139 (23.7) 20/63 (31.8) 0.231
PVD, n (%) 10/139 (7.2) 11/64 (17.2) 0.030
Immunosuppression, n (%) 8/139 (5.8) 5/64 (7.8) 0.553
Chronic lung disease, n (%) 21/139 (15.1) 8/64 (12.5) 0.622
Baseline mRS, median (IQR) 0 (0–1) 0 (0–2) 0.107
Antiplatelet use, n (%) 35/138 (25.4) 21/64 (32.8) 0.271
Anticoagulant use, n (%) 21/138 (15.2) 13/64 (20.3) 0.368
Smoking status, n (%) 0.827
None 93/137 (67.9) 42/64 (65.6)
Current 21/137 (15.3) 12/64 (18.8)
Former 23/137 (16.8) 10/64 (15.6)
NIHSS, median (IQR) 10 (3–19) 4 (1.5–9) <0.001
ASPECTS, median (IQR) 10 (9–10) 10 (10–10) 0.022
Tandem occlusion, n (%) 12/104 (11.5) 1/49 (2.0) 0.062
IV tPA, n (%) 28/136 (20.6) 10/61 (16.4) 0.490
Thrombectomy, (n%) 38/139 (27.3) 8/64 (12.5) 0.019
SD, standard deviation; CHF, congestive heart failure; PVD, Peripheral vascular disease; mRS, modified Rankin scale; IQR, Interquartile range; NIHSS, National Institutes of Health Stroke Scale score; ASPECTS, Alberta stroke program early computed tomography score; IV, intravenous; tPA, tissue plasminogen activator.
Comparison of Outcomes Between Fully Vaccinated and Unvaccinated COVID-19 Positive Patients Presenting with Strokes
At discharge, the mRS score was significantly lower in the vaccinated cohort compared to the unvaccinated cohort (3[1–4] vs. 4[2–5], OR = 0.508[0.301–0.859]; P = 0.011) before adjustment but was not significant after adjustment (OR = 0.490[0.211–1.139]; P = 0.098). This was further dissected in the secondary outcomes which showed a significantly higher rate of mRS (0–1) at discharge in the vaccinated cohort (37.5% vs. 23.9%, OR = 1.913[1.005–3.639]; P = 0.048), which did not remain significant after adjustment (OR = 1.733[0.583–5.149]; P = 0.322), and a significantly higher rate of mRS (0–2) in the vaccinated cohort before adjustment (45.3% vs. 30.6% OR = 1.879[1.017–3.473]; P = 0.044) but not after adjustment (OR = 2.278[0.757–6.858]; P = 0.143). Also, median NIHSS score was significantly higher in the unvaccinated cohort compared to the vaccinated cohort at 24 hours (6[1–19] vs. 3[1–8]; P = 0.008) and at discharge (6 [1–21] vs. 2[0–6]; P = 0.003) before adjustment, but did not remain significant after adjustment (P = 0.122) (P = 0.149), respectively. Interestingly, the mortality rate at discharge was significantly higher in the unvaccinated cohort (20.2% vs. 7.8%, OR = 0.336[0.123–0.918]; P = 0.033) before adjustment but did not remain significant after adjustment (OR = 0.556(0.069–4.479); P = 0.581). At 3 months of follow-up, the median mRS, median NIHSS, and mortality rates were comparable between both cohorts. Though not significant, the rate of decompressive hemicraniectomy was 4-times higher in the unvaccinated cohort (4.4% vs. 0.9%; P = 0.180). Lastly, the median length of stay was comparable between both cohorts (6[3–14] vs. 5[3–10]; P = 0.009) before adjustment; however, this did not remain significant after adjustment (P = 0.059) (Table 2 ).Table 2 Comparison of Outcomes Between COVID-19-Positive Unvaccinated Versus Fully Vaccinated Patients Presenting with Stroke
Unvaccinated Vaccinated Effect Variable Unadjusted Value (95% CI) Unadjusted P-value Adjusted Value (95% CI)∗ Adjusted P-Value∗
Primary Outcome
mRS at discharge, median (IQR) 4 (2–5) 3 (1–4) Common Odds Ratio 0.508 (0.301–0.859) 0.011 0.490 (0.211–1.139) 0.098
Secondary Outcomes
mRS 0–1 at discharge, n (%) 32/134 (23.9) 24/64 (37.5) Odds Ratio 1.913 (1.005–3.639) 0.048 1.733 (0.583–5.149) 0.322
mRS 0–2 at discharge, n (%) 41/134 (30.6) 29/64 (45.3) Odds Ratio 1.879 (1.017–3.473) 0.044 2.278 (0.757–6.858) 0.143
NIHSS at 24 hours, median (IQR) 6 (1–19) 3 (1–8) Beta −4.062 (−7.072–−1.053) 0.008 −1.919 (−4.362–0.524) 0.122
NIHSS at discharge, median (IQR) 6 (1–21) 2 (0–6) Beta −6.882 (−11.363–−2.401) 0.003 −3.589 (−8.490–1.312) 0.149
Mortality at discharge, n (%) 27/134 (20.2) 5/64 (7.8) Odds Ratio 0.336 (0.123–0.918) 0.033 0.556 (0.069–4.479) 0.581
mRS at 3 months, median (IQR) 6 (3–6) 6 (1–6) Common Odds Ratio 0.872 (0.251–3.035) 0.830 0.882 (0.110–7.085) 0.906
mRS 0–1 at 3 months, n (%) 10/55 (18.2) 3/11 (27.3) Odds Ratio 1.688 (0.379–7.513) 0.492 2.365 (0.167–33.495) 0.524
mRS 0–2 at 3 months, n (%) 12/55 (21.8) 4/11 (36.4) Odds Ratio 2.048 (0.512–8.181) 0.311 3.177 (0.246–40.962) 0.376
Mortality at 3 months, n (%) 28/55 (50.9) 6/11 (54.6) Odds Ratio 1.157 (0.316–4.243) 0.826 3.646 (0.184–72.275) 0.396
NIHSS at 3 months, median (IQR) 42 (5–42) 42 (0–42) Beta −1.756 (−15.762–12.251) 0.802 4.187 (−17.203–25.576) 0.690
Decompressive craniectomy, n (%) 6/136 (4.4) 0/62 (0) –– –– 0.180 –– ––
Length of hospital stay, median (IQR) days 6 (3–14) 5 (3–10) Beta −3.761 (−6.577–−0.944) 0.009 −4.364 (−8.895–0.168) 0.059
CI, confidence interval; mRS, modified Rankin scale; IQR, Interquartile range; NIHSS, National Institutes of Health Stroke Scale score.
∗ Adjusted for hypertension, atrial fibrillation, PVD, admission NIHSS, ASPECTS, tandem occlusion, and thrombectomy.
Discussion
Vaccines against SARS-CoV-2 have been approved and used worldwide with unprecedented speed and it has been established that these vaccines provided critical protection against SARS-CoV-2.5, 6, 7 While most studies assessed the overall efficacy and safety vaccination, reports on the effect of vaccination on stroke severity specifically remain scarce.7, 8, 9, 10 This study demonstrated that in general, the severity of strokes and large vessel occlusions (LVOs) has significantly increased after the pandemic in both COVID-19 positive and negative patients compared to the prepandemic. This corroborates what has been established in prior studies that COVID-19 is an independent predictor of poor functional outcome and mortality in stroke patients.3 , 4 , 11 , 12 As for prognosis, compared to fully vaccinated patients, unvaccinated patients with strokes not only had significantly worse functional outcomes at discharge compared to vaccinated patients (mRS 4 [2–5] vs. 3 [1–4], OR = 0.508 [0.301–0.859]; P = 0.011 and NIHSS at discharge 6 [1–21] vs. 2 [0–6]; P = 0.003), but also mortality rate was 3 times higher than vaccinated patients (20.2% vs. 7.8%, OR = 0.336 [0.123–0.918], P = 0.033). Although functional outcome at 3 months of follow-up was comparable between cohorts, this may be due to several reasons. First, because mortality rate in the unvaccinated cohort was significantly higher compared to the vaccinated cohort, this means that fewer patients from the unvaccinated cohort with severe symptoms survived, which may have underestimated the rate of unfavorable outcomes at 3 months for the unvaccinated cohort. Another cause may be the loss of patients to follow-up, which could be attributed to the high load of patients where priority was given to more severe cases. Lastly, because COVID-19 patients receive blood thinners as a part of their thromboprophylaxis, this may have protected them during the follow-up period from any new strokes or reocclusions that may have affected their functional outcome.13
Several factors play a role in the degree of recovery following a stroke, some of which are inherent to patients and others to stroke characteristics. Although adjusted for, our study demonstrated that unvaccinated patients had worse stroke characteristics on presentation compared to vaccinated patients in terms of tandem occlusion rates (11.5% vs. 2.0%; P = 0.062) which was comparable to the literature.3 , 11 With respect to stroke care, the unvaccinated cohort underwent more mechanical thrombectomies which may be due to the higher rate of LVOs.
By catalyzing the conversion of angiotensin I and II to angiotensin (1–7), Angiotensin-converting enzyme 2 (ACE-2) mitigates the pro-inflammatory and pro-thrombotic effects of angiotensin II.14 Because COVID-19 possess high tropism to ACE-2, its attachment via the spike protein to ACE-2 dysregulates the latter's effect on angiotensin II triggering inflammation, vasoconstriction, and a pro-coagulant state.15 As a result, this pro-thrombotic milieu confers more stroke, LVOs, tandem occlusions, and multiple vessel occlusions.16 At a macro level, the complexity of strokes due to the high clot burden and consistency requiring longer procedure times and more number of passes, in addition to the higher frequency and incidence of strokes during the pandemic ultimately took a toll on health care systems.17 , 18 To date, the Food and Drug Administration has approved the following 3 vaccines: 1) Comirnaty (BNT 162b2 mRNA vaccine) by Pfizer BionTech, 2) Ad26.COV2.S adenovirus vaccine by Johnson & Johnson/Janssen,3 and Spikevax (mRNA-1273 vaccine) by Moderna.19 All the approved vaccines are based on the full-length homotrimeric SARS-CoV-2 spike protein which plays a key role in the viral attachment to the ACE-2 receptor.20 , 21 The messenger ribonucleic acid (mRNA) based vaccines (Pfizer and Moderna) consist of a lipid-enclosed nucleoside-modified mRNA encoding a different mutated spike protein, while the DNA-based vaccine (Janssen) consists of both a chimpanzee nonreplicating adenovirus and a type 26 nonreplicating recombinant adenovirus vector.19 , 22 By creating a replica of COVID-19's spike protein, either by DNA or mRNA technology, vaccines promote the production of antibodies against COVID-19 and provide a strong immune reaction to destabilize the virus when an infection takes place.22 Thus, at a micro level, vaccines decrease stroke severity by minimizing the pro-thrombotic and pro-inflammatory milieu of COVID-19 through mounting an immune response that limits COVID-19's effect on the ACE-2 receptor. Effectiveness and safety of vaccines in targeting the virus, diminishing it's spread, and minimizing its side effects have been demonstrated in several studies.19 , 23, 24, 25, 26, 27
Limitations
Conclusions of our work were limited by the retrospective nature of this single center study and the absence of randomization. Also, the loss of patients to follow-up may be a limiting factor of our mid-term results regarding functional outcome and mortality. This may have been due to either the high patient load health care systems that were faced with during the pandemic where priority was given to more severe cases or the retrospective nature of the study design. Furthermore, the small number of vaccinated patients limited the ability to compare the dose-dependent influence of vaccines on stroke outcome. Lastly, significant differences in baseline characteristics between both cohorts may have affected our outcomes; however, this confounding bias was controlled by adjustment.
Conclusions
The risk of stroke complicating infections with COVID-19 has been widely reported since the beginning of the pandemic. Our study showed that compared to the prepandemic era, stroke severity measured by the NIHSS score increased during the pandemic. Although vaccination did not show any significant difference in stroke patient outcomes on follow-up, vaccines were associated with lower rates of morbidity and mortality at discharge among COVID-positive stroke patients during the pandemic.
CRediT authorship contribution statement
Kareem El Naamani: Conceptualization, Methodology, Acquisition of data, Formal analysis, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Abdelaziz Amllay: Conceptualization, Methodology, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Ching-Jen Chen: Conceptualization, Methodology, Formal analysis, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Stephen Capone: Acquisition of data, Agree to be accountable for all aspects of the work. Rawad Abbas: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Georgios S. Sioutas: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Alfredo Munoz: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Clifford J. Yudkoff: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Angeleah Carreras: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Abhijeet Sambangi: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Adam Hunt: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Paarth Jain: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Emily A. Stine: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Anish Sathe: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Rupert Smit: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Fouad Yazbeck: Acquisition of data, Interpretation of data, Writing – original draft, Agree to be accountable for all aspects of the work. Stavropoula I. Tjoumakaris: Writing – review & editing, Agree to be accountable for all aspects of the work. Michael R. Gooch: Writing – review & editing, Agree to be accountable for all aspects of the work. Nabeel A. Herial: Writing – review & editing, Agree to be accountable for all aspects of the work. Robert H. Rosenwasser: Writing – review & editing, Agree to be accountable for all aspects of the work. Hekmat Zarzour: Writing – review & editing, Agree to be accountable for all aspects of the work. Richard F. Schmidt: Writing – review & editing, Agree to be accountable for all aspects of the work. Mohammad El-Ghanem: Writing – review & editing, Agree to be accountable for all aspects of the work. Pascal M. Jabbour: Conceptualization, Methodology, Writing – review & editing, Agree to be accountable for all aspects of the work, Final approval of the version.
Conflict of interest statement: Dr. Jabbour is a consultant for Medtronic, MicroVention, Balt and Cerus Endovascular. Dr. Tjoumakaris is a consultant for Medtronic and MicroVention. Dr. Gooch is consultant for Stryker. The remaining authors have no conflicts to report.
==== Refs
References
1 Webster P. COVID-19 timeline of events Nat Med 27 2021 2054 2055
2 Carvalho T. Krammer F. Iwasaki A. The first 12 months of COVID-19: a timeline of immunological insights Nat Rev Immunol 21 2021 245 256
3 Jabbour P. Dmytriw A.A. Sweid A. Characteristics of a COVID-19 cohort with large vessel occlusion: a multicenter international study Neurosurgery 90 2022 725 733
4 Ntaios G. Michel P. Georgiopoulos G. Characteristics and outcomes in patients with COVID-19 and acute ischemic stroke: the Global COVID-19 stroke Registry Stroke 51 2020 e254 e258
5 Hodgson S.H. Mansatta K. Mallett G. Harris V. Emary K.R.W. Pollard A.J. What defines an efficacious COVID-19 vaccine? A review of the challenges assessing the clinical efficacy of vaccines against SARS-CoV-2 Lancet Infect Dis 21 2021 e26 e35
6 Izda V. Jeffries M.A. Sawalha A.H. COVID-19: a review of therapeutic strategies and vaccine candidates Clin Immunol 222 2021 108634 33217545
7 Voysey M. Clemens S.A.C. Madhi S.A. Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK Lancet 397 2021 99 111 33306989
8 Folegatti P.M. Ewer K.J. Aley P.K. Safety and immunogenicity of the ChAdOx1 nCoV-19 vaccine against SARS-CoV-2: a preliminary report of a phase 1/2, single-blind, randomised controlled trial Lancet 396 2020 467 478 32702298
9 Ramasamy M.N. Minassian A.M. Ewer K.J. Safety and immunogenicity of ChAdOx1 nCoV-19 vaccine administered in a prime-boost regimen in young and old adults (COV002): a single-blind, randomised, controlled, phase 2/3 trial Lancet 396 2021 1979 1993 33220855
10 Bartsch S.M. O'Shea K.J. Ferguson M.C. Vaccine efficacy needed for a COVID-19 coronavirus vaccine to prevent or stop an epidemic as the sole intervention Am J Prev Med 59 2020 493 503 32778354
11 Srivastava P.K. Zhang S. Xian Y. Acute ischemic stroke in patients with COVID-19: an analysis from Get with the Guidelines-stroke Stroke 52 2021 1826 1829 33728926
12 Kihira S. Schefflein J. Mahmoudi K. Association of coronavirus disease (COVID-19) with large vessel occlusion strokes: a case-control study AJR Am J Roentgenol 216 2021 150 156 32755225
13 Gerotziafas G.T. Catalano M. Colgan M.P. Guidance for the management of patients with vascular disease or cardiovascular risk factors and COVID-19: position paper from VAS-European independent foundation in angiology/vascular medicine Thromb Haemost 120 2020 1597 1628 32920811
14 Labò N. Ohnuki H. Tosato G. Vasculopathy and coagulopathy associated with SARS-CoV-2 infection Cells 9 2020 1583 32629875
15 Mecca A.P. Regenhardt R.W. O'Connor T.E. Cerebroprotection by angiotensin-(1-7) in endothelin-1-induced ischaemic stroke Exp Physiol 96 2011 1084 1096 21685445
16 Clinical spectrum of SARS-CoV-2 infection Available at: https://www.covid19treatmentguidelines.nih.gov/overview/clinical-spectrum/ 2021
17 Khandelwal P. Al-Mufti F. Tiwari A. Incidence, characteristics and outcomes of large vessel stroke in COVID-19 cohort: an international multicenter study Neurosurgery 89 2021 E35 E41 33734404
18 Altersberger V.L. Stolze L.J. Heldner M.R. Maintenance of acute stroke care service during the COVID-19 pandemic lockdown Stroke 52 2021 1693 1701 33793320
19 De Michele M. Kahan J. Berto I. Cerebrovascular complications of COVID-19 and COVID-19 vaccination Circ Res 130 2022 1187 1203 35420916
20 Yang Y. Du L. SARS-CoV-2 spike protein: a key target for eliciting persistent neutralizing antibodies Signal Transduct Target Ther 6 2021 95 33637679
21 Lan J. Ge J. Yu J. Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor Nature 581 2020 215 220 32225176
22 Abbasi J. COVID-19 and mRNA vaccines-first large test for a new approach JAMA 324 2020 1125 1127 32880613
23 Heath P.T. Galiza E.P. Baxter D.N. Safety and efficacy of NVX-CoV2373 covid-19 vaccine N Engl J Med 385 2021 1172 1183 34192426
24 Xing K. Tu X.Y. Liu M. Efficacy and safety of COVID-19 vaccines: a systematic review Zhongguo Dang Dai Er Ke Za Zhi 23 2021 221 228 33691913
25 Cai C. Peng Y. Shen E. A comprehensive analysis of the efficacy and safety of COVID-19 vaccines Mol Ther 29 2021 2794 2805 34365034
26 Fathizadeh H. Afshar S. Masoudi M.R. SARS-CoV-2 (Covid-19) vaccines structure, mechanisms and effectiveness: a review Int J Biol Macromol 188 2021 740 750 34403674
27 Vitiello A. Ferrara F. Troiano V. La Porta R. COVID-19 vaccines and decreased transmission of SARS-CoV-2 Inflammopharmacology 29 2021 1357 1360 34279767
| 36494068 | PMC9726206 | NO-CC CODE | 2022-12-16 23:16:03 | no | World Neurosurg. 2022 Dec 7; doi: 10.1016/j.wneu.2022.11.132 | utf-8 | World Neurosurg | 2,022 | 10.1016/j.wneu.2022.11.132 | oa_other |
==== Front
Sci Total Environ
Sci Total Environ
The Science of the Total Environment
0048-9697
1879-1026
Elsevier B.V.
S0048-9697(22)07803-2
10.1016/j.scitotenv.2022.160700
160700
Article
Polyaniline-based electrochemical immunosensor for the determination of antibodies against SARS-CoV-2 spike protein
Drobysh Maryia ab
Ramanavicius Arunas ab⁎
Baradoke Ausra a
a State Research Institute Center for Physical and Technological Sciences, Sauletekio ave. 3, Vilnius, Lithuania
b NanoTechnas—Center of Nanotechnology and Materials Science, Faculty of Chemistry and Geosciences, Vilnius University, Naugarduko str. 24, 03225 Vilnius, Lithuania
⁎ Corresponding author at: State Research Institute Center for Physical and Technological Sciences, Sauletekio ave. 3, Vilnius, Lithuania.
7 12 2022
7 12 2022
16070020 10 2022
30 11 2022
1 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
In this work, we report an impedimetric system for the detection of antibodies against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Spike protein. The sensing platform is based on recombinant Spike protein (SCoV2-rS) immobilized on the phytic acid doped polyaniline films (PANI-PA). The affinity interaction between immobilized SCoV2-rS protein and antibodies in the physiological range of concentrations was registered by electrochemical impedance spectroscopy. Analytical parameters of the sensing platform were tuned by the variation of electropolymerization times during the synthesis of PANI-PA films. The lowest limit of detection and quantitation were obtained for electropolymerization time of 20 min and equalled 8.00 ± 0.20 nM and 23.93 ± 0.60 nM with an equilibrium dissociation constant of 3 nM. The presented sensing system is label-free and suitable for the direct detection of antibodies against SARS-CoV-2 in real patient serum samples after coronavirus disease 2019 and/or vaccination.
Graphical abstract
Unlabelled Image
Keywords
SARS-CoV-2 Spike protein
Electrochemical impedance spectroscopy (EIS)
Immunosensor
Screen-printed electrodes
Phytic acid
Polyaniline
Editor: Damià Barceló
==== Body
pmcData availability
Data will be made available on request.
| 36493838 | PMC9726207 | NO-CC CODE | 2022-12-12 23:20:29 | no | Sci Total Environ. 2023 Mar 1; 862:160700 | utf-8 | Sci Total Environ | 2,022 | 10.1016/j.scitotenv.2022.160700 | oa_other |
==== Front
Sci Total Environ
Sci Total Environ
The Science of the Total Environment
0048-9697
1879-1026
Elsevier B.V.
S0048-9697(22)07870-6
10.1016/j.scitotenv.2022.160767
160767
Article
Radiative effects and feedbacks of anthropogenic aerosols on boundary layer meteorology and fine particulate matter during the COVID-19 lockdown over China
Liang Mingjie ab
Han Zhiwei ab⁎
Li Jiawei a
Sun Yele c
Liang Lin ab
Li Yue ab
a Key Laboratory of Regional Climate-Environment for Temperate East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China
b University of Chinese Academy of Sciences, Beijing 100049, China
c State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China
⁎ Corresponding author at: Key Laboratory of Regional Climate-Environment for Temperate East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China.
7 12 2022
7 12 2022
1607671 10 2022
19 11 2022
4 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The COVID-19 epidemic has exerted significant impacts on human health, social and economic activities, air quality and atmospheric chemistry, and potentially on climate change. In this study, an online coupled regional climate–chemistry–aerosol model (RIEMS-Chem) was applied to explore the direct, indirect, and feedback effects of anthropogenic aerosols on radiation, boundary layer meteorology, and fine particulate matter during the COVID-19 lockdown period from 23 January to 8 April 2020 over China. Model performance was validated against a variety of observations for meteorological variables, PM2.5 and its chemical components, aerosol optical properties, as well as shortwave radiation flux, which demonstrated that RIEMS-Chem was able to reproduce the spatial distribution and temporal variation of the above variables reasonably well. During the study period, direct radiative effect (DRE) of anthropogenic aerosols was stronger than indirect radiative effect (IRE) in most regions north of the Yangtze River, whereas IRE dominated over DRE in the Yangtze River regions and South China. In North China, DRE induced larger changes in meteorology and PM2.5 than those induced by IRE, whereas in South China, the changes by IRE were remarkably larger than those by DRE. Emission reduction alone during the COVID-19 lockdown reduced PM2.5 concentration by approximately 32 % on average over East China. As a result, DRE at the surface was weakened by 15 %, whereas IRE changed little over East China, leading to a decrease in total radiative effect (TRE) by approximately 7 % in terms of domain average. The DRE-induced changes in meteorology and PM2.5 were weakened due to emission reduction, whereas the IRE-induced changes were almost the same between the cases with and without emission reductions. By aerosol radiative and feedback effects, the COVID-19 emission reductions resulted in 0.06 °C and 0.04 °C surface warming, 1.6 and 4.0 μg m−3 PM2.5 decrease, 0.4 and 1.3 mm precipitation increase during the lockdown period in 2020 in terms of domain average over North China and South China, respectively, whereas the lockdown caused negligible changes on average over East Asia.
Graphical abstract
The model simulated changes in (a) total radiative effect of anthropogenic aerosols at the surface (W m−2), and changes in the radiative feedback-induced (b) surface air temperature decrease (°C), (c) PBLH decrease (m), (d) surface PM2.5 concentration increase (μg m−3) due to anthropogenic emission reductions during the COVID-19 lockdown from 23 January to 8 April 2020. The changes are derived from the BASE case minus EXP case. The numbers in the upper right corner of each panel denote averages over East China during the study period.Unlabelled Image
Keywords
Direct radiative effect
Indirect effect
Feedback
Anthropogenic aerosols
COVID-19
Boundary layer meteorology
Editor: Jianmin Chen
==== Body
pmcData availability
Data will be made available on request.
| 36493835 | PMC9726208 | NO-CC CODE | 2022-12-10 23:15:27 | no | Sci Total Environ. 2023 Mar 1; 862:160767 | utf-8 | Sci Total Environ | 2,022 | 10.1016/j.scitotenv.2022.160767 | oa_other |
==== Front
Thromb Res
Thromb Res
Thrombosis Research
0049-3848
1879-2472
Pergamon Press
S0049-3848(22)00477-7
10.1016/j.thromres.2022.11.028
Full Length Article
A novel interaction between extracellular vimentin and fibrinogen in fibrin formation
Martinez-Vargas Marina abe
Cebula Adrian abe
Brubaker Lisa S. ce
Seshadri Nitin abe
Lam Fong W. de
Loor Michele c
Rosengart Todd K. c
Yee Andrew de
Rumbaut Rolando E. be
Cruz Miguel A. abe⁎
a Section of Cardiovascular Research, Michael E. DeBakey VA Medical Center, Houston, TX 77030, United States of America
b Department of Medicine, Michael E. DeBakey VA Medical Center, Houston, TX 77030, United States of America
c Department of Surgery, Michael E. DeBakey VA Medical Center, Houston, TX 77030, United States of America
d Department of Pediatrics, Baylor College of Medicine, Michael E. DeBakey VA Medical Center, Houston, TX 77030, United States of America
e Center for Translational Research on Inflammatory Diseases (CTRID), Michael E. DeBakey VA Medical Center, Houston, TX 77030, United States of America
⁎ Corresponding author at: Section of Cardiovascular Research, Baylor College of Medicine and MEDVAMC, 2002 Holcombe, Bldg. 109, R-146, Houston, TX 77030, United States of America.
7 12 2022
7 12 2022
8 8 2022
7 11 2022
30 11 2022
Published by Elsevier Ltd.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Thrombosis is frequently manifested in critically ill patients with systemic inflammation, including sepsis and COVID-19. The coagulopathy in systemic inflammation is often associated with increased levels of fibrinogen and D-dimer. Because elevated levels of vimentin have been detected in sepsis, we sought to investigate the relationship between vimentin and the increased fibrin formation potential observed in these patients.
Materials and methods
This hypothesis was examined by using recombinant human vimentin, anti-vimentin antibodies, plasma derived from healthy and critically ill patients, confocal microscopy, co-immunoprecipitation assays, and size exclusion chromatography.
Results
The level of vimentin in plasma derived from critically ill subjects with systemic inflammation was on average two-fold higher than that of healthy volunteers. We determined that vimentin directly interacts with fibrinogen and enhances fibrin formation. Anti-vimentin antibody effectively blocked fibrin formation ex vivo and caused changes in the fibrin structure in plasma. Additionally, confocal imaging demonstrated plasma vimentin enmeshed in the fibrin fibrils. Size exclusion chromatography column and co-immunoprecipitation assays demonstrated a direct interaction between extracellular vimentin and fibrinogen in plasma from critically ill patients but not in healthy plasma.
Conclusions
The results describe that extracellular vimentin engages fibrinogen in fibrin formation. In addition, the data suggest that elevated levels of an apparent aberrant extracellular vimentin potentiate fibrin clot formation in critically ill patients with systemic inflammation; consistent with the notion that plasma vimentin contributes to the pathogenesis of thrombosis.
Keywords
Vimentin
Coagulopathy
Systemic inflammation
Fibrinogen
sepsis
COVID-19
==== Body
pmc1 Introduction
Thrombosis is frequently seen in critically ill patients with systemic inflammation including sepsis, trauma, burns, and infection with the SARS-CoV-2 virus that causes COVID-19 [1], [2], [3]. Similarities in the prothrombotic and antifibrinolytic state in these patients include an enhanced fibrin formation and widespread fibrin deposition in small to mid-size blood vessels that eventually cause organ ischemia and dysfunction [2], [4], [5]. The high levels of fibrinogen and D-dimer, a fibrin degradation product, reveal active fibrin formation and fibrinolysis [6], [7]. Moreover, autopsy findings from patients who succumbed to sepsis-induced coagulopathy [8] or COVID-19 [9], [10], [11] often reveal disseminated fibrin-rich microthrombi suggesting the association between widespread fibrin deposition and poor outcomes.
We and others have reported that vimentin, a cytoplasmic intermediate filament protein [12], is expressed on the cell surface of different cell types [13], [14], [15], [16] and is also detected in blood [16], [17]. An increasing number of studies have described cell surface vimentin as a receptor for viruses, bacteria and plasma proteins [13], [16], [18], [19], [20], [21], [22]. On the other hand, extracellular vimentin has been reported to be found in blood from healthy subjects and at elevated levels in plasma from patients with clinical conditions such as coronary artery disease [23], rheumatoid arthritis (citrullinated vimentin) [24], cancer [25], [26], and sepsis [27]. However, the function of extracellular plasma vimentin remains elusive.
In an effort to understand the mechanisms of systemic inflammation-associated coagulopathy, we performed a study that, to our knowledge, is the first one to characterize the potential relationship between vimentin and fibrinogen and the impact of this association to coagulation. The objective of our study was to explore the novel role of plasma vimentin in fibrin formation, particularly, in clinical conditions associated with systemic inflammation such as sepsis and COVID-19.
2 Materials and methods
2.1 Plasma from patients and healthy donors
The plasma samples from critically ill patients used in this study were obtained in a previous study focused on fibrin clot structure in critical illnesses [28]. To obtain healthy human blood, informed consent was provided based on the recommendations of the Declaration of Helsinki. Approval was attained from the Baylor College of Medicine institutional review board (IRB) for these studies.
2.2 Reagents
Purified recombinant human vimentin was purchased from SinoBiological (Wayne, PA). Human fibrinogen from Calbiochem (Gibbstown, NJ), and thrombin from Sigma (St. Louis, MO). Sheep anti-Vimentin antibody was obtained from Affinity Biologicals (Ancaster, Canada), and the rabbit anti-Vimentin antibody was purchased from Proteintech (Rosemont, IL). This rabbit anti-Vimentin antibody was validated by testing its reactivity to purified recombinant vimentin and plasma vimentin by ELISA and western blot. Anti-vimentin antibody V9 and isotype immunoglobulins (IgGs) were purchased from Sigma. Recombinant human vimentin rod-domain (rhRod) was expressed and purified as described [29].
2.3 Gel filtration chromatography
The plasma samples were subjected to size exclusion chromatography using a Superose 6™ 10/300 GL column (Cytiva) equilibrated with 25 mM Tris-HCl, 150 mM NaCl, pH 7.4 (TBS) and a constant flow rate of 0.5 ml/min [30]. The collected fractions were analyzed by immunoblotting to verify the presence of vimentin [16].
2.4 Bio-layer interferometry
We used bio-layer interferometry for kinetic binding measurements as previously described [29]. Briefly, we immobilized purified human fibrinogen 50 μg/mL onto amine reactive 2nd generation sensors (AR2G; Pall ForteBio LLC) and the sensors were used to assess binding kinetics to recombinant human vimentin at concentrations ranging from 0 to 2500 nM. We analyzed the data using the Octet Data Acquisition Software version 11.1.
2.5 ELISA to measure plasma vimentin
Sheep anti-vimentin antibody was diluted to 1.0 μg/ml with carbonate buffer (pH -9.6), added into microtiter wells (50 μl/well) and incubated overnight at 4 °C. After washed and blocked with 3 % BSA (150 μl/well) for 1 h at 37 °C, 50 μl of diluted plasma (1:20) or increasing concentrations of purified vimentin (for standard curve) in Tris-buffer saline-0.05 % Tween-20 (TBS-T) were incubated in the coated wells for 1 h at 37 °C. After washed the wells three times with TBS-T, they were incubated with rabbit anti-vimentin antibody (1:1000 in TBS-T- 50 μl/well) at 37 °C for 1 h. The wells were washed four times with TBS-T and incubated with a secondary goat anti-rabbit IgG antibody (1:10,000 in TBS-T) horseradish peroxidase (HRP) conjugate at 37 °C for 1 h.
2.6 Fibrin porosity
As described [28], [31], the fibrin clot was visualized with confocal microscopy and porosity was quantified using Image J software.
2.7 Fibrin polymerization assays
Fibrin formation and degradation were performed as we described [31]. Briefly, 20 % (16 % for fibrinolytic studies) plasma in Tris buffered saline (TBS) (50 mmol/l Tris, 0.15 mol/l NaCl, pH 7.4) was mixed with anti-vimentin antibody or isotype IgG in the presence or absence of tissue plasminogen activator (tPA, 250 ng/ml) (Cathflo Activase). Polymerization and/or fibrinolysis was initiated with the addition of 1.0 U of human thrombin. When using plasma, the enzymatic reaction was evaluated by tracking turbidity using a spectrophotometer set to 350 nm. In another set of experiments of fibrin formation and degradation, we adjusted the volume of plasma to load fibrinogen at a concentration (0.5 mg/ml) that was equivalent between healthy donors and patients. In addition, we also performed experiments of fibrin polymerization in which we adjusted the concentration of fibrinogen (3.4 mg/ml) to be equivalent between healthy donor and patient by adding purified human fibrinogen to healthy plasma.
The fibrin polymerization of a purified system (human fibrinogen, calcium, and thrombin) was performed using the recombinant human vimentin or buffer with 1.0 U of human thrombin and 20 mmol/l CaCl2 (final concentrations) [31]. The enzymatic reaction using the purified system was evaluated by tracking turbidity using a spectrophotometer set to 405 nm.
2.8 Imaging of the fibrin-clot structure
Briefly, plasma was supplemented with 2 % (w/w) of human fibrinogen conjugated to Alexa Fluor 488 (Thermo Scientific). Clot formation was initiated with the addition of 1 U thrombin (EMD Millipore) in the presence of 2.4 mM calcium as described [31].
For fibrin-vimentin interaction, plasma was supplemented with 2 % (w/w) of human fibrinogen conjugated to Alexa Fluor 647 (Thermo Scientific) and with mouse anti-human Vimentin (V9, Sigma-Aldrich) or mouse IgG (isotype control) prior to initiating clot formation with the addition of 1 U thrombin (EMD Millipore) in the presence of 2.4 mM calcium as described [31]. Fibrin clots were washed three times with phosphate buffered saline (PBS) and incubated immediately with a secondary anti-mouse IgG-CF488A (1:1000, Sigma) conjugate for 1 h. Then, fibrin-clot were washed three times with PBS and fixed with 4 % paraformaldehyde for 10 min and washed three times for visualization by confocal microscopy.
2.9 Statistical analysis
GraphPad Prism 8 software (San Diego, CA) was used to perform statistical analyses. Comparisons between groups were conducted by t-test and ANOVA. P values were 2-sided, and statistical significance was determined by a P value <0.05.
3 Results
3.1 Vimentin binds to fibrinogen and enhances the fibrin formation potential
We and others have reported the presence of circulating, extracellular vimentin [16], [17] whose levels become elevated in sepsis, a condition associated with marked inflammation [27]. Thus, we first determined the levels of vimentin in plasma from healthy donors (the median for fibrinogen level in healthy subjects used in this study: 2.86 (2.1–3.6) mg/ml) and critically ill patients (median: 3.54 (0.9–8.2) mg/ml), for further clinical characteristics see reference [28]. By using ELISA, we confirmed that critically ill [27], systemically inflamed patients had significantly elevated circulating vimentin levels as shown in Fig. S1 (patients 543.5 ± 253 ng/ml (n = 28) vs. healthy subjects 325.7 ± 103 ng/ml (n = 10), mean ± SD, ***p < 0.0006). High levels of plasma vimentin in parallel with hypercoagulability associated with aberrant fibrin structures [28] seen in critically ill patients opened the possibility of an interaction between fibrinogen and vimentin. Biolayer interferometry demonstrated binding of recombinant vimentin to purified fibrinogen in a concentration-dependent and saturable manner with a binding constant of KD = 580.0 ± 0.9 nM (Fig. 1A and B). Additionally, the blot overlay (Fig. S2) suggests that vimentin may preferentially bind the reduced/denatured form of fibrinogen alpha chain. This interaction of vimentin with fibrinogen may influence fibrin polymerization; thus, we tested this hypothesis using a purified system [31], with recombinant vimentin at concentrations comparable to those found in plasma from healthy subjects (Fig. S1). Recombinant vimentin significantly increased fibrin formation potential (maximum absorbance) in the purified system in a dose dependent manner (Fig. 2A). Note that the initial rate of change (slope) increased at the highest concentrations of vimentin (4.0 and 5.0 nM), although the protein did not have an apparent effect in the lag time (Supplemental Table 1). The initial rate of change was determined as described [31]. To examine whether the capacity of vimentin to potentiate fibrin formation depends on the levels of fibrinogen, we performed fibrin polymerization in a purified system by mixing increasing concentrations of purified human fibrinogen (0.3, 0.5, and 0.7 mg/ml) with buffer or a fixed concentration of recombinant vimentin (5.0 nM). Fig. 2B shows that, in comparison to samples without vimentin, vimentin positively affected fibrin formation (greater maximum absorbance) with increasing concentration of fibrinogen, suggesting an interaction between vimentin and fibrinogen. Importantly, it is also possible that the changes observed in the turbidity assays (Fig. 2A and B) represent changes in the resultant fibrin clot structure, which was examined below. We further tested our hypothesis of functional interaction between vimentin and fibrinogen by measuring fibrin formation in the presence of anti-vimentin antibodies. In comparison to the corresponding isotype IgG (Fig. S3A), fibrin formation was blocked with a rabbit-polyclonal anti-vimentin antibody and a mouse-monoclonal (V9) antibody by 60 % and 15 %, respectively, (Fig. S3B). The sheep anti-human vimentin antibody did not inhibit the effect of the recombinant vimentin on fibrin formation (Fig. S3B). These different outcomes may be due to the different epitopes recognized by each antibody in the vimentin structure (Fig. S4A), and suggest a putative binding site for fibrinogen within the rod domain of vimentin. We tested this postulate by characterizing the binding of our recombinant human vimentin rod domain (rROD) [29] to fibrinogen. Fig. S4B demonstrates that rROD bound to fibrinogen with a binding constant of KD = 260 ± 0.6 nM. These results point to a specific interaction between vimentin and fibrinogen and that the vimentin rod domain may interface with fibrinogen alpha chain.Fig. 1 Recombinant vimentin binds to fibrinogen. (A) BLI was performed with immobilized fibrinogen. The phases of measurement are shown as association phase and dissociation phase. (B) Steady state binding analysis curve of Req versus concentration of recombinant vimentin to fibrinogen. Vimentin interacted with a binding constant of KD = 580 nM ± 0.9 nM, mean ± SEM of three determinations.
Fig. 1
Fig. 2 Vimentin enhances fibrin formation potential. (A) Fibrin was formed from purified human fibrinogen (0.3 mg/ml), calcium, thrombin and recombinant vimentin protein was added at concentrations as indicated. Turbidity was measured at 405 nm. Each curve is the average of four separate experiments. Higher levels of recombinant vimentin (4.0 and 5.0 nM) resulted in significantly more fibrin compared to the absence of vimentin (**p < 0.001). (B) Fibrin formation was performed as described for A. Increasing concentrations of human fibrinogen (indicated) were mixed with buffer or recombinant vimentin (5.0 nM). Each curve represents the average of three separate experiments. The positive effect of vimentin on fibrin polymerization increments the maximum absorbance as the fibrinogen concentration increases (*p < 0.01; ***p < 0.0004).
Fig. 2
3.2 Extracellular plasma vimentin contributes to fibrin formation and is incorporated into the fibrin clot
Fibrinogen levels are elevated in systemic inflammation [7], [32]. As vimentin is more abundant in systemic inflammation and enhances fibrin formation in vitro, we hypothesized that the effect of extracellular vimentin on fibrin polymerization is greater in plasma from critically ill patients with systemic inflammation than in healthy plasma. To test this hypothesis, we used anti-vimentin antibody to block the effect of extracellular vimentin in fibrin polymerization ex vivo. The rabbit anti-vimentin antibody, which reduced fibrin formation in plasma in a dose dependent manner (Fig. S5), inhibited fibrin formation in plasma from patients by 50.1 ± 8.1 %, mean ± SD (n = 6) (Fig. 3A and C). Note that the antibody had a greater inhibitory effect in plasma from patients as compared to plasma from healthy donors (inhibited 20 ± 6.2 %, mean ± SD, n = 3) (Fig. 3B and C), indicating that vimentin may participate in clot formation even at low levels. To reduce the variability in fibrin formation among patients and healthy donors, we tested for differences in the effect of the antibody between healthy and patients using an equivalent fibrinogen concentration (3.4 mg/ml). Fig. S6B clearly demonstrates that the rabbit anti-vimentin antibody reduced fibrin formation (maximal absorbance) by 42 % in patient plasma and 27 % in healthy plasma containing the same fibrinogen concentration and volume of plasma. Note that the antibody had an inhibitory effect comparable to those shown in Fig. 3C using plasma with different levels of fibrinogen. This outcome indicates that apparently the antibody preferably interacts with plasma vimentin independent of the fibrinogen level.Fig. 3 Anti-vimentin antibody reduced fibrin formation in plasma. The tracings shown representative experiments of plasma from (A) patient or (B) healthy subject. Fibrin was formed from 20 % plasma. Plasma was mixed with rabbit (Rb) anti-vimentin antibody or isotype IgG (20 μg/ml). Turbidity was measured at 350 nm. (C) Anti-vimentin antibody blocked fibrin formation in plasma from critically ill patients by 50.8 ± 8.1 %, mean ± SD (n = 6 patients), while it blocked in healthy plasma by 20 ± 3.5 %, mean ± SD (n = 3 donors).
Fig. 3
Next, we investigated the effect of blocking plasma vimentin on the resultant fibrin network structure. Figs. 4A and S7 show representative clot structures generated from plasma of a healthy subject and critically ill patients. The antibody against vimentin provoked a change in the resultant fibrin network structure. It increased the area unoccupied by fibrin formed from plasma, suggesting an increase in porosity area (Fig. 4B). In comparison to isotype IgG, the anti-vimentin antibody at 2.5 μg/ml significantly increased the porosity area of the fibrin network in plasma. This change in the clot architecture also suggested that vimentin could be incorporated into the resultant network. To test our speculation that vimentin is enmeshed in the clot, we visualized fibrin-associated vimentin from plasma of critically ill patients by confocal immunofluorescence microscopy. Punctate staining for vimentin was found distributed along the length and branching points of the fibrin fibrils and at sites of fibrin fibrils branching (Fig. 5 , lower panels), illustrating the incorporation of vimentin into fibrin. Since the fibrin clot architecture influences fibrinolysis [33], we further examined the effect of vimentin on fibrinolytic susceptibility to plasmin. Fig. 6A depicts representative fibrin polymerization/proteolysis curves of a patient plasma in the presence of tissue plasminogen activator. In comparison to isotype IgG, analyses of the area under the curve (AUC) revealed that the anti-vimentin antibody significantly attenuated fibrin formation, consistent with fibrin polymerization studies described above (Fig. 3) and thereby reducing the fibrinolytic burden in plasma from critically ill patients (Fig. 6B) and healthy subjects (Fig. 6C). To reduce the variability in fibrin formation and degradation (AUC) among patients and healthy donors, we tested for differences in fibrinolytic susceptibility between healthy and patients using an equivalent fibrinogen concentration (0.5 mg/ml). Consistent with the fibrinolysis assays starting with the same volume of plasma, Fig. S8 clearly shows the inhibitory effect of the anti-vimentin antibody in the absence of any variation in plasma fibrinogen concentration. These results reveal a novel finding that extracellular vimentin in plasma plays a role in fibrin clot and structure formation, possibly impeding efficient fibrinolysis.Fig. 4 Anti-vimentin antibody induced changes in the fibrin clot structure. (A) Representative confocal microscopy images of fibrin clots at magnification of 120× formed in plasma from healthy subject or critically ill patient. The anti-vim antibody clearly changed the architecture of resultant clot network in the plasma from the healthy and patient at 2.5 μg/ml (right panel). (B) Anti-vimentin antibody changed the resultant fibrin network structure in plasma from healthy donors (blue symbols) and critically ill patients (black symbols), increasing the area of unoccupied by fibrin formed from plasma (increased porosity area). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 4
Fig. 5 Localization of vimentin in plasma fibrin clot. Representative of three experiments. The confocal microscopy images show fibrin clots at magnification of 177× formed from patient plasma. Plasma was supplemented with human fibrinogen conjugated to Alexa Fluor 647. The lower-left panel shows the optical section through the plasma clot stained for vimentin with the monoclonal mouse anti-vimentin antibody (V9), and detected with a FITC-conjugated secondary antibody. The upper-left panel shows fibrin stained with an isotype control to anti-vimentin and with the same fluorescent secondary as the bottom right panel. Note the staining pattern for vimentin along the fibrin fibrils and the lower-right panel shows the punctate foci for vimentin (white arrows). Scale bar, 5 μm.
Fig. 5
Fig. 6 Anti-vimentin antibody affected fibrin formation and fibrinolysis in plasma. (A) Fibrin formation and fibrinolysis were initiated with thrombin, calcium, and in the presence of recombinant tissue-type plasminogen activator, Rb anti-vim antibody or isotype IgG (20 μg/ml). Turbidity was measured at 350 nm. Representative curves for fibrin formation and fibrinolysis; each curve is the mean of duplicate reactions using plasma of a patient. B and C: Measurements of both fibrin formation and fibrinolysis are presented as the area under the curve (AUC). The Rb anti-vim antibody significantly reduced fibrin formation and thus diminishing the clot burden to lyse in plasma from patients (***p = 0.0003), and healthy donors (*p = 0.03).
Fig. 6
3.3 Circulating, extracellular vimentin is in complex with plasma fibrinogen
Compared to fibrin formation using purified proteins (Fig. 2A), plasma from critically ill patients robustly formed fibrin with almost twice the concentration of vimentin (recombinant 5 nM = ~285 ng/mL in Fig. 2A vs. endogenous 9.5 nM = ~540 ng/mL in Fig. S1), suggesting the possibility of a strong, heteromeric interaction between endogenous vimentin and fibrinogen in circulation. To investigate this possibility, we fractioned patient or healthy plasma by size-exclusion chromatography and analyzed the eluates by Western blot using an anti-vimentin antibody as we previously reported [16]. In contrast to healthy plasma, vimentin in plasma from critically ill patients co-eluted with fibrinogen from the column (Fig. 7A, inset, and Fig. S9A-C). A Coomasie-blue stained SDS-PAGE under reduced conditions demonstrated the presence of fibrinogen for both patient and healthy plasma in their corresponding elution peaks (Fig. S9D). Co-immunoprecipitation of fibrinogen with vimentin from eluates of peak 2 from critically ill patients but not healthy donors confirmed the interaction between circulating vimentin and fibrinogen (Fig. 7B) and suggested that this interaction may be part of the inflammatory spectrum.Fig. 7 Gel filtration and Western blot analyses of extracellular vimentin in plasma. (A) Plasma from a critically ill patient or a healthy subject was subjected to gel filtration to separate vimentin (expected MW = 57 kDa) from fibrinogen (expected MW = 340 kDa). Using a Superose 6 gel filtration column with a TBS mobile phase, the proteins were detected by their absorbance at 280 nm. The graph represents the elution profile for a patient plasma showing the elution time for fibrinogen (elution peak). mAu, milliabsorbance units. Inset, in sharp contrast to healthy plasma, the Western blot verified the presence of extracellular vimentin in the fibrinogen elution peak for a critically ill patient plasma. Figure represents one of four independents experiments. (B) Binding of extracellular vimentin to fibrinogen in solution (elution peak) was further demonstrated by co-immunoprecipitation. Vimentin was immunoprecipitated (IP) from gel-filtered fractions corresponding to the elution peak and immunoblotted for fibrinogen.
Fig. 7
4 Discussion
Coagulopathy is a clinical manifestation seen in critically ill patients with systemic inflammation, including sepsis (bacterial, fungal or viral), trauma, and burns [3], [8], [34]. Several factors have been proposed to be associated with the hypercoagulable state in systemic inflammation. Among the markers of coagulopathy in these patients are fibrinogen and D-dimer, a fibrin proteolytic product [5], [6]. Elevated levels of these markers are most likely indicative of an active fibrin formation and fibrinolysis. This study describes a novel role for extracellular vimentin in interacting and actively participating in fibrin formation, particularly in a greater magnitude in plasma from critically ill patients with systemic inflammation.
This is the first study to characterize the binding of vimentin to fibrinogen by several means. First, we used a purified recombinant human vimentin to demonstrate the interaction by binding kinetic and fibrin polymerization studies. Our data suggest that the vimentin rod domain and fibrinogen alpha chain comprise this interaction. Second, the interaction between endogenous vimentin and fibrinogen in plasma was revealed with co-immunoprecipitation assays, indicating this heterocomplex exits in solution. Lastly, the capacity of an anti-vimentin antibody to block fibrin polymerization in plasma validated the relationship between vimentin and fibrinogen in circulation. The vimentin-fibrinogen interaction seems to be stronger in the context of systemic inflammation because vimentin, which has a molecular mass of ~57 kDa co-eluted with fibrinogen (molecular mass of ~340 kDa) from the size exclusion chromatography. It is possible that the extracellular vimentin protein forms a complex with fibrinogen in plasma as implied by the co-immunoprecipitation assay of the co-eluted proteins. If the anti-vimentin antibody does not displace the fibrinogen-bound vimentin one can suggest that the antibody bound to vimentin in complex with fibrinogen perturbs the first step in fibrin fiber formation. The addition of the antibody in plasma apparently affected both lag time (protofibrils formation) and initial rate (lateral aggregation) of fibrin polymerization (Fig. 6). Future studies will be needed to address the underpinning mechanisms by which anti-vimentin antibody changes fibrin formation kinetics and structural features of the resultant fibrin network structure. A number of studies have reported that the functions described for vimentin in health and diseases are linked to posttranslational modifications (PTMs) in the protein [35], [36], [37]. One can argue that PTMs in vimentin cause structural modifications that change its binding properties for fibrinogen, consequently altering the kinetics of fibrin polymerization. In fact, the difference on the binding constant for fibrinogen between the commercially available insect cell-derived recombinant vimentin and our bacterial-derived rROD protein could be explained by the fact that the former is glycosylated. The mechanism for generating a vimentin-fibrinogen complex remains unknown.
This study reports elevated levels of extracellular vimentin in plasma from patients with systemic inflammation, similar to a previous study that reported elevated vimentin levels in sepsis [27]. Although others and we have detected vimentin in plasma from patients with other clinical conditions or healthy donors [16], [38], [39], normal ranges for plasma vimentin is undefined. Notwithstanding, the comparative analysis between patients and healthy donors in this report demonstrate that plasma levels of vimentin may vary in health and become elevated with systemic inflammation. Additionally, the source for the increased extracellular vimentin in the plasma remains elusive. The increment of plasma vimentin in sepsis may represent its involvement in the immune response to the infection or injury, possibly, being secreted by activated macrophages and/or endothelial cells [22], [40]. Further investigations are necessary to identify the cell types and the mechanisms by which pathogens or conditions associated with systemic inflammation induce the cells to secrete vimentin or a different form of vimentin.
Another fascinating and clinically relevant result obtained from this study is the role of extracellular vimentin in mediating both fibrin formation and fibrin clot structure, especially in plasma from critically ill patients. This novel function of vimentin in plasma was demonstrated by the inhibitory effect conferred by the anti-vimentin antibody in attenuating fibrin formation and reducing the fibrinolytic burden in plasma. Furthermore, the effect of the antibody in the architecture of the resultant fibrin clot structure (i.e. increase porosity area) seems to be similar in plasma from both healthy volunteers and critically ill patients. Consequently, the fibrinogen-bound vimentin alters the conversion of fibrinogen to fibrin after thrombin cleavage, causing the formation of a denser or aberrant clot structure as we and other recently reported for patients with severe COVID-19 [28], [41]. Thus, we reason that extracellular vimentin may be a potential procoagulant agent in critical illness with systemic inflammation, including severe cases of COVID-19. Further work is necessary to investigate the mechanism(s) by which vimentin and fibrinogen interact.
The anti-vimentin antibody was also effective at reducing (to a lower degree) fibrin formation in healthy plasma, suggesting a normal role for vimentin in coagulation. This outcome appears to be in line with our previous study in which we reported a modest prolonged tail bleeding time in the knock out mice as compared to wild type mice [16]. Despite the different phenotypes (>25) reported in vimentin null mice (review [17]), bleeding is not one of them. Vimentin is expressed in several cell types of mesenchymal origin, explaining its participation in many cell functions (review [20]), and therefore, the vimentin null mice may not be suitable to determine the source of the plasma vimentin that interacts with fibrinogen. Nonetheless, all the outcomes from our studies strongly provide evidence that the binding of extracellular vimentin to fibrinogen modulates fibrin formation.
The interaction of vimentin with fibrinogen could have other clinical implications. 1) Elevated levels of an aberrant extracellular vimentin in thromboinflammatory diseases could serve as a marker for an altered fibrin formation and thrombosis. 2) It is known that various types of cancer overexpress vimentin [42], [43], and elevated levels of extracellular vimentin detected in circulation could serve as a marker for cancer-associated thrombosis [44], [45].
In summary, the results of this study identify extracellular vimentin as an active mediator of fibrin formation. Additionally, the data suggest that elevated level of an apparent aberrant vimentin may increase the risk for thrombosis in patients with persistent systemic inflammation. One can propose that strategies to inhibit vimentin may represent a novel therapeutic approach to attenuate the prothrombotic state in critically ill patients with systemic inflammation.
CRediT authorship contribution statement
M. Martinez, L. Brubaker, performed experiments, analyzed data and contributed in writing the manuscript. N. Seshadri and A. Cebula performed experiments. M. Loor, L. Brubaker and T.K. Rosengart provided blood samples from patients as previously described [28]. R. Rumbaut, A. Yee and M. A. Cruz designed experiments, analyzed data, wrote, and edited the manuscript.
Declaration of competing interest
M.A. Cruz is the founder and CSO of A2 Therapeutics, Inc.
Appendix A Supplementary data
Supplementary material
Image 1
Acknowledgements
The Alkek Foundation, the 10.13039/100015158 Fondren Foundation (M.A.C. and A.C.), 10.13039/100001422 American Society of Hematology Scholar Award (A.Y.), and 10.13039/100000002 NIH -10.13039/100000057 NIGMS R01 GM112806 and 10.13039/100000002 NIH -10.13039/100000065 NINDS R01 NS094280 (M.A.C.). 10.13039/100000002 NIH -10.13039/100000050 NHLBI R01 HL154688 (M.A.C. and A.Y.) and T32 HL139425 (M.M-V. and L.S.B), T32 GM136554 (A.C.) and a Merit Review Award I01 BX002551 from the 10.13039/100000738 Department of Veterans Affairs Biomedical Laboratory Research & Development (R.E.R. and M.A.C.). The content is solely the responsibility of the authors and does not represent the official views of National Institutes of Health, Department of Veterans Affairs or the United States government.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.thromres.2022.11.028.
==== Refs
References
1 Katneni U.K. Alexaki A. Hunt R.C. Coagulopathy and thrombosis as a result of severe COVID-19 infection: a microvascular focus Thromb. Haemost. 2020
2 Simmons J. Pittet J.F. The coagulopathy of acute sepsis Curr. Opin. Anaesthesiol. 28 2 2015 227 236 25590467
3 Jackson S.P. Darbousset R. Schoenwaelder S.M. Thromboinflammation: challenges of therapeutically targeting coagulation and other host defense mechanisms Blood 133 9 2019 906 918 30642917
4 Bray M.A. Sartain S.E. Gollamudi J. Rumbaut R.E. Microvascular thrombosis: experimental and clinical implications Transl. Res. 225 2020 105 130 32454092
5 Bouck E.G. Denorme F. Holle L.A. COVID-19 and sepsis are associated with different abnormalities in plasma procoagulant and fibrinolytic activity Arterioscler. Thromb. Vasc. Biol. 2020 Atvbaha120315338
6 Toh J.M. Ken-Dror G. Downey C. Abrams S.T. The clinical utility of fibrin-related biomarkers in sepsis Blood Coagul. Fibrinolysis 24 8 2013 839 843 24030119
7 Iba T. Levy J.H. Levi M. Connors J.M. Thachil J. Coagulopathy of coronavirus disease 2019 Crit. Care Med. 2020
8 Iba T. Levi M. Levy J.H. Sepsis-induced coagulopathy and disseminated intravascular coagulation Semin. Thromb. Hemost. 46 1 2020 89 95 31443111
9 Yao X.H. Li T.Y. He Z.C. A pathological report of three COVID-19 cases by minimally invasive autopsies Zhonghua Bing Li Xue Za Zhi 49 2020 E009
10 Wang C. Xie J. Zhao L. Alveolar macrophage dysfunction and cytokine storm in the pathogenesis of two severe COVID-19 patients EBioMedicine 57 2020 102833
11 Ackermann M. Verleden S.E. Kuehnel M. Pulmonary vascular endothelialitis, thrombosis, and angiogenesis in Covid-19 N. Engl. J. Med. 383 2 2020 120 128 32437596
12 Ivaska J. Pallari H.M. Nevo J. Eriksson J.E. Novel functions of vimentin in cell adhesion, migration, and signaling Exp. Cell Res. 313 10 2007 2050 2062 17512929
13 Chi F. Jong T.D. Wang L. Vimentin-mediated signalling is required for IbeA+ E. Coli K1 invasion of human brain microvascular endothelial cells Biochem. J. 427 1 2010 79 90 20088823
14 Fasipe T.A. Hong S.H. Da Q. Extracellular Vimentin/VWF (von willebrand Factor) interaction contributes to VWF string formation and stroke pathology Stroke 49 10 2018 2536 2540 30355099
15 Podor T.J. Singh D. Chindemi P. Vimentin exposed on activated platelets and platelet microparticles localizes vitronectin and plasminogen activator inhibitor complexes on their surface J. Biol. Chem. 277 9 2002 7529 7539 11744725
16 Da Q. Behymer M. Correa J.I. Vijayan K.V. Cruz M.A. Platelet adhesion involves a novel interaction between vimentin and von willebrand factor under high shear stress Blood 123 17 2014 2715 2721 24642750
17 Danielsson F. Peterson M.K. Caldeira Araujo H. Lautenschlager F. Gad A.K.B. Vimentin diversity in health and disease Cells 7 10 2018
18 Das S. Ravi V. Desai A. Japanese encephalitis virus interacts with vimentin to facilitate its entry into porcine kidney cell line Virus Res. 160 1–2 2011 404 408 21798293
19 Du N. Cong H. Tian H. Cell surface vimentin is an attachment receptor for enterovirus 71 J. Virol. 2014
20 Ramos I. Stamatakis K. Oeste C.L. Perez-Sala D. Vimentin as a multifaceted player and potential therapeutic target in viral infections Int. J. Mol. Sci. 21 13 2020
21 Zhang Y. Wen Z. Shi X. Liu Y.J. Eriksson J.E. Jiu Y. The diverse roles and dynamic rearrangement of vimentin during viral infection J. Cell Sci. 134 5 2020
22 Pall T. Pink A. Kasak L. Soluble CD44 interacts with intermediate filament protein vimentin on endothelial cell surface PLoS One 6 12 2011 e29305
23 Gong D.H. Dai Y. Chen S. Secretory vimentin is associated with coronary artery disease in patients and induces atherogenesis in ApoE(-/-) mice Int. J. Cardiol. 283 2019 9 16 30808602
24 Bang H. Egerer K. Gauliard A. Mutation and citrullination modifies vimentin to a novel autoantigen for rheumatoid arthritis Arthritis Rheum. 56 8 2007 2503 2511 17665451
25 Sun S. Poon R.T. Lee N.P. Proteomics of hepatocellular carcinoma: serum vimentin as a surrogate marker for small tumors (<or=2 cm) J. Proteome Res. 9 4 2010 1923 1930 20121168
26 Aggarwal S. Singh B. Sharma S.C. Das S.N. Circulating vimentin over-expression in patients with oral sub mucosal fibrosis and oral squamous cell carcinoma Indian J. Otolaryngol. Head Neck Surg. 2022 1 6
27 Su L. Pan P. Yan P. Role of vimentin in modulating immune cell apoptosis and inflammatory responses in sepsis Sci. Rep. 9 1 2019 5747 30952998
28 Brubaker L.S. Saini A. Nguyen T.C. Aberrant fibrin clot structure visualized ex vivo in critically ill patients with severe acute respiratory syndrome coronavirus 2 infection Crit. Care Med. 2022
29 Lam F.W. Brown C.A. Valladolid C. Emebo D.C. Palzkill T.G. Cruz M.A. The vimentin rod domain blocks P-selectin-P-selectin glycoprotein ligand 1 interactions to attenuate leukocyte adhesion to inflamed endothelium PLoS One 15 10 2020 e0240164
30 Auton M. Sowa K.E. Smith S.M. Sedlak E. Vijayan K.V. Cruz M.A. Destabilization of the A1 domain in von willebrand factor dissociates the A1A2A3 tri-domain and provokes spontaneous binding to glycoprotein ibalpha and platelet activation under shear stress J. Biol. Chem. 285 30 2010 22831 22839 20498367
31 Valladolid C. Martinez-Vargas M. Sekhar N. Modulating the rate of fibrin formation and clot structure attenuates microvascular thrombosis in systemic inflammation Blood Adv. 4 7 2020 1340 1349 32259201
32 Iba T. Ito T. Maruyama I. Potential diagnostic markers for disseminated intravascular coagulation of sepsis Blood Rev. 30 2 2016 149 155 26574054
33 Abu-Fanne R. Stepanova V. Litvinov R.I. Neutrophil alpha-defensins promote thrombosis in vivo by altering fibrin formation, structure, and stability Blood 133 5 2019 481 493 30442678
34 Mucha S.R. Dugar S. McCrae K. Coagulopathy in COVID-19 Cleve. Clin. J. Med. 2020
35 Slawson C. Lakshmanan T. Knapp S. Hart G.W. A mitotic GlcNAcylation/phosphorylation signaling complex alters the posttranslational state of the cytoskeletal protein vimentin Mol. Biol. Cell 19 10 2008 4130 4140 18653473
36 Snider N.T. Omary M.B. Post-translational modifications of intermediate filament proteins: mechanisms and functions Nat. Rev. Mol. Cell Biol. 15 3 2014 163 177 24556839
37 Eriksson J.E. He T. Trejo-Skalli A.V. Specific in vivo phosphorylation sites determine the assembly dynamics of vimentin intermediate filaments J. Cell Sci. 117 Pt 6 2004 919 932 14762106
38 Bonotti A. Simonini S. Pantani E. Serum mesothelin, osteopontin and vimentin: useful markers for clinical monitoring of malignant pleural mesothelioma Int. J. Biol. Markers 32 1 2017 e126 e131 27646775
39 Kerget B. Afşin D.E. Kerget F. Aşkın S. Araz Ö. Akgün M. Is vimentin the cause or effect of obstructive sleep apnea development? Lung 198 2 2020 275 282 32088750
40 Mor-Vaknin N. Punturieri A. Sitwala K. Markovitz D.M. Vimentin is secreted by activated macrophages Nat. Cell Biol. 5 1 2003 59 63 12483219
41 Wygrecka M. Birnhuber A. Seeliger B. Altered fibrin clot structure and dysregulated fibrinolysis contribute to thrombosis risk in severe COVID-19 Blood Adv. 2021
42 Arko-Boham B. Lomotey J.T. Tetteh E.N. Higher serum concentrations of vimentin and DAKP1 are associated with aggressive breast tumour phenotypes in Ghanaian women Biomark. Res. 5 1 2017 21 28616237
43 Satelli A. Li S. Vimentin in cancer and its potential as a molecular target for cancer therapy Cell. Mol. Life Sci. 68 18 2011 3033 3046 21637948
44 Lip G.Y. Chin B.S. Blann A.D. Cancer and the prothrombotic state Lancet Oncol. 3 1 2002 27 34 11908507
45 Abdol Razak N.B. Jones G. Bhandari M. Berndt M.C. Metharom P. Cancer-associated thrombosis: an overview of mechanisms, risk factors, and treatment Cancers 10 10 2018 380 30314362
| 36495717 | PMC9726209 | NO-CC CODE | 2022-12-08 23:19:00 | no | Thromb Res. 2023 Jan 7; 221:97-104 | utf-8 | Thromb Res | 2,022 | 10.1016/j.thromres.2022.11.028 | oa_other |
==== Front
Comput Commun
Comput Commun
Computer Communications
0140-3664
1873-703X
Elsevier B.V.
S0140-3664(22)00453-4
10.1016/j.comcom.2022.12.002
Article
Road crash risk prediction during COVID-19 for flash crowd traffic prevention: The case of Los Angeles
Wang Junbo ab
Yang Xiusong ab
Yu Songcan ab
Yuan Qing ab
Lian Zhuotao c
Yang Qinglin ab⁎
a School of Intelligent Systems Engineering, Sun Yat-Sen University, Shenzhen, 518107, PR China
b Guangdong Provincial Key Laboratory of Intelligent Transportation System, Sun Yat-Sen University, Shenzhen, 510275, PR China
c Department of Computer Science and Engineering, the University of Aizu, Aizuwakamatsu, Japan
⁎ Corresponding author at: School of Intelligent Systems Engineering, Sun Yat-Sen University, Shenzhen, 518107, PR China.
7 12 2022
7 12 2022
8 6 2022
1 10 2022
1 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Road crashes are a major problem for traffic safety management, which usually causes flash crowd traffic with a profound influence on traffic management and communication systems. In 2020, the sudden outbreak of the novel coronavirus disease (COVID-19) pandemic led to significant changes in road traffic conditions. In this paper, by analyzing crash data from 2016 to 2020 and new COVID-19 case data in 2020, we find that the average crash severity and crash deaths during this period (a rapid increase of new COVID-19 cases in 2020) are higher than those in previous four years. Hence, it is necessary to exploit a novel road crash risk prediction model for such an emergency. We propose a novel data-adaptive fatigue focal loss (DA-FFL) method by fusing fatigue factors to establish a road crash risk prediction model under the scenario of large-scale emergencies. Finally, the experimental results demonstrate that DA-FFL performs better than the other typical methods in terms of area under curve (AUC) and false alarm rate (FAR) for imbalanced data. Furthermore, DA-FFL has better prediction performance in convolutional neural networks-long short-term memory (CNN-LSTM).
Keywords
COVID-19
Crash risk prediction
Flash crowd traffic
Data imbalance
Large-scale emergencies
==== Body
pmc1 Introduction
Traffic accidents have become the eighth leading cause of death in the world and cost three percent of Gross Domestic Product (GDP) in most of the countries [1]. Safety is the biggest concern in an intelligent transportation system. According to official statistics [1], around 1.35 million people died in traffic accidents every year. Furthermore, traffic accidents can cause traffic jams to form a flash crowd traffic that are supply–demand mismatching problems for communication resources and communication congestion problems. When a traffic jam occurs, vehicle-to-vehicle distance becomes very close, which leads to a significant increase in traffic density [2]. People in cars use their cell phones more frequently than normal during a traffic jam in a fixed field, which will increase the communication overhead of the roadside communication base stations.
To reduce the occurrence of traffic accidents, Chen et al. studied the crash risk prediction of road sections as an important factor for traffic management to control traffic accurately [3]. It can not only reduce the incidence of road traffic accidents but also provide much more auxiliary information for decision-making. However, Zhang et al. think that the road traffic environment has become more complex and variable due to the rapid increase in the number of vehicles each year [4].
On the other hand, Yang et al. [5] found that COVID-19 has strong contagiousness and uncontrollable variability and it has been developed into a global public health issue in a very short period. The COVID-19 pandemic has affected various industries, such as healthcare, the global economy, education, tourism, transportation [6], and so on. During the COVID-19 pandemic, people travel less by public transportation [7]. As a result, the proportion of car trips, destinations, and routes have changed more than the period before the COVID-19 pandemic. Due to the dramatically affection for the transportation systems [8], the methods for road crash risk prediction face new challenges. First, crash data becomes much more unbalanced [9] during COVID-19. Unbalanced data can cause models to be biased toward most classes, which makes risk prediction much harder.
In order to study in-depth the impact of the COVID-19 pandemic on traffic, Adanu et al. analyze crash data during the COVID-19 pandemic and found that traffic, vehicle mileage, and the number of crashes have all dropped significantly during COVID-19 in [10], and the number of fatal crashes has increased.
Li et al. analyzed empirical spatiotemporal road traffic congestion during the COVID-19 pandemic [11], which found that traffic congestion had an upward trend after the pandemic. Furthermore, Zhou et al. analyzed the response policy during the COVID-19 outbreak and found that the prevention policy would have a new impact on urban traffic [12]. However, the existing works about road crash risk prediction models can not meet the unique requirement of stable and accurate.
To address the above challenges, this research aims to fuse multi-source data to analyze the characteristics of road crash data during COVID-19 and exploit the intrinsic connections between COVID-19 and road traffic crashes.
Meanwhile, we propose an improved focal loss function to deal with imbalanced crash data by considering fatigue factors in the transportation system. The details of this proposal will be discussed in Section 4.3. The main contributions of this work can be summarized as follows,
• We analyzed Covid-19 data and crash data from 2016 to 2020 and find that the average severity and number of fatalities per crash increased during the COVID-19 pandemic compared to the pre-COVID-19 period.
• To the best of our knowledge, this is a pioneering study exploring road crash risk prediction models under specific emergencies (COVID-19).
• We propose to introduce a fatigue factor to improve the focal loss function, which further overcomes the problem of data imbalance.
• The proposed DA-FFL approach can reduce flash crowd traffic caused by road crashes and maintain road traffic safety and communication efficiency during large-scale emergencies.
The rest of this paper is organized as follows: In Section 2, the researches in road crash risk prediction and traffic safety during the COVID-19 pandemic are presented. In Section 3, we introduce the sources of experimental data and some new findings after data analysis. In Section 4, we illustrate the main techniques used in this study, and the fatigue factor and DA-FFL method proposed by this paper. In Section 5, we explain how to process the data and the results are analyzed. In Section 6, we discuss the differences from existing work, and implications of this study and the challenge faced. Finally, Section 7 concludes this paper.
2 Related work
To improve the safety of road traffic, researchers employ many different methods to predict road crash risk. In this section, we shall discuss the most related works from three aspects: traditional machine learning based methods, deep learning based methods, and traffic safety during COVID-19.
2.1 Machine learning based methods
In the field of road crash risk prediction, machine learning methods have emerged as a promising approach to train accurate and robust statistical models from data. support vector machines (SVM) is a traditional machine learning method with a solid theoretical foundation that performs well in solving small samples, nonlinearity, regression, and binary classification problems with high dimensionality. Yu et al. used the SVM model with radial basis kernel function (RBF) to achieve better results in crash risk prediction with a small sample size [13].
Logistic regression is also a classical machine learning method, which is simple and easy to understand and has good model interpretability. It can demonstrate the impact of different features on the final result from the weights of the features and also have a small memory resource consumption. Cheng et al. [14] proposed a crash risk prediction model based on the extended Logit method with real-time traffic flow data as input for urban freeways. Guo et al. [15] analyzed the relationship between traffic flow, risky driving behavior, and traffic crash risk. The authors also built a model to predict traffic crash risk based on logistic regression. Meanwhile, Alrukaibi et al. [16] proposed to predict freeway crash frequency by using a mixed logit model.
In addition, K-Nearest Neighbor (K-NN) is also good at dealing with classification problems and is suitable for classifying rare events without estimating parameters. Dimitrijevic et al. [17] used a comparative analysis of various machine learning methods such as Bayesian Logistics Regression, K-NN, and Random Forest to conclude that the use of driver behavior data can improve the predictive performance of the model. Random Forests, a widely used algorithm in machine learning, is an integrated learning method based on Bagging that can handle classification and regression well. Zhai et al. fused traffic monitoring data, foggy accident data, and road geometry data to identify and rank the most important variables by using the random forest. Then, the authors built a crash risk prediction model for freeways under foggy conditions based on Bayesian logistic regression [18].
Machine learning methods based on statistical learning methods have shown stable performance in road crash prediction, traditional machine learning methods are usually not the best choice due to the face of increasing amounts of data and more complex tasks [19].
2.2 Deep learning based methods
With the rapid development of deep learning (DL) technologies, many DL algorithms have been applied to crash risk prediction since deep neural networks have a strong ability for feature extraction.
Moosavi et al. [20] first created a large-scale public accident information database (US-Accidents) including traffic events, weather data, points of interest and time. Then, the authors proposed a deep accident prediction neural network (DAP) that significantly improves the prediction of rare crash cases. To learn the Spatio-temporal characteristics of the road traffic data, Bao et al. [21] proposed a Spatio-temporal convolution long-term and short-term memory network (STCL-net). This proposal fuses the Global Positioning System (GPS) data of taxis with other multi-source data training models, and well predicted the short-term crash risk in New York City.
Although the neural network model is able to predict the road crash risk, the real road crash data only occupies a small part of the massive traffic data. It shows that the highly unbalanced data set affects the learning ability of the model since the number of crash samples and non-crash samples in the data used for research often varies greatly. To solve the problem of data imbalance for the deep learning prediction model, Li et al. [22] used the synthetic minority over-sampling technique (SMOTE) to sample a few samples outside the model. They proposed an CNN-LSTM based crash risk prediction model for urban roads, which is superior to the traditional machine learning model in sensitivity and false alarm rate. Yu et al. [23] fused five minutes of multi-source data into time slices with spatial information, enhanced the weight of a few samples from the inside of the model by using the focus loss function [24]. They designed a convolutional neural network based crash risk prediction model.
2.3 Researches on traffic safety during COVID-19
Many works have studied the corresponding emergency plans based on the data during COVID-19.
Adanu et al. [10] showed that although the traffic volume, vehicle mileage, and the number of traffic accidents decreased significantly during COVID-19, the number of fatal traffic accidents increased. In addition, the work [25] researched the changes of bicycle crashes in Arlington, Virginia before and after COVID-19 and finds that the crash risk of bicycle travel increased during COVID-19. Meanwhile, after studying the data on road crashes, deaths and minor injuries in Greece during COVID-19, Sekadakis et al. found that the number of deaths and minor injuries in road crashes increased significantly in the first month of COVID-19 in [9].
In such situations, a sustainable and stable communication network is very important in the event of large-scale emergencies [26]. However, as aforementioned, most of the current studies have only analyzed the changes in traffic crashes during the COVID-19 pandemic without considering flash crowd traffic caused by crashes and the communication congestion on the road during large-scale emergencies. Hence, we explore road crash changes during large-scale emergencies and attempt to develop a crash risk prediction model for large-scale emergency scenarios.
The related literature is summarized in Table 1, including data types, models, and road types.
Table 1 Related work summary.
Author Year Data Prediction model Road type
Yu et al. 2013 Traffic data SVM Colorado Mountainous freeway
Cheng et al. 2022 Traffic flow EL Shanghai Urban freeway
Guo et al. 2021 Driving behavior LR China G15 freeway
Alrukaibi et al. 2021 Crash data MLM Kuwait highways
Dimitrijevic et al. 2022 Multi-source data BLR,K-NN New Jersey highway
Zhai et al. 2020 Multi-source data BLR California freeways
Moosavi et al. 2019 US-Accidents DAP The United States
Bao et al. 2019 Multi-source data STCL net New York City Roads
Li et al. 2020 Multi-source data CNN-LSTM Urban roads in Orlando
Yu et al. 2020 Multi-source data CNN Shanghai urban expressway
Adanu et al. 2021 Traffic data,COVID-19 ML,LC-MNL Alabama roadways
Monfort et al. 2021 COVID-19,Bicycle crash FMM Arlington bike lanes
Sekadakis et al. 2021 Traffic data,COVID-19 SARIMA Greece roadways
3 Dataset and analysis
3.1 Dataset
The data used in this research are illustrated as follows.
• Traffic Data
In this study, we select the Interstate 110 (I-110) freeway whose total length is 48.4 km, which has more accidents among the freeways in Los Angeles. As shown in Fig. 1, the visualization illustrates the data of southbound and northbound roads through the freeway Performance Measurement System (PeMS) maintained by the California Department of Transportation (CalTrans) [27], ranging from 22 and 23 traffic detection stations. It includes various features such as traffic flow, speed, road occupancy, vehicle miles traveled (VMT) and vehicle hours traveled (VHT). However, it is difficult to directly utilize the traffic data for each lane recorded by the detection station for the model [18] due to the random noise. Hence, the data are aggregated into 5-minute intervals instead of every 30 s.
• Weather Data
We obtain weather data from Los Angeles International Airport which is nearby the I-110 freeway, from the National Oceanic and Atmospheric Administration (NOAA) [28]. It includes features such as visibility, rainfall, temperature, humidity and air pressure. All the weather data are updated hourly. The details of the selection of the weather features will be explained later.
• Crash Data
Road crash data on the I-110 freeway is obtained in the Transportation Injury Mapping System (TIMS) [29] developed by the University of California, Berkeley. Road crash data include specific crash time, location, severity, number of casualties, crash type, etc. For an in-depth analysis of the crash data characteristics during COVID-19 in 2020, we collect all accident data from January to December 2016–2020.
• COVID-19 Case Data As shown in Fig. 1, we select 18 Census-Designated Places (CDPs) located along the I-110 freeway with a high population density, which are also the sections with the highest number of crashes. The selected data include new confirmed cases of COVID-19 per day from March 2020 to December 2020 in 18 CDPs. The COVID-19 case data in this study are publicly available [30], the detailed information on CDPs is shown in Table 2.
Fig. 1 Research I-110 freeway in Los Angeles.
Table 2 Census-designated place and population.
CDP Population CDP Population
Elysian Park 5712 Exposition Park 44917
Angelino Heights 2502 West Vernon 53644
Wholesale District 36129 South Park 37961
Chinatown 8021 Harvard Park 37935
Temple-Beaudry 39482 Florence-Firestone 47445
Westlake 59355 Vermont Knolls 17200
Downtown 27507 Vermont Vista 41186
Pico-Union 41842 Century Palms/Cove 33766
University Park 27456 Figueroa Park Square 8721
3.2 Data analysis and findings
First, we analyze the crash situation during the years from 2016 to 2020 in general and then summarize crash data in Table 3 where we calculate the number of crashes, the number of injured, and the number of deaths specifically for each year. It can be seen that there are more deaths caused by road crashes in 2020 than in the previous four years, with 11, 8, 1, and 12 more than from 2016 to 2019, respectively. Meanwhile, we calculate the average number of the above three items and compute the overall severity which means the accumulation of the severity value for each crash. The severity of the crash takes values from 1 to 4, and the smaller number indicates a more serious crash. In Table 3, we can see that the average number of injured people, death, and severity show a different pattern in 2020, compared with the cases in the other four years.
In 2020, the number of injured in the average of a crash peaked at 1.469, the death toll per average of a crash peaked at 0.02, and crash severity reached a five-year high of 3.631(the smaller the number, the more serious). From the above data, it can be seen that the internal characteristics of road traffic crashes have changed in 2020 under the influence of COVID-19.
Next, we exploit to compare the crash data from January to December for the five years from 2016 to 2020 since the number of cars and population do not change much in the nearly five years, so as to the road conditions which makes comparison results of the crash data meaningful. As depicted in Fig. 2, a square represents the crash situation for a day. The number of crashes for each day is represented by different colors, where the color near blue shows smaller cases and close to red represents more crash cases. The number of crashes counts the crash caused by normal automobiles, pedestrians, bicycles, motorcycles, and trucks, but do not include crashes related to alcohol and drugs. Fig. 2 shows that the crashes are more evenly distributed from 2016 to 2019 than in 2020. And it is obvious that the blue squares are mainly clustered in the months of March to July, November, and December of 2020, indicating that lower crashes occurred during this period than in the previous four years. Meanwhile, we know that the I-110 freeway is located in an area where the new cases of COVID-19 began to appear in March. Therefore, the drop-down of crashes during this period may have a strong correlation with COVID-19.Table 3 Comparison of crash data from 2016 to 2020.
Year 2016 2017 2018 2019 2020
NO. of crashes 730 901 842 869 795
Total injuries 1013 1290 1106 1196 1168
Average NO. of injured 1.388 1.432 1.314 1.376 1.469
Total deaths 5 8 15 4 16
Average NO. of deaths 0.007 0.009 0.018 0.005 0.020
Total severity 2697 3322 3074 3214 2287
Average severity 3.691 3.687 3.651 3.699 3.631
To further verify the above correlation, we count the number of new COVID-19 cases each day for the 18 CDPs from March 13 to December 31, 2020, and show the results in Fig. 3. Between March and July, there is a growing trend in the number of new cases each day over time. Then, it is followed by a gradual decrease from August to October. After that, the number of new COVID-19 cases per day exploded in November and December due to colder weather and several other factors. There are even 1,890 new cases on December 16. We find that the two intervals of rapid growth in Fig. 3 with an increasing trend of new daily cases (March to July, November to December) correspond strictly to the two intervals with a much smaller number of crashes in Fig. 2, and this phenomenon is most obvious at the beginning of COVID-19. It means that there is a negative correlation between the number of crashes and the number of new cases per day, which can confirm that the number of crashes has a strong correlation with COVID-19.Fig. 2 Number of crashes per day from 2016 to 2020.
To further explore the deeper impact of the COVID-19 pandemic on road crashes, we make a more detailed comparison of the crash data from 2016 to 2020. As illustrated in Fig. 4, the average of each crash’s severity and the number of deaths from January to December are plotted.Fig. 3 Total number of new COVID-19 cases per day in 2020 in 18 CDPs.
Firstly, Fig. 4(a) compares the average severity per crash for each month. In the vast majority of months during these five years, the average severity exceeds 3.5. In particular, crashes are more severe during periods from March to May and November to December 2020, in which the two lowest values occur in April and December at 3.4 and 3.42, respectively. Moreover, these two periods are not only the worst in 2020, but they are worse than these two periods in the previous four years. This is consistent with the findings in the previous section.Fig. 4 Average severity and deaths of a crash for January to December from 2016 to 2020.
Secondly, Fig. 4(b) represents the average number of deaths per crash for each month. The number of deaths in 2020 is concentrated in a period of rapid growth in COVID-19 cases, with these two periods (March to May, November to December) accounting for 56.25% of the year’s deaths. Specifically, March and November are the two months with the highest average number of deaths per crash in 2020 at 4.3% and 6.5%, respectively, which coincides with the beginning of two rapid growth intervals of new COVID-19 cases.
Based on the above analysis, we find that when the number of new cases of COVID-19 increases rapidly, the number of crashes decreases significantly, but on average the severity of each crash and the number of deaths are increasing. As a result, it can be determined that COVID-19 has a serious impact on road traffic crashes. And the crash data from March to May, the period most affected by COVID-19, is utilized for further study.
3.3 Motivation of this study
The COVID-19 pandemic has changed traffic patterns, including but not limited to a reduction in the number of trips and a change in the purpose and route of trips. Based on the above new findings, the most important impact on traffic safety is the apparent reduction in the number of crashes during COVID-19, and a significant increase in the average severity and number of deaths per crash, especially during the period of rapid growth in the number of new COVID-19 cases per day. In addition, communication resources are more important than usual during the COVID-19 pandemic.
Reducing traffic crashes can decrease flash crowd traffic caused by traffic jams, further diminish communication congestion, and maintain an efficient road communication environment during large-scale emergencies. Therefore, there is an urgent need to study the changes in road crashes in the context of the COVID-19 pandemic and develop crash risk prediction models based on data during COVID-19 to provide decision support for traffic management authority (TMA) to optimize traffic systems, reduce the occurrence of traffic crashes, prevent flash crowd traffic on roads, and address the root causes of communication problems due to traffic crashes.
4 Methodology
This section describes the main theories and techniques, including SMOTE and focal loss functions, as well as the proposed fatigue factor.
4.1 Synthetic minority oversampling technique
SMOTE as a data synthesis method is only applied to the training data-set. The test data will not be synthesized, so the test data can still reflect the real information [21]. There are several types of SMOTE, including regular SMOTE, ADASYN, Borderline-SMOTE, SVMSMOTE, and SMOTE+ENN [31]. The conventional SMOTE has been chosen mostly [21] since it shows good performance in road crash risk prediction.
Specifically, the entire data sample is defined as S, and the subset Smin∈S denotes the minority class sample. The definition of SMOTE can be written by Eq. (1): (1) xnew=xi+rand(0,1)×xˆi−xi,
where xi∈Smin denotes the ith minority class sample, xˆi∈Smin, and xˆi is one of the K-nearest neighbors for xi, xnew is the newly generated sample.
To create a synthetic sample xnew, we calculate the Euclidean distance between a minority class sample B and one of its random nearest neighbors. The second item at the right of Eq. (1) denotes the weighted Euclidean distance with a random number between(0,1). Therefore, the synthetic sample generated according to Eq. (1) is a point on a line segment along the minority class sample xi and a randomly selected K-nearest neighbor xˆi.
The specific process of the new sample synthesized by SMOTE is illustrated in Fig. 5.
Fig. 5 The illustration of SMOTE data generation based on Euclidean distance (K=6).
4.2 Fatigue factor
In this study, we analyze the crash data for each year and find that the crash severity and the number of crashes are very similar to the distribution of the 24-hour driver fatigue index mentioned in [32]. To verify this finding, we divide the crash number and crash severity data from March to May 2020 into 24 hourly segments for cumulative statistics. For comparison, we normalize the above two sets of data and the driver fatigue data from [32] by Min-Max normalization, respectively.
To depict the similarity between the crash number curve, crash severity curve, and driver fatigue curve, we introduce dynamic time warping (DTW) [33] to calculate the similarity.
Firstly, let Q denote the fatigue index, C denote the number of crashes, and S denote the total crash severity. Q={q1,q2,…,qn}; C={c1,c2,…,cm}; S={s1,s2,…,sm}. All three sequences are 24-hour values, so n=m. Mq,c is the Euclidean distance matrix between the points of Q and C. Mq,s is the Euclidean distance matrix between the points of Q and S.
Then search for a minimum path Rq,c from Mq,c(1,1) to Mq,c(n,m), and search for a minimum path Rq,s from Mq,s(1,1) to Mq,s(n,m). Rq,c and Rq,s are obtained by a dynamic programming algorithm. The dynamic programming algorithm is based on the following recurrence relation. (2) φ(i,j)=dqi,cj+min{φ(i−1,j−1),
φ(i−1,j),φ(i,j−1)}
(3) η(i,j)=dqi,sj+min{η(i−1,j−1),
η(i−1,j),η(i,j−1)},
where φ(i,j) and η(i,j) denote the cumulative distance from the start point to the current element. dqi,cj and dqi,sj denote the distance of the current element, which is the distance between qi and cj, qi and sj, respectively. The shortest path from the start point to the current element is the length of the shortest path from the start point to the previous element plus the value of the current element
Therefore, the DTW values of Q and C, Q and S are the minimum values of the cumulative distance of the path elements, respectively, which are calculated by Eq. (4), and the results are shown in Fig. 6. (4) DTW(Q,C)=min∑k=1KwkDTW(Q,S)=min∑k=1Kvk,
where wk=(i,j)k denote the kth element on Rq,c, vk=(i,j)k denote the kth element on Rq,s. Rq,c and Rq,s are the shortest path from Mq,c(1,1) to Mq,c(n,m) and from Mq,s(1,1) to Mq,s(n,m), respectively.
The DTW values calculated from the crash number curve and driver fatigue curve and from the crash severity curve and driver fatigue curve are 2.727 and 2.754, respectively. These results show that the two sets of curves are very similar and fully indicate that road crashes are highly correlated with driver fatigue. Therefore, it is necessary to take driver fatigue into account in road crash risk prediction. However, we did not collect data directly describing driver fatigue during the COVID-19 epidemic, so the crash severity curve with the smaller DTW value is selected instead of the fatigue curve and is denoted as the fatigue factor for subsequent studies.
Fig. 6 DTW-based curve similarity comparison.
4.3 Fatigue-FL and DA-FFL method
The focal loss function was originally proposed by Lin et al. in [24] and has been widely used in subsequent studies on data imbalance. (5) LFL=−∑i=1N[α1−piγyilogpi+(1−α)piγ1−yilog1−pi]
In this study, we use the fatigue factor to modify the focal loss function. Eq. (6) is the improved loss function denoted as the fatigue focal loss (Fatigue-FL) function, where each period time corresponds to a fatigue factor value. (6) LFatigue−FL=−∑i=1N[α1−piγyilogpi+(1−α)piγ1−yilog1−pi]1−e(f−1),
where N is the number of samples predicted, i represents the ith sample, yi indicates the label of the ith real sample, and yi∈{0,1}. yi=1 means crash occurrence, yi=0 indicates no crash occurrence. pi denotes the probability that the ith sample is predicted to be a crash, and pi∈(0,1]. Weighting factor α∈[0,1], focusing parameter γ follows γ≥0. The focal loss shall evolve into the α-weighted cross entropy when γ=0.
f is the fatigue factor that corresponds to each hour. In our proposed Fatigue-FL, the value of f is automatically adjusted according to the fatigue time feature of the crash sample, the values of f at different time intervals are shown in Fig. 6. We define a correction function Φf=1−e(f−1) to achieve adaptive adjustment of the f value. When the sample is in a fatigue-prone period the corresponding fatigue factor will achieve a higher value, and the loss value of the Fatigue-FL function will be reduced.
Fig. 7 shows the values of the fatigue factor f and the values after Φf processing, the blue and red dots indicate the original values. For the sake of observation, we used the Savitzky-Golay filter [34] for smoothing blue and red dashed lines. When the value of f is high, the lower value can be obtained after Φf processing, which makes the loss value lower and makes the model pay more attention to the crash sample. Similarly, when the value of f is low, it makes the model pay less attention to the crash sample. In brief, we incorporate the fatigue factor into the focal loss to obtain the Fatigue-FL proposed in this study, which is a loss function that can adaptively adjust the degree of attention to the crash sample as the fatigue level varies for each time interval. The specific performance is verified by subsequent experiments.Fig. 7 Fatigue factor and the correction function.
Furthermore, we propose a method DA-FFL for predicting road crash risk based on the Fatigue-FL function, as depicted in Fig. 8. When the ratio of positive to negative samples is greater than 0.05, the Fatigue-FL function is used in the neural network model. However, when the ratio of positive to negative samples is less than 0.05, the positive samples are first processed using SMOTE before using the fatigue focus loss function. This approach adaptively selects the optimal processing according to the degree of data imbalance to maintain the predictive performance of the model.
Fig. 8 Data adaptive fatigue-focal loss method based crash risk prediction model.
5 Experiments
5.1 Data preprocessing
Based on the results of the data analysis in the previous section, the data used in this research is for 92 days from March 1, 2020, to May 31, 2020. Because this is the first three months of the COVID-19 pandemic, where the impact of road traffic is most pronounced. The initial data needs to be cleaned before the experiment. Due to the uncontrollable factors of the detector technology, the initial data is often missing randomly. Traffic data has time continuity, the average of the two data before and after the missing time is filled in as the missing position.
There are many types of weather data, but not every type of data is suitable for model training. It is necessary to select the best weather data to enhance the model’s accuracy. We use the Pearson correlation coefficient [35] to perform correlation analysis on the weather data and use the Kolmogorov–Smirnov (K-S) [36] test to verify the normal distribution of the weather data. As depicted in Fig. 9, the Pearson correlation coefficient values close to −1 are negative correlations and close to 1 indicate positive correlations. If the correlation between the two features is high, it will be verified with the normal distribution, Table 4 presents the results of the normal distribution of weather features. In this table, Remove the feature data that do not satisfy the normal distribution or the normal distribution is not obvious. Because data close to the normal distribution will be more conducive to improving the training of the model. Therefore, we choose the four indicators that best represent the normal distribution as the standard deviation (SD), skewness, kurtosis, and K-S test, respectively. The SD of precipitation reaches a minimum of 0.087, the skewness of WetBulbTemperature (WBTemperature) is closest to 0, the kurtosis of humidity is closest to 0 and the K-S test of pressure has a minimum p-value of 0.067. Furthermore, we also consider the impact of actual weather conditions on the risk of road crashes, visibility is taken into account. After the above screening methods, we finally select the five weather characteristics of temperature, precipitation, humidity, pressure and visibility.
For the crash data, we first locate the crash position and then find the two nearest traffic detection stations upstream and downstream respectively. Then the weather data and the traffic data of the detection station are time-aligned. As depicted in Fig. 10, the traffic data of the two detection stations are integrated to train the AI model. As studied in [37], the data between 10 and 15 min before crash time can achieve better crash prediction ability than other time periods. Thus, we label the data of 15 min before the crash time as the crash data. Meanwhile, the crash can be caused by alcohol and drugs, and the crash data is cleaned since such crashes are caused by human factors and are not in the scope of this study. Furthermore, most of the crashes between alcohol and drugs are concentrated in the 00:00–6:00 time period, and the vehicles on the road during this time period are often very rare. Based on this, we remove the data from the time period 00:00–6:00, and the fatigue factor for 6:00–24:00 is retained. Traffic on nearby roads is briefly affected after each crash, and traffic data during this period tends to be highly volatile, so we also removed this data within 60 min after each crash.Fig. 9 Correlation analysis of weather features.
Table 4 Weather data normal distribution test.
Variable Median Mean SD Skewness Kurtosis K-S
DPTemperature 30 24.67 14.354 −0.546 −1.27 0.276
DBTemperature 16 15.93 10.690 0.322 −0.65 0.121
WBTemperature 57 56.99 5.048 −0.143 −0.20 0.067
Precipitation 0 0.01 0.087 12.841 193.73 0.473
Humidity 63 61.36 15.054 −0.680 0.21 0.080
Pressure 29.77 29.78 0.093 0.235 −0.48 0.067
Visibility 1 2.59 2.555 2.015 4.32 0.313
WindDirection 0 157.15 241.02 1.513 0.99 0.317
WindGustSpeed 0 0.28 2.278 8.234 67.32 0.534
WindSpeed 0 2.11 2.790 1.139 0.82 0.349
After the above processing, the shape of the experimental data is (893002, 3, 18), in which 892796 is the total number of samples, including 309 crash samples and 892693 non-crash samples. As shown in Fig. 11, the relationship between crash samples and non-crash samples for the entire data sample is illustrated in the time dimension. Fig. 12 shows the structure of each sample, which contains three 5-minute time slices, and each time slice is fused by 6 upstream traffic features, 6 downstream traffic features and 6 downstream traffic features, 5 weather features, and 1 fatigue time distribution feature.Fig. 10 Crash location and upstream&downstream detection stations.
Fig. 11 The relationship between positive and negative samples in the direction of time.
Fig. 12 Experimental data sample structure.
5.2 Method assessment
The conventional accuracy evaluation methods can not meet the requirements of the road crash prediction task since the crash and non-crash samples are very imbalanced. Hence, we exploit to evaluate our model using several methods as follows.
The confusion matrix summarizes the data samples according to the real category and the category predicted by the model. The confusion matrix of the binary classification is a table with two rows and two columns, as shown in Table 5. True negative (TN) means to correctly predict a negative sample, that is, to predict a negative sample as a negative sample, false positive (FP), false negative (FN) and true positive (TP) also have the same representation.
FAR represents the rate at which negative samples are predicted to be positive samples. True positive rate (TPR), means that the correct rate is predicted in all positive samples. The receiver operating characteristic (ROC) curve is often used to evaluate the prediction performance of the binary classification model [38], which can reflect the trend of FPR and TPR when the model selects different thresholds. The AUC value is used as the area under the ROC curve to evaluate the predictive performance of the model [39]. The closer the AUC value is to 1, the better the predictive performance of the model.Table 5 Confusion matrix.
True\Predicted Non_crashes Crashes
Non_crashes TN FP
Crashes FN TP
5.3 Experimental results
In this section, we use long short-term memory (LSTM) [40] to verify the effectiveness of Fatigue-FL, and then the Fatigue-FL is used in three models to verify the performance of different models for road crash risk prediction during COVID-19. The main structure of our model is CNN and LSTM, including the input layer, 1D convolutional layer, average pooling layer, LSTM layer and fully connected layer. The detail of the CNN-LSTM model architecture for DA-FFL is depicted in Table 6. In the Fatigue-FL function with α=0.25, γ=2, and the optimizer chooses adam.
Firstly, we take 6 methods of dealing with data imbalance for comparison experiments, which are none(No method is used), random oversampling, SMOTE, focal Loss, Fatigue-FL, Fatigue-FL&SMOTE (SMOTE and Fatigue-FL are used on the exterior and interior of the model, respectively).Table 6 CNN-LSTM model structure for DA-FFL.
Layer Output shape
Input layer 3,18
Convolution1D None, 3, 64
AveragePooling1D None, 3, 64
LSTM None, 64
Dense None, 20
Dense None, 2
The experimental results for FAR and AUC are shown in Table 7 and 8, respectively. The AUC values at all scales after random oversampling are lower than None’s, the reason might be that random oversampling generates a lot of duplicate data. After the experimental data are processed by SMOTE, the FAR results are significantly better than the none on all 6 proportions and better than random oversampling on most proportions. Meanwhile, the AUC values are improved to different degrees on all 6 proportions, which indicates that SMOTE can effectively enhance the features of the samples. Focal loss has high FAR values for all scales except at 1:10 where it performs better, but the AUC values are higher relative to the previous methods, which indicates that focal loss can effectively reduce the impact of unbalanced data in road crash risk prediction.
When using Fatigue-FL, the FAR is lowest at data ratios of 1:2, 1:10 and the AUC value is highest at 1:5, 1:100, indicating that it shows that the Fatigue-FL function further improves the performance compared to focal Loss. However, this advantage becomes gradually smaller as the proportion of negative samples increases. Finally, the overall performance of SMOTE&Fatigue-FL is slightly lower than Fatigue-FL when the data ratio is lower than 1:10, but the overall performance is slightly higher than Fatigue-FL starting from 1:20. This also suggests that using SMOTE and Fatigue-FL separately inside and outside the model will lead to further improvements in overall model performance as the proportion of positive and negative samples gradually increases. The above experimental results can prove that the proposed DA-FFL method can effectively solve the effect of data imbalance.
Next, we apply DA-FFL to LSTM, BiLSTM, and CNN-LSTM models and select the suitable model for DA-FFL through performance evaluation. Based on the proposed DA-FFL method, the model comparison section uses the Fatigue-FL loss function at data proportions of 1:2, 1:5, and 1:10, at data proportions of 1:20, 1:100 and 1:200 the data is first processed appropriately with SMOTE and then the Fatigue-FL loss function is used. The experimental results are shown in Fig. 13, where CNN-LSTM performs better than the other two models for different proportions of data with DA-FFL.Table 7 False alarm rate experimental results.
Methods Crash:Non-crash
1:2 1:5 1:10 1:20 1:100 1:200
None 0.144 0.491 0.436 0.528 0.475 0.599
Randon Oversampling 0.558 0.491 0.460 0.489 0.280 0.349
SMOTE 0.354 0.349 0.296 0.467 0.448 0.503
Focal Loss 0.500 0.730 0.230 0.525 0.550 0.582
Fatigue-FL 0.138 0.348 0.143 0.406 0.534 0.238
SMOTE+Fatigue-FL 0.612 0.058 0.273 0.402 0.029 0.114
Table 8 AUC value experimental results.
Methods Crash:Non-crash
1:2 1:5 1:10 1:20 1:100 1:200
None 0.671 0.661 0.638 0.624 0.599 0.578
Random Oversampling 0.652 0.639 0.602 0.616 0.598 0.603
SMOTE 0.678 0.671 0.635 0.647 0.638 0.614
Focal Loss 0.678 0.707 0.694 0.655 0.641 0.628
Fatigue-FL 0.686 0.712 0.710 0.692 0.674 0.654
SMOTE&Fatigue-FL 0.693 0.702 0.736 0.694 0.673 0.668
Fig. 13 AUC values of LSTM, BiLSTM and CNN-LSTM.
5.4 Application scenarios
Traffic crash is one of the major reasons for the occurrence of traffic jam, thereby causing flash crowd traffic, which further leads to a sudden high overhead of communication for the neighbor base stations, as depicted in Fig. 14.
Therefore, our proposal, DA-FFL, can be applied to the edge computing of roadside units (RSU) with the following reasons. First, edge servers are in charge of the prediction of crash risk information through real-time weather data and traffic data. Then, the predicted crash risk information is dispensed to nearby vehicles for future actions through road-vehicle communication. Finally, the crash risk information is uploaded to the cloud for precise regulation by the traffic management authority (TMA).
Therefore, real-time crash risk can be provided to help drivers or autonomous vehicles to make more rational decisions. Our approach can reduce traffic crashes and improves road safety by predicting road crash risk. Meanwhile, it can reduce the flash crowd traffic caused by crashes, thus reducing the incidence of local roadside communication base station congestion and ensuring safe and smooth road and communication during emergencies.
Fig. 14 Relationship between models and flash crowd traffic and communication congestion.
6 Discussion
COVID-19 is a global problem that has affected various industries. The most direct impact on traffic is that people travel less and fewer vehicles travel on the road, but we found a deeper intrinsic impact by analyzing traffic data and infection case data during the COVID-19 pandemic. Therefore, we propose a new method to predict the risk of road crashes under large-scale emergencies, which can help TMA to improve their management capabilities and also prevent the formation of flash crowd traffic, which can lead to local communication congestion along roads.
The theoretical significance lies in the discovery of changes in traffic safety during the COVID-19 pandemic, and we first integrate the fatigue factor into the crash risk prediction model and propose a crash risk prediction model under large-scale emergency scenarios. It provides ideas and directions for future researchers to pay more attention to the impact of large-scale emergency events on traffic. The practical significance is that the research results can be used in traffic management to reduce the flash crowd traffic caused by accidents, thereby reducing the incidence of local roadside communication base station congestion and ensuring the safe and smooth flow of roads during emergencies.
Currently, there are many new research hotspots and discoveries in the field of road crash risk prediction. With the development of vehicle-road cooperation technology and autonomous driving technology, the crash risk prediction of vehicle groups will become important. However, there is no practical application for the mobile vehicle population on the road, so our study focuses on the crash risk of the whole road section.
This paper proposes innovative approaches while facing some challenges. Traffic characteristics may vary from road to road at different times, and models developed for one road cannot be directly used to predict crash risk predictions for another road. Similarly, models built based on data during the COVID-19 pandemic may not yield the same results for ordinary periods, making it a challenge to use migration learning for emergency management to help with large-scale emergencies like COVID-19. Of course, our research also faces some other challenges, such as data imbalance, conducting a larger study, etc., which are worthy of continued in-depth research.
7 Conclusion
In this paper, we identify anomalies of the crash data during the 2020 COVID-19 pandemic through data analysis. By comparing the annual average data from 2016 to 2020, we find that there is the highest number of deaths and injuries of per crash in 2020. In further analysis, we find that when the number of daily new COVID-19 cases increases rapidly, the overall number of crashes decreases. During this period, however, the average crash severity is worse than in the previous four years, and the average number of deaths per crash is higher. This is particularly evident from March to May, which coincides with the first three months of the appearance of 18 CDPs for confirmed cases of COVID-19. Subsequently, we build a road crash risk prediction model for specific large-scale emergency scenarios. Based on the data from March to May 2020, we propose fatigue-factor to further construct Fatigue-FL and DA-FFL to improve the predictive performance of the model.
Through the comparison experiments of different models, CNN-LSTM is more suitable for the proposal DA-FFL compared to LSTM and BiLSTM. And the proposed method can alleviate the flash crowd traffic on road and communication congestion on communication systems caused by road crash. In future work, we will further improve the overall performance of the model.
CRediT authorship contribution statement
Junbo Wang: Conception and design of study, Writing – original draft, Writing – review & editing. Xiusong Yang: Conception and design of study, Acquisition of data, Analysis and/or interpretation of data, Writing – original draft, Writing – review & editing. Songcan Yu: Acquisition of data. Qing Yuan: Analysis and/or interpretation of data. Zhuotao Lian: Writing – review & editing. Qinglin Yang: Writing – review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
The data that has been used is confidential.
Acknowledgment
This work is supported with National Nature Science Foundation of China, No. 62072485, and Guangdong Basic and Applied Basic Research Foundation No. 2022A1515011294. All authors approved the version of the manuscript to be published.
==== Refs
References
1 World Health Organization Global Status Report on Road Safety 2018: Summary: Technical Report 2018 World Health Organization
2 Graham D.J. Variable returns to agglomeration and the effect of road traffic congestion J. Urban Econ. 62 1 2007 103 120
3 Chen Z. Qin X. A novel method for imminent crash prediction and prevention Accid. Anal. Prev. 125 2019 320 329 30007587
4 Zhang Y. Li C. Liu Q.E. Wu W. The socioeconomic characteristics, urban built environment and household car ownership in a rapidly growing city: evidence from Zhongshan, China J. Asian Archit. Build. Eng. 17 1 2018 133 140
5 Yang L. Liu S. Liu J. Zhang Z. Wan X. Huang B. Chen Y. Zhang Y. COVID-19: immunopathogenesis and immunotherapeutics Signal Transduct. Target. Ther. 5 1 2020 1 8 32296011
6 Yao Y. Geara T.G. Shi W. Impact of COVID-19 on city-scale transportation and safety: an early experience from Detroit Smart Health 22 2021 100218
7 Jenelius E. Cebecauer M. Impacts of COVID-19 on public transport ridership in Sweden: Analysis of ticket validations, sales and passenger counts Transp. Res. Interdiscip. Perspect. 8 2020 100242
8 Hu Y. Barbour W. Samaranayake S. Work D. Impacts of Covid-19 mode shift on road traffic 2020 arXiv preprint arXiv:2005.01610
9 Sekadakis M. Katrakazas C. Michelaraki E. Kehagia F. Yannis G. Analysis of the impact of COVID-19 on collisions, fatalities and injuries using time series forecasting: The case of Greece Accid. Anal. Prev. 162 2021 106391
10 Adanu E.K. Brown D. Jones S. Parrish A. How did the COVID-19 pandemic affect road crashes and crash outcomes in Alabama? Accid. Anal. Prev. 163 2021 106428
11 Li J. Xu P. Li W. Urban road congestion patterns under the COVID-19 pandemic: A case study in Shanghai Int. J. Transp. Sci. Technol. 10 2 2021 212 222
12 Zhou H. Wang Y. Huscroft J.R. Bai K. Impacts of COVID-19 and anti-pandemic policies on urban transport—an empirical study in China Transp. Policy 110 2021 135 149
13 Yu R. Abdel-Aty M. Utilizing support vector machine in real-time crash risk evaluation Accid. Anal. Prev. 51 2013 252 259 23287112
14 Cheng Z. Yuan J. Yu B. Lu J. Zhao Y. Crash risks evaluation of urban expressways: A case study in Shanghai IEEE Trans. Intell. Transp. Syst. 2022
15 Guo M. Zhao X. Yao Y. Yan P. Su Y. Bi C. Wu D. A study of freeway crash risk prediction and interpretation based on risky driving behavior and traffic flow data Accid. Anal. Prev. 160 2021 106328
16 Alrukaibi F. AlKheder S. Sayed T. Alburait A. Injury severity influence factors and collision prediction-A case study on Kuwait highways J. Transp. Health 20 2021 101025
17 Dimitrijevic B. Khales S.D. Asadi R. Lee J. Short-term segment-level crash risk prediction using advanced data modeling with proactive and reactive crash data Appl. Sci. 12 2 2022 856
18 Zhai B. Lu J. Wang Y. Wu B. Real-time prediction of crash risk on freeways under fog conditions Int. J. Transp. Sci. Technol. 9 4 2020 287 298
19 Janiesch C. Zschech P. Heinrich K. Machine learning and deep learning Electron. Mark. 31 3 2021 685 695
20 S. Moosavi, M.H. Samavatian, S. Parthasarathy, R. Teodorescu, R. Ramnath, Accident risk prediction based on heterogeneous sparse data: New dataset and insights, in: Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, 2019, pp. 33–42.
21 Bao J. Liu P. Ukkusuri S.V. A spatiotemporal deep learning approach for citywide short-term crash risk prediction with multi-source data Accid. Anal. Prev. 122 2019 239 254 30390519
22 Li P. Abdel-Aty M. Yuan J. Real-time crash risk prediction on arterials based on LSTM-CNN Accid. Anal. Prev. 135 2020 105371
23 Yu R. Wang Y. Zou Z. Wang L. Convolutional neural networks with refined loss functions for the real-time crash risk analysis Transp. Res. C 119 2020 102740
24 T.-Y. Lin, P. Goyal, R. Girshick, K. He, P. Dollár, Focal loss for dense object detection, in: Proceedings of the IEEE International Conference on Computer Vision, 2017, pp. 2980–2988.
25 Monfort S.S. Cicchino J.B. Patton D. Weekday bicycle traffic and crash rates during the COVID-19 pandemic J. Transp. Health 23 2021 101289
26 Wang J. Wu Y. Yen N. Guo S. Cheng Z. Big data analytics for emergency communication networks: A survey IEEE Commun. Surv. Tutor. 18 3 2016 1758 1778
27 California Department of Transportation J. Performance measurement system 2022 URL: https://pems.dot.ca.gov. (Accessed 2 June 2022)
28 National Centers for Environmental Information J. National oceanic and atmospheric administration 2022 URL: https://www.ncei.noaa.gov. (Accessed 2 June 2022)
29 Safe Transportation Research and Education Center, University of California, Berkeley J. Transportation injury mapping system 2022 URL: https://tims.berkeley.edu. (Accessed 2 June 2022)
30 Los Angeles Times Data and Graphics Department J. california-coronavirus-data 2022 URL: https://github.com/datadesk/california-coronavirus-data. Accessed: 2 June 2022)
31 He H. Garcia E.A. Learning from imbalanced data IEEE Trans. Knowl. Data Eng. 21 9 2009 1263 1284
32 Friswell R. Williamson A. Comparison of the fatigue experiences of short haul light and long distance heavy vehicle drivers Saf. Sci. 57 2013 203 213
33 Myers C. Rabiner L. Rosenberg A. Performance tradeoffs in dynamic time warping algorithms for isolated word recognition IEEE Trans. Acoust. Speech Signal Process. 28 6 1980 623 635
34 Press W.H. Teukolsky S.A. Savitzky-golay smoothing filters Comput. Phys. 4 6 1990 669 672
35 Benesty J. Chen J. Huang Y. Cohen I. Pearson correlation coefficient Noise Reduction in Speech Processing 2009 Springer 1 4
36 Massey F.J. Jr. The Kolmogorov-Smirnov test for goodness of fit J. Amer. Statist. Assoc. 46 253 1951 68 78
37 LeCun Y. Boser B. Denker J.S. Henderson D. Howard R.E. Hubbard W. Jackel L.D. Backpropagation applied to handwritten zip code recognition Neural Comput. 1 4 1989 541 551
38 Zweig M.H. Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine Clin. Chem. 39 4 1993 561 577 8472349
39 Bradley A.P. The use of the area under the ROC curve in the evaluation of machine learning algorithms Pattern Recognit. 30 7 1997 1145 1159
40 Graves A. Long short-term memory Supervised Sequence Labelling with Recurrent Neural Networks 2012 Springer 37 45
| 36506874 | PMC9726210 | NO-CC CODE | 2022-12-10 23:15:23 | no | Comput Commun. 2023 Jan 15; 198:195-205 | utf-8 | Comput Commun | 2,022 | 10.1016/j.comcom.2022.12.002 | oa_other |
==== Front
J Infect Public Health
J Infect Public Health
Journal of Infection and Public Health
1876-0341
1876-035X
Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
S1876-0341(22)00345-8
10.1016/j.jiph.2022.12.005
Original Article
Neurological Manifestations of hospitalized patients with mild to moderate infection with SARS-CoV-2 Omicron variant in Shanghai, China
Shen Xiaolei a1
Wang Ping a1
Shen Jun a
Jiang Yuhan a
Wu Li a
Nie Xin b
Liu Jianren a⁎
Chen Wei a⁎
a Department of Neurology, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, 200011
b Biostatistics Office of Clinical Research Unit, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, 200011
⁎ Corresponding authors.
1 contributed equally to this work.
7 12 2022
7 12 2022
25 8 2022
2 12 2022
5 12 2022
© 2022 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
Accumulating studies demonstrated that patients with coronavirus disease 2019(COVID-19) could develop a variety of neurological manifestations and long-term neurological sequelae, which may be different from the strains. At the peak of the Omicron variant outbreak in Shanghai, China, no relevant epidemiological data about neurological manifestations associated with this strain was reported.
Objective
To investigate neurological manifestations and related clinical features in patients with mild to moderate COVID-19 patients with Omicron variant.
Methods
A self-designed clinical information registration form was used to gather the neurological manifestations of mild to moderate COVID-19 patients admitted to Shanghai Hospital from April 18, 2022 to June 1, 2022. Demographics, clinical presentations, laboratory findings, treatments and clinical outcomes were compared between patients with and without neurological manifestations.
Results
One hundred sixty-nine(48.1%) of 351 patients diagnosed with mild to moderate COVID-19 exhibited neurological manifestations, the most common of which were fatigue/weakness(25.1%) and myalgia(20.7%), whereas acute cerebrovascular disease(0.9%), impaired consciousness(0.6%) and seizure(0.6%) were rare. Younger age(p=0.001), female gender(p=0.026) and without anticoagulant medication(p=0.042) were associated with increasing proportions of neurological manifestations as revealed by multivariate logistic regressions. Patients with neurological manifestations had lower creatine kinase and myoglobin levels, as well as higher proportion of patchy shadowing on chest scan. Vaccination status, clinical classification of COVID-19 and clinical outcomes were similar between the two groups.
Conclusions
Nearly half of the involved patients have neurological manifestations which were relatively subjective and closely associated with younger age, female gender and without anticoagulation. Patients with neurologic manifestations may be accompanied by increased lung patchy shadowing.
Keywords
COVID-19
Omicron
neurological manifestations
SARS-CoV-2
subjective symptoms
==== Body
pmc1 Introduction
Since the first case of coronavirus disease 2019(COVID-2019) was identified in December 2019 in Hubei province of China, the respiratory virus has swept all over the world. COVID-2019 is a highly contagious disease with asymptomatic infection during the incubation period, and can be transmitted through respiratory droplets, contact and aerosols. To date, multiple variants of severe acute respiratory syndrome coronavirus 2(SARS-CoV-2) have emerged, including Alpha, Beta, Gamma, Zeta, Delta, Lambda and Omicron. The Omicron variant was first reported by South Africa on November 24, 2021, and quickly spread across many countries, including China. Since early 2022, the fast spread of the SARS-CoV-2 Omicron variant has fueled a surge in newly-diagnosed cases across China, with the majority occurring in Shanghai. According to the Shanghai Municipal Health Commission, as of May 4, 2022, more than 600,000 people have been infected, most of them with the Omicron BA.2 variant.[1].
The primary target of SARS-CoV-2 is respiratory system, resulting in the most common clinical symptoms such as cough, dyspnea and fever[2]. Since the initial clinical descriptions of COVID-19, there has been accumulating evidence of the potential neurologic involvement of SARS-CoV-2, which is demonstrated by symptoms like fatigue/weakness, olfactory dysfunction, myalgia, headache and cerebrovascular events[3], [4]. The pathogenesis of neurological manifestation after COVID-19 is elusive. SARS-CoV-2 is known to have neurotropic properties. The SARS-CoV virus's spike protein S1 allows it to connect to the cell membrane by engaging with the host angiotensin-converting enzyme 2(ACE2) receptor. The nervous system has been shown to express ACE2 receptors, making it a possible target for COVID-19. Furthermore, there are various plausible mechanisms such as Blood-Brain Barrier(BBB) damage, direct nerve infection, hypoxia produced by micro-thrombus, and immune-mediated by cytokine storm.
Previous studies have indicated that although Omicron evolves towards being more transmissible, its pathogenicity and clinical severity appear to be weakened compared with previous strains [5], [6]. The frequency of neurological manifestations associated with COVID-19 had quite a large variation around the world during the pandemic. Mao et. al [7] first reported that 36.4% of hospitalized COVID-19 patients with the Alpha strain in Wuhan, China had neurological manifestations which were related to the severity of infection. At the same time, Yan et al. reported that 30.3% of non-critically ill patients in Fangcang Shelter Hospital had neurological manifestations[8]. Similar studies, however, were reported during the pandemic all over the world with a frequency variation of 34.7% to 69.3% from 2019 to 2021[9], [10], [11]. It is still debatable whether the variations are related to the different strains or the severity of the infection. Nonetheless, little is known about the neurological manifestations of patients infected with the Omicron variant, particularly in Shanghai, China.
According to China's zero-COVID policy, symptomatic patients should be treated in designated hospitals with more medical resources and attention in order to avoid exacerbation [12]. Since severe/critical cases were usually transferred to Intensive Care Unit (ICU) where it was hard to finish the anamnestic interview. In the present study, we aimed to investigate the incidence and clinical characteristics of neurological manifestations of mild to moderate COVID-19 patients hospitalized in general wards in our designated hospital. In addition, inflammatory and coagulation indicators in blood associated with the occurrence of neurological manifestations were explored.
2 Materials And Methods
2.1 Patients
The study was designed as a cross-sectional study carried out in patients affected by COVID-19 during the current Omicron wave in Shanghai. All patients were hospitalized in general wards of designated hospital in Shanghai for COVID-19, between 18 April 2022 and 01 June 2022. Eligible patients affected by the mild to moderate respiratory form of COVID-19 were defined as positive results for SARS-CoV-2 on real-time reverse-transcriptase-polymerase-chain-reaction(RT-PCR) assay of nasal and pharyngeal swab specimens and had to meet the following inclusion criteria: be over 18-years-old, conscious, cognitively and mentally conserved, and linguistically competent to respond to the anamnestic interview at discharge and one-month follow-up. Patients were consecutive enrolled in our general wards. Drop outs happened when the patients needed to be transferred to ICU.
2.2 Data collection
The medical records and extracted data on general characteristics, laboratory tests, chest computed tomography(CT) reports, treatment, prognosis, and RT-PCR dynamic detection of SARS-CoV-2 during hospitalization were recorded. Data regarding treatment and prognosis was updated until 01 June 2022. At the time of admission or during hospitalization, patients with COVID-19 fulfilling the entry criteria for the study underwent an anamnestic interview to investigate the presence of symptoms. The interview consisted of items including personal information, measured anthropometric parameters, vaccination status, and medical history. All symptoms were mainly subjectively expressed by patients and were further classified into nervous system(NS) symptoms, typical symptoms and gastrointestinal symptoms according to the previous publications. To be more specific, fatigue/weakness, headache, dizziness, emotional disorder, impaired consciousness, acute cerebrovascular disease and seizure were defined as central nervous system(CNS) symptoms; myalgia, taste impairment, smell impairment, vision impairment and neuralgia were defined as peripheral nervous system(PNS) symptoms. Symptoms like fatigue/weakness, myalgia and headache varied widely between studies due to their subjective nature and self-reporting. We selected to include them in the neurological manifestations based on the systematic review and meta-analysis[4]. These symptoms were also considered as persistent neurological manifestations in long COVID-19 syndrome according to recent review[13]. All anamnestic interviews were collected by clinicians from the general wards where patients were hospitalized, and reviewed and confirmed by two trained neurologists.
Chest CT scans were assessed in all of the patients. The clinical severity of COVID-19 was classified by reports of chest CT according to the latest version of the guidelines[12]. In addition, we classified and summarized the specific lesions described in the CT report.
Blood samples from most patients on admission were collected. Routine blood biochemistry including total white blood cell(WBC), lymphocyte and monocyte count, percentages of neutrophil, hemoglobin, platelet count, C-reactive protein(CRP), total bilirubin(TBil) and albumin; coagulation function including D-dimer, prothrombin time(PT), fibrinogen and activated partial thromboplastin time(APTT); myocardial enzymes including Creatine kinase(CK), lactate dehydrogenase and myohemoglobin.
Test reports and treatments were available in the medical history system. After patients' discharge, complications and prognosis(duration of positive to negative, treatment to negative and hospitalization) of each patient were summarized according to the discharge diagnosis along with the medical history and laboratory tests. The discharge criteria for patients were that the two consecutive RT-PCR test results, which were performed with an interval of at least 24 hours, in both N gene and ORF gene of SARS-CoV-2 were over 35 Cycle threshold(Ct).
One month after discharge, all patients received a phone call from the clinical researchers to ask questions pertaining to clinical symptoms, existing comorbidities, and issues related to mobility, self-care, and the ability to perform everyday activities.
2.3 Statistical analysis
Statistical Package for the Social Sciences for Windows(SPSS version 26.0; IBM Corp, Armonk, NY, USA) was used to perform the statistical analyses. Categorical data was expressed as absolute frequencies and percentages where appropriate. Continuous variables were expressed as the mean±SD or medians(interquartile ranges [IQR], Q1-Q3). Categorical data was compared using Pearson's chi-square test or Fisher's exact test. Mann-Whitney U test and logistic regression were used to compare the differences for continuous variables between groups. Multivariate analysis was carried out using binary logistic regressions(significant variables from univariate analyses and confounding factors were entered into the logistic regression analysis model). Odds ratios and 95% confidence intervals were estimated. A p-value (two-sided) less than 0.05 was considered significant.
3 Results
3.1 Demographic and Clinical Characteristics
Patients were consecutively enrolled in our general wards. Nine cases were classified as severe/critical COVID-19, and they were transferred to ICU and dropped out in the present study. One hundred and sixty-nine(48.1%) out of 351 patients with a confirmed SARS-CoV2 infection had neurological manifestations. The demographic and clinical characteristics were shown in Table 1. Demographically, COVID-19 patients with neurological manifestations were younger(67.4 vs 73.7, p＜0.001). Female(60.9% vs 48.9%, p=0.025) made up the majority of patients with neurological manifestations. Although the comorbidities(76.9% vs 88.5%, p=0.005), vaccination status(46.7% vs 34.1%, p=0.017), treatment of anticoagulation(11.2% vs 22.2%, p=0.007) and nutritional support(24.9% vs 35.2%, p=0.037) had significant difference in univariate analysis, there were only three independent relevant factors for neurological manifestations: younger age(p=0.001), female gender(p=0.026), and without anticoagulant medication(p=0.042) after multivariate regression. There was no statistically significant difference in the severity of COVID-19 and clinical outcomes between the two groups(Table 1).Table 1 Demographic and clinical characteristics in mild to moderate patients with SARS-CoV-2 Omicron based on neurological manifestations.
Table 1 n(%) or mean(SD) or median[IQR] p- value Multivariate regressiona
Total With
neurological manifestations Without
neurological manifestations B (95%CI) p-value
(n=351) (n=169) (n=182)
Characteristics
Age, y 70.6(15.1) 67.4(15.7) 73.7(14.0) ＜0.001⁎⁎⁎ -0.026(0.959, 0.989) 0.001⁎⁎
Gender
Female 192(54.7) 103(60.9) 89(48.9) 0.025⁎ 0.498(1.060, 2.555) 0.026⁎
Male 159(45.3) 66(39.1) 93(51.1)
BMI 23.1(3.4) 22.9(3.4) 23.2(3.4) 0.673
BMI＞24 126(36.8) 59(34.9) 67(38.7) 0.587
Comorbidities
Any 291(84.6) 130(76.9) 161(88.5) 0.005⁎⁎ -0.420(0.320, 1.350) 0.253
Hypertension 186(53.0) 91(53.8) 95(52.2) 0.831
Diabetes 67(19.1) 25(14.8) 42(23.1) 0.057 -0.286(0.405, 1.396) 0.366
Cardiac disease 109(31.1) 53(31.4) 56(30.8) 0.909
Cerebrovascular disease 73(20.8） 36(21.3) 37(20.3) 0.895
Chronic lung disease 51(14.5) 30(17.8) 21(11.5) 0.129
Comorbidities ≥ 2 201(57.3) 88(52.1) 113(62.1) 0.067 0.168(0.677, 2.068) 0.554
Vaccinated 141(40.2) 79(46.7) 62(34.1) 0.017⁎ 0.277(0.721, 2.413) 0.369
Booster injection 77(22.1) 46(27.4) 31(17.1) 0.028⁎ -0.180(0.414, 1.685) 0.615
Degree of severity
Mild 286(81.5) 132(78.1) 154(84.6) 0.131
Moderate 65(18.5) 37(21.9) 28(15.4)
Treatment
Traditional Chinese Medicine 328(93.4) 161(95.3） 167(91.8） 0.202
Antibiotics 78(22.3） 38(22.5） 40(22.1） 1.000
Anticoagulation 59(16.9） 19(11.2） 40(22.2） 0.007⁎⁎ 0.634(0.288, 0.978) 0.042⁎
Glucocorticoid 27(7.7） 9(5.3) 18(10.0) 0.113
Thymosin 73(20.9) 32(18.9) 41(22.7) 0.431
Paxlovid 221(63.0) 100(59.2) 121(66.5) 0.184
Nutritional support 106(30.2) 42(24.9) 64(35.2) 0.037⁎ -0.141(0.514, 1.465) 0.597
Clinical outcomes
Onset of positive to negative, d 9(7, 12,3) 9(7, 13) 9(7, 12) 0.381
Onset of treatment to negative, d 6(2, 8) 6(2, 8) 6(3, 8) 0.364
Length of hospitalization, d 8(5, 11) 8.5(5, 11) 8(5, 11） 0.752
Follow-up
Residual abnormalities 25(7.1) 16(9.5) 9(4.9) 0.145
Abbreviation: BMI, body mass index; y, year; d, day;
Note：⁎, p < 0.05; ⁎⁎, p < 0.01; ⁎⁎⁎, p < 0.001; a, Significant variables(p<0.1) from univariate analyses(comorbidities, diabetes, vaccinated, booster injection, anticoagulation and nutritional support) were entered into the logistic regression analysis model.
3.2 Clinical symptoms
In comparison to the typical symptoms(301, 85.8%), such as fever(157, 44.9%), cough(253, 72.1%), expectoration(196, 55.8%), sore throat(132, 37.6%), nasal obstruction(58, 16.5%), and runny nose(107, 30.5%), NS symptoms were the second most commonly symptoms(169, 48.1%). Fatigue/weakness(88, 25.1%) was the most common NS symptom. Other CNS(135, 38.5%) symptoms included headache(48, 13.7%), dizziness(47, 13.4%), emotional disorder(17, 4.9%), acute cerebrovascular disease(3, 0.9%), impaired consciousness(2, 0.6%) and seizure(2, 0.6%). 105 patients(29.9%) suffered PNS symptoms such as myalgia(73, 20.7%), taste impairment(20, 5.7%), smell impairment(22, 6.3%), vision impairment(19, 5.4%) and neuralgia(1, 0.4%). Of these 315 patients, 100 patients(28.5%) were found to present with gastrointestinal symptoms, including diarrhea(44, 12.5%), abdominal pain(13, 3.7%), nausea(22, 6.3%), vomiting(14, 4.0%), and poor appetite(66, 18.8%). Clinical symptoms are shown in Table 2.Table 2 Symptoms of Patients with SARS-CoV-2 Omicron(n=351).
Table 2Symptoms n(%)
Typical symptoms 301(85.8)
Fever 157(44.9)
Cough 253(72.1)
Expectoration 196(55.8)
Sore throat 132(37.6)
Nasal obstruction 58(16.5)
Runny nose 107(30.5)
Gastrointestinal symptoms 100(28.5)
Diarrhea 44(12.5)
Abdominal pain 13(3.7)
Nausea 22(6.3)
Vomiting 14(4.0)
Poor appetite 66(18.8)
NS symptoms
Any 169(48.1）
CNS 135(38.5）
Headache 48(13.7）
Dizziness 47(13.4）
Impaired consciousness 2(0.6）
Acute cerebrovascular disease 3(0.9）
Emotional Disorder 17(4.9)
Seizure 2(0.6）
Fatigue/weakness 88(25.1)
PNS 105(29.9)
Taste impairment 20(5.7)
Smell impairment 22(6.3)
Vision impairment 19(5.4)
Neuralgia 1(0.4)
Myalgia 73(20.7)
Abbreviation: NS, nervous system; CNS, central nervous system; PNS, peripheral nervous system
Subgroup analysis revealed that clinical symptoms were closely associated with age and gender( Fig. 1). In particular, PNS symptoms including myalgia and visual impairment were more prevalent in the female group and under the age of 70. Besides that, headache was more common in younger patients, whereas emotional disorders and taste impairment were more common in female patients. For other symptoms, younger patients were more likely to experience fever and sore throat, while female patients were more prone to experience cough, abdominal pain and nausea. Meanwhile, age and gender could also impact the clinical outcomes showed in Figure S1. Despite having more symptoms, older patients still required longer hospital stays. Additional symptoms may have an impact on female patients' hospitalization.Fig. 1 Symptoms of mild to moderate Patients with SARS-CoV-2 Omicron stratified by sex and age. Abbreviation: NS, neurological symptoms; CNS, central nervous symptoms; PNS, peripheral nervous symptoms Note: The age categories were divided based on the average age of enrollment.; ⁎, p < 0.05; ⁎⁎, p < 0.01; ⁎⁎⁎, p < 0.001.
Fig. 1
3.3 Laboratory findings
When performing univariate regression, blood tests showed that participants with neurological manifestations had higher albumin(40 vs. 39, p=0.008) and somewhat lower CRP(3.86 vs. 4.76, p=0.013), CK(81.5 vs. 95.5, p=0.005) and myohemoglobin (24.3 vs. 36.35, p<0.001). However, after multivariate analysis eliminated confounding variables, only CK(p=0.039) and myohemoglobin(p=0.006) appeared to be decreased in the neurological manifestations group( Table 3). The findings appeared to be at odds with PNS symptoms like myalgia related to muscular damage.Table 3 Laboratory and chest findings on admission of the mild to moderate Patients with SARS-CoV-2 Omicron based on neurological manifestations.
Table 3Items With
neurological manifestations Without
neurological manifestations Univariate regression Multivariate regressiona
(n= 169) (n= 182) B(95%CI) p-value B(95%CI) p-value
Blood test, median(25th, 75th)
WBC count, ⁎10^9/L 5.20(3.83, 6.30) 5.10(4.00, 6.20) -0.004(0.887, 1.118) 0.947 -0.006(0.880, 1.124) 0.929
Monocyte cell count, ⁎10^9/L 0.52(0.39, 0.68) 0.49(0.41, 0.68) -0.196(0.391, 1.729) 0.605 -0.022(0.452, 2.119) 0.956
Lymphocyte count, ⁎10^9/L 1.35(1.00, 1.80) 1.30(1.00, 2.00) 0.274(0.904, 1.913) 0.152 0.056(0.708, 1.580) 0.786
Neutrophil, % 58.25(49.93, 66.58) 59.40(41.86, 68.40) -0.005(0.979, 1.011) 0.534 0.000(0.983, 1.017) 0.982
Hemoglobin, g/L 131.00(123.25, 142.00) 128.00(116.00, 141.00) 0.012(0.999, 1.024) 0.062 0.012(0.997, 1.026) 0.114
Platelet count, ⁎10^9/L 189.00(154.00, 231.50) 174.00(138.00, 230.00) 0.003(1.000, 1.006) 0.081 0.001(0.998, 1.004) 0.552
C-reactive protein, mg/L 3.86(1.69, 7.49) 4.76(1.61, 15.46) -0.024(0.958, 0.995) 0.013⁎ -0.016(0.996, 1.002) 0.083
TBil, mmol/L 10.40(8.25, 13.85) 10.00(8.00, 12.90) 0.013(0.997, 1.052) 0.476 0.028(0.988, 1.070) 0.174
Albumin, g/L 40.00(38.00, 43.00) 39.00(37.00, 43.00) 0.071(1.019,1.131) 0.008⁎⁎ 0.032(0.974, 1.094) 0.286
D-dimer, mg/L 0.36(0.19, 0.73)) 0.46(0.25, 0.90) -0.202(0.639, 1.044) 0.106 -0.053(0.729, 1.238) 0.699
PT, s 10.80(10.40, 11.30) 11.00(10.50, 11.40) -0.023(0.920, 1.038) 0.449 -0.014(0.941, 1.033) 0.558
Fibrinogen, g/L 2.92(2.53, 3.52) 3.10(2.62, 3.68) -0.021(0.922, 1.040) 0.492 0.022(0.985, 1.092) 0.506
APTT, s 28.50(26.70, 30.20) 28.60(26.50, 30.80) -0.021(0.922, 1.040) 0.492 0.022(0.958, 1.092) 0.506
Creatine kinase, U/L 81.50(57.25, 108.75) 95.50(64.00, 139.00) -0.004(0.994, 0.999) 0.005⁎⁎ -0.003(0.995, 1.000) 0.039⁎
Lactate dehydrogenase, U/L 203.00(175.5, 229.75) 210.50(188.75, 238.00) -0.006(0.988, 1.000) 0.058 -0.005(0.989, 1.002) 0.146
Myohemoglobin, μg/L 24.30(16.55, 35.70) 36.35(24.10, 54.08) -0.023(0.967, 0.988) <0.001⁎⁎⁎ -0.016(0.974, 0.996) 0.006⁎⁎
Chest CT findings,n(%)
Bilateral lung involvement 69(40.8) 72(39.6) 0.053(0.688, 1.616) 0.809 0.272(0.827, 2.083) 0.249
Patchy shadowing 84(49.7) 79(43.4) 0.253(0.846, 1.962) 0.238 0.497(1.048, 2.577) 0.030⁎
Effusion shadowing 3(1.8) 17(9.3) -1.741(0.050, 0.610) 0.006⁎⁎ -1.268(0.077, 1.025) 0.054
Ground-glass opacity 10(5.9) 12(6.6) -0.115(0.375, 2.120) 0.794 -0.270(0.310, 1.879) 0.557
Interstitial abnormalities 13(7.7) 18(9.9) -0.275(0.360, 1.601) 0.469 -0.025(0.450, 2.113) 0.949
Consolidation 3(1.8) 8(4.4) -0.934(0.103, 1.507) 0.173 -0.540(0.142, 2.390) 0.453
Abbreviation: WBC, White blood cell; TBil, total bilirubin; PT, prothrombin time; APTT, activated partial thromboplastin time
Note: CI, confidence interval; ⁎, p < 0.05; ⁎⁎, p < 0.01; ⁎⁎⁎, p < 0.001; a, confounding factors(age, gender and anticoagulation) from Table 1 were entered into the logistic regression analysis model.
In a univariate analysis of the chest CT results, the proportion of effusion shadowing(1.8% vs. 9.3%, p=0.006) was increased in neurological manifestations group. Patchy shadowing( Fig. 2 a) on chest CT was still an independent associated factor of patients with neurological manifestations as revealed in multivariate analysis (p=0.030) (Table 3). The typical CT sign, ground-glass opacity(Fig. 2 b) showed no difference between the two groups.Fig. 2 Chest CT images of a COVID-19 patient with neurological manifestations Note: Axis chest CT scan showed atypical patchy shadowing(a) and patchy ground-glass opacity consistent with typical moderate COVID-19(b).
Fig. 2
4 Discussion
In the research of COVID-19, neurological manifestations have always been a hot topic. According to our knowledge, this is the first study to be conducted in China on the epidemiology of neurological symptoms following Omicron variant infection. 351 patients with the mild to moderate COVID-19 Omicron variant enrolled in a designated hospital revealed that: (1) nearly half of COVID-19 patients infected with the Omicron strain had neurological manifestations; (2) younger age, female gender and anticoagulants medication had an impact on the presence of neurological manifestations; (3) mismatched blood test results with the symptoms indicated most neurological symptoms were highly subjective and unspecific; and (4) patients with neurologic manifestations may be accompanied by increased lung patchy shadowing.
Our study showed 48.1% patient with neurological manifestations, which is higher than the study in Wuhan with the frequency of 30.3%-36.5% [7], [8] back in 2019. In the USA and Europe, similar series with an overall frequency of 34.7%-69.3%[9], [10], [11]. These contrasting results could be related to the definition of neurological manifestations, the different severity of infection and the variants of SARS-CoV-2. Definitions of neurological symptoms have varied considerably during the pandemic. The early studies of COVID-19 always focus on severe neurological impairments, such as cerebrovascular events, impaired consciousness and epilepsy, as neurological manifestations because patients had more severe symptoms during the first wave of the pandemic. However, as the virus mutates, concern about non-specific neurological symptoms like fatigue/weakness, myalgia, headaches, and dizziness have been increasing in recent studies of COVID-19 related neurological manifestations [4]. Based on all these studies, our study was designed as a cross-sectional study. To fully collect all relevant symptoms and detect neurologic manifestations that may otherwise be missed throughout the entire hospitalization period, an anamnestic interview was prospectively designed as a Yes-or-No question, which could be a reason that our data showed an outstanding increase of non-specific neurological symptoms. Non-specific neurological symptoms like fatigue/weakness and myalgia are common symptoms of any viral or bacterial infection that make researchers consider them as systemic symptoms at the beginning of the pandemic. But recently studies show that they are also common COVID-19 sequelas even after 6-months follow-up which challenge the theory that they are caused by acute viral infection and inflammation [13], indicating they may be the results of the virus's effect on the muscles and mental health. Some studies showed that COVID-19-associated inflammation might lead to neurotransmitter impairment, possibly representing the basis of fatigue and explaining mental disorder[14]. But the pathology of long COVID syndrome is still unclear. And the timing of risk for acute, subacute, and long-term neurologic manifestations remains unknown, implying that all clinical data collected at different time points should be consistent in order to obtain comparable data. Meanwhile, the severe neurological impairments such as impaired consciousness and acute cerebrovascular disease were decreased compared with the data back in Wuhan, which was consistent with the reduced rate of severe patients and pathogenicity of the virus strain[15], [16]. According to the research published on Nature[17], Omicron replication is ACE2 dependent and the binding of the spike of Omicron spike to ACE2 is enhanced compared with that of the WT virus which indicate Omicron is shown to have little effect on receptor binding affinity, but the efficiency of entry into host cells is reduced in cells expressing the TMPRSS2 protease. This mutation increased the ability of Omicron to enter the body, leading to increased infectivity and transmission and a new wave of infection around the world[18], [19]. ACE2 is widely distributed in the nervous system[20], and Omicron's reduced dependence on TMPRSS2 makes virus more easily to entry into the nervous system. Thus, compared to studies in the early stage of the pandemic, the proportion of neurological manifestations reported is getting higher.
Our study suggests that both age and gender may have an impact on the development of neurological manifestations in mild to moderate patients. Study from Liotta EM et al. [21]. suggested that younger patients (mean age of 57.9), without or had a long time from COVID-19 onset to hospitalization, were more likely to present any neurologic manifestations than older people (mean age 62.9 years). While in most studies, encephalopathy such as altered mental status and cerebrovascular events, the most frequent CNS manifestation in COVID-19, had been manifested more likely in older patients[22], [23]. However, the majority of these studies included severe patients, indicating that neurological symptoms were more likely to occur in such people. Previous research had shown that neurological symptoms[24], [25], particularly the PNS, were inversely correlated with age in mild to moderate patients, which was what we focused on. With regard to the role of gender on resistance and disease severity, there may be some gender differences in the neuropathological events caused by COVID-19, even though multiple publications found that women are at lower risk of COVID-19 severity than males in various cohorts[26], [27]. In COVID-19-positive females compared to males, chemosensory dysfunctions and subjective neurological symptoms are more prevalent[28], [29]. Although these studies have shown that gender does affect the infection of SARS-CoV-2, there are still no consensual studies that can explain this difference. Some[30], [31] believe that women acquire stronger immune responses than men, which results in faster pathogen clearance but also contributes to their increased susceptibility to strong symptoms.
The only treatment in our trial that appeared to be related to the onset of neurological symptoms was anticoagulant medication. Patients with COVID-19 frequently have a pro-coagulative condition as a result of complement cascade hyper-activation, cytokine storm, and endothelial dysfunction brought on by the virus[32]. Diffuse microvascular thrombi are frequently seen in several organs, which appear to be directly connected to the severity of the condition[33]. After review, the majority of the patients in our study who used anticoagulants were high-risk patients, and the sample size was small. Once the confounding element of age, which was most likely to change the results, was eliminated, anticoagulation still had impact on the development of neurological symptoms. Theoretical support is still lacking in mild cases, though.
In our investigation, the most of the blood tests results were within normal ranges. Inflammatory and coagulation indicators did not differ between groups. However, previous researches[3], [7], [8], [9], [10], [11] have revealed neurological manifestations that can be related to both indicators. Surprisingly, the indicators that could directly represent muscular injury also showed a negative connection between groups, indicating that symptoms like myalgia were highly subjective and had little bearing on the outcome of the disease. According to Liguori’s research[34], COVID-19 infection could cause subjective neurological symptoms, and women were more likely to experience these symptoms than men. This outcome seems to be consistent with what we discovered. But in fact, although the neurological symptoms in this study were classified as subjective symptoms, no further evidence indicated whether their occurrence was related to an objective infection or injury in nervous system. As shown in a discovery published in Nature[35], SARS-CoV-2 is linked to alterations in brain structure even in cases of mild to moderate infection. The SARS-CoV-2 can cross the Blood-Brain Barrier(BBB) and cause pathological changes including hypoxia, ischemia, micro-bleeds, and inflammation[36], [37]. These changes in the structure of the nervous system are insidious and persistent, but can result in a variety of non-specific neurological manifestations. These changes in brain could lead to the persistent neurological manifestations in long COVID syndrome as well, for the most common symptoms in COVID-19 sequelae were also highly subjective like fatigue and brain fog[13]. Further research is still needed to determine whether neurological manifestations in the early stages of infection can signal these modifications.
Pulmonary infection is an important marker of the severity of COVID-19 infection. Ground-glass opacities are characteristic CT signs of the lungs. In our study, there was no statistical difference between the two groups in terms of disease severity and the presentation of typical pulmonary CT signs. Atypical patchy shadows, however, were more common in patients with neurological manifestations. After searching the literatures, this was found for the first time in COVID-19 patients that lacks generalizability and theoretical justification. Whereas it is concerning that the Post-COVID-19 Syndrome is generally accompanied by atypical pulmonary symptoms like cough and non-specific neurological manifestations like fatigue[38], [39], [40]. Song et al. reviewed studies showed that cough could often be accompanied by chronic fatigue, cognitive impairment, or pain collection of long-term effects referred to as the post-COVID syndrome or long COVID, which persisted for weeks or months after SARS-CoV-2 infection. They suggested that neurotropism, neuroinflammation, and neuroimmunomodulation were all involved in pulmonary symptoms[41]. But there are still gaps in understanding of the mechanisms. More epidemiological and fundamental science research is required to explain the connection.
Our research represents the first cross-sectional survey of COVID-19-related neurological manifestations during the Omicron wave in Shanghai, China. In contrast to earlier studies, ours utilized a prospective design to fully collect all relevant data throughout the entire hospitalization period as well as the short-term follow-up of mild to moderate patients. The neurologists performed a face-to-face assessment of neurological manifestations, which ensured the reliability and professionalism of diagnosis. Nevertheless, we have a few limitations. This study is only a single-center study with a limited sample size. It was challenging to consistently assess the psychological status of the patients in order to identify subjective symptoms in such situations due to the personal protective equipment (PPE). No adequate neuropsychiatric scales were used in our study to formally categorize the non-specific neurological symptoms. Non-specific symptoms such as fatigue could not necessarily be CNS mediated.and it was hard to distinguish between mental/cognitive and physical fatigue in the acute phase of infection that could also make the results rather subjective. Furthermore, patients could not receive additional auxillary examinations like magnetic resonance imaging (MRI) and electroencephalography(EEG) since they were at the peak of the pandemic.
5 Conclusions
The epidemiological information and clinical traits of neurological manifestations connected to Omicron are expanded by our study based on the Chinese population. Nearly half of COVID-19 patients infected with the Omicron strain have neurological manifestations, mostly in young women. The clinical symptoms and the results from blood test do not correspond, indicating that the majority of neurological symptoms are highly subjective. Subjects with neurological manifestations may be accompanied by increased lung patchy shadowing. More mechanism research and long-term follow-up of neurological manifestations in COVID-19 are warranted in the future.
Appendix A Supplementary material
Supplementary material.
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jiph.2022.12.005.
==== Refs
References
1 Zhang X. Zhang W. Chen S. Shanghai's life-saving efforts against the current Omicron wave of the COVID-19 pandemic Lancet. 399 10340 2022 May 28 2011 2012 10.1016/S0140-6736(22)00838-8 35533708
2 Zhang H. Zhou P. Wei Y. Histopathologic Changes and SARS-CoV-2 Immunostaining in the Lung of a Patient With COVID-19 Ann Intern Med 172 9 2020 May 5 629 632 10.7326/M20-0533 32163542
3 Collantes M.E.V. Espiritu A.I. Sy M.C.C. Neurological Manifestations in COVID-19 Infection: A Systematic Review and Meta-Analysis Can J Neurol Sci 48 1 2021 Jan 66 76 10.1017/cjn.2020.146 32665054
4 Misra S. Kolappa K. Prasad M. Frequency of Neurologic Manifestations in COVID-19: A Systematic Review and Meta-analysis Neurology. 97 23 2021 Dec 7 e2269 e2281 10.1212/WNL.0000000000012930 34635561
5 Tian D. Sun Y. Xu H. Ye Q. The emergence and epidemic characteristics of the highly mutated SARS-CoV-2 Omicron variant J Med Virol 94 6 2022 Jun 2376 2383 10.1002/jmv.27643 35118687
6 Meo S.A. Meo A.S. Al-Jassir F.F. Klonoff D.C. Omicron SARS-CoV-2 new variant: global prevalence and biological and clinical characteristics Eur Rev Med Pharmacol Sci 25 24 2021 Dec 8012 8018 10.26355/eurrev_202112_27652 34982465
7 Mao L. Jin H. Wang M. Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease 2019 Wuhan, China. JAMA Neurol 77 6 2020 Jun 1 683 690 10.1001/jamaneurol.2020.1127 32275288
8 Yan N. Xu Z. Mei B. Neurological Implications of Non-critically Ill Patients With Coronavirus Disease 2019 in a Fangcang Shelter Hospital in Wuhan, China Front Neurol 11 2020 Aug 26 895 10.3389/fneur.2020.00895 32982925
9 Flores-Silva F.D. García-Grimshaw M. Valdés-Ferrer S.I. Neurologic manifestations in hospitalized patients with COVID-19 in Mexico City PLoS One 16 4 2021 Apr 8 e0247433 10.1371/journal.pone.0247433
10 Romero-Sánchez C.M. Díaz-Maroto I. Fernández-Díaz E. Neurologic manifestations in hospitalized patients with COVID-19: The ALBACOVID registry Neurology 95 8 2020 Aug 25 e1060 e1070 10.1212/WNL.0000000000009937 32482845
11 Luigetti M. Iorio R. Bentivoglio A.R. Assessment of neurological manifestations in hospitalized patients with COVID-19 Eur J Neurol 27 11 2020 Nov 2322 2328 10.1111/ene.14444 32681611
12 Wang, Haijuan Diagnosis and Treatment Protocol for COVID-19 Patients (Tentative 9th Version). Infectious Diseases & Immunity: July 2022 - Volume 2 - Issue 3 - p 135-144. http://doi.org/10.1097/ID9.0000000000000059
13 Pinzon R.T. Wijaya V.O. Jody A.A. Persistent neurological manifestations in long COVID-19 syndrome: A systematic review and meta-analysis J Infect Public Health 15 8 2022 Aug 856 869 10.1016/j.jiph.2022.06.013 35785594
14 Ortelli P. Ferrazzoli D. Sebastianelli L. Neuropsychological and neurophysiological correlates of fatigue in post-acute patients with neurological manifestations of COVID-19: Insights into a challenging symptom J Neurol Sci 420 2021 Jan 15 117271 10.1016/j.jns.2020.117271
15 Nyberg T. Ferguson N.M. Nash S.G. Comparative analysis of the risks of hospitalisation and death associated with SARS-CoV-2 omicron (B.1.1.529) and delta (B.1.617.2) variants in England: a cohort study Lancet. 399 10332 2022 Apr 2 1303 1312 10.1016/S0140-6736(22)00462-7 35305296
16 Menni C. Valdes A.M. Polidori L. Symptom prevalence, duration, and risk of hospital admission in individuals infected with SARS-CoV-2 during periods of omicron and delta variant dominance: a prospective observational study from the ZOE COVID Study Lancet. 399 10335 2022 Apr 23 1618 1624 10.1016/S0140-6736(22)00327-0 35397851
17 Hui K.P.Y. Ho J.C.W. Cheung M.C. SARS-CoV-2 Omicron variant replication in human bronchus and lung ex vivo Nature 603 7902 2022 Mar 715 720 10.1038/s41586-022-04479-6 35104836
18 Karim S.S.A. Karim Q.A. Omicron SARS-CoV-2 variant: a new chapter in the COVID-19 pandemic Lancet. 398 10317 2021 Dec 11 2126 2128 10.1016/S0140-6736(21)02758-6 34871545
19 Long B. Carius B.M. Chavez S. Clinical update on COVID-19 for the emergency clinician: Presentation and evaluation Am J Emerg Med 54 2022 Apr 46 57 10.1016/j.ajem.2022.01.028 35121478
20 Dewanjee S. Vallamkondu J. Kalra R.S. Emerging COVID-19 Neurological Manifestations: Present Outlook and Potential Neurological Challenges in COVID-19 Pandemic Mol Neurobiol 58 9 2021 Sep 4694 4715 10.1007/s12035-021-02450-6 34169443
21 Liotta E.M. Batra A. Clark J.R. Frequent neurologic manifestations and encephalopathy-associated morbidity in Covid-19 patients Ann Clin Transl Neurol 7 11 2020 Nov 2221 2230 10.1002/acn3.51210 33016619
22 Varatharaj A. Thomas N. Ellul M.A. Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study Lancet Psychiatry 7 10 2020 Oct 875 882 10.1016/S2215-0366(20)30287-X 32593341
23 Ellul M.A. Benjamin L. Singh B. Neurological associations of COVID-19 Lancet Neurol 19 9 2020 Sep 767 783 10.1016/S1474-4422(20)30221-0 32622375
24 Nouchi A. Chastang J. Miyara M. Prevalence of hyposmia and hypogeusia in 390 COVID-19 hospitalized patients and outpatients: a cross-sectional study Eur J Clin Microbiol Infect Dis 40 4 2021 Apr 691 697 10.1007/s10096-020-04056-7 33033955
25 Ferraro S. Tuccori M. Convertino I. Olfactory and gustatory impairments in COVID-19 patients: Role in early diagnosis and interferences by concomitant drugs Br J Clin Pharmacol 87 5 2021 May 2186 2188 10.1111/bcp.14634 33185930
26 Jin J.M. Bai P. He W. Gender Differences in Patients With COVID-19: Focus on Severity and Mortality Front Public Health 8 2020 Apr 29 152 10.3389/fpubh.2020.00152 32411652
27 Onder G. Rezza G. Brusaferro S. Case-Fatality Rate and Characteristics of Patients Dying in Relation to COVID-19 in Italy JAMA 323 18 2020 May 12 1775 1776 10.1001/jama.2020.4683 32203977
28 Ferraro S. Tuccori M. Convertino I. Olfactory and gustatory impairments in COVID-19 patients: Role in early diagnosis and interferences by concomitant drugs Br J Clin Pharmacol 87 5 2021 May 2186 2188 10.1111/bcp.14634 33185930
29 Giacomelli A. Pezzati L. Conti F. Self-reported Olfactory and Taste Disorders in Patients With Severe Acute Respiratory Coronavirus 2 Infection: A Cross-sectional Study Clin Infect Dis 71 15 2020 Jul 28 889 890 10.1093/cid/ciaa330 32215618
30 Conti P. Younes A. Coronavirus COV-19/SARS-CoV-2 affects women less than men: clinical response to viral infection J Biol Regul Homeost Agents 34 2 2020 March-April 339 343 10.23812/Editorial-Conti-3 32253888
31 Pilotto A. Masciocchi S. Volonghi I. Clinical Presentation and Outcomes of Severe Acute Respiratory Syndrome Coronavirus 2-Related Encephalitis: The ENCOVID Multicenter Study J Infect Dis 223 1 2021 Jan 4 28 37 10.1093/infdis/jiaa609 32986824
32 Ali M.A.M. Spinler S.A. COVID-19 and thrombosis: From bench to bedside Trends Cardiovasc Med 31 3 2021 Apr 143 160 10.1016/j.tcm.2020.12.004 33338635
33 Carfora V. Spiniello G. Ricciolino R. Anticoagulant treatment in COVID-19: a narrative review J Thromb Thrombolysis 51 3 2021 Apr 642 648 10.1007/s11239-020-02242-0 32809158
34 Liguori C. Pierantozzi M. Spanetta M. Subjective neurological symptoms frequently occur in patients with SARS-CoV2 infection Brain Behav Immun 88 2020 Aug 11 16 10.1016/j.bbi.2020.05.037 32416289
35 Douaud G. Lee S. Alfaro-Almagro F. SARS-CoV-2 is associated with changes in brain structure in UK Biobank Nature 604 7907 2022 Apr 697 707 10.1038/s41586-022-04569-5 35255491
36 Maiese A. Manetti A.C. Bosetti C. SARS-CoV-2 and the brain: A review of the current knowledge on neuropathology in COVID-19 Brain Pathol 31 6 2021 Nov e13013 10.1111/bpa.13013
37 Erickson M.A. Rhea E.M. Knopp R.C. Banks W.A. Interactions of SARS-CoV-2 with the Blood-Brain Barrier Int J Mol Sci 22 5 2021 Mar 6 2681 10.3390/ijms22052681 33800954
38 Moghimi N. Di Napoli M. Biller J. The Neurological Manifestations of Post-Acute Sequelae of SARS-CoV-2 infection Curr Neurol Neurosci Rep 21 9 2021 Jun 28 44 10.1007/s11910-021-01130-1 34181102
39 Ceban F. Ling S. Lui L.M.W. Fatigue and cognitive impairment in Post-COVID-19 Syndrome: A systematic review and meta-analysis Brain Behav Immun 101 2022 Mar 93 135 10.1016/j.bbi.2021.12.020 34973396
40 Esendağli D. Yilmaz A. Akçay Ş. Özlü T. Post-COVID syndrome: pulmonary complications Turk J Med Sci 51 SI-1 2021 Dec 17 3359 3371 10.3906/sag-2106-238 34284532
41 Song W.J. Hui C.K.M. Hull J.H. Confronting COVID-19-associated cough and the post-COVID syndrome: role of viral neurotropism, neuroinflammation, and neuroimmune responses Lancet Respir Med 9 5 2021 May 533 544 10.1016/S2213-2600(21)00125-9 33857435
| 0 | PMC9726211 | NO-CC CODE | 2022-12-08 23:18:16 | no | J Infect Public Health. 2023 Feb 7; 16(2):155-162 | utf-8 | J Infect Public Health | 2,022 | 10.1016/j.jiph.2022.12.005 | oa_other |
==== Front
Appl Soft Comput
Appl Soft Comput
Applied Soft Computing
1568-4946
1872-9681
Elsevier B.V.
S1568-4946(22)00955-3
10.1016/j.asoc.2022.109906
109906
Article
Detection of Covid-19 and other pneumonia cases from CT and X-ray chest images using deep learning based on feature reuse residual block and depthwise dilated convolutions neural network
Celik Gaffari
Agri Ibrahim Cecen University, Department of Computer Technology, Agri, Turkey
7 12 2022
1 2023
7 12 2022
133 109906109906
26 3 2022
29 11 2022
1 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Covid-19 has become a worldwide epidemic which has caused the death of millions in a very short time. This disease, which is transmitted rapidly, has mutated and different variations have emerged. Early diagnosis is important to prevent the spread of this disease. In this study, a new deep learning-based architecture is proposed for rapid detection of Covid-19 and other symptoms using CT and X-ray chest images. This method, called CovidDWNet, is based on a structure based on feature reuse residual block (FRB) and depthwise dilated convolutions (DDC) units. The FRB and DDC units efficiently acquired various features in the chest scan images and it was seen that the proposed architecture significantly improved its performance. In addition, the feature maps obtained with the CovidDWNet architecture were estimated with the Gradient boosting (GB) algorithm. With the CovidDWNet+GB architecture, which is a combination of CovidDWNet and GB, a performance increase of approximately 7% in CT images and between 3% and 4% in X-ray images has been achieved. The CovidDWNet+GB architecture achieved the highest success compared to other architectures, with 99.84% and 100% accuracy rates, respectively, on different datasets containing binary class (Covid-19 and Normal) CT images. Similarly, the proposed architecture showed the highest success with 96.81% accuracy in multi-class (Covid-19, Lung Opacity, Normal and Viral Pneumonia) X-ray images and 96.32% accuracy in the dataset containing X-ray and CT images. When the time to predict the disease in CT or X-ray images is examined, it is possible to say that it has a high speed because the CovidDWNet+GB method predicts thousands of images within seconds.
Keywords
Covid-19 diagnosis
Deep learning
Depthwise dilated convolutions
Feature reuse residual block
Gradient boosting
==== Body
pmcCode metadata
Permanent link to reproducible Capsule: https://doi.org/10.24433/CO.2183919.v1.
1 Introduction
Coronavirus (Covid-19) is a disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-COV-2). After emerging in Wuhan, China in December 2019, it soon spread around the world and became a global pandemic [1]. According to the data of the World Health Organization (WHO), it has been determined that more than 410 million cases have been seen so far, and close to 6 million people have died [2]. WHO declared the coronavirus infection as a Covid-19 pandemic in March 2020 due to the increasing number of deaths and cases. Due to the increasing cases and deaths, many states have had to close their borders to prevent the spread of the pandemic. In addition, many countries have imposed curfews for a certain period as a precaution [3].
This disease usually affects the respiratory system, such as the lungs, and also appears to cause pneumonia-like symptoms [4]. Patients commonly experience symptoms such as fever, cough, sneezing, and shortness of breath. It spreads rapidly through respiratory droplets produced by the cough or sneeze of an infected person. Elderly people and people with chronic illnesses appear to be more prone to Covid-19 infection [5].
One of the most common methods used to diagnose Covid-19 is reverse transcription-polymerase chain reaction (RT-PCR) tests. These tests are performed to determine whether individuals have been infected with SARS-COV-2, the virus that causes Covid-19 disease, momentarily or in the past. The disadvantages of these tests are that test results take time, the number of available RT-PCR test kits is low, and the risk of health personnel contracting the disease during the test is high [6]. It is also costly in that special equipment, materials, and tools are often required for RT-PCR examinations. Therefore, many countries have difficulties in procuring test kits due to budgetary and technical constraints [7]. At the same time, the sensitivity of the RT-PCR test is a cause for concern because of sample and laboratory errors that may occur [8], [9]. Liu et al. [10] have expressed their opinion on the poor performance of RT-PCR in its sensitivity. Similarly, in a study conducted by Drame et al. [11], they expressed their reservations about the use of RT-PCR to determine the viral load in the diagnosis of 2019 coronavirus disease (Covid-19). In addition, it was stated in another study that the sensitivity of these tests could be as low as 38% [12].
Covid-19, which manifests itself as a lung infection, computed tomography (CT) and chest X-ray (X-ray) images are other methods used for the detection of this disease [5]. Typical radiographic features can be reliably detected in patients with pneumonia caused by this disease with CT imaging. Although these methods have some advantages over RT-PCR testing in terms of early detection of Covid-19, specialist physicians are needed to understand and make sense of images. Considering the disadvantages of RT-PCR tests, CT and X-ray imaging techniques used in the diagnosis of Covid-19, Artificial Intelligence (AI), and Deep Learning (DL) based methods are seen as alternative methods. AI and DL methods can help the early diagnosis of this disease and make the treatment process faster by leading experts to reach a fast and accurate diagnosis through CT and X-ray images in the detection process of Covid-19 [13], [14], [15].
Artificial intelligence and deep learning methods are widely used by researchers for the detection of Covid-19 infection from X-ray and CT images. Due to the improved performance of deep learning methods, they are widely used compared to traditional methods. One of the most important reasons that led researchers to this field is that, unlike machine learning and traditional methods, in deep learning architectures, there is no need for feature extraction in the data during the preprocessing stage. Deep learning architectures can be trained with the help of the hyperparameters of the convolutional neural network (CNN) architecture to learn the best features according to the dataset used [3]. Researchers used deep learning methods in many areas classification of white blood cells [16], segmentation of brain MRI images [17], synthetic image generation [18], generating images from EEG signals [19], skin cancer classification [20], fundus image segmentation [21], diagnose different types of Otitis media [22], breast cancer detection [23], breast lymph node segmentation [24], brain disease classification [25], lung segmentation [26], [27], detection of arrhythmia [28], [29], [30] and detecting pneumonia from chest X-ray images [31]. With the pandemic, the use of deep learning methods for the detection of coronavirus symptoms from X-ray and CT images has increased significantly [3].
In literature, it can be seen that many deep learning-based studies have been carried out for the diagnosis of Covid-19 with the help of radiological images [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42].
In the study by Leracitano et al. [32], The authors proposed a fuzzy logic-based deep learning approach to differentiate X-ray images of patients with Covid-19 pneumonia and non-Covid-19-related interstitial pneumonia. The model developed here, called CovNNet, uses the blurry edge detection algorithm together with the blurry images to extract some relevant features from the X-ray images. This study [32] for the detection of Covid-19 from binary class (Covid, Non-Covid) X-ray images, performed poorly compared to many other studies in the literature, with an accuracy rate of 81%. Ahamed et al. [33] proposed a deep learning-based Covid-19 case detection model trained with a dataset of chest CT scans and X-ray images. A modified ResNet50V2 architecture is used as a deep learning architecture in the proposed model. High performance was achieved in this study using two, three, and four classes CT and X-ray images. However, a complex architecture with high processing power was used. Verma et al. [34] proposed different models such as vanilla (vanilla) LSTM, stacked LSTM, ED_LSTM, BiLSTM, CNN, and hybrid CNN+LSTM model to capture the complex trend of the COVID-19 outbreak and perform the Covid-19 prediction. In another study by Khan et al. [35], two new deep learning-based models named deep hybrid learning (DHL) and deep boosted hybrid learning (DBHL) are proposed for effective Covid-19 detection in X-ray datasets. In the proposed DHL architecture, the representation learning capability of the two developed COVID-RENet-1 & 2 models and a machine learning classifier is used separately. In the Covid-RENet model, region and edge-based attention mechanisms were applied to extract boundary features and learn region homogeneity. In addition, the transfer learning method was used in chest X-rays in the proposed architectures. In this study, in which two-class (Covid, Non-Covid) X-ray images are used, it is seen that it has an accuracy of 98.53%. In this study, performance evaluation with only binary class X-ray images is seen as a disadvantage in terms of the performance of the architectures. Because it is important to use different data sets for Covid-19 detection. The success of CNN architectures may vary according to the number of classes and image type.
In the study by Loey et al. [36], a bayesian optimization-based CNN model was proposed for the classification of chest X-ray images. In the proposed model, CNN architecture is used to extract and learn deep features. In addition, CNN hyperparameters are adjusted according to an objective function using a Bayes-based optimizer method. In another study by Lahsaini et al. [37], they used a dataset of Covid and non-Covid CT images validated by RT-PCR tests at Tlemcen hospital in Algeria. A comparative study was carried out on Inception, Resnet-V2, VGG16, VGG19, DenseNet121, DenseNet201, ImageNet, and Xception deep models using the transfer learning method. Also, a model based on DenseNet201 architecture and the GradCam algorithm is proposed. In another study by Toğaçar et al. [38], images were preprocessed using the fuzzy color technique to classify X-ray images. Then, the features obtained with MobileNetV2, and SqueezeNet models were processed with the help of the social mimic optimization method. The productive features obtained were classified using support vector machines (SVM). The DarkCovidNet method developed by Ozturk et al. [39] was used as a classifier for the YOLO real-time object detection system. By applying seventeen convolution layers and adding different filtering to each layer. As in previous studies for the detection of Covid-19, only CT in [37] and only X-ray images were used in [38], [39].
In addition, when the studies were examined, different models were developed by the researchers, defined by the names CoroNet [40], CovidXrayNet [41], and CovXNet [42]. The CoroNet [40] model, which is proposed as a deep CNN model, is based on the Xception architecture pre-trained on the ImageNet dataset. The CovidXrayNet [41] method, based on the EfficientNet-B0 model and based on the optimization method, is proposed. In this study [41], the data augmentation method is used to increase accuracy and CNN hyperparameters are optimized. In the CovXNet [42] technique, deep CNN-based architecture and a model that uses depthwise convolution and varying dilation rates to extract features efficiently are proposed. In the proposed method, different forms of CovXNets are designed and trained with X-ray images of various resolutions. In addition, a stacking algorithm was used to increase the performance rate, and abnormal regions of X-ray images were distinguished by integrating a gradient-based discriminative localization. Looking at the time complexity of the CovXNets architecture (Fig. 10), it was seen that it had a higher time complexity compared to the other architectures studied. This shows that the CovXNets architecture has a complex structure.Fig. 1 Example images are included in the datasets.
When the studies for the detection of Covid-19 disease with deep learning methods are examined, it is seen that researchers generally use X-ray [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53] or CT [54], [55], [56], [57], [58] images, but few studies use both X-ray and CT images [59], [60], [61], [62]. At the same time, it has been seen that the researchers only examined the performance of their architectures on the dataset used or compared them only with traditional architectures. In addition, it has been determined that no performance evaluation has been made according to the training and test times in the literature reviews. In our study, contrary to these studies, CT and X-ray images were used, and a performance evaluation was carried out on the same dataset, including traditional architectures as well as different current architectures. In addition, performance evaluation was made by considering the training and test times of the architectures. The proposed model was developed to reduce the workload of specialist physicians by providing effective, efficient, and rapid detection of Covid-19 and similar cases. With the thought that the CovidDWNet+GB architecture will guide different studies, it has been opened to everyone’s access on the Github page (https://github.com/GaffariCelik/Covid-19).
Our main contributions to this work are listed as follows:
• A new deep learning-based model (CovidDWNet) has been proposed for the detection of Covid-19 and other pneumonia cases.
• The performance of the CovidDWNet architecture has been increased by using multiple feature reuse residual blocks and depthwise dilated convolutions neural networks. In addition, the success rate has been increased by performing the disease prediction process with the Gradient boosting algorithm of the feature vectors obtained with the CovidDWNet architecture.
• By using different CT and X-ray datasets, a real performance evaluation was made among the current architectures in the literature.
2 Material
In this study, three datasets were used: Covid-CT [63] and Sars-Cov-2 [64] datasets containing CT images, and Dataset-X-ray [65] dataset containing X-ray images. These datasets have been made publicly available for researchers to carry out their work. Example images in datasets are given in Fig. 1.
The Covid-CT [63] consists of 812 CT images, 349 of which are Covid-19 and 463 normal, taken from 216 patients. This dataset has been validated by a senior radiologist at Tongji Hospital in Wuhan, China, who diagnosed and treated a large number of Covid-19 patients at the time of emergence of this disease between January and April 2019. The Sars-Cov-2 [64] dataset contains a total of 2482 CT chest scan images, of which 1252 are Covid-19 and 1230 are normal. This dataset was obtained from different hospitals in Sao Paulo, Brazil. The Database-X-ray [65] dataset was created for COVID-19 positive cases with the collaboration of a team of researchers from the University of Qatar, Dhaka University, Bangladesh, and medical doctors in Pakistan and Malaysia. This dataset includes X-ray images of Covid-19, normal and other lung infection diseases. It consists of 21165 X-ray images in total, including 3616 Covid-19, 10192 normal, 6012 lung opacity (Non-Covid lung infection), and 1345 Viral pneumonia.
3 Method
As a method, a CNN-based architecture has been proposed for the detection of Covid-19 and other pneumonia symptoms. This architecture is a method based on feature reuse residual block (FRB) and depthwise dilated convolutions (DDC) units.
Convolutional Neural Networks (CNNs) are models that provide high classification performance in multi-class problems and have self-learning capabilities. CNNs are coordinated combinations of multilayer perceptrons, in which every neuron in one layer is associated with all neurons in the next. A convolutional network consists of a convolutional layer and a rectified linear unit (ReLu). Convolution layers form the basic structure of CNN models. Inputs are convolutionalized and applied with nuclei across the entire visual field with convolution filters. Thus, simpler, small patterns are obtained with more complex, detailed patterns. In this way, the hierarchical network structure enables the extraction of the highest feature maps, enhanced generalization capability, and reduced computational complexity [15], [53], [66], [67]. The basic convolution operation can be written mathematically as [33]: (1) Fi,j=I∗Ki,j=∑m∑nI(i+m,j+n)K(m,n)
Here, I represents the input matrix (may be an image), mxn filter size, and K the two-dimensional filter.
3.1 Feature reuse residual block (FRB)
The feature reuse block (FRB) is a widely used technique in computer vision. In this method, feature maps of previous layers are given as input to all subsequent layers. Thus, the performance of the network is highly increased by reusing the features of the previous layers in all subsequent layers [68], [69]. The mathematical formulation of the FRB technique used in this study is defined as follows: (2) FRB=y=Fx,wiox
Here, x and y represent the inputs and outputs of the considered layers, wi weights, and the combination of those features. The architecture of the FRB technique is presented in Fig. 2(a).
Fig. 2 Feature reuse residual block architecture-(a) and depthwise dilated convolutions architecture-(b), which constitute the basic structure of the proposed architecture.
3.2 Depthwise dilated convolutions (DDC)
As shown in Fig. 3, dilated convolutions can be expanded in the scope of the convolution kernel by changing the dilation ratio compared to standard convolution. By expanding the scope of the convolution kernel, multi-scale features (information) can be obtained. Choosing a dilation rate of one captures the same properties as standard convolution. However, in dilated convolution, when the dilation ratio is selected as greater than one, more detailed features can be obtained than in standard convolution [70], [71], [72].
In depthwise dilated convolutions operation, the convolution operation is applied to each input channel separately. With point convolution (conventional convolution with 1 × 1 window), inter-channel features are projected into a new space. More efficient features are obtained by using a combination of 1 × 1 convolution and 3 × 3 deep convolution instead of 3 × 3 standard convolution. Therefore, various spatial information is extracted from local information to broader generalized information. In this way, different features extracted with varying dilation ratios with different convolution operations will result in greater diversity in the feature extraction process [42], [73], [74]. The mathematical formulation of depthwise convolution is as follows: (3) DeptwiseConv(W,y)(i,j)=∑k,lK,LW(k,l)⊙y(i+k,j+l)
Fig. 3 Dilated Convolution, covering different areas for different dilation ratios when the kernel size (3×3) is selected.
Here, y represents layer, K, and L layer size, i and j layer index, and k, and l filter respectively. W denotes a learnable convolution filter and ⊙ an element-wise multiplication operator.
In addition, if the pneumonia disease has spread over a larger area rather than just one region in X-ray images, it may be necessary to combine features from different observation levels. Therefore, the depthwise dilated convolution technique can be used effectively in the diagnosis of pneumonia [42] (see Fig. 4).
Fig. 4 (a) Normal convolution and (b) depthwise convolutions operations. In depthwise convolutions, the number of filters is equal to the number of channels of the input [75].
3.3 Gradient boosting (GB)
Gradient boosting (GB) is a machine learning algorithm used for classification and regression problems [76]. GB aims to combine strong learner models to obtain a weak learner with high prediction accuracy [77], [78]. The GB method tries to minimize the cost function to find an additive model. Therefore, the GB algorithm iteratively adds weak learners (a new decision tree) to the model, reducing the cost function at the highest rate at each step [77]. The steps of the GB algorithm are given below mathematically [76], [77]:
1. Input variable (x) and target variable (y) are determined. The cost function (L(y,f(x))) is defined.
2. A simple decision tree (DT0) is initialized that establishes the relationship f(x) between x and y. Here, it is aimed to minimize the cost function (f(x) = DT0).
3. A pseudo-residue is defined to obtain a new target variable. The defined pseudo-residue is used as the new target variable (ri=yi−fxi,yi=ri,i=i+1).
4. A new decision tree (DTi) suitable for the pseudo-residual is developed. By including DTi in the model, f(x) is updated ((f(x) = ∑DT).
5. Step 3 and step 4 are repeated for the specified number of cycles.
6. Finally, all decision trees are combined and the result of the GB model is obtained (f(x) = ∑DT).
3.4 Proposed architecture
In this study, a new architecture named CovidDWNet+GB is proposed. This architecture consists of four blocks as shown in detail in Fig. 5. First, the input image is represented in the larger area of the input information by two successive convolution operations. In the second stage, new features are extracted after the obtained information FRB and then DDC operations. The same operations were repeated four times with different filter (f) numbers and dilation ratios (maxd) to increase the depth of the mesh. The depth of the DDC unit was determined by reducing the dilation rate by five at the beginning and by one in the next steps. Then, the obtained feature map is applied to the global average pooling layer (GAP) and three fully connected layers, respectively. In addition, after each convolution operation, Relu was used as the activation function and BatchNormalization was used as the normalization operation. Finally, after the CovidDWNet architecture was trained, the feature vectors obtained from the second fully connected (FC(64, R)) layer were estimated using Gradient Boosting (GB) machines. The proposed architecture is detailed in Fig. 5 and its methodology is shown in Eq. (4)–(5). (4) CovidDWNetx,y=CL−CL−FRBi−DDCi−CLii=1n−GAP−FC1−FC2−Zk
(5) prediction=yout=GB(FC2)
Here, CL stands for convolution layer, GAP global average pooling, FC fully connected, Zk softmax activation function, and GB gradient boosting classifier. {..}i=1n represents the number of repetitions of the operation (n> = 1).
Blocks in the proposed architecture include DDC, FRB units, and Convolution layer. The FRB unit (Fig. 2(a)) consists of four interconnected Convolution layers. The features obtained in the last step are combined with the input features. In this way, it adds depth to the architecture by reusing previous features. The FRB unit is mathematically shown in Eq. (6). (6) x1=Φw1f∗x
x2=Φw2f∗x1
x3=Φw2f∗x2
x4=Φw3f∗x3
FRBk=[x4,x]
Here, wif weights represent xi outputs. * indicates the convolution operation. Φ means applying the Relu and BatchNormalization (BN) operations of the layer output, respectively. [ ] denotes the merge operation. FRBk refers to the FRB operation of the kth block.
The DDC unit, detailed in Fig. 2(b), is expanded with varying dilatation rates and the receptive area of X-ray and CT images. In this way, distinctive features are obtained effectively and more diversity is provided in the feature extraction process. The mathematical notation of the DDC unit used in the proposed architecture is given in Eq. (7). (7) o1=Φ(w1d⊙FRBk)
o2=Φ(w2d⊙o1)
⋮
on=Φ(wnd⊙on−1)
DDCk=[o1,o2…,on]
Here, wid shows weights, oi outputs, and ⊙ DeptwiseConv operation. [o1,o2…,on] means the join operation. DDCk refers to the DDC operation of the kth block.
Finally, the features obtained by FRB and DCC operations in blocks are given to the Convolution layer. In this way, important properties are obtained. The mathematical expression required for this operation is given in Eq. (8). (8) CLk=Φ(wk∗DDCk)
Here wk weights, * denotes the convolution operation.
Fig. 5 Proposed architecture (CovidDWNet+GB).
3.5 Training and optimization of the proposed architecture
The CovidDWNet architecture is trained with a backpropagation algorithm using Cross-Entropy Eq. (9) for training multi-class datasets and binary cross-entropy Eq. (10) cost functions for training two-class datasets. These cost functions can be expressed mathematically as: (9) Lyˆ,y=−∑yilogyˆi
(10) LBCE=−1n∑((yi.log(yˆi))+(1−yi).log(1−yˆi))
Here, n is the number of samples, y is the actual value, and yˆ is the predicted value.
Adam optimization [42], [76] algorithm is used as the optimization algorithm for updating the weights in the architecture. Adam optimization algorithm with learning coefficient η at time t: (11) wt+1j=wtj−ηvtst+ɛ−×gt
(12) vt=β1×vt−1−1−β1×gt
(13) st=β2×st−1−1−β2×gt2
Here w stands for weights, hyperparameters β1 and β2 time t η learning rate coefficient. gt represents the gradient at time t.vt and st represent the exponential mean of gradients and squares of gradients along, wt. In the proposed architecture, the Relu activation function used after each convolution operation is given in Eq. (14) [34]. (14) f(x)Relu=max{0,x}
Fully connected layers (FC) form a fundamental part of CNN architectures, where all neurons in the previous layer connect to all neurons in the next layer and calculate how much each value matches the class. As the last layer, the output of the FC is combined with the activation functions of sigmoid, SVM, softmax, etc. for class prediction. Softmax activation function used for classification in this study, a probability distribution of n number of output categories is calculated according to Eq. (15) [33], [79]. (15) Zk=exk∑i=1nexn
Here, x is the input vector, n is the number of classes, up to, k=1….n, and Z is the output vector. The sum of all Z values is equal to 1.
4 Experimental results and discussion
For the detection of Covid-19, some literature studies based on deep learning using X-ray and CT images are presented in Table 1. It can be said that there are differences in success rates according to the datasets that researchers use by developing different architectures. In general, it is seen that the success rates of studies with two classes are higher than those with multiple classes. Marques et al. [80] performed binary and triple classification on X-ray images with the CNN-based architecture they developed using the EfficientNet architecture. This method has shown the highest success in the classification of binary class (Covid-19, normal) X-ray images with an accuracy rate of 99.6% compared to other architectures. On four-class X-ray datasets, Umer et al. [81] showed the lowest performance with 85.0% accuracy, while the proposed architecture achieved the highest performance with 96.8% accuracy. In addition, when studies using two-class CT images containing Covid-19 and Normal images were examined, Gifani et al. [82] showed the lowest performance with 85% accuracy using CNNs-based architecture, while the recommended architecture showed the highest performance with 100% accuracy. However, when the results here are examined, it is seen that the datasets used by the researchers affect the success rates of the methods. At the same time, the number of samples and the number of classes in the datasets are the factors affecting the success of the architectures.
To evaluate the performance of the architectures developed by the researchers more fairly, it is important to conduct experimental studies using common datasets. Therefore, in this study, four different experimental applications were carried out using three different datasets containing CT and X-ray images for the diagnosis of Covid-19. To objectively evaluate the performance of current architectures mentioned in the literature with CovidDWNet+GB (Our model), training was carried out on the same dataset by keeping certain parameters the same. The results obtained according to different metrics by training each model 200 epochs are presented in Table 4, 5, 6, and 7. Commonly used metrics accuracy, precision, recall, F1-Score, specificity, and AUC were used to evaluate the results. These metrics are: (16) Accuracy=TP+TNTP+TN+FP+FN
(17) Precision/PPV=TPTP+FP
(18) Recall/Sensiviy=TPTP+FN
(19) F1−Score=2∗Precision∗RecallPrecision+Recall
(20) Specificity=TNTN+FP
Table 1 Some deep learning approaches and success results for Covid-19 diagnosis from X-ray and CT images.
Study Architecture Class Scanning Accuracy (%)
Sethy et al. [83] ResNet50 plus 2-class (Covid-19, noncovid-19) CT 95.38
Li et al. [84] Stacked-autoencoder 2-class (Covid-19, Pneumonia, normal) CT 94.7
Gifani et al. [82] CNNs models 2-class( Covid-19, noncovid-19) CT 85.0
Xu et al. [85] ResNet + Loc-ation Attention 3-class (Influenza-A, Normal, covid-19) CT 86.7
Heidarian et al. [86] COVID-FACT 3-class (Covid-19, Pneumonia, normal) CT 90.82
Mukherjee et al. [87] Tailored Deep NN 2-class (Covid-19, noncovid-19) CT 95.83
Mukherjee et al. [87] Tailored Deep NN 2-class (Covid-19, noncovid) X-ray 96.13
Wang et al. [88] COVID-Net 3-class (Covid-19, pneumonia, normal) X-ray 93.3
Heidari et al. [89] VGG16-based CNN 3-class (Covid-19, pneumonia, normal) X-ray 94.5
Chakraborty [90] Corona-Nidaan 3-class (Covid-19, normal, pneumonia,) X-ray 95.0
Umer et al. [81] COVINet 2-class (Covid-19, normal) X-ray 97.0
Umer et al. [81] COVINet 3-class (Covid-19, normal, virus pneumonia) X-ray 90.0
Umer et al. [81] COVINet 4-class (Covid-19, normal, virus pneumonia, bacterial pneumonia) X-ray 85.0
Babukarthik et al. [91] Genetic deep CNN 2- class (Covid-19, normal) X-ray 98.8
Apostolopoulos
et al. [92] Pretrained CNNs 3-class (Covid-19, nonCovid-19 pneumonia, normal) X-ray 96.7
Ismael et al [93] Deep CNNs 2- class (Covid-19, normal) X-ray 92.6
Oh et al. [94] ResNet-18 4-class (Covid-19+viral pneumonia, bacterial pneumonia, tuberculosis, normal) X-ray 91.9
Ezzat et al. [95] GSA-DenseNet121 2-class (Covid-19, pneumonia) X-ray 93.4
Marques et al. [80] CNN + EfficientNet 2- class (Covid-19, normal) X-ray 99.6
Marques et al. [80] CNN + EfficientNet 3- class (Covid-19, nonCovid-19 pneunomia, normal) X-ray 96.7
Hussain et al. [96] CoroDet 2- class (Covid-19, normal) X-ray 99.1
Hussain et al. [96] CoroDet 3- class (Covid-19, normal, pneumonia) X-ray 94.2
Hussain et al. [96] CoroDet 4- class (Covid-19, normal, non-Covid-19 pneumonia, non-Covid-19 bacterial pneumonia) X-ray 91.2
Proposed CovidDWNet 2- class (Covid-19, normal) CT 100.0
proposed CovidDWNet 4-class (Covid-19, Lung Opacity, Normal, Viral pneumonia) X-ray 96.8
Here, TP (True Positives) denotes correctly classified diseased cases, TN (True Negatives) correctly defined healthy cases, FP (False Positives) misclassified diseased cases, FN (False Negatives) misclassified healthy cases [3], [97]. The receiver operating characteristic (ROC) curve is used in classification problems to evaluate the performance of models by plotting the true positive rate (TPR) versus the false positive rate (FPR). The area under the curve (AUC) indicates the area under the ROC, which is a probability curve [3]. (21) FPR=FPFP+TN
(22) TPR=TPTP+FN
The hyperparameters of the architectures during the training phase are presented in Table 2. CovidDWNet architecture, developed on Keras/Tensorflow platform, takes images as input by scaling 128 × 128; Adam (Learning_rate = 0.001) optimization function and batch size value 32 are given.
The image distribution of the datasets used in experimental applications for the detection of Covid-19 and other pneumonia diseases according to training and test sets is given in Table 3 in detail. At the same time, the number of images according to the types of diseases in each application is presented in this table. Datasets are reserved for approximately 80% training and 20% testing. In the first application, the Sars-Cov-2 [64] dataset was used. This dataset is divided into two datasets, training and testing. The training dataset contains a total of 1986 images, 1002 Covid, and 984 Normal. The test dataset contains 495 images, 250 of which are Covid and 245 are normal. Similarly, a second application was performed using the Covid-CT [63] and Sars-Cov-2 [64] datasets containing CT images. In this application, there are 2589 images (1288 Covid, 1301 Normal) in the training dataset and 645 images (320 Covid, 325 Normal) in the test dataset. In the third application, Dataset-X-ray [65] dataset containing X-ray images was used. In this application, there are 16933 images (2893 Covid, 8154 Normal, 4810 Lung Opacity, and 1079 Viral Pneumonia) in the training dataset and 4232 images (723 Covid, 2038 Normal, 1202 Lung Opacity, and 269 Viral Pneumonia) in the test dataset. The fourth application was performed by combining all datasets containing X-ray and CT images. In this application, a total of 19515 images, including 4174 Covid, 9455 Normal, 4810 Lung Opacity, and 1079 Viral Pneumonia, in the training dataset; In the test dataset, there are a total of 4877 images, including 1043 Covid, 2363 Normal, 1202 Lung Opacity and 269 Viral Pneumonia.Table 2 Hyperparameters of architectures for Covid-19 detection.
Model Data augmentation Software Input size Optimizer Learning rate Batch size
DenseNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
AlexNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
ResNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
CspNet [98] No Keras, TensorFlow 224 × 224 Adam 0.0001 32
VGG16 No Keras, TensorFlow 224 × 224 Adam 0.0001 32
VGG19 No Keras, TensorFlow 224 × 224 Adam 0.0001 32
CovXNet [42] Yes Keras, TensorFlow 128 × 128 Adam 0.001 16
CoroNet [40] Yes Keras, TensorFlow 150 × 150 Adam 0.0001 10
CovidXrayNet [41] Yes Fastai, PyTorch 256 × 256 Adam – 32
DarkCovidNet [39] No Fastai, PyTorch 256 × 256 Adam 0.003 32
Proposed (No DDC) No Keras, TensorFlow 128 × 128 Adam 0.001 32
Proposed No Keras, TensorFlow 128 × 128 Adam 0.001 32
The results of the experimental study with the SARS-COV-2 [64] dataset containing CT images for Covid-19 detection are given in Table 4. The success rate has been increased by adding a DDC module to the proposed architecture. In addition, high performance has been achieved by adding GB classifier to the proposed architecture. However, when the data augmentation method is applied to the proposed architecture, it is seen that the success decreases. When the results obtained in this application are examined in a general way, we can say that our model exhibits the highest performance in all metrics with a 100% success rate. Also, the confusion matrix results of the proposed architecture for this application are given in Fig. 6(a). When the results are examined, it is seen that the proposed architecture correctly predicts Covid-19 patients and non-patients with 100% high performance.Table 3 Number of records in datasets used in applications.
Application (s) Data set(s) Image (s) Train/Test Covid Normal Lung Opacity Viral Pneumonia Total
Application1 Sars-Cov-2 [64] CT Train Set 1002 984 – – 1986
Test Set 250 245 – – 495
Application2 Covid-CT [63] and
Sars-Cov-2 [64] CT Train Set 1288 1301 2589
Test Set 320 325 645
Application3 Dataset-X-ray [65] X-ray Train Set 2893 8154 4810 1076 16933
Test Set 723 2038 1202 269 4232
Application4 All datasets X-ray + CT Train Set 4174 9455 4810 1076 19515
Test Set 1043 2363 1202 269 4877
In the second application, an experimental study was performed by combining the Covid-CT [63] and Sars-Cov-2 [64] datasets containing Covid-19 and Normal CT images, and the results are presented in Table 5. When the results are examined, the proposed architecture (CovidDWNet+GB) showed the highest success with 99.84% according to the accuracy metric and 100% (1.00) according to the precision, recall, and F1-Score metrics. Similarly, the CovidDWNet architecture achieved the highest success with 99.85% performance according to specificity and AUC metrics. In addition, when the confusion matrix results are examined in Fig. 6(b), It is seen that the CovidDWNet+GB architecture detects Covid-19 patients and normal people who are not sick with 100% accuracy.Table 4 The success of the models in detecting Covid-19 on the Sars-Cov-2 dataset containing CT images.
Model Accuracy (%) Precision Recall F1-Score Specificity(%) AUC (%)
DenseNet 97.37 0.97 0.97 0.97 98.57 97.37
AlexNet 94.14 0.94 0.94 0.94 94.12 94.12
ResNet 95.96 0.96 0.96 0.96 95.97 95.97
CspNet [98] 95.15 0.95 0.95 0.95 95.14 95.14
VGG16 50.51 0.50 0.34 0.25 50.00 50.00
VGG19 50.51 0.50 0.34 0.25 50.00 50.00
CovXNet [42] 98.18 0.98 0.98 0.98 98.18 98.18
CoroNet [40] 98.59 0.99 0.99 0.99 98.57 98.57
CovidXrayNet [41] 97.97 0.98 0.98 0.98 98.00 98.00
DarkCovidNet [39] 95.35 0.95 0.95 0.95 95.35 95.35
Proposed (No DDC) 98.38 0.98 0.98 0.98 98.39 98.39
Proposed+ DataAug. 97.78 0.98 0.98 0.98 97.78 97.78
Proposed (No GB) 98.59 0.99 0.99 0.99 98.58 98.58
Proposed(CovidDWNet+GB) 100.0 1.00 1.00 1.00 100.0 100.0
In the third experimental study for Covid-19 detection, the four-class Dataset-X-ray [65] dataset was used. The results of the experimental study are shown in Table 6. When the results of the application are examined, our model (CovidDWNet+GB) achieved the highest performance with 96.81% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-Score, 95.54% specificity, and 97.98% AUC. At the same time, when the training and testing times of the third application (Application3) are examined (in Table 8), it is seen that the proposed architecture is faster than the CovidXrayNet and DarkCovidNet architectures, which are the closest to the success rate. In addition, when the success distribution of the CovidDWNet+GB architecture according to classes is analyzed in Fig. 6(c), we can say that it correctly predicts Covid-19 patients 99%, Lung Opacity patients 92%, people who are not sick 98% and Viral Pneumonia patients 100%.Table 5 Success rates of models according to different metrics in detecting Covid-19 on the Covid-CT and Sars-Cov-2 datasets containing CT images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 92.09 0.92 0.92 0.92 92.09 92.09
AlexNet 86.51 0.87 0.86 0.87 86.49 86.49
ResNet 87.44 0.88 0.97 0.87 87.47 87.47
CspNet [98] 85.58 0.86 0.86 0.86 85.53 85.53
VGG16 50.39 0.25 0.50 0.34 50.00 50.00
VGG19 50.39 0.25 0.50 0.34 50.00 50.00
CovXNet [42] 88.99 0.89 0.89 0.89 89.03 89.03
CoroNet [40] 92.25 0.92 0.92 0.92 92.26 92.26
CovidXrayNet [41] 91.16 0.92 0.91 0.91 91.12 91.12
DarkCovidNet [39] 88.92 0.88 0.87 0.87 88.92 88.92
Proposed (No DDC) 91.63 0.92 0.92 0.92 91.62 91.62
Proposed+ DataAug. 86.36 0.86 0.86 0.86 86.33 86.33
Proposed (No GB) 93.33 0.93 0.93 0.93 93.31 93.31
Proposed(CovidDWNet+GB) 99.84 1.0 1.00 1.00 99.85 99.85
In the last experimental study for Covid-19 detection, an application was performed by combining all datasets (Covid-CT, Sars-Cov-2, and Dataset-X-ray). The results obtained according to different metrics are given in Table 7. When the results are examined, we can say that CovidDWNet+GB, 96.32% accuracy, 0.97 precision, 0.97 recall, 0.97 F1-Score, 95.17% specificity, and 97.67% AUC showed the highest success. Also, the confusion matrix results for the CovidDWNet+GB architecture of this application are given in Fig. 6(d). When the results are examined, it is seen that he predicted Covid 19 patients at 97%, Lung Opacity patients at 93%, non-sick people at 97%, and Viral Pneumonia patients at 100% correct.Table 6 Results of architectures for Covid-19 detection according to different metrics on the Dataset-X-ray dataset containing X-ray images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 87.10 0.88 0.89 0.88 87.66 92.12
AlexNet 90.62 0.91 0.91 0.91 88.12 93.72
ResNet 92.70 0.94 0.93 0.94 90.79 95.14
CspNet [98] 82.63 0.79 0.84 0.81 80.09 88.84
VGG16 91.09 0.92 0.91 0.92 87.59 93.88
VGG19 91.33 0.93 0.92 0.92 87.55 93.96
CovXNet [42] 92.44 0.95 0.89 0.91 89.28 92.76
CoroNet [40] 92.11 0.93 0.92 0.92 90.52 94.40
CovidXrayNet [41] 95.39 0.96 0.96 0.96 94.33 96.88
DarkCovidNet [39] 94.33 0.96 0.94 0.95 92.17 95.67
Proposed (No DDC) 93.30 0.94 0.94 0.94 90.75 95.50
Proposed+ DataAug. 93.19 0.95 0.92 0.94 89.72 94.54
Proposed (No GB) 93.76 0.95 0.94 0.94 90.91 95.67
Proposed(CovidDWNet+GB) 96.81 0.98 0.97 0.98 95.54 97.98
According to the experimental application results, the class performances (confusion matrix) of the proposed architecture (CovidDWNet+GB) for Covid-19 detection are given in Fig. 6. It shows the results of the binary classification in (a) and (b), and multi-classification in (c) and (d). (a) gives the results of the first application, (b) the second application, (c) the third application, and (d) the fourth application. In applications containing the proposed architectural CT images (Fig. 6 (a–b)), it appears to predict Covid-19 and Normal images extremely successfully with 100% success rates.Table 7 Performance of Architectures for Covid-19 detection by different metrics on all datasets containing X-ray and CT images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 93.05 0.94 0.93 0.94 90.41 95.11
AlexNet 90.98 0.92 0.91 0.91 86.91 93.66
ResNet 92.02 0.93 0.93 0.93 90.30 94.70
CspNet [98] 90.71 0.92 0.91 0.91 87.14 93.67
VGG16 89.65 0.91 0.90 0.90 84.90 92.88
VGG19 89.24 0.91 0.89 0.90 83.64 92.27
CovXNet [42] 92.45 0.94 0.91 0.93 88.85 93.86
CoroNet [40] 92.23 0.94 0.92 0.93 87.28 94.43
CovidXrayNet [41] 95.30 0.96 0.96 0.96 92.93 96.85
DarkCovidNet [39] 90.59 0.92 0.91 0.91 88.05 93.06
Proposed (No DDC) 92.19 0.93 0.92 0.93 87.69 94.40
Proposed+ DataAug. 91.92 0.94 0.91 0.90 89.90 93.10
Proposed (No GB) 93.36 0.94 0.94 0.94 91.73 95.50
Proposed(CovidDWNet+GB) 96.32 0.97 0.97 0.97 95.17 97.67
Similarly, multiple classification performances of the CovidDWNet architecture are given in Fig. 6-(c) and (d). In the third experimental study including X-ray images, it is seen that the proposed architecture, Covid-19, Lung Opacity, Normal and Viral Pneumonia images were estimated with success rates of 99%, 92%, 98%, and 100%, respectively. In addition, in the fourth application containing all datasets, it was observed that he predicted Covid-19 images with 97%, Lung Opacity images with 93%, Normal images with 97%, and Viral pneumonia images with a rate of 100%. It can be said that it performs extremely satisfactorily in classes other than the Lung Opacity class. It is thought that its lower success in images containing Lung Opacity is due to its overlapping features with other classes.
The ROC curve of our proposed model is shown in Fig. 7. The ROC curve is a graphical representation of the classification performance of the network. The closer the curve is to its upper left limit, the higher the performance. Fig. 7 (a–b) shows the results of the CT images, (c) the results of the X-ray images, and (d) the results of the X-ray and CT images. In CT images, it is seen that AUC values of 99.85% and 100% results are obtained. We can say that AUC values of 97.98% and 97.67% were obtained in X-ray and all images, respectively.Fig. 6 Performance results of the proposed architecture in binary and multiple classes. (a) The first application, (b) second application, (c) third application, (d) fourth application results.
Gradient-based class activation mapping (Grad-CAM) algorithm [99] is used to highlight important points on X-ray and CT images that affect the performance of CNN architectures. The main purpose of this algorithm was developed to create stronger deep networks. The last convolutional layer is considered to be the stage where the best balance is achieved between important spatial information and the highest semantics [100]. Grad-CAM generates heatmap heat zones to highlight key locations from features derived from the final convolution layer. This information indicates which regions the algorithm pays more attention to. In Fig. 8, heatmap and Grad-CAM images obtained for sample Covid-19 images with the Grad-CAM algorithm are given. Green and yellow areas on heatmaps highlight key regions where the CovidDWNet architecture is concentrated. Regions with dark yellow in heatmaps and red in Grad-CAM indicate important regions with high distinctiveness.Fig. 7 ROC analysis of the proposed model. (a) First application, (b) second application, (c) third application, (d) fourth application results.
CNNs are used for classification and recognition problems by making use of fully connected layers of feature maps obtained as a result of the convolution process [101]. Feature maps are obtained with filters defined by convolution operations on the input image. Feature maps obtained for a particular input image are used to understand which features of the input are detected or preserved. It is expected to detect small or fine details from the image given as input to the models. However, the models will capture more general feature maps close to the output [102]. In Fig. 9, an example of tens of feature maps obtained from the images given as an introduction to the CovidDWNet architecture is given. It is seen that the different features of the images are emphasized in the first two convolution layers. These images appear to be understandable images. We can say that the feature maps obtained from the last convolution layer of the next blocks (Block1-4) capture more fine details. These attributes are meaningful features that are not understood by humans but can be understood by CNN models. At the same time, it is possible to say that the feature maps show fewer and fewer details as they go deeper and that these details are meaningful features in the decision-making process by CNN models.Fig. 8 From CovidDWNet architecture using Grad-CAM resulting sample heatmaps and Covid-19 visuals.
Also, the training and test times of the applications are given in Table 8. Training times in hours and minutes; Test times are shown in seconds. Training times, architectures 200 epoch training time; test times represent the time elapsed during the estimation of all samples in the test dataset. When the training and testing times are examined, we can say that the AlexNet architecture has a higher speed compared to other architectures. However, when the overall success of the AlexNet architecture is examined in the experimental applications, it has been observed that it exhibits a low performance.Fig. 9 Feature maps were obtained from sample CT and X-ray images.
The time complexity of the architectures according to the training and test times is given in Fig. 10. When the time complexity diagram is carefully examined, it is seen that the AlexNet architecture has the smallest time complexity. We can also say that the CovXNet architecture has the highest time complexity. It is possible to say that the proposed architecture has moderate time complexity.Table 8 Training and testing times of architectures. Training times in hours and minutes; Test times are shown in seconds.
Model Application1 Application2 Application3 Application4
Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.)
DenseNet 0.53 1 1.18 2 5.00 6 6.10 7
AlexNet 0.53 1 1.16 1 4.50 3 6.04 4
ResNet 1.10 2 1.50 3 6.56 8 7.36 10
CspNet [80] 1.30 2 1.46 3 6.00 7 8.20 9
VGG16 1.07 3 1.30 3 7.50 18 9.13 19
VGG19 1.14 3 1.33 4 8.16 19 9.23 21
CovXNet [42] 5.05 7 6.05 8 15.33 26 18.33 29
CoroNet [40] 1.20 2 1.43 2 6.33 5 7.50 7
CovidXrayNet [41] 1.10 2 1.43 3 8.33 12 9.13 15
DarkCovidNet [39] 1.18 2 1.48 2 7.54 13 8.43 15
Proposed (CovidDWNet) 1.16 2 1.45 3 7.24 8 8.32 11
When the results of experimental studies are examined in general, it is seen that it predicts X-ray and CT images with high performance. A higher success was achieved with CT images compared to X-ray images. We can say that this is due to the more sensitive and finer detailed structures of CT images [13], [42].Fig. 10 The time complexity of architectures: (a) Time complexity based on training times (in hours), (b) Time complexity based on test times (in seconds).
In addition, a higher performance has been achieved by integrating the DDC module into the CovidDWNet architecture, providing different expansion rates and deepening the feature map with depthwise convolution. However, when the data augmentation method is applied to the proposed architecture, it has been observed that it affects success negatively. The hyperparameters and values of the applied data augmentation method are given in Table 9.
Table 9 Data augmentation hyperparameters for the proposed architecture.
Parameters Value
Width shift 0.2
Height shift 0.2
Shear 0.25
Zoom 0.2
Rotation 30
Horizontal flip True
Vertical flip True
5 Conclusion
Covid-19 pandemic cases are increasing day by day and cause the death of many people. It has caused millions of cases and the death of millions of people so far. This disease, which brings with it different health problems, poses serious threats to human health with the emergence of new variations. Many states are taking many measures to prevent the spread of the disease and reduce deaths. RT-PCR tests are generally used to detect this disease. However, considering the inadequacy of RT-PCR tests, the risk of transmission to healthcare personnel, pain to patients, and cost, it brings with it many problems. In this sense, different researches are carried out and different solutions are offered. Deep learning architectures with high performance are one of these studies. When the literature is examined, it is possible to see many studies with deep learning. In these studies, it is seen that only one of the CT or X-ray datasets is used mostly. At the same time, it was seen that the performance evaluation of the studies was limited in themselves.
In this study for the detection of Covid-19 and similar symptoms, datasets containing CT or X-ray images were used. A new architecture is proposed, called CovidDWNet, based on feature reuse residual block (FRB) and Depthwise dilated convolutions (DDC) units. High performance has been achieved by providing the combination of the proposed architecture and the Gradient boosting (GB) algorithm (CovidDWNet+GB). In addition, the current architectures in the literature were examined, the architectures were trained on the same data sets and performance evaluation was made accordingly.
It has been observed that CovidDWNet+GB exhibits the highest success with 99.84% and 100% accuracy rates in applications performed on CT datasets with two classes (Covid-19, and non Covid-19). In addition, it has been observed that it provides the highest success according to precision, recall, F1-Score, specificity, and AUC metrics. The proposed architecture showed the highest success in the application using four classes (Covid-19, Lung Opacity, Normal and Viral Pneumonia) X-ray images, with 96.81% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-Score, 95.54% specificity, and 97.98% AUC. Similarly, we can say that the CovidDWNet+GB architecture showed the highest success in the experimental study using X-ray and CT images, with 96.32% accuracy, 0.97 precision, 0.97 recall, 0.97 F1-Score, 95.17% specificity, and 97.67% AUC. Also, it has been observed that the proposed architecture predicts 4877 images in the test dataset with a high speed of 11 s.
As a result, when the performances of different architectures are examined by keeping certain parameters constant on the same datasets, it is possible to say that the proposed architecture exhibits a respectable success in the literature and shows a remarkable performance among current architectures.
CRediT authorship contribution statement
Gaffari Celik: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Visualization, Investigation, Validation.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
No data was used for the research described in the article.
The code (and data) in this article has been certified as Reproducible by Code Ocean: (https://codeocean.com/). More information on the Reproducibility Badge Initiative is available at https://www.elsevier.com/physical-sciences-and-engineering/computer-science/journals.
==== Refs
References
1 Wu F. A new coronavirus associated with human respiratory disease in China Nature 579 7798 2020 265 269 10.1038/s41586-020-2008-3 32015508
2 Coronavirus disease (COVID-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus-2019.
3 Subramanian N. Elharrouss O. Al-Maadeed S. Chowdhury M. A review of deep learning-based detection methods for COVID-19 Comput. Biol. Med. 143 2022 105233 10.1016/j.compbiomed.2022.105233
4 Rubin G.D. The role of chest imaging in patient management during the COVID-19 pandemic Chest 158 1 2020 106 116 10.1016/j.chest.2020.04.003 32275978
5 Singh R. Corona virus (COVID-19) symptoms prevention and treatment: A short review J. Drug Deliv. Ther. 11 2-S 2021 118 120 10.22270/jddt.v11i2-S.4644
6 R S. An efficient hardware architecture based on an ensemble of deep learning models for COVID -19 prediction Sustain. Cities Soc. 2022 103713 10.1016/j.scs.2022.103713
7 Heidari A. Jafari Navimipour N. Unal M. Toumaj S. The COVID-19 epidemic analysis and diagnosis using deep learning: A systematic literature review and future directions Comput. Biol. Med. 141 2021 2022 105141 10.1016/j.compbiomed.2021.105141
8 Sharfstein J.M. Becker S.J. Mello M.M. Diagnostic testing for the novel coronavirus JAMA 323 15 2020 1437 10.1001/jama.2020.3864 32150622
9 Stephanie S. Determinants of chest radiography sensitivity for COVID-19: A multi-institutional study in the United States Radiol. Cardiothorac. Imaging 2 5 2020 e200337 10.1148/ryct.2020200337
10 Liu R. Clinica Chimica Acta positive rate of RT-PCR detection of SARS-CoV-2 infection in 4880 cases from one hospital in Wuhan, China, from Jan to 2020 Clin. Chim. Acta 505 March 2020 172 175 10.1016/j.cca.2020.03.009 32156607
11 Dramé M. Should RT-PCR be considered a gold standard in the diagnosis of COVID-19? J. Med. Virol. 92 11 2020 2312 2313 10.1002/jmv.25996 32383182
12 Xie J. Characteristics of patients with coronavirus disease (COVID-19) confirmed using an IgM-IgG antibody test J. Med. Virol. 92 10 2020 2004 2010 10.1002/jmv.25930 32330303
13 Hassan H. Supervised and weakly supervised deep learning models for COVID-19 CT diagnosis: A systematic review Comput. Methods Programs Biomed. 218 2022 106731 10.1016/j.cmpb.2022.106731
14 Gaur P. Malaviya V. Gupta A. Bhatia G. Pachori R.B. Sharma D. COVID-19 disease identification from chest CT images using empirical wavelet transformation and transfer learning Biomed. Signal Process. Control 71 PA 2022 103076 10.1016/j.bspc.2021.103076
15 Ucar F. Korkmaz D. COVIDiagnosis-Net: Deep Bayes-SqueezeNet based diagnosis of the coronavirus disease, 2019 (COVID-19) from X-ray images Med. Hypotheses 140 April 2020 109761 10.1016/j.mehy.2020.109761
16 Başaran E. Classification of white blood cells with SVM by selecting SqueezeNet and LIME properties by mRMR method Signal, Image Video Process 2022 10.1007/s11760-022-02141-2
17 Çelik G. Talu M.F. A new 3D MRI segmentation method based on generative adversarial network and atrous convolution Biomed. Signal Process. Control 71 PA 2022 103155 10.1016/j.bspc.2021.103155
18 Goodfellow I. Generative adversarial networks Commun. ACM 63 11 2020 139 144 10.1145/3422622
19 Çelik G. Talu M.F. Generating the image viewed from EEG signals Pamukkale Univ. J. Eng. Sci. 27 2 2021 129 138 10.5505/pajes.2020.76399
20 Esteva A. Dermatologist-level classification of skin cancer with deep neural networks Nature 542 7639 2017 115 118 10.1038/nature21056 28117445
21 Tan J.H. Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network Inf. Sci. (Ny) 420 2017 66 76 10.1016/j.ins.2017.08.050
22 Başaran E. Cömert Z. Çelik Y. Neighbourhood component analysis and deep feature-based diagnosis model for middle ear otoscope images Neural Comput. Appl. 2022 10.1007/s00521-021-06810-0
23 Celik Y. Talo M. Yildirim O. Karabatak M. Acharya U.R. Automated invasive ductal carcinoma detection based using deep transfer learning with whole-slide images Pattern Recognit. Lett. 133 2020 232 239 10.1016/j.patrec.2020.03.011
24 Bozdag Z. Talu F.M. Pyramidal nonlocal network for histopathological image of breast lymph node segmentation Int. J. Comput. Intell. Syst. 14 1 2021 122 131 10.2991/ijcis.d.201030.001
25 Talo M. Yildirim O. Baloglu U.B. Aydin G. Acharya U.R. Convolutional neural networks for multi-class brain disease detection using MRI images Comput. Med. Imaging Graph. 78 2019 101673 10.1016/j.compmedimag.2019.101673
26 Gaál G. Maga B. Lukács A. Attention U-net based adversarial architectures for chest X-ray lung segmentation CEUR Workshop Proc. 2692 2020 1 7
27 Souza J.C. Bandeira Diniz J.O. Ferreira J.L. França da Silva G.L. Corrêa Silva A. de Paiva A.C. An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks Comput. Methods Programs Biomed. 177 2019 285 296 10.1016/j.cmpb.2019.06.005 31319957
28 Yıldırım Ö. Pławiak P. Tan R.S. Acharya U.R. Arrhythmia detection using deep convolutional neural network with long duration ECG signals Comput. Biol. Med. 102 September 2018 411 420 10.1016/j.compbiomed.2018.09.009 30245122
29 Hannun A.Y. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network Nature Med. 25 1 2019 65 69 10.1038/s41591-018-0268-3 30617320
30 Acharya U.R. A deep convolutional neural network model to classify heartbeats Comput. Biol. Med. 89 August 2017 389 396 10.1016/j.compbiomed.2017.08.022 28869899
31 Rajpurkar P. CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning 2017 3 9 [Online]. Available: http://arxiv.org/abs/1711.05225
32 Ieracitano C. A fuzzy-enhanced deep learning approach for early detection of Covid-19 pneumonia from portable chest X-ray images Neurocomputing 481 2022 202 215 10.1016/j.neucom.2022.01.055 35079203
33 Ahamed K.U. A deep learning approach using effective preprocessing techniques to detect COVID-19 from chest CT-scan and X-ray images Comput. Biol. Med. 139 October 2021 105014 10.1016/j.compbiomed.2021.105014
34 Verma H. Mandal S. Gupta A. Temporal deep learning architecture for prediction of COVID-19 cases in India Expert Syst. Appl. 195 January 2021 116611 10.1016/j.eswa.2022.116611
35 Khan S.H. COVID-19 detection in chest X-ray images using deep boosted hybrid learning Comput. Biol. Med. 137 August 2021 104816 10.1016/j.compbiomed.2021.104816
36 Loey M. El-Sappagh S. Mirjalili S. Bayesian-based optimized deep learning model to detect COVID-19 patients using chest X-ray image data Comput. Biol. Med. 142 January 2022 105213 10.1016/j.compbiomed.2022.105213
37 Lahsaini I. El Habib Daho M. Chikh M.A. Deep transfer learning based classification model for COVID-19 using chest CT-scans Pattern Recognit. Lett. 152 2021 122 128 10.1016/j.patrec.2021.08.035 34566222
38 Toğaçar M. Ergen B. Cömert Z. COVID-19 detection using deep learning models to exploit social mimic optimization and structured chest X-ray images using fuzzy color and stacking approaches Comput. Biol. Med. 121 March 2020 10.1016/j.compbiomed.2020.103805
39 Ozturk T. Talo M. Yildirim E.A. Baloglu U.B. Yildirim O. Rajendra Acharya U. Automated detection of COVID-19 cases using deep neural networks with X-ray images Comput. Biol. Med. 121 April 2020 103792 10.1016/j.compbiomed.2020.103792
40 Khan A.I. Shah J.L. Bhat M.M. CoroNet: A deep neural network for detection and diagnosis of COVID-19 from chest x-ray images Comput. Methods Programs Biomed. 196 2020 105581 10.1016/j.cmpb.2020.105581
41 Monshi M.M.A. Poon J. Chung V. Monshi F.M. CovidXrayNet: Optimizing data augmentation and CNN hyperparameters for improved COVID-19 detection from CXR Comput. Biol. Med. 133 March 2021 104375 10.1016/j.compbiomed.2021.104375
42 Mahmud T. Rahman M.A. Fattah S.A. CovXNet: A multi-dilation convolutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization Comput. Biol. Med. 122 June 2020 103869 10.1016/j.compbiomed.2020.103869
43 Calderon-Ramirez S. Yang S. Elizondo D. Moemeni A. Dealing with distribution mismatch in semi-supervised deep learning for COVID-19 detection using chest X-ray images: A novel approach using feature densities Appl. Soft Comput. 123 2022 108983 10.1016/j.asoc.2022.108983
44 Gupta A. Anjum Gupta S. Katarya R. InstaCovNet-19: A deep learning classification model for the detection of COVID-19 patients using Chest X-ray Appl. Soft Comput. 99 2021 106859 10.1016/j.asoc.2020.106859
45 Feki I. Ammar S. Kessentini Y. Muhammad K. Federated learning for COVID-19 screening from Chest X-ray images Appl. Soft Comput. 106 2021 107330 10.1016/j.asoc.2021.107330
46 de Moura J. Novo J. Ortega M. Fully automatic deep convolutional approaches for the analysis of COVID-19 using chest X-ray images Appl. Soft Comput. 115 2022 108190 10.1016/j.asoc.2021.108190
47 Shankar K. An optimal cascaded recurrent neural network for intelligent COVID-19 detection using Chest X-ray images Appl. Soft Comput. 113 2021 107878 10.1016/j.asoc.2021.107878
48 Albahli S. Ayub N. Shiraz M. Coronavirus disease (COVID-19) detection using X-ray images and enhanced DenseNet Appl. Soft Comput. 110 2021 107645 10.1016/j.asoc.2021.107645
49 Elazab A. Elfattah M.A. Zhang Y. Novel multi-site graph convolutional network with supervision mechanism for COVID-19 diagnosis from X-ray radiographs Appl. Soft Comput. 114 2022 108041 10.1016/j.asoc.2021.108041
50 Ozcan T. A new composite approach for COVID-19 detection in X-ray images using deep features Appl. Soft Comput. 111 2021 107669 10.1016/j.asoc.2021.107669
51 Calderon-Ramirez S. Correcting data imbalance for semi-supervised COVID-19 detection using X-ray chest images Appl. Soft Comput. 111 2021 107692 10.1016/j.asoc.2021.107692
52 Karthik R. Menaka R. Hariharan M. Learning distinctive filters for COVID-19 detection from chest X-ray using shuffled residual CNN Appl. Soft Comput. 99 2021 106744 10.1016/j.asoc.2020.106744
53 Demir F. DeepCoroNet: A deep LSTM approach for automated detection of COVID-19 cases from chest X-ray images Appl. Soft Comput. 103 2021 107160 10.1016/j.asoc.2021.107160
54 Zhou T. Lu H. Yang Z. Qiu S. Huo B. Dong Y. The ensemble deep learning model for novel COVID-19 on CT images Appl. Soft Comput. 98 2021 106885 10.1016/j.asoc.2020.106885
55 Bandyopadhyay R. Basu A. Cuevas E. Sarkar R. Harris Hawks optimisation with simulated annealing as a deep feature selection method for screening of COVID-19 CT-scans Appl. Soft Comput. 111 2021 107698 10.1016/j.asoc.2021.107698
56 Ye Q. Robust weakly supervised learning for COVID-19 recognition using multi-center CT images Appl. Soft Comput. 116 2022 108291 10.1016/j.asoc.2021.108291
57 Song L. A deep fuzzy model for diagnosis of COVID-19 from CT images Appl. Soft Comput. 122 2022 108883 10.1016/j.asoc.2022.108883
58 Liang S. Nie R. Cao J. Wang X. Zhang G. FCF: Feature complement fusion network for detecting COVID-19 through CT scan images Appl. Soft Comput. 125 2022 109111 10.1016/j.asoc.2022.109111
59 Saygılı A. A new approach for computer-aided detection of coronavirus (COVID-19) from CT and X-ray images using machine learning methods Appl. Soft Comput. 105 2021 107323 10.1016/j.asoc.2021.107323
60 Naeem H. Bin-Salem A.A. A CNN-LSTM network with multi-level feature extraction-based approach for automated detection of coronavirus from CT scan and X-ray images Appl. Soft Comput. 113 2021 107918 10.1016/j.asoc.2021.107918
61 Vinod D.N. Jeyavadhanam B.R. Zungeru A.M. Prabaharan S.R.S. Fully automated unified prognosis of Covid-19 chest X-ray/CT scan images using Deep Covix-Net model Comput. Biol. Med. 136 August 2021 104729 10.1016/j.compbiomed.2021.104729
62 Li J. Multi-task contrastive learning for automatic CT and X-ray diagnosis of COVID-19 Pattern Recognit. 114 2021 107848 10.1016/j.patcog.2021.107848
63 Yang X. He X. Zhao J. Zhang Y. Zhang S. Xie P. COVID-CT-dataset: A CT scan dataset about COVID-19 2020 1 14 [Online]. Available: http://arxiv.org/abs/2003.13865
64 Soares E. Angelov P. Biaso S. Froes M.H. Abe D.K. SARS-CoV-2 CT-scan dataset: A large dataset of real patients CT scans for SARS-CoV-2 identification 2020 medRxiv, p. 2020.04.24.20078584, [Online]. Available: https://www.medrxiv.org/content/10.1101/2020.04.24.20078584v3%0Ahttps://www.medrxiv.org/content/10.1101/2020.04.24.20078584v3.abstract
65 Chowdhury M.E.H. Can AI help in screening viral and COVID-19 pneumonia? IEEE Access 8 July 2020 132665 132676 10.1109/ACCESS.2020.3010287
66 Gu J. Recent advances in convolutional neural networks Pattern Recognit. 77 2018 354 377 10.1016/j.patcog.2017.10.013
67 Budak Ü. Cömert Z. Çıbuk M. Şengür A. DCCMED-Net: Densely connected and concatenated multi Encoder-Decoder CNNs for retinal vessel extraction from fundus images Med. Hypotheses 134 2019 2020 10.1016/j.mehy.2019.109426
68 Ren F. Liu W. Wu G. Feature reuse residual networks for insect pest recognition IEEE Access 7 2019 122758 122768 10.1109/ACCESS.2019.2938194
69 He K. Zhang X. Ren S. Sun J. Deep residual learning for image recognition 2015 [Online]. Available: http://arxiv.org/abs/1512.03385
70 Kim S. Park I. Kwon S. Han J. Motion retargetting based on dilated convolutions and skeleton-specific loss functions Comput. Graph. Forum 39 2 2020 497 507 10.1111/cgf.13947
71 Sooksatra S. Kondo T. Bunnun P. Yoshitaka A. Redesigned skip-network for crowd counting with dilated convolution and backward connection J. Imaging 6 5 2020 10.3390/JIMAGING6050028
72 Li X. Zhai M. Sun J. DDCNNC: Dilated and depthwise separable convolutional neural network for diagnosis COVID-19 via chest X-ray images Int. J. Cogn. Comput. Eng. 2 April 2021 71 82 10.1016/j.ijcce.2021.04.001
73 Chollet F. Xception: Deep learning with depthwise separable convolutions Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, 2017-Janua 2017 1800 1807 10.1109/CVPR.2017.195
74 Ma Y. Wang C. SdcNet for object recognition Comput. Vis. Image Underst. 215 2020 103332 10.1016/j.cviu.2021.103332 2022
75 Wang C.-F. A basic introduction to separable convolutions 2018 https://l24.im/hrH8qwp. (Accessed 22 Nov. 2021)
76 Friedman J.H. Greedy function approximation: A gradient boosting machine Ann. Statist. 29 5 2001 1189 1232 10.1214/aos/1013203451
77 Chen H. Shen Z. Wang L. Qi C. Tian Y. Prediction of undrained failure envelopes of skirted circular foundations using gradient boosting machine algorithm Ocean Eng. 258 May 2022 111767 10.1016/j.oceaneng.2022.111767
78 Touzani S. Granderson J. Fernandes S. Gradient boosting machine for modeling the energy consumption of commercial buildings Energy Build. 158 2018 1533 1543 10.1016/j.enbuild.2017.11.039
79 Gao B. Pavel L. On the properties of the softmax function with application in game theory and reinforcement learning 2017 1 10 [Online]. Available: http://arxiv.org/abs/1704.00805
80 Marques G. Agarwal D. de la Torre Díez I. Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network Appl. Soft Comput. 96 2020 106691 10.1016/j.asoc.2020.106691
81 Umer M. Ashraf I. Ullah S. Mehmood A. Choi G.S. COVINet: a convolutional neural network approach for predicting COVID-19 from chest X-ray images J. Ambient Intell. Humaniz. Comput. 13 1 2022 535 547 10.1007/s12652-021-02917-3 33527000
82 gifani P. Shalbaf A. Vafaeezadeh M. Automated detection of COVID-19 using ensemble of transfer learning with deep convolutional neural network based on CT scans Int. J. Comput. Assist. Radiol. Surg. 16 1 2021 115 123 10.1007/s11548-020-02286-w 33191476
83 Sethy P.K. Behera S.K. Ratha P.K. Biswas P. Detection of coronavirus disease (COVID-19) based on deep features Int. J. Math. Eng. Manag. Sci. 5 4 2020 643 651 10.20944/preprints202003.0300.v1
84 Li D. Fu Z. Xu J. Stacked-autoencoder-based model for COVID-19 diagnosis on CT images Appl. Intell. 51 5 2021 2805 2817 10.1007/s10489-020-02002-w
85 Xu X. A deep learning system to screen novel coronavirus disease 2019 pneumonia Engineering 6 10 2020 1122 1129 10.1016/j.eng.2020.04.010 32837749
86 Heidarian S. COVID-FACT: A fully-automated capsule network-based framework for identification of COVID-19 cases from chest CT scans Front. Artif. Intell. 4 May 2021 1 13 10.3389/frai.2021.598932
87 Mukherjee H. Ghosh S. Dhar A. Obaidullah S.M. Santosh K.C. Roy K. Deep neural network to detect COVID-19: one architecture for both CT scans and chest X-rays Appl. Intell. 51 5 2021 2777 2789 10.1007/s10489-020-01943-6
88 Wang L. Lin Z.Q. Wong A. COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images Sci. Rep. 10 1 2020 19549 10.1038/s41598-020-76550-z 33177550
89 Heidari M. Mirniaharikandehei S. Khuzani A.Z. Danala G. Qiu Y. Zheng B. Improving the performance of CNN to predict the likelihood of COVID-19 using chest X-ray images with preprocessing algorithms Int. J. Med. Inform. 144 June 2020 104284 10.1016/j.ijmedinf.2020.104284
90 Chakraborty M. Dhavale S.V. Ingole J. Corona-Nidaan: lightweight deep convolutional neural network for chest X-ray based COVID-19 infection detection Appl. Intell. 51 5 2021 3026 3043 10.1007/s10489-020-01978-9
91 Babukarthik R.G. Ananth Krishna Adiga V. Sambasivam G. Chandramohan D. Amudhavel A.J. Prediction of COVID-19 using genetic deep learning convolutional neural network (GDCNN) IEEE Access 8 2020 177647 177666 10.1109/ACCESS.2020.3025164 34786292
92 Apostolopoulos I.D. Mpesiana T.A. COVID-19: automatic detection from X-ray images utilizing transfer learning with convolutional neural networks Phys. Eng. Sci. Med. 43 2 2020 635 640 10.1007/s13246-020-00865-4 32524445
93 Ismael A.M. Şengür A. Deep learning approaches for COVID-19 detection based on chest X-ray images Expert Syst. Appl. 164 2020 2021 10.1016/j.eswa.2020.114054
94 Oh Y. Park S. Ye J.C. Deep learning COVID-19 features on CXR using limited training data sets IEEE Trans. Med. Imaging 39 8 2020 2688 2700 10.1109/TMI.2020.2993291 32396075
95 Ezzat D. Hassanien A.E. Ella H.A. An optimized deep learning architecture for the diagnosis of COVID-19 disease based on gravitational search optimization Appl. Soft Comput. 98 2021 106742 10.1016/j.asoc.2020.106742
96 Hussain E. Hasan M. Rahman M.A. Lee I. Tamanna T. Parvez M.Z. CoroDet: A deep learning based classification for COVID-19 detection using chest X-ray images Chaos Solitons Fractals 142 2021 110495 10.1016/j.chaos.2020.110495
97 Başaran E. Cömert Z. Çelik Y. Convolutional neural network approach for automatic tympanic membrane detection and classification Biomed. Signal Process. Control 56 2020 10.1016/j.bspc.2019.101734
98 Wang C.Y. Mark Liao H.Y. Wu Y.H. Chen P.Y. Hsieh J.W. Yeh I.H. CSPNet: A new backbone that can enhance learning capability of CNN IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. Work., 2020-June 2020 1571 1580 10.1109/CVPRW50498.2020.00203
99 Selvaraju R.R. Cogswell M. Das A. Vedantam R. Parikh D. Batra D. Grad-CAM: Visual explanations from deep networks via gradient-based localization Proc. IEEE Int. Conf. Comput. Vis., 2017-Octob 2017 618 626 10.1109/ICCV.2017.74
100 Abbasniya M.R. Sheikholeslamzadeh S.A. Nasiri H. Emami S. Classification of Breast Tumours Based on Histopathology Images using Deep Features and Ensemble of Gradient Boosting Methods, Vol. 103 2022 10.1016/j.compeleceng.2022.108382 arXiv Prepr. arXiv2209.01380. June. 108382
101 Kim H. Jung W.K. Park Y.C. Lee J.W. Ahn S.H. Broken stitch detection method for sewing operation using CNN feature map and image-processing techniques Expert Syst. Appl. 188 2022 116014 10.1016/j.eswa.2021.116014
102 Brownlee J. How to visualize filters and feature maps in convolutional neural networks 2019 https://3c5.com/w7im4. (Accessed 25 Sep. 2022)
| 36504726 | PMC9726212 | NO-CC CODE | 2022-12-15 23:17:44 | no | Appl Soft Comput. 2023 Jan 7; 133:109906 | utf-8 | Appl Soft Comput | 2,022 | 10.1016/j.asoc.2022.109906 | oa_other |
==== Front
Appl Soft Comput
Appl Soft Comput
Applied Soft Computing
1568-4946
1872-9681
Elsevier B.V.
S1568-4946(22)00955-3
10.1016/j.asoc.2022.109906
109906
Article
Detection of Covid-19 and other pneumonia cases from CT and X-ray chest images using deep learning based on feature reuse residual block and depthwise dilated convolutions neural network
Celik Gaffari
Agri Ibrahim Cecen University, Department of Computer Technology, Agri, Turkey
7 12 2022
1 2023
7 12 2022
133 109906109906
26 3 2022
29 11 2022
1 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Covid-19 has become a worldwide epidemic which has caused the death of millions in a very short time. This disease, which is transmitted rapidly, has mutated and different variations have emerged. Early diagnosis is important to prevent the spread of this disease. In this study, a new deep learning-based architecture is proposed for rapid detection of Covid-19 and other symptoms using CT and X-ray chest images. This method, called CovidDWNet, is based on a structure based on feature reuse residual block (FRB) and depthwise dilated convolutions (DDC) units. The FRB and DDC units efficiently acquired various features in the chest scan images and it was seen that the proposed architecture significantly improved its performance. In addition, the feature maps obtained with the CovidDWNet architecture were estimated with the Gradient boosting (GB) algorithm. With the CovidDWNet+GB architecture, which is a combination of CovidDWNet and GB, a performance increase of approximately 7% in CT images and between 3% and 4% in X-ray images has been achieved. The CovidDWNet+GB architecture achieved the highest success compared to other architectures, with 99.84% and 100% accuracy rates, respectively, on different datasets containing binary class (Covid-19 and Normal) CT images. Similarly, the proposed architecture showed the highest success with 96.81% accuracy in multi-class (Covid-19, Lung Opacity, Normal and Viral Pneumonia) X-ray images and 96.32% accuracy in the dataset containing X-ray and CT images. When the time to predict the disease in CT or X-ray images is examined, it is possible to say that it has a high speed because the CovidDWNet+GB method predicts thousands of images within seconds.
Keywords
Covid-19 diagnosis
Deep learning
Depthwise dilated convolutions
Feature reuse residual block
Gradient boosting
==== Body
pmcCode metadata
Permanent link to reproducible Capsule: https://doi.org/10.24433/CO.2183919.v1.
1 Introduction
Coronavirus (Covid-19) is a disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-COV-2). After emerging in Wuhan, China in December 2019, it soon spread around the world and became a global pandemic [1]. According to the data of the World Health Organization (WHO), it has been determined that more than 410 million cases have been seen so far, and close to 6 million people have died [2]. WHO declared the coronavirus infection as a Covid-19 pandemic in March 2020 due to the increasing number of deaths and cases. Due to the increasing cases and deaths, many states have had to close their borders to prevent the spread of the pandemic. In addition, many countries have imposed curfews for a certain period as a precaution [3].
This disease usually affects the respiratory system, such as the lungs, and also appears to cause pneumonia-like symptoms [4]. Patients commonly experience symptoms such as fever, cough, sneezing, and shortness of breath. It spreads rapidly through respiratory droplets produced by the cough or sneeze of an infected person. Elderly people and people with chronic illnesses appear to be more prone to Covid-19 infection [5].
One of the most common methods used to diagnose Covid-19 is reverse transcription-polymerase chain reaction (RT-PCR) tests. These tests are performed to determine whether individuals have been infected with SARS-COV-2, the virus that causes Covid-19 disease, momentarily or in the past. The disadvantages of these tests are that test results take time, the number of available RT-PCR test kits is low, and the risk of health personnel contracting the disease during the test is high [6]. It is also costly in that special equipment, materials, and tools are often required for RT-PCR examinations. Therefore, many countries have difficulties in procuring test kits due to budgetary and technical constraints [7]. At the same time, the sensitivity of the RT-PCR test is a cause for concern because of sample and laboratory errors that may occur [8], [9]. Liu et al. [10] have expressed their opinion on the poor performance of RT-PCR in its sensitivity. Similarly, in a study conducted by Drame et al. [11], they expressed their reservations about the use of RT-PCR to determine the viral load in the diagnosis of 2019 coronavirus disease (Covid-19). In addition, it was stated in another study that the sensitivity of these tests could be as low as 38% [12].
Covid-19, which manifests itself as a lung infection, computed tomography (CT) and chest X-ray (X-ray) images are other methods used for the detection of this disease [5]. Typical radiographic features can be reliably detected in patients with pneumonia caused by this disease with CT imaging. Although these methods have some advantages over RT-PCR testing in terms of early detection of Covid-19, specialist physicians are needed to understand and make sense of images. Considering the disadvantages of RT-PCR tests, CT and X-ray imaging techniques used in the diagnosis of Covid-19, Artificial Intelligence (AI), and Deep Learning (DL) based methods are seen as alternative methods. AI and DL methods can help the early diagnosis of this disease and make the treatment process faster by leading experts to reach a fast and accurate diagnosis through CT and X-ray images in the detection process of Covid-19 [13], [14], [15].
Artificial intelligence and deep learning methods are widely used by researchers for the detection of Covid-19 infection from X-ray and CT images. Due to the improved performance of deep learning methods, they are widely used compared to traditional methods. One of the most important reasons that led researchers to this field is that, unlike machine learning and traditional methods, in deep learning architectures, there is no need for feature extraction in the data during the preprocessing stage. Deep learning architectures can be trained with the help of the hyperparameters of the convolutional neural network (CNN) architecture to learn the best features according to the dataset used [3]. Researchers used deep learning methods in many areas classification of white blood cells [16], segmentation of brain MRI images [17], synthetic image generation [18], generating images from EEG signals [19], skin cancer classification [20], fundus image segmentation [21], diagnose different types of Otitis media [22], breast cancer detection [23], breast lymph node segmentation [24], brain disease classification [25], lung segmentation [26], [27], detection of arrhythmia [28], [29], [30] and detecting pneumonia from chest X-ray images [31]. With the pandemic, the use of deep learning methods for the detection of coronavirus symptoms from X-ray and CT images has increased significantly [3].
In literature, it can be seen that many deep learning-based studies have been carried out for the diagnosis of Covid-19 with the help of radiological images [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42].
In the study by Leracitano et al. [32], The authors proposed a fuzzy logic-based deep learning approach to differentiate X-ray images of patients with Covid-19 pneumonia and non-Covid-19-related interstitial pneumonia. The model developed here, called CovNNet, uses the blurry edge detection algorithm together with the blurry images to extract some relevant features from the X-ray images. This study [32] for the detection of Covid-19 from binary class (Covid, Non-Covid) X-ray images, performed poorly compared to many other studies in the literature, with an accuracy rate of 81%. Ahamed et al. [33] proposed a deep learning-based Covid-19 case detection model trained with a dataset of chest CT scans and X-ray images. A modified ResNet50V2 architecture is used as a deep learning architecture in the proposed model. High performance was achieved in this study using two, three, and four classes CT and X-ray images. However, a complex architecture with high processing power was used. Verma et al. [34] proposed different models such as vanilla (vanilla) LSTM, stacked LSTM, ED_LSTM, BiLSTM, CNN, and hybrid CNN+LSTM model to capture the complex trend of the COVID-19 outbreak and perform the Covid-19 prediction. In another study by Khan et al. [35], two new deep learning-based models named deep hybrid learning (DHL) and deep boosted hybrid learning (DBHL) are proposed for effective Covid-19 detection in X-ray datasets. In the proposed DHL architecture, the representation learning capability of the two developed COVID-RENet-1 & 2 models and a machine learning classifier is used separately. In the Covid-RENet model, region and edge-based attention mechanisms were applied to extract boundary features and learn region homogeneity. In addition, the transfer learning method was used in chest X-rays in the proposed architectures. In this study, in which two-class (Covid, Non-Covid) X-ray images are used, it is seen that it has an accuracy of 98.53%. In this study, performance evaluation with only binary class X-ray images is seen as a disadvantage in terms of the performance of the architectures. Because it is important to use different data sets for Covid-19 detection. The success of CNN architectures may vary according to the number of classes and image type.
In the study by Loey et al. [36], a bayesian optimization-based CNN model was proposed for the classification of chest X-ray images. In the proposed model, CNN architecture is used to extract and learn deep features. In addition, CNN hyperparameters are adjusted according to an objective function using a Bayes-based optimizer method. In another study by Lahsaini et al. [37], they used a dataset of Covid and non-Covid CT images validated by RT-PCR tests at Tlemcen hospital in Algeria. A comparative study was carried out on Inception, Resnet-V2, VGG16, VGG19, DenseNet121, DenseNet201, ImageNet, and Xception deep models using the transfer learning method. Also, a model based on DenseNet201 architecture and the GradCam algorithm is proposed. In another study by Toğaçar et al. [38], images were preprocessed using the fuzzy color technique to classify X-ray images. Then, the features obtained with MobileNetV2, and SqueezeNet models were processed with the help of the social mimic optimization method. The productive features obtained were classified using support vector machines (SVM). The DarkCovidNet method developed by Ozturk et al. [39] was used as a classifier for the YOLO real-time object detection system. By applying seventeen convolution layers and adding different filtering to each layer. As in previous studies for the detection of Covid-19, only CT in [37] and only X-ray images were used in [38], [39].
In addition, when the studies were examined, different models were developed by the researchers, defined by the names CoroNet [40], CovidXrayNet [41], and CovXNet [42]. The CoroNet [40] model, which is proposed as a deep CNN model, is based on the Xception architecture pre-trained on the ImageNet dataset. The CovidXrayNet [41] method, based on the EfficientNet-B0 model and based on the optimization method, is proposed. In this study [41], the data augmentation method is used to increase accuracy and CNN hyperparameters are optimized. In the CovXNet [42] technique, deep CNN-based architecture and a model that uses depthwise convolution and varying dilation rates to extract features efficiently are proposed. In the proposed method, different forms of CovXNets are designed and trained with X-ray images of various resolutions. In addition, a stacking algorithm was used to increase the performance rate, and abnormal regions of X-ray images were distinguished by integrating a gradient-based discriminative localization. Looking at the time complexity of the CovXNets architecture (Fig. 10), it was seen that it had a higher time complexity compared to the other architectures studied. This shows that the CovXNets architecture has a complex structure.Fig. 1 Example images are included in the datasets.
When the studies for the detection of Covid-19 disease with deep learning methods are examined, it is seen that researchers generally use X-ray [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53] or CT [54], [55], [56], [57], [58] images, but few studies use both X-ray and CT images [59], [60], [61], [62]. At the same time, it has been seen that the researchers only examined the performance of their architectures on the dataset used or compared them only with traditional architectures. In addition, it has been determined that no performance evaluation has been made according to the training and test times in the literature reviews. In our study, contrary to these studies, CT and X-ray images were used, and a performance evaluation was carried out on the same dataset, including traditional architectures as well as different current architectures. In addition, performance evaluation was made by considering the training and test times of the architectures. The proposed model was developed to reduce the workload of specialist physicians by providing effective, efficient, and rapid detection of Covid-19 and similar cases. With the thought that the CovidDWNet+GB architecture will guide different studies, it has been opened to everyone’s access on the Github page (https://github.com/GaffariCelik/Covid-19).
Our main contributions to this work are listed as follows:
• A new deep learning-based model (CovidDWNet) has been proposed for the detection of Covid-19 and other pneumonia cases.
• The performance of the CovidDWNet architecture has been increased by using multiple feature reuse residual blocks and depthwise dilated convolutions neural networks. In addition, the success rate has been increased by performing the disease prediction process with the Gradient boosting algorithm of the feature vectors obtained with the CovidDWNet architecture.
• By using different CT and X-ray datasets, a real performance evaluation was made among the current architectures in the literature.
2 Material
In this study, three datasets were used: Covid-CT [63] and Sars-Cov-2 [64] datasets containing CT images, and Dataset-X-ray [65] dataset containing X-ray images. These datasets have been made publicly available for researchers to carry out their work. Example images in datasets are given in Fig. 1.
The Covid-CT [63] consists of 812 CT images, 349 of which are Covid-19 and 463 normal, taken from 216 patients. This dataset has been validated by a senior radiologist at Tongji Hospital in Wuhan, China, who diagnosed and treated a large number of Covid-19 patients at the time of emergence of this disease between January and April 2019. The Sars-Cov-2 [64] dataset contains a total of 2482 CT chest scan images, of which 1252 are Covid-19 and 1230 are normal. This dataset was obtained from different hospitals in Sao Paulo, Brazil. The Database-X-ray [65] dataset was created for COVID-19 positive cases with the collaboration of a team of researchers from the University of Qatar, Dhaka University, Bangladesh, and medical doctors in Pakistan and Malaysia. This dataset includes X-ray images of Covid-19, normal and other lung infection diseases. It consists of 21165 X-ray images in total, including 3616 Covid-19, 10192 normal, 6012 lung opacity (Non-Covid lung infection), and 1345 Viral pneumonia.
3 Method
As a method, a CNN-based architecture has been proposed for the detection of Covid-19 and other pneumonia symptoms. This architecture is a method based on feature reuse residual block (FRB) and depthwise dilated convolutions (DDC) units.
Convolutional Neural Networks (CNNs) are models that provide high classification performance in multi-class problems and have self-learning capabilities. CNNs are coordinated combinations of multilayer perceptrons, in which every neuron in one layer is associated with all neurons in the next. A convolutional network consists of a convolutional layer and a rectified linear unit (ReLu). Convolution layers form the basic structure of CNN models. Inputs are convolutionalized and applied with nuclei across the entire visual field with convolution filters. Thus, simpler, small patterns are obtained with more complex, detailed patterns. In this way, the hierarchical network structure enables the extraction of the highest feature maps, enhanced generalization capability, and reduced computational complexity [15], [53], [66], [67]. The basic convolution operation can be written mathematically as [33]: (1) Fi,j=I∗Ki,j=∑m∑nI(i+m,j+n)K(m,n)
Here, I represents the input matrix (may be an image), mxn filter size, and K the two-dimensional filter.
3.1 Feature reuse residual block (FRB)
The feature reuse block (FRB) is a widely used technique in computer vision. In this method, feature maps of previous layers are given as input to all subsequent layers. Thus, the performance of the network is highly increased by reusing the features of the previous layers in all subsequent layers [68], [69]. The mathematical formulation of the FRB technique used in this study is defined as follows: (2) FRB=y=Fx,wiox
Here, x and y represent the inputs and outputs of the considered layers, wi weights, and the combination of those features. The architecture of the FRB technique is presented in Fig. 2(a).
Fig. 2 Feature reuse residual block architecture-(a) and depthwise dilated convolutions architecture-(b), which constitute the basic structure of the proposed architecture.
3.2 Depthwise dilated convolutions (DDC)
As shown in Fig. 3, dilated convolutions can be expanded in the scope of the convolution kernel by changing the dilation ratio compared to standard convolution. By expanding the scope of the convolution kernel, multi-scale features (information) can be obtained. Choosing a dilation rate of one captures the same properties as standard convolution. However, in dilated convolution, when the dilation ratio is selected as greater than one, more detailed features can be obtained than in standard convolution [70], [71], [72].
In depthwise dilated convolutions operation, the convolution operation is applied to each input channel separately. With point convolution (conventional convolution with 1 × 1 window), inter-channel features are projected into a new space. More efficient features are obtained by using a combination of 1 × 1 convolution and 3 × 3 deep convolution instead of 3 × 3 standard convolution. Therefore, various spatial information is extracted from local information to broader generalized information. In this way, different features extracted with varying dilation ratios with different convolution operations will result in greater diversity in the feature extraction process [42], [73], [74]. The mathematical formulation of depthwise convolution is as follows: (3) DeptwiseConv(W,y)(i,j)=∑k,lK,LW(k,l)⊙y(i+k,j+l)
Fig. 3 Dilated Convolution, covering different areas for different dilation ratios when the kernel size (3×3) is selected.
Here, y represents layer, K, and L layer size, i and j layer index, and k, and l filter respectively. W denotes a learnable convolution filter and ⊙ an element-wise multiplication operator.
In addition, if the pneumonia disease has spread over a larger area rather than just one region in X-ray images, it may be necessary to combine features from different observation levels. Therefore, the depthwise dilated convolution technique can be used effectively in the diagnosis of pneumonia [42] (see Fig. 4).
Fig. 4 (a) Normal convolution and (b) depthwise convolutions operations. In depthwise convolutions, the number of filters is equal to the number of channels of the input [75].
3.3 Gradient boosting (GB)
Gradient boosting (GB) is a machine learning algorithm used for classification and regression problems [76]. GB aims to combine strong learner models to obtain a weak learner with high prediction accuracy [77], [78]. The GB method tries to minimize the cost function to find an additive model. Therefore, the GB algorithm iteratively adds weak learners (a new decision tree) to the model, reducing the cost function at the highest rate at each step [77]. The steps of the GB algorithm are given below mathematically [76], [77]:
1. Input variable (x) and target variable (y) are determined. The cost function (L(y,f(x))) is defined.
2. A simple decision tree (DT0) is initialized that establishes the relationship f(x) between x and y. Here, it is aimed to minimize the cost function (f(x) = DT0).
3. A pseudo-residue is defined to obtain a new target variable. The defined pseudo-residue is used as the new target variable (ri=yi−fxi,yi=ri,i=i+1).
4. A new decision tree (DTi) suitable for the pseudo-residual is developed. By including DTi in the model, f(x) is updated ((f(x) = ∑DT).
5. Step 3 and step 4 are repeated for the specified number of cycles.
6. Finally, all decision trees are combined and the result of the GB model is obtained (f(x) = ∑DT).
3.4 Proposed architecture
In this study, a new architecture named CovidDWNet+GB is proposed. This architecture consists of four blocks as shown in detail in Fig. 5. First, the input image is represented in the larger area of the input information by two successive convolution operations. In the second stage, new features are extracted after the obtained information FRB and then DDC operations. The same operations were repeated four times with different filter (f) numbers and dilation ratios (maxd) to increase the depth of the mesh. The depth of the DDC unit was determined by reducing the dilation rate by five at the beginning and by one in the next steps. Then, the obtained feature map is applied to the global average pooling layer (GAP) and three fully connected layers, respectively. In addition, after each convolution operation, Relu was used as the activation function and BatchNormalization was used as the normalization operation. Finally, after the CovidDWNet architecture was trained, the feature vectors obtained from the second fully connected (FC(64, R)) layer were estimated using Gradient Boosting (GB) machines. The proposed architecture is detailed in Fig. 5 and its methodology is shown in Eq. (4)–(5). (4) CovidDWNetx,y=CL−CL−FRBi−DDCi−CLii=1n−GAP−FC1−FC2−Zk
(5) prediction=yout=GB(FC2)
Here, CL stands for convolution layer, GAP global average pooling, FC fully connected, Zk softmax activation function, and GB gradient boosting classifier. {..}i=1n represents the number of repetitions of the operation (n> = 1).
Blocks in the proposed architecture include DDC, FRB units, and Convolution layer. The FRB unit (Fig. 2(a)) consists of four interconnected Convolution layers. The features obtained in the last step are combined with the input features. In this way, it adds depth to the architecture by reusing previous features. The FRB unit is mathematically shown in Eq. (6). (6) x1=Φw1f∗x
x2=Φw2f∗x1
x3=Φw2f∗x2
x4=Φw3f∗x3
FRBk=[x4,x]
Here, wif weights represent xi outputs. * indicates the convolution operation. Φ means applying the Relu and BatchNormalization (BN) operations of the layer output, respectively. [ ] denotes the merge operation. FRBk refers to the FRB operation of the kth block.
The DDC unit, detailed in Fig. 2(b), is expanded with varying dilatation rates and the receptive area of X-ray and CT images. In this way, distinctive features are obtained effectively and more diversity is provided in the feature extraction process. The mathematical notation of the DDC unit used in the proposed architecture is given in Eq. (7). (7) o1=Φ(w1d⊙FRBk)
o2=Φ(w2d⊙o1)
⋮
on=Φ(wnd⊙on−1)
DDCk=[o1,o2…,on]
Here, wid shows weights, oi outputs, and ⊙ DeptwiseConv operation. [o1,o2…,on] means the join operation. DDCk refers to the DDC operation of the kth block.
Finally, the features obtained by FRB and DCC operations in blocks are given to the Convolution layer. In this way, important properties are obtained. The mathematical expression required for this operation is given in Eq. (8). (8) CLk=Φ(wk∗DDCk)
Here wk weights, * denotes the convolution operation.
Fig. 5 Proposed architecture (CovidDWNet+GB).
3.5 Training and optimization of the proposed architecture
The CovidDWNet architecture is trained with a backpropagation algorithm using Cross-Entropy Eq. (9) for training multi-class datasets and binary cross-entropy Eq. (10) cost functions for training two-class datasets. These cost functions can be expressed mathematically as: (9) Lyˆ,y=−∑yilogyˆi
(10) LBCE=−1n∑((yi.log(yˆi))+(1−yi).log(1−yˆi))
Here, n is the number of samples, y is the actual value, and yˆ is the predicted value.
Adam optimization [42], [76] algorithm is used as the optimization algorithm for updating the weights in the architecture. Adam optimization algorithm with learning coefficient η at time t: (11) wt+1j=wtj−ηvtst+ɛ−×gt
(12) vt=β1×vt−1−1−β1×gt
(13) st=β2×st−1−1−β2×gt2
Here w stands for weights, hyperparameters β1 and β2 time t η learning rate coefficient. gt represents the gradient at time t.vt and st represent the exponential mean of gradients and squares of gradients along, wt. In the proposed architecture, the Relu activation function used after each convolution operation is given in Eq. (14) [34]. (14) f(x)Relu=max{0,x}
Fully connected layers (FC) form a fundamental part of CNN architectures, where all neurons in the previous layer connect to all neurons in the next layer and calculate how much each value matches the class. As the last layer, the output of the FC is combined with the activation functions of sigmoid, SVM, softmax, etc. for class prediction. Softmax activation function used for classification in this study, a probability distribution of n number of output categories is calculated according to Eq. (15) [33], [79]. (15) Zk=exk∑i=1nexn
Here, x is the input vector, n is the number of classes, up to, k=1….n, and Z is the output vector. The sum of all Z values is equal to 1.
4 Experimental results and discussion
For the detection of Covid-19, some literature studies based on deep learning using X-ray and CT images are presented in Table 1. It can be said that there are differences in success rates according to the datasets that researchers use by developing different architectures. In general, it is seen that the success rates of studies with two classes are higher than those with multiple classes. Marques et al. [80] performed binary and triple classification on X-ray images with the CNN-based architecture they developed using the EfficientNet architecture. This method has shown the highest success in the classification of binary class (Covid-19, normal) X-ray images with an accuracy rate of 99.6% compared to other architectures. On four-class X-ray datasets, Umer et al. [81] showed the lowest performance with 85.0% accuracy, while the proposed architecture achieved the highest performance with 96.8% accuracy. In addition, when studies using two-class CT images containing Covid-19 and Normal images were examined, Gifani et al. [82] showed the lowest performance with 85% accuracy using CNNs-based architecture, while the recommended architecture showed the highest performance with 100% accuracy. However, when the results here are examined, it is seen that the datasets used by the researchers affect the success rates of the methods. At the same time, the number of samples and the number of classes in the datasets are the factors affecting the success of the architectures.
To evaluate the performance of the architectures developed by the researchers more fairly, it is important to conduct experimental studies using common datasets. Therefore, in this study, four different experimental applications were carried out using three different datasets containing CT and X-ray images for the diagnosis of Covid-19. To objectively evaluate the performance of current architectures mentioned in the literature with CovidDWNet+GB (Our model), training was carried out on the same dataset by keeping certain parameters the same. The results obtained according to different metrics by training each model 200 epochs are presented in Table 4, 5, 6, and 7. Commonly used metrics accuracy, precision, recall, F1-Score, specificity, and AUC were used to evaluate the results. These metrics are: (16) Accuracy=TP+TNTP+TN+FP+FN
(17) Precision/PPV=TPTP+FP
(18) Recall/Sensiviy=TPTP+FN
(19) F1−Score=2∗Precision∗RecallPrecision+Recall
(20) Specificity=TNTN+FP
Table 1 Some deep learning approaches and success results for Covid-19 diagnosis from X-ray and CT images.
Study Architecture Class Scanning Accuracy (%)
Sethy et al. [83] ResNet50 plus 2-class (Covid-19, noncovid-19) CT 95.38
Li et al. [84] Stacked-autoencoder 2-class (Covid-19, Pneumonia, normal) CT 94.7
Gifani et al. [82] CNNs models 2-class( Covid-19, noncovid-19) CT 85.0
Xu et al. [85] ResNet + Loc-ation Attention 3-class (Influenza-A, Normal, covid-19) CT 86.7
Heidarian et al. [86] COVID-FACT 3-class (Covid-19, Pneumonia, normal) CT 90.82
Mukherjee et al. [87] Tailored Deep NN 2-class (Covid-19, noncovid-19) CT 95.83
Mukherjee et al. [87] Tailored Deep NN 2-class (Covid-19, noncovid) X-ray 96.13
Wang et al. [88] COVID-Net 3-class (Covid-19, pneumonia, normal) X-ray 93.3
Heidari et al. [89] VGG16-based CNN 3-class (Covid-19, pneumonia, normal) X-ray 94.5
Chakraborty [90] Corona-Nidaan 3-class (Covid-19, normal, pneumonia,) X-ray 95.0
Umer et al. [81] COVINet 2-class (Covid-19, normal) X-ray 97.0
Umer et al. [81] COVINet 3-class (Covid-19, normal, virus pneumonia) X-ray 90.0
Umer et al. [81] COVINet 4-class (Covid-19, normal, virus pneumonia, bacterial pneumonia) X-ray 85.0
Babukarthik et al. [91] Genetic deep CNN 2- class (Covid-19, normal) X-ray 98.8
Apostolopoulos
et al. [92] Pretrained CNNs 3-class (Covid-19, nonCovid-19 pneumonia, normal) X-ray 96.7
Ismael et al [93] Deep CNNs 2- class (Covid-19, normal) X-ray 92.6
Oh et al. [94] ResNet-18 4-class (Covid-19+viral pneumonia, bacterial pneumonia, tuberculosis, normal) X-ray 91.9
Ezzat et al. [95] GSA-DenseNet121 2-class (Covid-19, pneumonia) X-ray 93.4
Marques et al. [80] CNN + EfficientNet 2- class (Covid-19, normal) X-ray 99.6
Marques et al. [80] CNN + EfficientNet 3- class (Covid-19, nonCovid-19 pneunomia, normal) X-ray 96.7
Hussain et al. [96] CoroDet 2- class (Covid-19, normal) X-ray 99.1
Hussain et al. [96] CoroDet 3- class (Covid-19, normal, pneumonia) X-ray 94.2
Hussain et al. [96] CoroDet 4- class (Covid-19, normal, non-Covid-19 pneumonia, non-Covid-19 bacterial pneumonia) X-ray 91.2
Proposed CovidDWNet 2- class (Covid-19, normal) CT 100.0
proposed CovidDWNet 4-class (Covid-19, Lung Opacity, Normal, Viral pneumonia) X-ray 96.8
Here, TP (True Positives) denotes correctly classified diseased cases, TN (True Negatives) correctly defined healthy cases, FP (False Positives) misclassified diseased cases, FN (False Negatives) misclassified healthy cases [3], [97]. The receiver operating characteristic (ROC) curve is used in classification problems to evaluate the performance of models by plotting the true positive rate (TPR) versus the false positive rate (FPR). The area under the curve (AUC) indicates the area under the ROC, which is a probability curve [3]. (21) FPR=FPFP+TN
(22) TPR=TPTP+FN
The hyperparameters of the architectures during the training phase are presented in Table 2. CovidDWNet architecture, developed on Keras/Tensorflow platform, takes images as input by scaling 128 × 128; Adam (Learning_rate = 0.001) optimization function and batch size value 32 are given.
The image distribution of the datasets used in experimental applications for the detection of Covid-19 and other pneumonia diseases according to training and test sets is given in Table 3 in detail. At the same time, the number of images according to the types of diseases in each application is presented in this table. Datasets are reserved for approximately 80% training and 20% testing. In the first application, the Sars-Cov-2 [64] dataset was used. This dataset is divided into two datasets, training and testing. The training dataset contains a total of 1986 images, 1002 Covid, and 984 Normal. The test dataset contains 495 images, 250 of which are Covid and 245 are normal. Similarly, a second application was performed using the Covid-CT [63] and Sars-Cov-2 [64] datasets containing CT images. In this application, there are 2589 images (1288 Covid, 1301 Normal) in the training dataset and 645 images (320 Covid, 325 Normal) in the test dataset. In the third application, Dataset-X-ray [65] dataset containing X-ray images was used. In this application, there are 16933 images (2893 Covid, 8154 Normal, 4810 Lung Opacity, and 1079 Viral Pneumonia) in the training dataset and 4232 images (723 Covid, 2038 Normal, 1202 Lung Opacity, and 269 Viral Pneumonia) in the test dataset. The fourth application was performed by combining all datasets containing X-ray and CT images. In this application, a total of 19515 images, including 4174 Covid, 9455 Normal, 4810 Lung Opacity, and 1079 Viral Pneumonia, in the training dataset; In the test dataset, there are a total of 4877 images, including 1043 Covid, 2363 Normal, 1202 Lung Opacity and 269 Viral Pneumonia.Table 2 Hyperparameters of architectures for Covid-19 detection.
Model Data augmentation Software Input size Optimizer Learning rate Batch size
DenseNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
AlexNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
ResNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
CspNet [98] No Keras, TensorFlow 224 × 224 Adam 0.0001 32
VGG16 No Keras, TensorFlow 224 × 224 Adam 0.0001 32
VGG19 No Keras, TensorFlow 224 × 224 Adam 0.0001 32
CovXNet [42] Yes Keras, TensorFlow 128 × 128 Adam 0.001 16
CoroNet [40] Yes Keras, TensorFlow 150 × 150 Adam 0.0001 10
CovidXrayNet [41] Yes Fastai, PyTorch 256 × 256 Adam – 32
DarkCovidNet [39] No Fastai, PyTorch 256 × 256 Adam 0.003 32
Proposed (No DDC) No Keras, TensorFlow 128 × 128 Adam 0.001 32
Proposed No Keras, TensorFlow 128 × 128 Adam 0.001 32
The results of the experimental study with the SARS-COV-2 [64] dataset containing CT images for Covid-19 detection are given in Table 4. The success rate has been increased by adding a DDC module to the proposed architecture. In addition, high performance has been achieved by adding GB classifier to the proposed architecture. However, when the data augmentation method is applied to the proposed architecture, it is seen that the success decreases. When the results obtained in this application are examined in a general way, we can say that our model exhibits the highest performance in all metrics with a 100% success rate. Also, the confusion matrix results of the proposed architecture for this application are given in Fig. 6(a). When the results are examined, it is seen that the proposed architecture correctly predicts Covid-19 patients and non-patients with 100% high performance.Table 3 Number of records in datasets used in applications.
Application (s) Data set(s) Image (s) Train/Test Covid Normal Lung Opacity Viral Pneumonia Total
Application1 Sars-Cov-2 [64] CT Train Set 1002 984 – – 1986
Test Set 250 245 – – 495
Application2 Covid-CT [63] and
Sars-Cov-2 [64] CT Train Set 1288 1301 2589
Test Set 320 325 645
Application3 Dataset-X-ray [65] X-ray Train Set 2893 8154 4810 1076 16933
Test Set 723 2038 1202 269 4232
Application4 All datasets X-ray + CT Train Set 4174 9455 4810 1076 19515
Test Set 1043 2363 1202 269 4877
In the second application, an experimental study was performed by combining the Covid-CT [63] and Sars-Cov-2 [64] datasets containing Covid-19 and Normal CT images, and the results are presented in Table 5. When the results are examined, the proposed architecture (CovidDWNet+GB) showed the highest success with 99.84% according to the accuracy metric and 100% (1.00) according to the precision, recall, and F1-Score metrics. Similarly, the CovidDWNet architecture achieved the highest success with 99.85% performance according to specificity and AUC metrics. In addition, when the confusion matrix results are examined in Fig. 6(b), It is seen that the CovidDWNet+GB architecture detects Covid-19 patients and normal people who are not sick with 100% accuracy.Table 4 The success of the models in detecting Covid-19 on the Sars-Cov-2 dataset containing CT images.
Model Accuracy (%) Precision Recall F1-Score Specificity(%) AUC (%)
DenseNet 97.37 0.97 0.97 0.97 98.57 97.37
AlexNet 94.14 0.94 0.94 0.94 94.12 94.12
ResNet 95.96 0.96 0.96 0.96 95.97 95.97
CspNet [98] 95.15 0.95 0.95 0.95 95.14 95.14
VGG16 50.51 0.50 0.34 0.25 50.00 50.00
VGG19 50.51 0.50 0.34 0.25 50.00 50.00
CovXNet [42] 98.18 0.98 0.98 0.98 98.18 98.18
CoroNet [40] 98.59 0.99 0.99 0.99 98.57 98.57
CovidXrayNet [41] 97.97 0.98 0.98 0.98 98.00 98.00
DarkCovidNet [39] 95.35 0.95 0.95 0.95 95.35 95.35
Proposed (No DDC) 98.38 0.98 0.98 0.98 98.39 98.39
Proposed+ DataAug. 97.78 0.98 0.98 0.98 97.78 97.78
Proposed (No GB) 98.59 0.99 0.99 0.99 98.58 98.58
Proposed(CovidDWNet+GB) 100.0 1.00 1.00 1.00 100.0 100.0
In the third experimental study for Covid-19 detection, the four-class Dataset-X-ray [65] dataset was used. The results of the experimental study are shown in Table 6. When the results of the application are examined, our model (CovidDWNet+GB) achieved the highest performance with 96.81% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-Score, 95.54% specificity, and 97.98% AUC. At the same time, when the training and testing times of the third application (Application3) are examined (in Table 8), it is seen that the proposed architecture is faster than the CovidXrayNet and DarkCovidNet architectures, which are the closest to the success rate. In addition, when the success distribution of the CovidDWNet+GB architecture according to classes is analyzed in Fig. 6(c), we can say that it correctly predicts Covid-19 patients 99%, Lung Opacity patients 92%, people who are not sick 98% and Viral Pneumonia patients 100%.Table 5 Success rates of models according to different metrics in detecting Covid-19 on the Covid-CT and Sars-Cov-2 datasets containing CT images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 92.09 0.92 0.92 0.92 92.09 92.09
AlexNet 86.51 0.87 0.86 0.87 86.49 86.49
ResNet 87.44 0.88 0.97 0.87 87.47 87.47
CspNet [98] 85.58 0.86 0.86 0.86 85.53 85.53
VGG16 50.39 0.25 0.50 0.34 50.00 50.00
VGG19 50.39 0.25 0.50 0.34 50.00 50.00
CovXNet [42] 88.99 0.89 0.89 0.89 89.03 89.03
CoroNet [40] 92.25 0.92 0.92 0.92 92.26 92.26
CovidXrayNet [41] 91.16 0.92 0.91 0.91 91.12 91.12
DarkCovidNet [39] 88.92 0.88 0.87 0.87 88.92 88.92
Proposed (No DDC) 91.63 0.92 0.92 0.92 91.62 91.62
Proposed+ DataAug. 86.36 0.86 0.86 0.86 86.33 86.33
Proposed (No GB) 93.33 0.93 0.93 0.93 93.31 93.31
Proposed(CovidDWNet+GB) 99.84 1.0 1.00 1.00 99.85 99.85
In the last experimental study for Covid-19 detection, an application was performed by combining all datasets (Covid-CT, Sars-Cov-2, and Dataset-X-ray). The results obtained according to different metrics are given in Table 7. When the results are examined, we can say that CovidDWNet+GB, 96.32% accuracy, 0.97 precision, 0.97 recall, 0.97 F1-Score, 95.17% specificity, and 97.67% AUC showed the highest success. Also, the confusion matrix results for the CovidDWNet+GB architecture of this application are given in Fig. 6(d). When the results are examined, it is seen that he predicted Covid 19 patients at 97%, Lung Opacity patients at 93%, non-sick people at 97%, and Viral Pneumonia patients at 100% correct.Table 6 Results of architectures for Covid-19 detection according to different metrics on the Dataset-X-ray dataset containing X-ray images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 87.10 0.88 0.89 0.88 87.66 92.12
AlexNet 90.62 0.91 0.91 0.91 88.12 93.72
ResNet 92.70 0.94 0.93 0.94 90.79 95.14
CspNet [98] 82.63 0.79 0.84 0.81 80.09 88.84
VGG16 91.09 0.92 0.91 0.92 87.59 93.88
VGG19 91.33 0.93 0.92 0.92 87.55 93.96
CovXNet [42] 92.44 0.95 0.89 0.91 89.28 92.76
CoroNet [40] 92.11 0.93 0.92 0.92 90.52 94.40
CovidXrayNet [41] 95.39 0.96 0.96 0.96 94.33 96.88
DarkCovidNet [39] 94.33 0.96 0.94 0.95 92.17 95.67
Proposed (No DDC) 93.30 0.94 0.94 0.94 90.75 95.50
Proposed+ DataAug. 93.19 0.95 0.92 0.94 89.72 94.54
Proposed (No GB) 93.76 0.95 0.94 0.94 90.91 95.67
Proposed(CovidDWNet+GB) 96.81 0.98 0.97 0.98 95.54 97.98
According to the experimental application results, the class performances (confusion matrix) of the proposed architecture (CovidDWNet+GB) for Covid-19 detection are given in Fig. 6. It shows the results of the binary classification in (a) and (b), and multi-classification in (c) and (d). (a) gives the results of the first application, (b) the second application, (c) the third application, and (d) the fourth application. In applications containing the proposed architectural CT images (Fig. 6 (a–b)), it appears to predict Covid-19 and Normal images extremely successfully with 100% success rates.Table 7 Performance of Architectures for Covid-19 detection by different metrics on all datasets containing X-ray and CT images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 93.05 0.94 0.93 0.94 90.41 95.11
AlexNet 90.98 0.92 0.91 0.91 86.91 93.66
ResNet 92.02 0.93 0.93 0.93 90.30 94.70
CspNet [98] 90.71 0.92 0.91 0.91 87.14 93.67
VGG16 89.65 0.91 0.90 0.90 84.90 92.88
VGG19 89.24 0.91 0.89 0.90 83.64 92.27
CovXNet [42] 92.45 0.94 0.91 0.93 88.85 93.86
CoroNet [40] 92.23 0.94 0.92 0.93 87.28 94.43
CovidXrayNet [41] 95.30 0.96 0.96 0.96 92.93 96.85
DarkCovidNet [39] 90.59 0.92 0.91 0.91 88.05 93.06
Proposed (No DDC) 92.19 0.93 0.92 0.93 87.69 94.40
Proposed+ DataAug. 91.92 0.94 0.91 0.90 89.90 93.10
Proposed (No GB) 93.36 0.94 0.94 0.94 91.73 95.50
Proposed(CovidDWNet+GB) 96.32 0.97 0.97 0.97 95.17 97.67
Similarly, multiple classification performances of the CovidDWNet architecture are given in Fig. 6-(c) and (d). In the third experimental study including X-ray images, it is seen that the proposed architecture, Covid-19, Lung Opacity, Normal and Viral Pneumonia images were estimated with success rates of 99%, 92%, 98%, and 100%, respectively. In addition, in the fourth application containing all datasets, it was observed that he predicted Covid-19 images with 97%, Lung Opacity images with 93%, Normal images with 97%, and Viral pneumonia images with a rate of 100%. It can be said that it performs extremely satisfactorily in classes other than the Lung Opacity class. It is thought that its lower success in images containing Lung Opacity is due to its overlapping features with other classes.
The ROC curve of our proposed model is shown in Fig. 7. The ROC curve is a graphical representation of the classification performance of the network. The closer the curve is to its upper left limit, the higher the performance. Fig. 7 (a–b) shows the results of the CT images, (c) the results of the X-ray images, and (d) the results of the X-ray and CT images. In CT images, it is seen that AUC values of 99.85% and 100% results are obtained. We can say that AUC values of 97.98% and 97.67% were obtained in X-ray and all images, respectively.Fig. 6 Performance results of the proposed architecture in binary and multiple classes. (a) The first application, (b) second application, (c) third application, (d) fourth application results.
Gradient-based class activation mapping (Grad-CAM) algorithm [99] is used to highlight important points on X-ray and CT images that affect the performance of CNN architectures. The main purpose of this algorithm was developed to create stronger deep networks. The last convolutional layer is considered to be the stage where the best balance is achieved between important spatial information and the highest semantics [100]. Grad-CAM generates heatmap heat zones to highlight key locations from features derived from the final convolution layer. This information indicates which regions the algorithm pays more attention to. In Fig. 8, heatmap and Grad-CAM images obtained for sample Covid-19 images with the Grad-CAM algorithm are given. Green and yellow areas on heatmaps highlight key regions where the CovidDWNet architecture is concentrated. Regions with dark yellow in heatmaps and red in Grad-CAM indicate important regions with high distinctiveness.Fig. 7 ROC analysis of the proposed model. (a) First application, (b) second application, (c) third application, (d) fourth application results.
CNNs are used for classification and recognition problems by making use of fully connected layers of feature maps obtained as a result of the convolution process [101]. Feature maps are obtained with filters defined by convolution operations on the input image. Feature maps obtained for a particular input image are used to understand which features of the input are detected or preserved. It is expected to detect small or fine details from the image given as input to the models. However, the models will capture more general feature maps close to the output [102]. In Fig. 9, an example of tens of feature maps obtained from the images given as an introduction to the CovidDWNet architecture is given. It is seen that the different features of the images are emphasized in the first two convolution layers. These images appear to be understandable images. We can say that the feature maps obtained from the last convolution layer of the next blocks (Block1-4) capture more fine details. These attributes are meaningful features that are not understood by humans but can be understood by CNN models. At the same time, it is possible to say that the feature maps show fewer and fewer details as they go deeper and that these details are meaningful features in the decision-making process by CNN models.Fig. 8 From CovidDWNet architecture using Grad-CAM resulting sample heatmaps and Covid-19 visuals.
Also, the training and test times of the applications are given in Table 8. Training times in hours and minutes; Test times are shown in seconds. Training times, architectures 200 epoch training time; test times represent the time elapsed during the estimation of all samples in the test dataset. When the training and testing times are examined, we can say that the AlexNet architecture has a higher speed compared to other architectures. However, when the overall success of the AlexNet architecture is examined in the experimental applications, it has been observed that it exhibits a low performance.Fig. 9 Feature maps were obtained from sample CT and X-ray images.
The time complexity of the architectures according to the training and test times is given in Fig. 10. When the time complexity diagram is carefully examined, it is seen that the AlexNet architecture has the smallest time complexity. We can also say that the CovXNet architecture has the highest time complexity. It is possible to say that the proposed architecture has moderate time complexity.Table 8 Training and testing times of architectures. Training times in hours and minutes; Test times are shown in seconds.
Model Application1 Application2 Application3 Application4
Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.)
DenseNet 0.53 1 1.18 2 5.00 6 6.10 7
AlexNet 0.53 1 1.16 1 4.50 3 6.04 4
ResNet 1.10 2 1.50 3 6.56 8 7.36 10
CspNet [80] 1.30 2 1.46 3 6.00 7 8.20 9
VGG16 1.07 3 1.30 3 7.50 18 9.13 19
VGG19 1.14 3 1.33 4 8.16 19 9.23 21
CovXNet [42] 5.05 7 6.05 8 15.33 26 18.33 29
CoroNet [40] 1.20 2 1.43 2 6.33 5 7.50 7
CovidXrayNet [41] 1.10 2 1.43 3 8.33 12 9.13 15
DarkCovidNet [39] 1.18 2 1.48 2 7.54 13 8.43 15
Proposed (CovidDWNet) 1.16 2 1.45 3 7.24 8 8.32 11
When the results of experimental studies are examined in general, it is seen that it predicts X-ray and CT images with high performance. A higher success was achieved with CT images compared to X-ray images. We can say that this is due to the more sensitive and finer detailed structures of CT images [13], [42].Fig. 10 The time complexity of architectures: (a) Time complexity based on training times (in hours), (b) Time complexity based on test times (in seconds).
In addition, a higher performance has been achieved by integrating the DDC module into the CovidDWNet architecture, providing different expansion rates and deepening the feature map with depthwise convolution. However, when the data augmentation method is applied to the proposed architecture, it has been observed that it affects success negatively. The hyperparameters and values of the applied data augmentation method are given in Table 9.
Table 9 Data augmentation hyperparameters for the proposed architecture.
Parameters Value
Width shift 0.2
Height shift 0.2
Shear 0.25
Zoom 0.2
Rotation 30
Horizontal flip True
Vertical flip True
5 Conclusion
Covid-19 pandemic cases are increasing day by day and cause the death of many people. It has caused millions of cases and the death of millions of people so far. This disease, which brings with it different health problems, poses serious threats to human health with the emergence of new variations. Many states are taking many measures to prevent the spread of the disease and reduce deaths. RT-PCR tests are generally used to detect this disease. However, considering the inadequacy of RT-PCR tests, the risk of transmission to healthcare personnel, pain to patients, and cost, it brings with it many problems. In this sense, different researches are carried out and different solutions are offered. Deep learning architectures with high performance are one of these studies. When the literature is examined, it is possible to see many studies with deep learning. In these studies, it is seen that only one of the CT or X-ray datasets is used mostly. At the same time, it was seen that the performance evaluation of the studies was limited in themselves.
In this study for the detection of Covid-19 and similar symptoms, datasets containing CT or X-ray images were used. A new architecture is proposed, called CovidDWNet, based on feature reuse residual block (FRB) and Depthwise dilated convolutions (DDC) units. High performance has been achieved by providing the combination of the proposed architecture and the Gradient boosting (GB) algorithm (CovidDWNet+GB). In addition, the current architectures in the literature were examined, the architectures were trained on the same data sets and performance evaluation was made accordingly.
It has been observed that CovidDWNet+GB exhibits the highest success with 99.84% and 100% accuracy rates in applications performed on CT datasets with two classes (Covid-19, and non Covid-19). In addition, it has been observed that it provides the highest success according to precision, recall, F1-Score, specificity, and AUC metrics. The proposed architecture showed the highest success in the application using four classes (Covid-19, Lung Opacity, Normal and Viral Pneumonia) X-ray images, with 96.81% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-Score, 95.54% specificity, and 97.98% AUC. Similarly, we can say that the CovidDWNet+GB architecture showed the highest success in the experimental study using X-ray and CT images, with 96.32% accuracy, 0.97 precision, 0.97 recall, 0.97 F1-Score, 95.17% specificity, and 97.67% AUC. Also, it has been observed that the proposed architecture predicts 4877 images in the test dataset with a high speed of 11 s.
As a result, when the performances of different architectures are examined by keeping certain parameters constant on the same datasets, it is possible to say that the proposed architecture exhibits a respectable success in the literature and shows a remarkable performance among current architectures.
CRediT authorship contribution statement
Gaffari Celik: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Visualization, Investigation, Validation.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
No data was used for the research described in the article.
The code (and data) in this article has been certified as Reproducible by Code Ocean: (https://codeocean.com/). More information on the Reproducibility Badge Initiative is available at https://www.elsevier.com/physical-sciences-and-engineering/computer-science/journals.
==== Refs
References
1 Wu F. A new coronavirus associated with human respiratory disease in China Nature 579 7798 2020 265 269 10.1038/s41586-020-2008-3 32015508
2 Coronavirus disease (COVID-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus-2019.
3 Subramanian N. Elharrouss O. Al-Maadeed S. Chowdhury M. A review of deep learning-based detection methods for COVID-19 Comput. Biol. Med. 143 2022 105233 10.1016/j.compbiomed.2022.105233
4 Rubin G.D. The role of chest imaging in patient management during the COVID-19 pandemic Chest 158 1 2020 106 116 10.1016/j.chest.2020.04.003 32275978
5 Singh R. Corona virus (COVID-19) symptoms prevention and treatment: A short review J. Drug Deliv. Ther. 11 2-S 2021 118 120 10.22270/jddt.v11i2-S.4644
6 R S. An efficient hardware architecture based on an ensemble of deep learning models for COVID -19 prediction Sustain. Cities Soc. 2022 103713 10.1016/j.scs.2022.103713
7 Heidari A. Jafari Navimipour N. Unal M. Toumaj S. The COVID-19 epidemic analysis and diagnosis using deep learning: A systematic literature review and future directions Comput. Biol. Med. 141 2021 2022 105141 10.1016/j.compbiomed.2021.105141
8 Sharfstein J.M. Becker S.J. Mello M.M. Diagnostic testing for the novel coronavirus JAMA 323 15 2020 1437 10.1001/jama.2020.3864 32150622
9 Stephanie S. Determinants of chest radiography sensitivity for COVID-19: A multi-institutional study in the United States Radiol. Cardiothorac. Imaging 2 5 2020 e200337 10.1148/ryct.2020200337
10 Liu R. Clinica Chimica Acta positive rate of RT-PCR detection of SARS-CoV-2 infection in 4880 cases from one hospital in Wuhan, China, from Jan to 2020 Clin. Chim. Acta 505 March 2020 172 175 10.1016/j.cca.2020.03.009 32156607
11 Dramé M. Should RT-PCR be considered a gold standard in the diagnosis of COVID-19? J. Med. Virol. 92 11 2020 2312 2313 10.1002/jmv.25996 32383182
12 Xie J. Characteristics of patients with coronavirus disease (COVID-19) confirmed using an IgM-IgG antibody test J. Med. Virol. 92 10 2020 2004 2010 10.1002/jmv.25930 32330303
13 Hassan H. Supervised and weakly supervised deep learning models for COVID-19 CT diagnosis: A systematic review Comput. Methods Programs Biomed. 218 2022 106731 10.1016/j.cmpb.2022.106731
14 Gaur P. Malaviya V. Gupta A. Bhatia G. Pachori R.B. Sharma D. COVID-19 disease identification from chest CT images using empirical wavelet transformation and transfer learning Biomed. Signal Process. Control 71 PA 2022 103076 10.1016/j.bspc.2021.103076
15 Ucar F. Korkmaz D. COVIDiagnosis-Net: Deep Bayes-SqueezeNet based diagnosis of the coronavirus disease, 2019 (COVID-19) from X-ray images Med. Hypotheses 140 April 2020 109761 10.1016/j.mehy.2020.109761
16 Başaran E. Classification of white blood cells with SVM by selecting SqueezeNet and LIME properties by mRMR method Signal, Image Video Process 2022 10.1007/s11760-022-02141-2
17 Çelik G. Talu M.F. A new 3D MRI segmentation method based on generative adversarial network and atrous convolution Biomed. Signal Process. Control 71 PA 2022 103155 10.1016/j.bspc.2021.103155
18 Goodfellow I. Generative adversarial networks Commun. ACM 63 11 2020 139 144 10.1145/3422622
19 Çelik G. Talu M.F. Generating the image viewed from EEG signals Pamukkale Univ. J. Eng. Sci. 27 2 2021 129 138 10.5505/pajes.2020.76399
20 Esteva A. Dermatologist-level classification of skin cancer with deep neural networks Nature 542 7639 2017 115 118 10.1038/nature21056 28117445
21 Tan J.H. Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network Inf. Sci. (Ny) 420 2017 66 76 10.1016/j.ins.2017.08.050
22 Başaran E. Cömert Z. Çelik Y. Neighbourhood component analysis and deep feature-based diagnosis model for middle ear otoscope images Neural Comput. Appl. 2022 10.1007/s00521-021-06810-0
23 Celik Y. Talo M. Yildirim O. Karabatak M. Acharya U.R. Automated invasive ductal carcinoma detection based using deep transfer learning with whole-slide images Pattern Recognit. Lett. 133 2020 232 239 10.1016/j.patrec.2020.03.011
24 Bozdag Z. Talu F.M. Pyramidal nonlocal network for histopathological image of breast lymph node segmentation Int. J. Comput. Intell. Syst. 14 1 2021 122 131 10.2991/ijcis.d.201030.001
25 Talo M. Yildirim O. Baloglu U.B. Aydin G. Acharya U.R. Convolutional neural networks for multi-class brain disease detection using MRI images Comput. Med. Imaging Graph. 78 2019 101673 10.1016/j.compmedimag.2019.101673
26 Gaál G. Maga B. Lukács A. Attention U-net based adversarial architectures for chest X-ray lung segmentation CEUR Workshop Proc. 2692 2020 1 7
27 Souza J.C. Bandeira Diniz J.O. Ferreira J.L. França da Silva G.L. Corrêa Silva A. de Paiva A.C. An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks Comput. Methods Programs Biomed. 177 2019 285 296 10.1016/j.cmpb.2019.06.005 31319957
28 Yıldırım Ö. Pławiak P. Tan R.S. Acharya U.R. Arrhythmia detection using deep convolutional neural network with long duration ECG signals Comput. Biol. Med. 102 September 2018 411 420 10.1016/j.compbiomed.2018.09.009 30245122
29 Hannun A.Y. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network Nature Med. 25 1 2019 65 69 10.1038/s41591-018-0268-3 30617320
30 Acharya U.R. A deep convolutional neural network model to classify heartbeats Comput. Biol. Med. 89 August 2017 389 396 10.1016/j.compbiomed.2017.08.022 28869899
31 Rajpurkar P. CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning 2017 3 9 [Online]. Available: http://arxiv.org/abs/1711.05225
32 Ieracitano C. A fuzzy-enhanced deep learning approach for early detection of Covid-19 pneumonia from portable chest X-ray images Neurocomputing 481 2022 202 215 10.1016/j.neucom.2022.01.055 35079203
33 Ahamed K.U. A deep learning approach using effective preprocessing techniques to detect COVID-19 from chest CT-scan and X-ray images Comput. Biol. Med. 139 October 2021 105014 10.1016/j.compbiomed.2021.105014
34 Verma H. Mandal S. Gupta A. Temporal deep learning architecture for prediction of COVID-19 cases in India Expert Syst. Appl. 195 January 2021 116611 10.1016/j.eswa.2022.116611
35 Khan S.H. COVID-19 detection in chest X-ray images using deep boosted hybrid learning Comput. Biol. Med. 137 August 2021 104816 10.1016/j.compbiomed.2021.104816
36 Loey M. El-Sappagh S. Mirjalili S. Bayesian-based optimized deep learning model to detect COVID-19 patients using chest X-ray image data Comput. Biol. Med. 142 January 2022 105213 10.1016/j.compbiomed.2022.105213
37 Lahsaini I. El Habib Daho M. Chikh M.A. Deep transfer learning based classification model for COVID-19 using chest CT-scans Pattern Recognit. Lett. 152 2021 122 128 10.1016/j.patrec.2021.08.035 34566222
38 Toğaçar M. Ergen B. Cömert Z. COVID-19 detection using deep learning models to exploit social mimic optimization and structured chest X-ray images using fuzzy color and stacking approaches Comput. Biol. Med. 121 March 2020 10.1016/j.compbiomed.2020.103805
39 Ozturk T. Talo M. Yildirim E.A. Baloglu U.B. Yildirim O. Rajendra Acharya U. Automated detection of COVID-19 cases using deep neural networks with X-ray images Comput. Biol. Med. 121 April 2020 103792 10.1016/j.compbiomed.2020.103792
40 Khan A.I. Shah J.L. Bhat M.M. CoroNet: A deep neural network for detection and diagnosis of COVID-19 from chest x-ray images Comput. Methods Programs Biomed. 196 2020 105581 10.1016/j.cmpb.2020.105581
41 Monshi M.M.A. Poon J. Chung V. Monshi F.M. CovidXrayNet: Optimizing data augmentation and CNN hyperparameters for improved COVID-19 detection from CXR Comput. Biol. Med. 133 March 2021 104375 10.1016/j.compbiomed.2021.104375
42 Mahmud T. Rahman M.A. Fattah S.A. CovXNet: A multi-dilation convolutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization Comput. Biol. Med. 122 June 2020 103869 10.1016/j.compbiomed.2020.103869
43 Calderon-Ramirez S. Yang S. Elizondo D. Moemeni A. Dealing with distribution mismatch in semi-supervised deep learning for COVID-19 detection using chest X-ray images: A novel approach using feature densities Appl. Soft Comput. 123 2022 108983 10.1016/j.asoc.2022.108983
44 Gupta A. Anjum Gupta S. Katarya R. InstaCovNet-19: A deep learning classification model for the detection of COVID-19 patients using Chest X-ray Appl. Soft Comput. 99 2021 106859 10.1016/j.asoc.2020.106859
45 Feki I. Ammar S. Kessentini Y. Muhammad K. Federated learning for COVID-19 screening from Chest X-ray images Appl. Soft Comput. 106 2021 107330 10.1016/j.asoc.2021.107330
46 de Moura J. Novo J. Ortega M. Fully automatic deep convolutional approaches for the analysis of COVID-19 using chest X-ray images Appl. Soft Comput. 115 2022 108190 10.1016/j.asoc.2021.108190
47 Shankar K. An optimal cascaded recurrent neural network for intelligent COVID-19 detection using Chest X-ray images Appl. Soft Comput. 113 2021 107878 10.1016/j.asoc.2021.107878
48 Albahli S. Ayub N. Shiraz M. Coronavirus disease (COVID-19) detection using X-ray images and enhanced DenseNet Appl. Soft Comput. 110 2021 107645 10.1016/j.asoc.2021.107645
49 Elazab A. Elfattah M.A. Zhang Y. Novel multi-site graph convolutional network with supervision mechanism for COVID-19 diagnosis from X-ray radiographs Appl. Soft Comput. 114 2022 108041 10.1016/j.asoc.2021.108041
50 Ozcan T. A new composite approach for COVID-19 detection in X-ray images using deep features Appl. Soft Comput. 111 2021 107669 10.1016/j.asoc.2021.107669
51 Calderon-Ramirez S. Correcting data imbalance for semi-supervised COVID-19 detection using X-ray chest images Appl. Soft Comput. 111 2021 107692 10.1016/j.asoc.2021.107692
52 Karthik R. Menaka R. Hariharan M. Learning distinctive filters for COVID-19 detection from chest X-ray using shuffled residual CNN Appl. Soft Comput. 99 2021 106744 10.1016/j.asoc.2020.106744
53 Demir F. DeepCoroNet: A deep LSTM approach for automated detection of COVID-19 cases from chest X-ray images Appl. Soft Comput. 103 2021 107160 10.1016/j.asoc.2021.107160
54 Zhou T. Lu H. Yang Z. Qiu S. Huo B. Dong Y. The ensemble deep learning model for novel COVID-19 on CT images Appl. Soft Comput. 98 2021 106885 10.1016/j.asoc.2020.106885
55 Bandyopadhyay R. Basu A. Cuevas E. Sarkar R. Harris Hawks optimisation with simulated annealing as a deep feature selection method for screening of COVID-19 CT-scans Appl. Soft Comput. 111 2021 107698 10.1016/j.asoc.2021.107698
56 Ye Q. Robust weakly supervised learning for COVID-19 recognition using multi-center CT images Appl. Soft Comput. 116 2022 108291 10.1016/j.asoc.2021.108291
57 Song L. A deep fuzzy model for diagnosis of COVID-19 from CT images Appl. Soft Comput. 122 2022 108883 10.1016/j.asoc.2022.108883
58 Liang S. Nie R. Cao J. Wang X. Zhang G. FCF: Feature complement fusion network for detecting COVID-19 through CT scan images Appl. Soft Comput. 125 2022 109111 10.1016/j.asoc.2022.109111
59 Saygılı A. A new approach for computer-aided detection of coronavirus (COVID-19) from CT and X-ray images using machine learning methods Appl. Soft Comput. 105 2021 107323 10.1016/j.asoc.2021.107323
60 Naeem H. Bin-Salem A.A. A CNN-LSTM network with multi-level feature extraction-based approach for automated detection of coronavirus from CT scan and X-ray images Appl. Soft Comput. 113 2021 107918 10.1016/j.asoc.2021.107918
61 Vinod D.N. Jeyavadhanam B.R. Zungeru A.M. Prabaharan S.R.S. Fully automated unified prognosis of Covid-19 chest X-ray/CT scan images using Deep Covix-Net model Comput. Biol. Med. 136 August 2021 104729 10.1016/j.compbiomed.2021.104729
62 Li J. Multi-task contrastive learning for automatic CT and X-ray diagnosis of COVID-19 Pattern Recognit. 114 2021 107848 10.1016/j.patcog.2021.107848
63 Yang X. He X. Zhao J. Zhang Y. Zhang S. Xie P. COVID-CT-dataset: A CT scan dataset about COVID-19 2020 1 14 [Online]. Available: http://arxiv.org/abs/2003.13865
64 Soares E. Angelov P. Biaso S. Froes M.H. Abe D.K. SARS-CoV-2 CT-scan dataset: A large dataset of real patients CT scans for SARS-CoV-2 identification 2020 medRxiv, p. 2020.04.24.20078584, [Online]. Available: https://www.medrxiv.org/content/10.1101/2020.04.24.20078584v3%0Ahttps://www.medrxiv.org/content/10.1101/2020.04.24.20078584v3.abstract
65 Chowdhury M.E.H. Can AI help in screening viral and COVID-19 pneumonia? IEEE Access 8 July 2020 132665 132676 10.1109/ACCESS.2020.3010287
66 Gu J. Recent advances in convolutional neural networks Pattern Recognit. 77 2018 354 377 10.1016/j.patcog.2017.10.013
67 Budak Ü. Cömert Z. Çıbuk M. Şengür A. DCCMED-Net: Densely connected and concatenated multi Encoder-Decoder CNNs for retinal vessel extraction from fundus images Med. Hypotheses 134 2019 2020 10.1016/j.mehy.2019.109426
68 Ren F. Liu W. Wu G. Feature reuse residual networks for insect pest recognition IEEE Access 7 2019 122758 122768 10.1109/ACCESS.2019.2938194
69 He K. Zhang X. Ren S. Sun J. Deep residual learning for image recognition 2015 [Online]. Available: http://arxiv.org/abs/1512.03385
70 Kim S. Park I. Kwon S. Han J. Motion retargetting based on dilated convolutions and skeleton-specific loss functions Comput. Graph. Forum 39 2 2020 497 507 10.1111/cgf.13947
71 Sooksatra S. Kondo T. Bunnun P. Yoshitaka A. Redesigned skip-network for crowd counting with dilated convolution and backward connection J. Imaging 6 5 2020 10.3390/JIMAGING6050028
72 Li X. Zhai M. Sun J. DDCNNC: Dilated and depthwise separable convolutional neural network for diagnosis COVID-19 via chest X-ray images Int. J. Cogn. Comput. Eng. 2 April 2021 71 82 10.1016/j.ijcce.2021.04.001
73 Chollet F. Xception: Deep learning with depthwise separable convolutions Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, 2017-Janua 2017 1800 1807 10.1109/CVPR.2017.195
74 Ma Y. Wang C. SdcNet for object recognition Comput. Vis. Image Underst. 215 2020 103332 10.1016/j.cviu.2021.103332 2022
75 Wang C.-F. A basic introduction to separable convolutions 2018 https://l24.im/hrH8qwp. (Accessed 22 Nov. 2021)
76 Friedman J.H. Greedy function approximation: A gradient boosting machine Ann. Statist. 29 5 2001 1189 1232 10.1214/aos/1013203451
77 Chen H. Shen Z. Wang L. Qi C. Tian Y. Prediction of undrained failure envelopes of skirted circular foundations using gradient boosting machine algorithm Ocean Eng. 258 May 2022 111767 10.1016/j.oceaneng.2022.111767
78 Touzani S. Granderson J. Fernandes S. Gradient boosting machine for modeling the energy consumption of commercial buildings Energy Build. 158 2018 1533 1543 10.1016/j.enbuild.2017.11.039
79 Gao B. Pavel L. On the properties of the softmax function with application in game theory and reinforcement learning 2017 1 10 [Online]. Available: http://arxiv.org/abs/1704.00805
80 Marques G. Agarwal D. de la Torre Díez I. Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network Appl. Soft Comput. 96 2020 106691 10.1016/j.asoc.2020.106691
81 Umer M. Ashraf I. Ullah S. Mehmood A. Choi G.S. COVINet: a convolutional neural network approach for predicting COVID-19 from chest X-ray images J. Ambient Intell. Humaniz. Comput. 13 1 2022 535 547 10.1007/s12652-021-02917-3 33527000
82 gifani P. Shalbaf A. Vafaeezadeh M. Automated detection of COVID-19 using ensemble of transfer learning with deep convolutional neural network based on CT scans Int. J. Comput. Assist. Radiol. Surg. 16 1 2021 115 123 10.1007/s11548-020-02286-w 33191476
83 Sethy P.K. Behera S.K. Ratha P.K. Biswas P. Detection of coronavirus disease (COVID-19) based on deep features Int. J. Math. Eng. Manag. Sci. 5 4 2020 643 651 10.20944/preprints202003.0300.v1
84 Li D. Fu Z. Xu J. Stacked-autoencoder-based model for COVID-19 diagnosis on CT images Appl. Intell. 51 5 2021 2805 2817 10.1007/s10489-020-02002-w
85 Xu X. A deep learning system to screen novel coronavirus disease 2019 pneumonia Engineering 6 10 2020 1122 1129 10.1016/j.eng.2020.04.010 32837749
86 Heidarian S. COVID-FACT: A fully-automated capsule network-based framework for identification of COVID-19 cases from chest CT scans Front. Artif. Intell. 4 May 2021 1 13 10.3389/frai.2021.598932
87 Mukherjee H. Ghosh S. Dhar A. Obaidullah S.M. Santosh K.C. Roy K. Deep neural network to detect COVID-19: one architecture for both CT scans and chest X-rays Appl. Intell. 51 5 2021 2777 2789 10.1007/s10489-020-01943-6
88 Wang L. Lin Z.Q. Wong A. COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images Sci. Rep. 10 1 2020 19549 10.1038/s41598-020-76550-z 33177550
89 Heidari M. Mirniaharikandehei S. Khuzani A.Z. Danala G. Qiu Y. Zheng B. Improving the performance of CNN to predict the likelihood of COVID-19 using chest X-ray images with preprocessing algorithms Int. J. Med. Inform. 144 June 2020 104284 10.1016/j.ijmedinf.2020.104284
90 Chakraborty M. Dhavale S.V. Ingole J. Corona-Nidaan: lightweight deep convolutional neural network for chest X-ray based COVID-19 infection detection Appl. Intell. 51 5 2021 3026 3043 10.1007/s10489-020-01978-9
91 Babukarthik R.G. Ananth Krishna Adiga V. Sambasivam G. Chandramohan D. Amudhavel A.J. Prediction of COVID-19 using genetic deep learning convolutional neural network (GDCNN) IEEE Access 8 2020 177647 177666 10.1109/ACCESS.2020.3025164 34786292
92 Apostolopoulos I.D. Mpesiana T.A. COVID-19: automatic detection from X-ray images utilizing transfer learning with convolutional neural networks Phys. Eng. Sci. Med. 43 2 2020 635 640 10.1007/s13246-020-00865-4 32524445
93 Ismael A.M. Şengür A. Deep learning approaches for COVID-19 detection based on chest X-ray images Expert Syst. Appl. 164 2020 2021 10.1016/j.eswa.2020.114054
94 Oh Y. Park S. Ye J.C. Deep learning COVID-19 features on CXR using limited training data sets IEEE Trans. Med. Imaging 39 8 2020 2688 2700 10.1109/TMI.2020.2993291 32396075
95 Ezzat D. Hassanien A.E. Ella H.A. An optimized deep learning architecture for the diagnosis of COVID-19 disease based on gravitational search optimization Appl. Soft Comput. 98 2021 106742 10.1016/j.asoc.2020.106742
96 Hussain E. Hasan M. Rahman M.A. Lee I. Tamanna T. Parvez M.Z. CoroDet: A deep learning based classification for COVID-19 detection using chest X-ray images Chaos Solitons Fractals 142 2021 110495 10.1016/j.chaos.2020.110495
97 Başaran E. Cömert Z. Çelik Y. Convolutional neural network approach for automatic tympanic membrane detection and classification Biomed. Signal Process. Control 56 2020 10.1016/j.bspc.2019.101734
98 Wang C.Y. Mark Liao H.Y. Wu Y.H. Chen P.Y. Hsieh J.W. Yeh I.H. CSPNet: A new backbone that can enhance learning capability of CNN IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. Work., 2020-June 2020 1571 1580 10.1109/CVPRW50498.2020.00203
99 Selvaraju R.R. Cogswell M. Das A. Vedantam R. Parikh D. Batra D. Grad-CAM: Visual explanations from deep networks via gradient-based localization Proc. IEEE Int. Conf. Comput. Vis., 2017-Octob 2017 618 626 10.1109/ICCV.2017.74
100 Abbasniya M.R. Sheikholeslamzadeh S.A. Nasiri H. Emami S. Classification of Breast Tumours Based on Histopathology Images using Deep Features and Ensemble of Gradient Boosting Methods, Vol. 103 2022 10.1016/j.compeleceng.2022.108382 arXiv Prepr. arXiv2209.01380. June. 108382
101 Kim H. Jung W.K. Park Y.C. Lee J.W. Ahn S.H. Broken stitch detection method for sewing operation using CNN feature map and image-processing techniques Expert Syst. Appl. 188 2022 116014 10.1016/j.eswa.2021.116014
102 Brownlee J. How to visualize filters and feature maps in convolutional neural networks 2019 https://3c5.com/w7im4. (Accessed 25 Sep. 2022)
| 0 | PMC9726415 | NO-CC CODE | 2022-12-08 23:18:16 | no | Air Med J. 2022 Dec 7 November-December; 41(6):571 | latin-1 | Air Med J | 2,022 | 10.1016/j.amj.2022.10.006 | oa_other |
==== Front
Appl Soft Comput
Appl Soft Comput
Applied Soft Computing
1568-4946
1872-9681
Elsevier B.V.
S1568-4946(22)00955-3
10.1016/j.asoc.2022.109906
109906
Article
Detection of Covid-19 and other pneumonia cases from CT and X-ray chest images using deep learning based on feature reuse residual block and depthwise dilated convolutions neural network
Celik Gaffari
Agri Ibrahim Cecen University, Department of Computer Technology, Agri, Turkey
7 12 2022
1 2023
7 12 2022
133 109906109906
26 3 2022
29 11 2022
1 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Covid-19 has become a worldwide epidemic which has caused the death of millions in a very short time. This disease, which is transmitted rapidly, has mutated and different variations have emerged. Early diagnosis is important to prevent the spread of this disease. In this study, a new deep learning-based architecture is proposed for rapid detection of Covid-19 and other symptoms using CT and X-ray chest images. This method, called CovidDWNet, is based on a structure based on feature reuse residual block (FRB) and depthwise dilated convolutions (DDC) units. The FRB and DDC units efficiently acquired various features in the chest scan images and it was seen that the proposed architecture significantly improved its performance. In addition, the feature maps obtained with the CovidDWNet architecture were estimated with the Gradient boosting (GB) algorithm. With the CovidDWNet+GB architecture, which is a combination of CovidDWNet and GB, a performance increase of approximately 7% in CT images and between 3% and 4% in X-ray images has been achieved. The CovidDWNet+GB architecture achieved the highest success compared to other architectures, with 99.84% and 100% accuracy rates, respectively, on different datasets containing binary class (Covid-19 and Normal) CT images. Similarly, the proposed architecture showed the highest success with 96.81% accuracy in multi-class (Covid-19, Lung Opacity, Normal and Viral Pneumonia) X-ray images and 96.32% accuracy in the dataset containing X-ray and CT images. When the time to predict the disease in CT or X-ray images is examined, it is possible to say that it has a high speed because the CovidDWNet+GB method predicts thousands of images within seconds.
Keywords
Covid-19 diagnosis
Deep learning
Depthwise dilated convolutions
Feature reuse residual block
Gradient boosting
==== Body
pmcCode metadata
Permanent link to reproducible Capsule: https://doi.org/10.24433/CO.2183919.v1.
1 Introduction
Coronavirus (Covid-19) is a disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-COV-2). After emerging in Wuhan, China in December 2019, it soon spread around the world and became a global pandemic [1]. According to the data of the World Health Organization (WHO), it has been determined that more than 410 million cases have been seen so far, and close to 6 million people have died [2]. WHO declared the coronavirus infection as a Covid-19 pandemic in March 2020 due to the increasing number of deaths and cases. Due to the increasing cases and deaths, many states have had to close their borders to prevent the spread of the pandemic. In addition, many countries have imposed curfews for a certain period as a precaution [3].
This disease usually affects the respiratory system, such as the lungs, and also appears to cause pneumonia-like symptoms [4]. Patients commonly experience symptoms such as fever, cough, sneezing, and shortness of breath. It spreads rapidly through respiratory droplets produced by the cough or sneeze of an infected person. Elderly people and people with chronic illnesses appear to be more prone to Covid-19 infection [5].
One of the most common methods used to diagnose Covid-19 is reverse transcription-polymerase chain reaction (RT-PCR) tests. These tests are performed to determine whether individuals have been infected with SARS-COV-2, the virus that causes Covid-19 disease, momentarily or in the past. The disadvantages of these tests are that test results take time, the number of available RT-PCR test kits is low, and the risk of health personnel contracting the disease during the test is high [6]. It is also costly in that special equipment, materials, and tools are often required for RT-PCR examinations. Therefore, many countries have difficulties in procuring test kits due to budgetary and technical constraints [7]. At the same time, the sensitivity of the RT-PCR test is a cause for concern because of sample and laboratory errors that may occur [8], [9]. Liu et al. [10] have expressed their opinion on the poor performance of RT-PCR in its sensitivity. Similarly, in a study conducted by Drame et al. [11], they expressed their reservations about the use of RT-PCR to determine the viral load in the diagnosis of 2019 coronavirus disease (Covid-19). In addition, it was stated in another study that the sensitivity of these tests could be as low as 38% [12].
Covid-19, which manifests itself as a lung infection, computed tomography (CT) and chest X-ray (X-ray) images are other methods used for the detection of this disease [5]. Typical radiographic features can be reliably detected in patients with pneumonia caused by this disease with CT imaging. Although these methods have some advantages over RT-PCR testing in terms of early detection of Covid-19, specialist physicians are needed to understand and make sense of images. Considering the disadvantages of RT-PCR tests, CT and X-ray imaging techniques used in the diagnosis of Covid-19, Artificial Intelligence (AI), and Deep Learning (DL) based methods are seen as alternative methods. AI and DL methods can help the early diagnosis of this disease and make the treatment process faster by leading experts to reach a fast and accurate diagnosis through CT and X-ray images in the detection process of Covid-19 [13], [14], [15].
Artificial intelligence and deep learning methods are widely used by researchers for the detection of Covid-19 infection from X-ray and CT images. Due to the improved performance of deep learning methods, they are widely used compared to traditional methods. One of the most important reasons that led researchers to this field is that, unlike machine learning and traditional methods, in deep learning architectures, there is no need for feature extraction in the data during the preprocessing stage. Deep learning architectures can be trained with the help of the hyperparameters of the convolutional neural network (CNN) architecture to learn the best features according to the dataset used [3]. Researchers used deep learning methods in many areas classification of white blood cells [16], segmentation of brain MRI images [17], synthetic image generation [18], generating images from EEG signals [19], skin cancer classification [20], fundus image segmentation [21], diagnose different types of Otitis media [22], breast cancer detection [23], breast lymph node segmentation [24], brain disease classification [25], lung segmentation [26], [27], detection of arrhythmia [28], [29], [30] and detecting pneumonia from chest X-ray images [31]. With the pandemic, the use of deep learning methods for the detection of coronavirus symptoms from X-ray and CT images has increased significantly [3].
In literature, it can be seen that many deep learning-based studies have been carried out for the diagnosis of Covid-19 with the help of radiological images [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42].
In the study by Leracitano et al. [32], The authors proposed a fuzzy logic-based deep learning approach to differentiate X-ray images of patients with Covid-19 pneumonia and non-Covid-19-related interstitial pneumonia. The model developed here, called CovNNet, uses the blurry edge detection algorithm together with the blurry images to extract some relevant features from the X-ray images. This study [32] for the detection of Covid-19 from binary class (Covid, Non-Covid) X-ray images, performed poorly compared to many other studies in the literature, with an accuracy rate of 81%. Ahamed et al. [33] proposed a deep learning-based Covid-19 case detection model trained with a dataset of chest CT scans and X-ray images. A modified ResNet50V2 architecture is used as a deep learning architecture in the proposed model. High performance was achieved in this study using two, three, and four classes CT and X-ray images. However, a complex architecture with high processing power was used. Verma et al. [34] proposed different models such as vanilla (vanilla) LSTM, stacked LSTM, ED_LSTM, BiLSTM, CNN, and hybrid CNN+LSTM model to capture the complex trend of the COVID-19 outbreak and perform the Covid-19 prediction. In another study by Khan et al. [35], two new deep learning-based models named deep hybrid learning (DHL) and deep boosted hybrid learning (DBHL) are proposed for effective Covid-19 detection in X-ray datasets. In the proposed DHL architecture, the representation learning capability of the two developed COVID-RENet-1 & 2 models and a machine learning classifier is used separately. In the Covid-RENet model, region and edge-based attention mechanisms were applied to extract boundary features and learn region homogeneity. In addition, the transfer learning method was used in chest X-rays in the proposed architectures. In this study, in which two-class (Covid, Non-Covid) X-ray images are used, it is seen that it has an accuracy of 98.53%. In this study, performance evaluation with only binary class X-ray images is seen as a disadvantage in terms of the performance of the architectures. Because it is important to use different data sets for Covid-19 detection. The success of CNN architectures may vary according to the number of classes and image type.
In the study by Loey et al. [36], a bayesian optimization-based CNN model was proposed for the classification of chest X-ray images. In the proposed model, CNN architecture is used to extract and learn deep features. In addition, CNN hyperparameters are adjusted according to an objective function using a Bayes-based optimizer method. In another study by Lahsaini et al. [37], they used a dataset of Covid and non-Covid CT images validated by RT-PCR tests at Tlemcen hospital in Algeria. A comparative study was carried out on Inception, Resnet-V2, VGG16, VGG19, DenseNet121, DenseNet201, ImageNet, and Xception deep models using the transfer learning method. Also, a model based on DenseNet201 architecture and the GradCam algorithm is proposed. In another study by Toğaçar et al. [38], images were preprocessed using the fuzzy color technique to classify X-ray images. Then, the features obtained with MobileNetV2, and SqueezeNet models were processed with the help of the social mimic optimization method. The productive features obtained were classified using support vector machines (SVM). The DarkCovidNet method developed by Ozturk et al. [39] was used as a classifier for the YOLO real-time object detection system. By applying seventeen convolution layers and adding different filtering to each layer. As in previous studies for the detection of Covid-19, only CT in [37] and only X-ray images were used in [38], [39].
In addition, when the studies were examined, different models were developed by the researchers, defined by the names CoroNet [40], CovidXrayNet [41], and CovXNet [42]. The CoroNet [40] model, which is proposed as a deep CNN model, is based on the Xception architecture pre-trained on the ImageNet dataset. The CovidXrayNet [41] method, based on the EfficientNet-B0 model and based on the optimization method, is proposed. In this study [41], the data augmentation method is used to increase accuracy and CNN hyperparameters are optimized. In the CovXNet [42] technique, deep CNN-based architecture and a model that uses depthwise convolution and varying dilation rates to extract features efficiently are proposed. In the proposed method, different forms of CovXNets are designed and trained with X-ray images of various resolutions. In addition, a stacking algorithm was used to increase the performance rate, and abnormal regions of X-ray images were distinguished by integrating a gradient-based discriminative localization. Looking at the time complexity of the CovXNets architecture (Fig. 10), it was seen that it had a higher time complexity compared to the other architectures studied. This shows that the CovXNets architecture has a complex structure.Fig. 1 Example images are included in the datasets.
When the studies for the detection of Covid-19 disease with deep learning methods are examined, it is seen that researchers generally use X-ray [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53] or CT [54], [55], [56], [57], [58] images, but few studies use both X-ray and CT images [59], [60], [61], [62]. At the same time, it has been seen that the researchers only examined the performance of their architectures on the dataset used or compared them only with traditional architectures. In addition, it has been determined that no performance evaluation has been made according to the training and test times in the literature reviews. In our study, contrary to these studies, CT and X-ray images were used, and a performance evaluation was carried out on the same dataset, including traditional architectures as well as different current architectures. In addition, performance evaluation was made by considering the training and test times of the architectures. The proposed model was developed to reduce the workload of specialist physicians by providing effective, efficient, and rapid detection of Covid-19 and similar cases. With the thought that the CovidDWNet+GB architecture will guide different studies, it has been opened to everyone’s access on the Github page (https://github.com/GaffariCelik/Covid-19).
Our main contributions to this work are listed as follows:
• A new deep learning-based model (CovidDWNet) has been proposed for the detection of Covid-19 and other pneumonia cases.
• The performance of the CovidDWNet architecture has been increased by using multiple feature reuse residual blocks and depthwise dilated convolutions neural networks. In addition, the success rate has been increased by performing the disease prediction process with the Gradient boosting algorithm of the feature vectors obtained with the CovidDWNet architecture.
• By using different CT and X-ray datasets, a real performance evaluation was made among the current architectures in the literature.
2 Material
In this study, three datasets were used: Covid-CT [63] and Sars-Cov-2 [64] datasets containing CT images, and Dataset-X-ray [65] dataset containing X-ray images. These datasets have been made publicly available for researchers to carry out their work. Example images in datasets are given in Fig. 1.
The Covid-CT [63] consists of 812 CT images, 349 of which are Covid-19 and 463 normal, taken from 216 patients. This dataset has been validated by a senior radiologist at Tongji Hospital in Wuhan, China, who diagnosed and treated a large number of Covid-19 patients at the time of emergence of this disease between January and April 2019. The Sars-Cov-2 [64] dataset contains a total of 2482 CT chest scan images, of which 1252 are Covid-19 and 1230 are normal. This dataset was obtained from different hospitals in Sao Paulo, Brazil. The Database-X-ray [65] dataset was created for COVID-19 positive cases with the collaboration of a team of researchers from the University of Qatar, Dhaka University, Bangladesh, and medical doctors in Pakistan and Malaysia. This dataset includes X-ray images of Covid-19, normal and other lung infection diseases. It consists of 21165 X-ray images in total, including 3616 Covid-19, 10192 normal, 6012 lung opacity (Non-Covid lung infection), and 1345 Viral pneumonia.
3 Method
As a method, a CNN-based architecture has been proposed for the detection of Covid-19 and other pneumonia symptoms. This architecture is a method based on feature reuse residual block (FRB) and depthwise dilated convolutions (DDC) units.
Convolutional Neural Networks (CNNs) are models that provide high classification performance in multi-class problems and have self-learning capabilities. CNNs are coordinated combinations of multilayer perceptrons, in which every neuron in one layer is associated with all neurons in the next. A convolutional network consists of a convolutional layer and a rectified linear unit (ReLu). Convolution layers form the basic structure of CNN models. Inputs are convolutionalized and applied with nuclei across the entire visual field with convolution filters. Thus, simpler, small patterns are obtained with more complex, detailed patterns. In this way, the hierarchical network structure enables the extraction of the highest feature maps, enhanced generalization capability, and reduced computational complexity [15], [53], [66], [67]. The basic convolution operation can be written mathematically as [33]: (1) Fi,j=I∗Ki,j=∑m∑nI(i+m,j+n)K(m,n)
Here, I represents the input matrix (may be an image), mxn filter size, and K the two-dimensional filter.
3.1 Feature reuse residual block (FRB)
The feature reuse block (FRB) is a widely used technique in computer vision. In this method, feature maps of previous layers are given as input to all subsequent layers. Thus, the performance of the network is highly increased by reusing the features of the previous layers in all subsequent layers [68], [69]. The mathematical formulation of the FRB technique used in this study is defined as follows: (2) FRB=y=Fx,wiox
Here, x and y represent the inputs and outputs of the considered layers, wi weights, and the combination of those features. The architecture of the FRB technique is presented in Fig. 2(a).
Fig. 2 Feature reuse residual block architecture-(a) and depthwise dilated convolutions architecture-(b), which constitute the basic structure of the proposed architecture.
3.2 Depthwise dilated convolutions (DDC)
As shown in Fig. 3, dilated convolutions can be expanded in the scope of the convolution kernel by changing the dilation ratio compared to standard convolution. By expanding the scope of the convolution kernel, multi-scale features (information) can be obtained. Choosing a dilation rate of one captures the same properties as standard convolution. However, in dilated convolution, when the dilation ratio is selected as greater than one, more detailed features can be obtained than in standard convolution [70], [71], [72].
In depthwise dilated convolutions operation, the convolution operation is applied to each input channel separately. With point convolution (conventional convolution with 1 × 1 window), inter-channel features are projected into a new space. More efficient features are obtained by using a combination of 1 × 1 convolution and 3 × 3 deep convolution instead of 3 × 3 standard convolution. Therefore, various spatial information is extracted from local information to broader generalized information. In this way, different features extracted with varying dilation ratios with different convolution operations will result in greater diversity in the feature extraction process [42], [73], [74]. The mathematical formulation of depthwise convolution is as follows: (3) DeptwiseConv(W,y)(i,j)=∑k,lK,LW(k,l)⊙y(i+k,j+l)
Fig. 3 Dilated Convolution, covering different areas for different dilation ratios when the kernel size (3×3) is selected.
Here, y represents layer, K, and L layer size, i and j layer index, and k, and l filter respectively. W denotes a learnable convolution filter and ⊙ an element-wise multiplication operator.
In addition, if the pneumonia disease has spread over a larger area rather than just one region in X-ray images, it may be necessary to combine features from different observation levels. Therefore, the depthwise dilated convolution technique can be used effectively in the diagnosis of pneumonia [42] (see Fig. 4).
Fig. 4 (a) Normal convolution and (b) depthwise convolutions operations. In depthwise convolutions, the number of filters is equal to the number of channels of the input [75].
3.3 Gradient boosting (GB)
Gradient boosting (GB) is a machine learning algorithm used for classification and regression problems [76]. GB aims to combine strong learner models to obtain a weak learner with high prediction accuracy [77], [78]. The GB method tries to minimize the cost function to find an additive model. Therefore, the GB algorithm iteratively adds weak learners (a new decision tree) to the model, reducing the cost function at the highest rate at each step [77]. The steps of the GB algorithm are given below mathematically [76], [77]:
1. Input variable (x) and target variable (y) are determined. The cost function (L(y,f(x))) is defined.
2. A simple decision tree (DT0) is initialized that establishes the relationship f(x) between x and y. Here, it is aimed to minimize the cost function (f(x) = DT0).
3. A pseudo-residue is defined to obtain a new target variable. The defined pseudo-residue is used as the new target variable (ri=yi−fxi,yi=ri,i=i+1).
4. A new decision tree (DTi) suitable for the pseudo-residual is developed. By including DTi in the model, f(x) is updated ((f(x) = ∑DT).
5. Step 3 and step 4 are repeated for the specified number of cycles.
6. Finally, all decision trees are combined and the result of the GB model is obtained (f(x) = ∑DT).
3.4 Proposed architecture
In this study, a new architecture named CovidDWNet+GB is proposed. This architecture consists of four blocks as shown in detail in Fig. 5. First, the input image is represented in the larger area of the input information by two successive convolution operations. In the second stage, new features are extracted after the obtained information FRB and then DDC operations. The same operations were repeated four times with different filter (f) numbers and dilation ratios (maxd) to increase the depth of the mesh. The depth of the DDC unit was determined by reducing the dilation rate by five at the beginning and by one in the next steps. Then, the obtained feature map is applied to the global average pooling layer (GAP) and three fully connected layers, respectively. In addition, after each convolution operation, Relu was used as the activation function and BatchNormalization was used as the normalization operation. Finally, after the CovidDWNet architecture was trained, the feature vectors obtained from the second fully connected (FC(64, R)) layer were estimated using Gradient Boosting (GB) machines. The proposed architecture is detailed in Fig. 5 and its methodology is shown in Eq. (4)–(5). (4) CovidDWNetx,y=CL−CL−FRBi−DDCi−CLii=1n−GAP−FC1−FC2−Zk
(5) prediction=yout=GB(FC2)
Here, CL stands for convolution layer, GAP global average pooling, FC fully connected, Zk softmax activation function, and GB gradient boosting classifier. {..}i=1n represents the number of repetitions of the operation (n> = 1).
Blocks in the proposed architecture include DDC, FRB units, and Convolution layer. The FRB unit (Fig. 2(a)) consists of four interconnected Convolution layers. The features obtained in the last step are combined with the input features. In this way, it adds depth to the architecture by reusing previous features. The FRB unit is mathematically shown in Eq. (6). (6) x1=Φw1f∗x
x2=Φw2f∗x1
x3=Φw2f∗x2
x4=Φw3f∗x3
FRBk=[x4,x]
Here, wif weights represent xi outputs. * indicates the convolution operation. Φ means applying the Relu and BatchNormalization (BN) operations of the layer output, respectively. [ ] denotes the merge operation. FRBk refers to the FRB operation of the kth block.
The DDC unit, detailed in Fig. 2(b), is expanded with varying dilatation rates and the receptive area of X-ray and CT images. In this way, distinctive features are obtained effectively and more diversity is provided in the feature extraction process. The mathematical notation of the DDC unit used in the proposed architecture is given in Eq. (7). (7) o1=Φ(w1d⊙FRBk)
o2=Φ(w2d⊙o1)
⋮
on=Φ(wnd⊙on−1)
DDCk=[o1,o2…,on]
Here, wid shows weights, oi outputs, and ⊙ DeptwiseConv operation. [o1,o2…,on] means the join operation. DDCk refers to the DDC operation of the kth block.
Finally, the features obtained by FRB and DCC operations in blocks are given to the Convolution layer. In this way, important properties are obtained. The mathematical expression required for this operation is given in Eq. (8). (8) CLk=Φ(wk∗DDCk)
Here wk weights, * denotes the convolution operation.
Fig. 5 Proposed architecture (CovidDWNet+GB).
3.5 Training and optimization of the proposed architecture
The CovidDWNet architecture is trained with a backpropagation algorithm using Cross-Entropy Eq. (9) for training multi-class datasets and binary cross-entropy Eq. (10) cost functions for training two-class datasets. These cost functions can be expressed mathematically as: (9) Lyˆ,y=−∑yilogyˆi
(10) LBCE=−1n∑((yi.log(yˆi))+(1−yi).log(1−yˆi))
Here, n is the number of samples, y is the actual value, and yˆ is the predicted value.
Adam optimization [42], [76] algorithm is used as the optimization algorithm for updating the weights in the architecture. Adam optimization algorithm with learning coefficient η at time t: (11) wt+1j=wtj−ηvtst+ɛ−×gt
(12) vt=β1×vt−1−1−β1×gt
(13) st=β2×st−1−1−β2×gt2
Here w stands for weights, hyperparameters β1 and β2 time t η learning rate coefficient. gt represents the gradient at time t.vt and st represent the exponential mean of gradients and squares of gradients along, wt. In the proposed architecture, the Relu activation function used after each convolution operation is given in Eq. (14) [34]. (14) f(x)Relu=max{0,x}
Fully connected layers (FC) form a fundamental part of CNN architectures, where all neurons in the previous layer connect to all neurons in the next layer and calculate how much each value matches the class. As the last layer, the output of the FC is combined with the activation functions of sigmoid, SVM, softmax, etc. for class prediction. Softmax activation function used for classification in this study, a probability distribution of n number of output categories is calculated according to Eq. (15) [33], [79]. (15) Zk=exk∑i=1nexn
Here, x is the input vector, n is the number of classes, up to, k=1….n, and Z is the output vector. The sum of all Z values is equal to 1.
4 Experimental results and discussion
For the detection of Covid-19, some literature studies based on deep learning using X-ray and CT images are presented in Table 1. It can be said that there are differences in success rates according to the datasets that researchers use by developing different architectures. In general, it is seen that the success rates of studies with two classes are higher than those with multiple classes. Marques et al. [80] performed binary and triple classification on X-ray images with the CNN-based architecture they developed using the EfficientNet architecture. This method has shown the highest success in the classification of binary class (Covid-19, normal) X-ray images with an accuracy rate of 99.6% compared to other architectures. On four-class X-ray datasets, Umer et al. [81] showed the lowest performance with 85.0% accuracy, while the proposed architecture achieved the highest performance with 96.8% accuracy. In addition, when studies using two-class CT images containing Covid-19 and Normal images were examined, Gifani et al. [82] showed the lowest performance with 85% accuracy using CNNs-based architecture, while the recommended architecture showed the highest performance with 100% accuracy. However, when the results here are examined, it is seen that the datasets used by the researchers affect the success rates of the methods. At the same time, the number of samples and the number of classes in the datasets are the factors affecting the success of the architectures.
To evaluate the performance of the architectures developed by the researchers more fairly, it is important to conduct experimental studies using common datasets. Therefore, in this study, four different experimental applications were carried out using three different datasets containing CT and X-ray images for the diagnosis of Covid-19. To objectively evaluate the performance of current architectures mentioned in the literature with CovidDWNet+GB (Our model), training was carried out on the same dataset by keeping certain parameters the same. The results obtained according to different metrics by training each model 200 epochs are presented in Table 4, 5, 6, and 7. Commonly used metrics accuracy, precision, recall, F1-Score, specificity, and AUC were used to evaluate the results. These metrics are: (16) Accuracy=TP+TNTP+TN+FP+FN
(17) Precision/PPV=TPTP+FP
(18) Recall/Sensiviy=TPTP+FN
(19) F1−Score=2∗Precision∗RecallPrecision+Recall
(20) Specificity=TNTN+FP
Table 1 Some deep learning approaches and success results for Covid-19 diagnosis from X-ray and CT images.
Study Architecture Class Scanning Accuracy (%)
Sethy et al. [83] ResNet50 plus 2-class (Covid-19, noncovid-19) CT 95.38
Li et al. [84] Stacked-autoencoder 2-class (Covid-19, Pneumonia, normal) CT 94.7
Gifani et al. [82] CNNs models 2-class( Covid-19, noncovid-19) CT 85.0
Xu et al. [85] ResNet + Loc-ation Attention 3-class (Influenza-A, Normal, covid-19) CT 86.7
Heidarian et al. [86] COVID-FACT 3-class (Covid-19, Pneumonia, normal) CT 90.82
Mukherjee et al. [87] Tailored Deep NN 2-class (Covid-19, noncovid-19) CT 95.83
Mukherjee et al. [87] Tailored Deep NN 2-class (Covid-19, noncovid) X-ray 96.13
Wang et al. [88] COVID-Net 3-class (Covid-19, pneumonia, normal) X-ray 93.3
Heidari et al. [89] VGG16-based CNN 3-class (Covid-19, pneumonia, normal) X-ray 94.5
Chakraborty [90] Corona-Nidaan 3-class (Covid-19, normal, pneumonia,) X-ray 95.0
Umer et al. [81] COVINet 2-class (Covid-19, normal) X-ray 97.0
Umer et al. [81] COVINet 3-class (Covid-19, normal, virus pneumonia) X-ray 90.0
Umer et al. [81] COVINet 4-class (Covid-19, normal, virus pneumonia, bacterial pneumonia) X-ray 85.0
Babukarthik et al. [91] Genetic deep CNN 2- class (Covid-19, normal) X-ray 98.8
Apostolopoulos
et al. [92] Pretrained CNNs 3-class (Covid-19, nonCovid-19 pneumonia, normal) X-ray 96.7
Ismael et al [93] Deep CNNs 2- class (Covid-19, normal) X-ray 92.6
Oh et al. [94] ResNet-18 4-class (Covid-19+viral pneumonia, bacterial pneumonia, tuberculosis, normal) X-ray 91.9
Ezzat et al. [95] GSA-DenseNet121 2-class (Covid-19, pneumonia) X-ray 93.4
Marques et al. [80] CNN + EfficientNet 2- class (Covid-19, normal) X-ray 99.6
Marques et al. [80] CNN + EfficientNet 3- class (Covid-19, nonCovid-19 pneunomia, normal) X-ray 96.7
Hussain et al. [96] CoroDet 2- class (Covid-19, normal) X-ray 99.1
Hussain et al. [96] CoroDet 3- class (Covid-19, normal, pneumonia) X-ray 94.2
Hussain et al. [96] CoroDet 4- class (Covid-19, normal, non-Covid-19 pneumonia, non-Covid-19 bacterial pneumonia) X-ray 91.2
Proposed CovidDWNet 2- class (Covid-19, normal) CT 100.0
proposed CovidDWNet 4-class (Covid-19, Lung Opacity, Normal, Viral pneumonia) X-ray 96.8
Here, TP (True Positives) denotes correctly classified diseased cases, TN (True Negatives) correctly defined healthy cases, FP (False Positives) misclassified diseased cases, FN (False Negatives) misclassified healthy cases [3], [97]. The receiver operating characteristic (ROC) curve is used in classification problems to evaluate the performance of models by plotting the true positive rate (TPR) versus the false positive rate (FPR). The area under the curve (AUC) indicates the area under the ROC, which is a probability curve [3]. (21) FPR=FPFP+TN
(22) TPR=TPTP+FN
The hyperparameters of the architectures during the training phase are presented in Table 2. CovidDWNet architecture, developed on Keras/Tensorflow platform, takes images as input by scaling 128 × 128; Adam (Learning_rate = 0.001) optimization function and batch size value 32 are given.
The image distribution of the datasets used in experimental applications for the detection of Covid-19 and other pneumonia diseases according to training and test sets is given in Table 3 in detail. At the same time, the number of images according to the types of diseases in each application is presented in this table. Datasets are reserved for approximately 80% training and 20% testing. In the first application, the Sars-Cov-2 [64] dataset was used. This dataset is divided into two datasets, training and testing. The training dataset contains a total of 1986 images, 1002 Covid, and 984 Normal. The test dataset contains 495 images, 250 of which are Covid and 245 are normal. Similarly, a second application was performed using the Covid-CT [63] and Sars-Cov-2 [64] datasets containing CT images. In this application, there are 2589 images (1288 Covid, 1301 Normal) in the training dataset and 645 images (320 Covid, 325 Normal) in the test dataset. In the third application, Dataset-X-ray [65] dataset containing X-ray images was used. In this application, there are 16933 images (2893 Covid, 8154 Normal, 4810 Lung Opacity, and 1079 Viral Pneumonia) in the training dataset and 4232 images (723 Covid, 2038 Normal, 1202 Lung Opacity, and 269 Viral Pneumonia) in the test dataset. The fourth application was performed by combining all datasets containing X-ray and CT images. In this application, a total of 19515 images, including 4174 Covid, 9455 Normal, 4810 Lung Opacity, and 1079 Viral Pneumonia, in the training dataset; In the test dataset, there are a total of 4877 images, including 1043 Covid, 2363 Normal, 1202 Lung Opacity and 269 Viral Pneumonia.Table 2 Hyperparameters of architectures for Covid-19 detection.
Model Data augmentation Software Input size Optimizer Learning rate Batch size
DenseNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
AlexNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
ResNet No Keras, TensorFlow 224 × 224 Adam 0.0001 32
CspNet [98] No Keras, TensorFlow 224 × 224 Adam 0.0001 32
VGG16 No Keras, TensorFlow 224 × 224 Adam 0.0001 32
VGG19 No Keras, TensorFlow 224 × 224 Adam 0.0001 32
CovXNet [42] Yes Keras, TensorFlow 128 × 128 Adam 0.001 16
CoroNet [40] Yes Keras, TensorFlow 150 × 150 Adam 0.0001 10
CovidXrayNet [41] Yes Fastai, PyTorch 256 × 256 Adam – 32
DarkCovidNet [39] No Fastai, PyTorch 256 × 256 Adam 0.003 32
Proposed (No DDC) No Keras, TensorFlow 128 × 128 Adam 0.001 32
Proposed No Keras, TensorFlow 128 × 128 Adam 0.001 32
The results of the experimental study with the SARS-COV-2 [64] dataset containing CT images for Covid-19 detection are given in Table 4. The success rate has been increased by adding a DDC module to the proposed architecture. In addition, high performance has been achieved by adding GB classifier to the proposed architecture. However, when the data augmentation method is applied to the proposed architecture, it is seen that the success decreases. When the results obtained in this application are examined in a general way, we can say that our model exhibits the highest performance in all metrics with a 100% success rate. Also, the confusion matrix results of the proposed architecture for this application are given in Fig. 6(a). When the results are examined, it is seen that the proposed architecture correctly predicts Covid-19 patients and non-patients with 100% high performance.Table 3 Number of records in datasets used in applications.
Application (s) Data set(s) Image (s) Train/Test Covid Normal Lung Opacity Viral Pneumonia Total
Application1 Sars-Cov-2 [64] CT Train Set 1002 984 – – 1986
Test Set 250 245 – – 495
Application2 Covid-CT [63] and
Sars-Cov-2 [64] CT Train Set 1288 1301 2589
Test Set 320 325 645
Application3 Dataset-X-ray [65] X-ray Train Set 2893 8154 4810 1076 16933
Test Set 723 2038 1202 269 4232
Application4 All datasets X-ray + CT Train Set 4174 9455 4810 1076 19515
Test Set 1043 2363 1202 269 4877
In the second application, an experimental study was performed by combining the Covid-CT [63] and Sars-Cov-2 [64] datasets containing Covid-19 and Normal CT images, and the results are presented in Table 5. When the results are examined, the proposed architecture (CovidDWNet+GB) showed the highest success with 99.84% according to the accuracy metric and 100% (1.00) according to the precision, recall, and F1-Score metrics. Similarly, the CovidDWNet architecture achieved the highest success with 99.85% performance according to specificity and AUC metrics. In addition, when the confusion matrix results are examined in Fig. 6(b), It is seen that the CovidDWNet+GB architecture detects Covid-19 patients and normal people who are not sick with 100% accuracy.Table 4 The success of the models in detecting Covid-19 on the Sars-Cov-2 dataset containing CT images.
Model Accuracy (%) Precision Recall F1-Score Specificity(%) AUC (%)
DenseNet 97.37 0.97 0.97 0.97 98.57 97.37
AlexNet 94.14 0.94 0.94 0.94 94.12 94.12
ResNet 95.96 0.96 0.96 0.96 95.97 95.97
CspNet [98] 95.15 0.95 0.95 0.95 95.14 95.14
VGG16 50.51 0.50 0.34 0.25 50.00 50.00
VGG19 50.51 0.50 0.34 0.25 50.00 50.00
CovXNet [42] 98.18 0.98 0.98 0.98 98.18 98.18
CoroNet [40] 98.59 0.99 0.99 0.99 98.57 98.57
CovidXrayNet [41] 97.97 0.98 0.98 0.98 98.00 98.00
DarkCovidNet [39] 95.35 0.95 0.95 0.95 95.35 95.35
Proposed (No DDC) 98.38 0.98 0.98 0.98 98.39 98.39
Proposed+ DataAug. 97.78 0.98 0.98 0.98 97.78 97.78
Proposed (No GB) 98.59 0.99 0.99 0.99 98.58 98.58
Proposed(CovidDWNet+GB) 100.0 1.00 1.00 1.00 100.0 100.0
In the third experimental study for Covid-19 detection, the four-class Dataset-X-ray [65] dataset was used. The results of the experimental study are shown in Table 6. When the results of the application are examined, our model (CovidDWNet+GB) achieved the highest performance with 96.81% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-Score, 95.54% specificity, and 97.98% AUC. At the same time, when the training and testing times of the third application (Application3) are examined (in Table 8), it is seen that the proposed architecture is faster than the CovidXrayNet and DarkCovidNet architectures, which are the closest to the success rate. In addition, when the success distribution of the CovidDWNet+GB architecture according to classes is analyzed in Fig. 6(c), we can say that it correctly predicts Covid-19 patients 99%, Lung Opacity patients 92%, people who are not sick 98% and Viral Pneumonia patients 100%.Table 5 Success rates of models according to different metrics in detecting Covid-19 on the Covid-CT and Sars-Cov-2 datasets containing CT images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 92.09 0.92 0.92 0.92 92.09 92.09
AlexNet 86.51 0.87 0.86 0.87 86.49 86.49
ResNet 87.44 0.88 0.97 0.87 87.47 87.47
CspNet [98] 85.58 0.86 0.86 0.86 85.53 85.53
VGG16 50.39 0.25 0.50 0.34 50.00 50.00
VGG19 50.39 0.25 0.50 0.34 50.00 50.00
CovXNet [42] 88.99 0.89 0.89 0.89 89.03 89.03
CoroNet [40] 92.25 0.92 0.92 0.92 92.26 92.26
CovidXrayNet [41] 91.16 0.92 0.91 0.91 91.12 91.12
DarkCovidNet [39] 88.92 0.88 0.87 0.87 88.92 88.92
Proposed (No DDC) 91.63 0.92 0.92 0.92 91.62 91.62
Proposed+ DataAug. 86.36 0.86 0.86 0.86 86.33 86.33
Proposed (No GB) 93.33 0.93 0.93 0.93 93.31 93.31
Proposed(CovidDWNet+GB) 99.84 1.0 1.00 1.00 99.85 99.85
In the last experimental study for Covid-19 detection, an application was performed by combining all datasets (Covid-CT, Sars-Cov-2, and Dataset-X-ray). The results obtained according to different metrics are given in Table 7. When the results are examined, we can say that CovidDWNet+GB, 96.32% accuracy, 0.97 precision, 0.97 recall, 0.97 F1-Score, 95.17% specificity, and 97.67% AUC showed the highest success. Also, the confusion matrix results for the CovidDWNet+GB architecture of this application are given in Fig. 6(d). When the results are examined, it is seen that he predicted Covid 19 patients at 97%, Lung Opacity patients at 93%, non-sick people at 97%, and Viral Pneumonia patients at 100% correct.Table 6 Results of architectures for Covid-19 detection according to different metrics on the Dataset-X-ray dataset containing X-ray images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 87.10 0.88 0.89 0.88 87.66 92.12
AlexNet 90.62 0.91 0.91 0.91 88.12 93.72
ResNet 92.70 0.94 0.93 0.94 90.79 95.14
CspNet [98] 82.63 0.79 0.84 0.81 80.09 88.84
VGG16 91.09 0.92 0.91 0.92 87.59 93.88
VGG19 91.33 0.93 0.92 0.92 87.55 93.96
CovXNet [42] 92.44 0.95 0.89 0.91 89.28 92.76
CoroNet [40] 92.11 0.93 0.92 0.92 90.52 94.40
CovidXrayNet [41] 95.39 0.96 0.96 0.96 94.33 96.88
DarkCovidNet [39] 94.33 0.96 0.94 0.95 92.17 95.67
Proposed (No DDC) 93.30 0.94 0.94 0.94 90.75 95.50
Proposed+ DataAug. 93.19 0.95 0.92 0.94 89.72 94.54
Proposed (No GB) 93.76 0.95 0.94 0.94 90.91 95.67
Proposed(CovidDWNet+GB) 96.81 0.98 0.97 0.98 95.54 97.98
According to the experimental application results, the class performances (confusion matrix) of the proposed architecture (CovidDWNet+GB) for Covid-19 detection are given in Fig. 6. It shows the results of the binary classification in (a) and (b), and multi-classification in (c) and (d). (a) gives the results of the first application, (b) the second application, (c) the third application, and (d) the fourth application. In applications containing the proposed architectural CT images (Fig. 6 (a–b)), it appears to predict Covid-19 and Normal images extremely successfully with 100% success rates.Table 7 Performance of Architectures for Covid-19 detection by different metrics on all datasets containing X-ray and CT images.
Model Accuracy Precision Recall F1-Score Specificity AUC (%)
DenseNet 93.05 0.94 0.93 0.94 90.41 95.11
AlexNet 90.98 0.92 0.91 0.91 86.91 93.66
ResNet 92.02 0.93 0.93 0.93 90.30 94.70
CspNet [98] 90.71 0.92 0.91 0.91 87.14 93.67
VGG16 89.65 0.91 0.90 0.90 84.90 92.88
VGG19 89.24 0.91 0.89 0.90 83.64 92.27
CovXNet [42] 92.45 0.94 0.91 0.93 88.85 93.86
CoroNet [40] 92.23 0.94 0.92 0.93 87.28 94.43
CovidXrayNet [41] 95.30 0.96 0.96 0.96 92.93 96.85
DarkCovidNet [39] 90.59 0.92 0.91 0.91 88.05 93.06
Proposed (No DDC) 92.19 0.93 0.92 0.93 87.69 94.40
Proposed+ DataAug. 91.92 0.94 0.91 0.90 89.90 93.10
Proposed (No GB) 93.36 0.94 0.94 0.94 91.73 95.50
Proposed(CovidDWNet+GB) 96.32 0.97 0.97 0.97 95.17 97.67
Similarly, multiple classification performances of the CovidDWNet architecture are given in Fig. 6-(c) and (d). In the third experimental study including X-ray images, it is seen that the proposed architecture, Covid-19, Lung Opacity, Normal and Viral Pneumonia images were estimated with success rates of 99%, 92%, 98%, and 100%, respectively. In addition, in the fourth application containing all datasets, it was observed that he predicted Covid-19 images with 97%, Lung Opacity images with 93%, Normal images with 97%, and Viral pneumonia images with a rate of 100%. It can be said that it performs extremely satisfactorily in classes other than the Lung Opacity class. It is thought that its lower success in images containing Lung Opacity is due to its overlapping features with other classes.
The ROC curve of our proposed model is shown in Fig. 7. The ROC curve is a graphical representation of the classification performance of the network. The closer the curve is to its upper left limit, the higher the performance. Fig. 7 (a–b) shows the results of the CT images, (c) the results of the X-ray images, and (d) the results of the X-ray and CT images. In CT images, it is seen that AUC values of 99.85% and 100% results are obtained. We can say that AUC values of 97.98% and 97.67% were obtained in X-ray and all images, respectively.Fig. 6 Performance results of the proposed architecture in binary and multiple classes. (a) The first application, (b) second application, (c) third application, (d) fourth application results.
Gradient-based class activation mapping (Grad-CAM) algorithm [99] is used to highlight important points on X-ray and CT images that affect the performance of CNN architectures. The main purpose of this algorithm was developed to create stronger deep networks. The last convolutional layer is considered to be the stage where the best balance is achieved between important spatial information and the highest semantics [100]. Grad-CAM generates heatmap heat zones to highlight key locations from features derived from the final convolution layer. This information indicates which regions the algorithm pays more attention to. In Fig. 8, heatmap and Grad-CAM images obtained for sample Covid-19 images with the Grad-CAM algorithm are given. Green and yellow areas on heatmaps highlight key regions where the CovidDWNet architecture is concentrated. Regions with dark yellow in heatmaps and red in Grad-CAM indicate important regions with high distinctiveness.Fig. 7 ROC analysis of the proposed model. (a) First application, (b) second application, (c) third application, (d) fourth application results.
CNNs are used for classification and recognition problems by making use of fully connected layers of feature maps obtained as a result of the convolution process [101]. Feature maps are obtained with filters defined by convolution operations on the input image. Feature maps obtained for a particular input image are used to understand which features of the input are detected or preserved. It is expected to detect small or fine details from the image given as input to the models. However, the models will capture more general feature maps close to the output [102]. In Fig. 9, an example of tens of feature maps obtained from the images given as an introduction to the CovidDWNet architecture is given. It is seen that the different features of the images are emphasized in the first two convolution layers. These images appear to be understandable images. We can say that the feature maps obtained from the last convolution layer of the next blocks (Block1-4) capture more fine details. These attributes are meaningful features that are not understood by humans but can be understood by CNN models. At the same time, it is possible to say that the feature maps show fewer and fewer details as they go deeper and that these details are meaningful features in the decision-making process by CNN models.Fig. 8 From CovidDWNet architecture using Grad-CAM resulting sample heatmaps and Covid-19 visuals.
Also, the training and test times of the applications are given in Table 8. Training times in hours and minutes; Test times are shown in seconds. Training times, architectures 200 epoch training time; test times represent the time elapsed during the estimation of all samples in the test dataset. When the training and testing times are examined, we can say that the AlexNet architecture has a higher speed compared to other architectures. However, when the overall success of the AlexNet architecture is examined in the experimental applications, it has been observed that it exhibits a low performance.Fig. 9 Feature maps were obtained from sample CT and X-ray images.
The time complexity of the architectures according to the training and test times is given in Fig. 10. When the time complexity diagram is carefully examined, it is seen that the AlexNet architecture has the smallest time complexity. We can also say that the CovXNet architecture has the highest time complexity. It is possible to say that the proposed architecture has moderate time complexity.Table 8 Training and testing times of architectures. Training times in hours and minutes; Test times are shown in seconds.
Model Application1 Application2 Application3 Application4
Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.) Train time (hr.min.) Test time (s.)
DenseNet 0.53 1 1.18 2 5.00 6 6.10 7
AlexNet 0.53 1 1.16 1 4.50 3 6.04 4
ResNet 1.10 2 1.50 3 6.56 8 7.36 10
CspNet [80] 1.30 2 1.46 3 6.00 7 8.20 9
VGG16 1.07 3 1.30 3 7.50 18 9.13 19
VGG19 1.14 3 1.33 4 8.16 19 9.23 21
CovXNet [42] 5.05 7 6.05 8 15.33 26 18.33 29
CoroNet [40] 1.20 2 1.43 2 6.33 5 7.50 7
CovidXrayNet [41] 1.10 2 1.43 3 8.33 12 9.13 15
DarkCovidNet [39] 1.18 2 1.48 2 7.54 13 8.43 15
Proposed (CovidDWNet) 1.16 2 1.45 3 7.24 8 8.32 11
When the results of experimental studies are examined in general, it is seen that it predicts X-ray and CT images with high performance. A higher success was achieved with CT images compared to X-ray images. We can say that this is due to the more sensitive and finer detailed structures of CT images [13], [42].Fig. 10 The time complexity of architectures: (a) Time complexity based on training times (in hours), (b) Time complexity based on test times (in seconds).
In addition, a higher performance has been achieved by integrating the DDC module into the CovidDWNet architecture, providing different expansion rates and deepening the feature map with depthwise convolution. However, when the data augmentation method is applied to the proposed architecture, it has been observed that it affects success negatively. The hyperparameters and values of the applied data augmentation method are given in Table 9.
Table 9 Data augmentation hyperparameters for the proposed architecture.
Parameters Value
Width shift 0.2
Height shift 0.2
Shear 0.25
Zoom 0.2
Rotation 30
Horizontal flip True
Vertical flip True
5 Conclusion
Covid-19 pandemic cases are increasing day by day and cause the death of many people. It has caused millions of cases and the death of millions of people so far. This disease, which brings with it different health problems, poses serious threats to human health with the emergence of new variations. Many states are taking many measures to prevent the spread of the disease and reduce deaths. RT-PCR tests are generally used to detect this disease. However, considering the inadequacy of RT-PCR tests, the risk of transmission to healthcare personnel, pain to patients, and cost, it brings with it many problems. In this sense, different researches are carried out and different solutions are offered. Deep learning architectures with high performance are one of these studies. When the literature is examined, it is possible to see many studies with deep learning. In these studies, it is seen that only one of the CT or X-ray datasets is used mostly. At the same time, it was seen that the performance evaluation of the studies was limited in themselves.
In this study for the detection of Covid-19 and similar symptoms, datasets containing CT or X-ray images were used. A new architecture is proposed, called CovidDWNet, based on feature reuse residual block (FRB) and Depthwise dilated convolutions (DDC) units. High performance has been achieved by providing the combination of the proposed architecture and the Gradient boosting (GB) algorithm (CovidDWNet+GB). In addition, the current architectures in the literature were examined, the architectures were trained on the same data sets and performance evaluation was made accordingly.
It has been observed that CovidDWNet+GB exhibits the highest success with 99.84% and 100% accuracy rates in applications performed on CT datasets with two classes (Covid-19, and non Covid-19). In addition, it has been observed that it provides the highest success according to precision, recall, F1-Score, specificity, and AUC metrics. The proposed architecture showed the highest success in the application using four classes (Covid-19, Lung Opacity, Normal and Viral Pneumonia) X-ray images, with 96.81% accuracy, 0.98 precision, 0.97 recall, 0.98 F1-Score, 95.54% specificity, and 97.98% AUC. Similarly, we can say that the CovidDWNet+GB architecture showed the highest success in the experimental study using X-ray and CT images, with 96.32% accuracy, 0.97 precision, 0.97 recall, 0.97 F1-Score, 95.17% specificity, and 97.67% AUC. Also, it has been observed that the proposed architecture predicts 4877 images in the test dataset with a high speed of 11 s.
As a result, when the performances of different architectures are examined by keeping certain parameters constant on the same datasets, it is possible to say that the proposed architecture exhibits a respectable success in the literature and shows a remarkable performance among current architectures.
CRediT authorship contribution statement
Gaffari Celik: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Visualization, Investigation, Validation.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
No data was used for the research described in the article.
The code (and data) in this article has been certified as Reproducible by Code Ocean: (https://codeocean.com/). More information on the Reproducibility Badge Initiative is available at https://www.elsevier.com/physical-sciences-and-engineering/computer-science/journals.
==== Refs
References
1 Wu F. A new coronavirus associated with human respiratory disease in China Nature 579 7798 2020 265 269 10.1038/s41586-020-2008-3 32015508
2 Coronavirus disease (COVID-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus-2019.
3 Subramanian N. Elharrouss O. Al-Maadeed S. Chowdhury M. A review of deep learning-based detection methods for COVID-19 Comput. Biol. Med. 143 2022 105233 10.1016/j.compbiomed.2022.105233
4 Rubin G.D. The role of chest imaging in patient management during the COVID-19 pandemic Chest 158 1 2020 106 116 10.1016/j.chest.2020.04.003 32275978
5 Singh R. Corona virus (COVID-19) symptoms prevention and treatment: A short review J. Drug Deliv. Ther. 11 2-S 2021 118 120 10.22270/jddt.v11i2-S.4644
6 R S. An efficient hardware architecture based on an ensemble of deep learning models for COVID -19 prediction Sustain. Cities Soc. 2022 103713 10.1016/j.scs.2022.103713
7 Heidari A. Jafari Navimipour N. Unal M. Toumaj S. The COVID-19 epidemic analysis and diagnosis using deep learning: A systematic literature review and future directions Comput. Biol. Med. 141 2021 2022 105141 10.1016/j.compbiomed.2021.105141
8 Sharfstein J.M. Becker S.J. Mello M.M. Diagnostic testing for the novel coronavirus JAMA 323 15 2020 1437 10.1001/jama.2020.3864 32150622
9 Stephanie S. Determinants of chest radiography sensitivity for COVID-19: A multi-institutional study in the United States Radiol. Cardiothorac. Imaging 2 5 2020 e200337 10.1148/ryct.2020200337
10 Liu R. Clinica Chimica Acta positive rate of RT-PCR detection of SARS-CoV-2 infection in 4880 cases from one hospital in Wuhan, China, from Jan to 2020 Clin. Chim. Acta 505 March 2020 172 175 10.1016/j.cca.2020.03.009 32156607
11 Dramé M. Should RT-PCR be considered a gold standard in the diagnosis of COVID-19? J. Med. Virol. 92 11 2020 2312 2313 10.1002/jmv.25996 32383182
12 Xie J. Characteristics of patients with coronavirus disease (COVID-19) confirmed using an IgM-IgG antibody test J. Med. Virol. 92 10 2020 2004 2010 10.1002/jmv.25930 32330303
13 Hassan H. Supervised and weakly supervised deep learning models for COVID-19 CT diagnosis: A systematic review Comput. Methods Programs Biomed. 218 2022 106731 10.1016/j.cmpb.2022.106731
14 Gaur P. Malaviya V. Gupta A. Bhatia G. Pachori R.B. Sharma D. COVID-19 disease identification from chest CT images using empirical wavelet transformation and transfer learning Biomed. Signal Process. Control 71 PA 2022 103076 10.1016/j.bspc.2021.103076
15 Ucar F. Korkmaz D. COVIDiagnosis-Net: Deep Bayes-SqueezeNet based diagnosis of the coronavirus disease, 2019 (COVID-19) from X-ray images Med. Hypotheses 140 April 2020 109761 10.1016/j.mehy.2020.109761
16 Başaran E. Classification of white blood cells with SVM by selecting SqueezeNet and LIME properties by mRMR method Signal, Image Video Process 2022 10.1007/s11760-022-02141-2
17 Çelik G. Talu M.F. A new 3D MRI segmentation method based on generative adversarial network and atrous convolution Biomed. Signal Process. Control 71 PA 2022 103155 10.1016/j.bspc.2021.103155
18 Goodfellow I. Generative adversarial networks Commun. ACM 63 11 2020 139 144 10.1145/3422622
19 Çelik G. Talu M.F. Generating the image viewed from EEG signals Pamukkale Univ. J. Eng. Sci. 27 2 2021 129 138 10.5505/pajes.2020.76399
20 Esteva A. Dermatologist-level classification of skin cancer with deep neural networks Nature 542 7639 2017 115 118 10.1038/nature21056 28117445
21 Tan J.H. Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network Inf. Sci. (Ny) 420 2017 66 76 10.1016/j.ins.2017.08.050
22 Başaran E. Cömert Z. Çelik Y. Neighbourhood component analysis and deep feature-based diagnosis model for middle ear otoscope images Neural Comput. Appl. 2022 10.1007/s00521-021-06810-0
23 Celik Y. Talo M. Yildirim O. Karabatak M. Acharya U.R. Automated invasive ductal carcinoma detection based using deep transfer learning with whole-slide images Pattern Recognit. Lett. 133 2020 232 239 10.1016/j.patrec.2020.03.011
24 Bozdag Z. Talu F.M. Pyramidal nonlocal network for histopathological image of breast lymph node segmentation Int. J. Comput. Intell. Syst. 14 1 2021 122 131 10.2991/ijcis.d.201030.001
25 Talo M. Yildirim O. Baloglu U.B. Aydin G. Acharya U.R. Convolutional neural networks for multi-class brain disease detection using MRI images Comput. Med. Imaging Graph. 78 2019 101673 10.1016/j.compmedimag.2019.101673
26 Gaál G. Maga B. Lukács A. Attention U-net based adversarial architectures for chest X-ray lung segmentation CEUR Workshop Proc. 2692 2020 1 7
27 Souza J.C. Bandeira Diniz J.O. Ferreira J.L. França da Silva G.L. Corrêa Silva A. de Paiva A.C. An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks Comput. Methods Programs Biomed. 177 2019 285 296 10.1016/j.cmpb.2019.06.005 31319957
28 Yıldırım Ö. Pławiak P. Tan R.S. Acharya U.R. Arrhythmia detection using deep convolutional neural network with long duration ECG signals Comput. Biol. Med. 102 September 2018 411 420 10.1016/j.compbiomed.2018.09.009 30245122
29 Hannun A.Y. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network Nature Med. 25 1 2019 65 69 10.1038/s41591-018-0268-3 30617320
30 Acharya U.R. A deep convolutional neural network model to classify heartbeats Comput. Biol. Med. 89 August 2017 389 396 10.1016/j.compbiomed.2017.08.022 28869899
31 Rajpurkar P. CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning 2017 3 9 [Online]. Available: http://arxiv.org/abs/1711.05225
32 Ieracitano C. A fuzzy-enhanced deep learning approach for early detection of Covid-19 pneumonia from portable chest X-ray images Neurocomputing 481 2022 202 215 10.1016/j.neucom.2022.01.055 35079203
33 Ahamed K.U. A deep learning approach using effective preprocessing techniques to detect COVID-19 from chest CT-scan and X-ray images Comput. Biol. Med. 139 October 2021 105014 10.1016/j.compbiomed.2021.105014
34 Verma H. Mandal S. Gupta A. Temporal deep learning architecture for prediction of COVID-19 cases in India Expert Syst. Appl. 195 January 2021 116611 10.1016/j.eswa.2022.116611
35 Khan S.H. COVID-19 detection in chest X-ray images using deep boosted hybrid learning Comput. Biol. Med. 137 August 2021 104816 10.1016/j.compbiomed.2021.104816
36 Loey M. El-Sappagh S. Mirjalili S. Bayesian-based optimized deep learning model to detect COVID-19 patients using chest X-ray image data Comput. Biol. Med. 142 January 2022 105213 10.1016/j.compbiomed.2022.105213
37 Lahsaini I. El Habib Daho M. Chikh M.A. Deep transfer learning based classification model for COVID-19 using chest CT-scans Pattern Recognit. Lett. 152 2021 122 128 10.1016/j.patrec.2021.08.035 34566222
38 Toğaçar M. Ergen B. Cömert Z. COVID-19 detection using deep learning models to exploit social mimic optimization and structured chest X-ray images using fuzzy color and stacking approaches Comput. Biol. Med. 121 March 2020 10.1016/j.compbiomed.2020.103805
39 Ozturk T. Talo M. Yildirim E.A. Baloglu U.B. Yildirim O. Rajendra Acharya U. Automated detection of COVID-19 cases using deep neural networks with X-ray images Comput. Biol. Med. 121 April 2020 103792 10.1016/j.compbiomed.2020.103792
40 Khan A.I. Shah J.L. Bhat M.M. CoroNet: A deep neural network for detection and diagnosis of COVID-19 from chest x-ray images Comput. Methods Programs Biomed. 196 2020 105581 10.1016/j.cmpb.2020.105581
41 Monshi M.M.A. Poon J. Chung V. Monshi F.M. CovidXrayNet: Optimizing data augmentation and CNN hyperparameters for improved COVID-19 detection from CXR Comput. Biol. Med. 133 March 2021 104375 10.1016/j.compbiomed.2021.104375
42 Mahmud T. Rahman M.A. Fattah S.A. CovXNet: A multi-dilation convolutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization Comput. Biol. Med. 122 June 2020 103869 10.1016/j.compbiomed.2020.103869
43 Calderon-Ramirez S. Yang S. Elizondo D. Moemeni A. Dealing with distribution mismatch in semi-supervised deep learning for COVID-19 detection using chest X-ray images: A novel approach using feature densities Appl. Soft Comput. 123 2022 108983 10.1016/j.asoc.2022.108983
44 Gupta A. Anjum Gupta S. Katarya R. InstaCovNet-19: A deep learning classification model for the detection of COVID-19 patients using Chest X-ray Appl. Soft Comput. 99 2021 106859 10.1016/j.asoc.2020.106859
45 Feki I. Ammar S. Kessentini Y. Muhammad K. Federated learning for COVID-19 screening from Chest X-ray images Appl. Soft Comput. 106 2021 107330 10.1016/j.asoc.2021.107330
46 de Moura J. Novo J. Ortega M. Fully automatic deep convolutional approaches for the analysis of COVID-19 using chest X-ray images Appl. Soft Comput. 115 2022 108190 10.1016/j.asoc.2021.108190
47 Shankar K. An optimal cascaded recurrent neural network for intelligent COVID-19 detection using Chest X-ray images Appl. Soft Comput. 113 2021 107878 10.1016/j.asoc.2021.107878
48 Albahli S. Ayub N. Shiraz M. Coronavirus disease (COVID-19) detection using X-ray images and enhanced DenseNet Appl. Soft Comput. 110 2021 107645 10.1016/j.asoc.2021.107645
49 Elazab A. Elfattah M.A. Zhang Y. Novel multi-site graph convolutional network with supervision mechanism for COVID-19 diagnosis from X-ray radiographs Appl. Soft Comput. 114 2022 108041 10.1016/j.asoc.2021.108041
50 Ozcan T. A new composite approach for COVID-19 detection in X-ray images using deep features Appl. Soft Comput. 111 2021 107669 10.1016/j.asoc.2021.107669
51 Calderon-Ramirez S. Correcting data imbalance for semi-supervised COVID-19 detection using X-ray chest images Appl. Soft Comput. 111 2021 107692 10.1016/j.asoc.2021.107692
52 Karthik R. Menaka R. Hariharan M. Learning distinctive filters for COVID-19 detection from chest X-ray using shuffled residual CNN Appl. Soft Comput. 99 2021 106744 10.1016/j.asoc.2020.106744
53 Demir F. DeepCoroNet: A deep LSTM approach for automated detection of COVID-19 cases from chest X-ray images Appl. Soft Comput. 103 2021 107160 10.1016/j.asoc.2021.107160
54 Zhou T. Lu H. Yang Z. Qiu S. Huo B. Dong Y. The ensemble deep learning model for novel COVID-19 on CT images Appl. Soft Comput. 98 2021 106885 10.1016/j.asoc.2020.106885
55 Bandyopadhyay R. Basu A. Cuevas E. Sarkar R. Harris Hawks optimisation with simulated annealing as a deep feature selection method for screening of COVID-19 CT-scans Appl. Soft Comput. 111 2021 107698 10.1016/j.asoc.2021.107698
56 Ye Q. Robust weakly supervised learning for COVID-19 recognition using multi-center CT images Appl. Soft Comput. 116 2022 108291 10.1016/j.asoc.2021.108291
57 Song L. A deep fuzzy model for diagnosis of COVID-19 from CT images Appl. Soft Comput. 122 2022 108883 10.1016/j.asoc.2022.108883
58 Liang S. Nie R. Cao J. Wang X. Zhang G. FCF: Feature complement fusion network for detecting COVID-19 through CT scan images Appl. Soft Comput. 125 2022 109111 10.1016/j.asoc.2022.109111
59 Saygılı A. A new approach for computer-aided detection of coronavirus (COVID-19) from CT and X-ray images using machine learning methods Appl. Soft Comput. 105 2021 107323 10.1016/j.asoc.2021.107323
60 Naeem H. Bin-Salem A.A. A CNN-LSTM network with multi-level feature extraction-based approach for automated detection of coronavirus from CT scan and X-ray images Appl. Soft Comput. 113 2021 107918 10.1016/j.asoc.2021.107918
61 Vinod D.N. Jeyavadhanam B.R. Zungeru A.M. Prabaharan S.R.S. Fully automated unified prognosis of Covid-19 chest X-ray/CT scan images using Deep Covix-Net model Comput. Biol. Med. 136 August 2021 104729 10.1016/j.compbiomed.2021.104729
62 Li J. Multi-task contrastive learning for automatic CT and X-ray diagnosis of COVID-19 Pattern Recognit. 114 2021 107848 10.1016/j.patcog.2021.107848
63 Yang X. He X. Zhao J. Zhang Y. Zhang S. Xie P. COVID-CT-dataset: A CT scan dataset about COVID-19 2020 1 14 [Online]. Available: http://arxiv.org/abs/2003.13865
64 Soares E. Angelov P. Biaso S. Froes M.H. Abe D.K. SARS-CoV-2 CT-scan dataset: A large dataset of real patients CT scans for SARS-CoV-2 identification 2020 medRxiv, p. 2020.04.24.20078584, [Online]. Available: https://www.medrxiv.org/content/10.1101/2020.04.24.20078584v3%0Ahttps://www.medrxiv.org/content/10.1101/2020.04.24.20078584v3.abstract
65 Chowdhury M.E.H. Can AI help in screening viral and COVID-19 pneumonia? IEEE Access 8 July 2020 132665 132676 10.1109/ACCESS.2020.3010287
66 Gu J. Recent advances in convolutional neural networks Pattern Recognit. 77 2018 354 377 10.1016/j.patcog.2017.10.013
67 Budak Ü. Cömert Z. Çıbuk M. Şengür A. DCCMED-Net: Densely connected and concatenated multi Encoder-Decoder CNNs for retinal vessel extraction from fundus images Med. Hypotheses 134 2019 2020 10.1016/j.mehy.2019.109426
68 Ren F. Liu W. Wu G. Feature reuse residual networks for insect pest recognition IEEE Access 7 2019 122758 122768 10.1109/ACCESS.2019.2938194
69 He K. Zhang X. Ren S. Sun J. Deep residual learning for image recognition 2015 [Online]. Available: http://arxiv.org/abs/1512.03385
70 Kim S. Park I. Kwon S. Han J. Motion retargetting based on dilated convolutions and skeleton-specific loss functions Comput. Graph. Forum 39 2 2020 497 507 10.1111/cgf.13947
71 Sooksatra S. Kondo T. Bunnun P. Yoshitaka A. Redesigned skip-network for crowd counting with dilated convolution and backward connection J. Imaging 6 5 2020 10.3390/JIMAGING6050028
72 Li X. Zhai M. Sun J. DDCNNC: Dilated and depthwise separable convolutional neural network for diagnosis COVID-19 via chest X-ray images Int. J. Cogn. Comput. Eng. 2 April 2021 71 82 10.1016/j.ijcce.2021.04.001
73 Chollet F. Xception: Deep learning with depthwise separable convolutions Proc. - 30th IEEE Conf. Comput. Vis. Pattern Recognition, CVPR 2017, 2017-Janua 2017 1800 1807 10.1109/CVPR.2017.195
74 Ma Y. Wang C. SdcNet for object recognition Comput. Vis. Image Underst. 215 2020 103332 10.1016/j.cviu.2021.103332 2022
75 Wang C.-F. A basic introduction to separable convolutions 2018 https://l24.im/hrH8qwp. (Accessed 22 Nov. 2021)
76 Friedman J.H. Greedy function approximation: A gradient boosting machine Ann. Statist. 29 5 2001 1189 1232 10.1214/aos/1013203451
77 Chen H. Shen Z. Wang L. Qi C. Tian Y. Prediction of undrained failure envelopes of skirted circular foundations using gradient boosting machine algorithm Ocean Eng. 258 May 2022 111767 10.1016/j.oceaneng.2022.111767
78 Touzani S. Granderson J. Fernandes S. Gradient boosting machine for modeling the energy consumption of commercial buildings Energy Build. 158 2018 1533 1543 10.1016/j.enbuild.2017.11.039
79 Gao B. Pavel L. On the properties of the softmax function with application in game theory and reinforcement learning 2017 1 10 [Online]. Available: http://arxiv.org/abs/1704.00805
80 Marques G. Agarwal D. de la Torre Díez I. Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network Appl. Soft Comput. 96 2020 106691 10.1016/j.asoc.2020.106691
81 Umer M. Ashraf I. Ullah S. Mehmood A. Choi G.S. COVINet: a convolutional neural network approach for predicting COVID-19 from chest X-ray images J. Ambient Intell. Humaniz. Comput. 13 1 2022 535 547 10.1007/s12652-021-02917-3 33527000
82 gifani P. Shalbaf A. Vafaeezadeh M. Automated detection of COVID-19 using ensemble of transfer learning with deep convolutional neural network based on CT scans Int. J. Comput. Assist. Radiol. Surg. 16 1 2021 115 123 10.1007/s11548-020-02286-w 33191476
83 Sethy P.K. Behera S.K. Ratha P.K. Biswas P. Detection of coronavirus disease (COVID-19) based on deep features Int. J. Math. Eng. Manag. Sci. 5 4 2020 643 651 10.20944/preprints202003.0300.v1
84 Li D. Fu Z. Xu J. Stacked-autoencoder-based model for COVID-19 diagnosis on CT images Appl. Intell. 51 5 2021 2805 2817 10.1007/s10489-020-02002-w
85 Xu X. A deep learning system to screen novel coronavirus disease 2019 pneumonia Engineering 6 10 2020 1122 1129 10.1016/j.eng.2020.04.010 32837749
86 Heidarian S. COVID-FACT: A fully-automated capsule network-based framework for identification of COVID-19 cases from chest CT scans Front. Artif. Intell. 4 May 2021 1 13 10.3389/frai.2021.598932
87 Mukherjee H. Ghosh S. Dhar A. Obaidullah S.M. Santosh K.C. Roy K. Deep neural network to detect COVID-19: one architecture for both CT scans and chest X-rays Appl. Intell. 51 5 2021 2777 2789 10.1007/s10489-020-01943-6
88 Wang L. Lin Z.Q. Wong A. COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images Sci. Rep. 10 1 2020 19549 10.1038/s41598-020-76550-z 33177550
89 Heidari M. Mirniaharikandehei S. Khuzani A.Z. Danala G. Qiu Y. Zheng B. Improving the performance of CNN to predict the likelihood of COVID-19 using chest X-ray images with preprocessing algorithms Int. J. Med. Inform. 144 June 2020 104284 10.1016/j.ijmedinf.2020.104284
90 Chakraborty M. Dhavale S.V. Ingole J. Corona-Nidaan: lightweight deep convolutional neural network for chest X-ray based COVID-19 infection detection Appl. Intell. 51 5 2021 3026 3043 10.1007/s10489-020-01978-9
91 Babukarthik R.G. Ananth Krishna Adiga V. Sambasivam G. Chandramohan D. Amudhavel A.J. Prediction of COVID-19 using genetic deep learning convolutional neural network (GDCNN) IEEE Access 8 2020 177647 177666 10.1109/ACCESS.2020.3025164 34786292
92 Apostolopoulos I.D. Mpesiana T.A. COVID-19: automatic detection from X-ray images utilizing transfer learning with convolutional neural networks Phys. Eng. Sci. Med. 43 2 2020 635 640 10.1007/s13246-020-00865-4 32524445
93 Ismael A.M. Şengür A. Deep learning approaches for COVID-19 detection based on chest X-ray images Expert Syst. Appl. 164 2020 2021 10.1016/j.eswa.2020.114054
94 Oh Y. Park S. Ye J.C. Deep learning COVID-19 features on CXR using limited training data sets IEEE Trans. Med. Imaging 39 8 2020 2688 2700 10.1109/TMI.2020.2993291 32396075
95 Ezzat D. Hassanien A.E. Ella H.A. An optimized deep learning architecture for the diagnosis of COVID-19 disease based on gravitational search optimization Appl. Soft Comput. 98 2021 106742 10.1016/j.asoc.2020.106742
96 Hussain E. Hasan M. Rahman M.A. Lee I. Tamanna T. Parvez M.Z. CoroDet: A deep learning based classification for COVID-19 detection using chest X-ray images Chaos Solitons Fractals 142 2021 110495 10.1016/j.chaos.2020.110495
97 Başaran E. Cömert Z. Çelik Y. Convolutional neural network approach for automatic tympanic membrane detection and classification Biomed. Signal Process. Control 56 2020 10.1016/j.bspc.2019.101734
98 Wang C.Y. Mark Liao H.Y. Wu Y.H. Chen P.Y. Hsieh J.W. Yeh I.H. CSPNet: A new backbone that can enhance learning capability of CNN IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. Work., 2020-June 2020 1571 1580 10.1109/CVPRW50498.2020.00203
99 Selvaraju R.R. Cogswell M. Das A. Vedantam R. Parikh D. Batra D. Grad-CAM: Visual explanations from deep networks via gradient-based localization Proc. IEEE Int. Conf. Comput. Vis., 2017-Octob 2017 618 626 10.1109/ICCV.2017.74
100 Abbasniya M.R. Sheikholeslamzadeh S.A. Nasiri H. Emami S. Classification of Breast Tumours Based on Histopathology Images using Deep Features and Ensemble of Gradient Boosting Methods, Vol. 103 2022 10.1016/j.compeleceng.2022.108382 arXiv Prepr. arXiv2209.01380. June. 108382
101 Kim H. Jung W.K. Park Y.C. Lee J.W. Ahn S.H. Broken stitch detection method for sewing operation using CNN feature map and image-processing techniques Expert Syst. Appl. 188 2022 116014 10.1016/j.eswa.2021.116014
102 Brownlee J. How to visualize filters and feature maps in convolutional neural networks 2019 https://3c5.com/w7im4. (Accessed 25 Sep. 2022)
| 0 | PMC9726416 | NO-CC CODE | 2022-12-08 23:18:16 | no | Air Med J. 2022 Dec 7 November-December; 41(6):570 | latin-1 | Air Med J | 2,022 | 10.1016/j.amj.2022.10.004 | oa_other |
==== Front
Environ Health Perspect
Environ Health Perspect
EHP
Environmental Health Perspectives
0091-6765
1552-9924
Environmental Health Perspectives
EHP11068
10.1289/EHP11068
Research
Predicting the Effects of Climate Change on Dengue Vector Densities in Southeast Asia through Process-Based Modeling
https://orcid.org/0000-0002-5674-1911
Bonnin Lucas 1
Tran Annelise 2 3 4 5
Herbreteau Vincent 6 7
Marcombe Sébastien 8
Boyer Sébastien 9
Mangeas Morgan 1
Menkes Christophe 1
1 ENTROPIE (UMR 9220), IRD, Université de la Réunion, CNRS, Ifremer, Université de Nouvelle Calédonie, Nouméa, Nouvelle-Calédonie
2 CIRAD, UMR TETIS, Sainte-Clotilde, Reunion Island, France
3 TETIS, Université Montpellier, AgroParisTech, CIRAD, CNRS, INRAE, Montpellier, France
4 CIRAD, UMR ASTRE, Sainte-Clotilde, Reunion Island, France
5 ASTRE, Université Montpellier, CIRAD, INRAE, Montpellier, France
6 ESPACE-DEV, IRD, Université Antilles, Université Guyane, Université Montpellier, Université de la Réunion, Montpellier, France
7 ESPACE-DEV, IRD, Université Antilles, Université Guyane, Université Montpellier, Université de la Réunion, Phnom Penh, Cambodia
8 Medical Entomology and Vector-Borne Disease Laboratory, Institut Pasteur du Laos, Vientiane, Lao PDR
9 Medical and Veterinary Entomology Unit, Institut Pasteur du Cambodge, Phnom Penh, Cambodia
Address correspondence to Lucas Bonnin, ENTROPIE (UMR 9220), IRD, Université de la Réunion, CNRS, Ifremer, Université de Nouvelle Calédonie, Nouméa, Nouvelle-Calédonie. Email: [email protected]
6 12 2022
12 2022
130 12 12700208 2 2022
19 9 2022
21 10 2022
https://ehp.niehs.nih.gov/about-ehp/license EHP is an open-access journal published with support from the National Institute of Environmental Health Sciences, National Institutes of Health. All content is public domain unless otherwise noted.
Background:
Aedes aegypti and Ae. albopictus mosquitoes are major vectors for several human diseases of global importance, such as dengue and yellow fever. Their life cycles and hosted arboviruses are climate sensitive and thus expected to be impacted by climate change. Most studies investigating climate change impacts on Aedes at global or continental scales focused on their future global distribution changes, whereas a single study focused on its effects on Ae. aegypti densities regionally.
Objectives:
A process-based approach was used to model densities of Ae. aegypti and Ae. albopictus and their potential evolution with climate change using a panel of nine CMIP6 climate models and climate scenarios ranging from strong to low mitigation measures at the Southeast Asian scale and for the next 80 y.
Methods:
The process-based model described, through a system of ordinary differential equations, the variations of mosquito densities in 10 compartments, corresponding to 10 different stages of mosquito life cycle, in response to temperature and precipitation variations. Local field data were used to validate model outputs.
Results:
We show that both species densities will globally increase due to future temperature increases. In Southeast Asia by the end of the century, Ae. aegypti densities are expected to increase from 25% with climate mitigation measures to 46% without; Ae. albopictus densities are expected to increase from 13%–21%, respectively. However, we find spatially contrasted responses at the seasonal scales with a significant decrease in Ae. albopictus densities in lowlands during summer in the future.
Discussion:
These results contrast with previous results, which brings new insight on the future impacts of climate change on Aedes densities. Major sources of uncertainties, such as mosquito model parametrization and climate model uncertainties, were addressed to explore the limits of such modeling. https://doi.org/10.1289/EHP11068
Supplemental Material is available online (https://doi.org/10.1289/EHP11068).
The authors declare they have nothing to disclose.
Note to readers with disabilities: EHP strives to ensure that all journal content is accessible to all readers. However, some figures and Supplemental Material published in EHP articles may not conform to 508 standards due to the complexity of the information being presented. If you need assistance accessing journal content, please contact [email protected]. Our staff will work with you to assess and meet your accessibility needs within 3 working days.
==== Body
pmcIntroduction
Mosquitoes from the genus Aedes are major vectors for several human arboviruses of global importance, such as dengue, Zika, and Chikungunya viruses.1–3 These viruses can cause severe febrile diseases with long-term physical and cognitive consequences. Due to their widespread distribution and the large populations at risk, they represent a major burden on global health and are estimated to be responsible annually for millions of disability-adjusted life years (DALY4) Because no specific treatment exists for such viral diseases, with the exception of dengue, for which the World Health Organization recommends the use of vaccine in very specific cases only,5 vector surveillance and control represent the only available and effective strategy to mitigate the outbreaks of these diseases.6–9 Indeed, vector density is a major component of vectorial capacity10 and potentially plays a key role in the transmission of arboviruses, notably for dengue fever.11–16 Mosquito life cycles are highly dependent on weather conditions because temperature drives reproduction, maturation and mortality,17 whereas rainfall drives the availability of breeding sites18 and juvenile mortality (washing off of aquatic stages). As a result, mosquito densities are expected to be greatly impacted by global climate change, but it remains unclear how and where vector risk will change in the future.
Predicting the future evolution of vector risk relies on large spatial scale modeling of Aedes population dynamics in response to climate spatial and temporal variations. Species distribution models have been developed for the two main dengue vector species, Ae. aegypti and Ae. albopictus, to predict the future response of these species’ distribution to climate change at global or continental scales.19–22 Such statistical models bring valuable knowledge about the potential ecological niche of the associated vectorborne diseases, yet they cannot bring insight about the level of vector risk in regions where these species are already present all year long. Statistical modeling of mosquito population densities, however, poses a greater challenge due to the difficulty to gather large data sets of spatially consistent density observations to effectively build the models.
Although data are key to constructing statistical models, mechanistic approaches on the other hand are less impacted by data availability because data are usually only used for validation. Such mechanistic models can simulate mosquitoes’ life cycles based on quantified knowledge of the response of their different life stages to climate variations. Process-based models have successfully been employed to describe Aedes mosquitoes’ density variations at local and global scales.23–25 However, they were seldom applied to climate projections to predict the future evolution of vector risk. To date, a single study used mechanistic models to predict the future evolution of densities of one Aedes species, Ae. aegypti, at a global scale,25 but with no validation of temporal dynamics. Because such models rely on strong hypotheses about mechanistic processes, confidence in future predictions can only arise from the comparison and confrontation of results from independently built models or independent data.
In this study, a compartment process-based model was used to describe seasonal variations of densities of two Aedes mosquito species, Ae. aegypti and Ae. albopictus, in response to contemporary and future climates. Aedes densities were modeled on a 10 by 10 arc-minutes grid at a regional scale in Southeast Asia (SEA), where both species are ubiquitous and where Aedes-borne diseases (especially dengue fever) display frequent outbreaks and pose crucial public health concerns.26–30 To validate the model seasonal dynamics, we used original density data from temporal entomological surveys from two Southeast Asian cities: Phnom Penh, Cambodia, and Vientiane, Lao People’s Democratic Republic (here after Lao PDR). Based on climate projections from a range of climate models and scenarios of greenhouse gas emissions (Worldclim database31; https://www.worldclim.org/data/index.html), predictions about the evolution of densities of both species were mapped over selected future periods of the 21st century.
Our results were used to both discuss the potential vector evolutions with the diversity of future potential climates but also the advantages and flaws of mechanistic approaches in comparison with other statistical modeling. Overall, the framework allows us to discuss the magnitudes of predictions uncertainties, which are key to identify future improvements.
Methods
Study Area
The study area includes the Southeast Asian countries of Myanmar, Lao PDR, Thailand, Cambodia, and Vietnam, whose latitudes varies between 5.6° and 28.5° North. This region is characterized by a contrasting topography, with lowlands under 300m in altitude (in central Myanmar, Thailand, Cambodia, and southern and northern Vietnam) and highlands in the north part of the region (in eastern Myanmar, northern Thailand, Lao PDR, and northern Vietnam) (Figure S1). The coldest months in the region (December, January, February) are characterized by average temperatures between 16°C and 24°C, whereas during the warmest months (April, May, June), they vary between 24°C and 28°C. Rains show a strong seasonality in the region, with the rainy season roughly occurring between May and October, with precipitations averaging 8mm/d and reaching 40mm/d in some parts of the region (e.g., western Myanmar).
Process-Based Climate-Driven Model
Mosquito population dynamics were modeled through a process-based approach, using the general mechanistic framework proposed by Cailly et al.32 This framework describes, through a system of ordinary differential equations (ODE), variations of mosquito abundance in 10 compartments, corresponding to different stages of mosquito life cycle. Aquatic stages are divided into eggs (E), larvae (L), and pupae (P); adult stages are divided into emerging (Aem), nulliparous (A1), and parous females (A2). Nulliparous and parous females are then subdivided according to their behavior: host seeking (Ah), transition from engorged to gravid (Ag), and oviposition site seeking (Ao). This framework was later used to develop models of Ae. albopictus population dynamics in temperate33 and tropical climates.34 In this study, the ODE system from Tran et al.34 was used to model population dynamics of both species separately. {E˙ = γAo(β1A1o+β2A2o)−(mE+fE)EL˙=fEE−(mL(1+L/kL)+fL)LP˙=fLL−(mP+fP)PA˙em=fPPσe−μem(1+P/kP)−(mA+γAem)AemA˙1h=γAemAem−(mA+μr+γAh)A1hA˙1g=γAhA1h−(mA+fAg)A1gA˙1o=fAgA1g−(mA+μr+fAo)A1oA˙2h=fAo(A1o+A2o)−(mA+μr+γAh)A2hA˙2g=γAhA2h−(mA+fAg)A2gA˙2o=fAgA2g−(mA+μr+γAo)A2o
The diagram of the model is available in Figure S2. All parameters and functions are derived from results of experimental studies.35,36 In this study, we used the Ae. albopictus parameters and functions described in Tran et al.34 and Ae. aegypti parameters and functions gathered from literature.23,37
Parameter values are described in Table 1, adapted from Tran et al.34: for each stage X, γX is the transition rate to the next compartment, μX is the mortality rate, βX is the egg laying rate, and σ is the sex ratio at the emergence; μr is an additional adult mortality rate related to seeking behavior.
Table 1 Process-based model parameters.
Notation Definition Value
Ae. albopictus Ae. aegypti
β1 Number of eggs laid/ovipositing nulliparous female 60.0
β2 Number of eggs laid/ovipositing parous female 80.0
σ Sex ratio at emergence 0.5
γAem Development rate of emerging adults (per day) 0.4
γAh Transition rate from host-seeking to engorged adults (per day) 0.2
γAo Minimum transition rate from ovipositing to host-seeking adults (per day) 0.2
μE Minimum egg mortality rate (per day) 0.05 0.01
μem Mortality rate during emergence (per day) 0.1
μr Mortality rate related to seeking behavior (per day) 0.08
TE Minimal temperature needed for egg development 10.0
TDDE Total number of degree days necessary for egg development (°C.day) 110
TAg Minimal temperature needed for egg maturation in females (°C) 10.0
TDDAg Total number of degree days necessary for egg maturation in females (°C.day) 77.0
Note: Adapted from Table 2 from Tran et al.34
Functions are used to describe the temperature- and precipitation-dependent transition rates (fX), mortality rates (mX), and carrying capacities (kX), which vary over time (Table 2). Carrying capacity is driven by precipitation and is used to regulate the density-dependent mortality of aquatic stages (larvae and pupae): As rainfall fills breeding sites, higher precipitation enables higher larval and pupal densities before high mortality.
Table 2 Process-based model functions.
Notation Definition Expression
Ae. Albopictus Ae. Aegypti
fE Transition function from egg to larva {T−TETDDE if T>TE 0otherwise
24ρTK298eΔHAR(1298−1TK)1+eΔHHR(1T1/2H−1TK)
With ρ=0.01066; ΔHA=10,798.18;
ΔHH=100,000; T1/2H=14,184.5
fL Transition function from larva to pupa q1T2+q2T+q3
with q1=−0.0007; q2=0.0392; q3=−0.3911 24ρTK298eΔHAR(1298−1TK)1+eΔHHR(1T1/2H−1TK)
With ρ=0.00873; ΔHA=26,018.51;
ΔHH=55,990; T1/2H=304.58
fP Transition function from pupa to emerging adult q1T2+q2T+q3
with q1=−0.0008; q2=0.0051; q3=−0.319 24ρTK298eΔHAR(1298−1TK)1+eΔHHR(1T1/2H−1TK)
With ρ=0.0161; ΔHA=14,931.94;
ΔHH=−472,379; T1/2H=148.45
fAg Transition function from engorged adult to ovipositing site-seeking adult T−TAgTDDAg
24ρTK298eΔHAR(1298−1TK)1+eΔHHR(1T1/2H−1TK)
With ρ=0.00898; ΔHA=15,725.23;
ΔHH=1,756,481.07; T1/2H=447.17
fAo Transition function from ovipositing to host-seeking adult γAo×(1+ Pnorm)
mE Egg mortality μE+{0.1 if P>800 otherwise
mL Larva mortality 0.02+0.0007e0.1838(T−10)+{0.5 if P>800 otherwise
mP Pupa mortality 0.02+0.0003e0.2228(T−10)+{0.5 if P>800 otherwise
mA Adult mortality 0.025+0.0003e0.1745(T−10)
kL Carrying capacity for larvae kL=Pnorm
kP Carrying capacity for pupae kP=Pnorm
Note: Adapted from Table 3 in Tran et al.34 Transition and mortality rates and carrying capacity as a function of daily temperature (T) and precipitation (P). Pnorm is the precipitation amount summed over a 1-wk period and normalized to vary between 0 and 1.
The population dynamics of both species were modeled on a 10 arc-minutes grid (about 16.3 km in the north of Myanmar and 18.4 km in the south of Thailand), based on the spatial resolution of climate data. Each grid cell was considered as independent from the others.
Climate Data
Population dynamics were modeled using seasonal variations of climate variables (surface air temperatures and precipitation) from contemporary data and from climate models using future projections. These data were obtained from the WorldClim initiative31 (https://www.worldclim.org/) at a monthly scale. As the model uses daily inputs, climate data were interpolated linearly to obtain daily values.
The contemporary seasonal climate corresponded to the 1970–2000 period. Future projections were based on nine climate models (BCC-CSM2-MR,38 CNRM-CM6-1,39 CNRM-ESM2-1,40 CanESM5,41 GFDL-ESM4,42 IPSL-CM6A-LR,43 MIROC-ES2L,44 MIROC6,45 and MRI-ESM2-046) for four scenarios of greenhouse gas emissions (Shared Socioeconomic Pathways: SSP1 2.6, SSP2 4.5, SSP3 7.0, and SSP5 8.5, based on CMIP6) and for four periods (2021–2040, 2041–2060, 2061–2080, 2081–2100) at a 10 arc-minutes spatial resolution. Special focus was given on climate projections following SSP1 2.6 and SSP3 7.0 because they correspond respectively to the objectives set by the Paris Agreements and to a most realistic scenario of greenhouse gas emissions in the absence of strong mitigation, as currently experienced. To be more specific, SSP1 2.6 corresponds to a zero-emission target reached after 2050, whereas SSP3 7.0 corresponds to a steady increase in greenhouse gas emissions throughout the century, with CO2 emissions in the year 2100 twice as high as contemporary levels. In Southeast Asia, at the end of the century, SSP1 2.6 and SSP3 7.0 are predicted to result in an average increase of temperature of 1.5°C and 3.5°C, respectively, and an average increase of precipitation of 0.2 and 0.1mm/d, respectively, although with high seasonal and spatial heterogeneity (Figure S7). Note that results for other scenarios are also given but discussed more briefly.
All simulations were run during 2 successive climatological years, but only the last year was analyzed to discard the model spin-up occurring during the first year. Monthly mean outputs were considered for the rest of the analysis.
Analysis and Mapping
To assess the relative increase or decrease of adult female densities associated with each climate projection, a percentage change was computed as the ratio of future density over contemporary density.
To assess the concordance between projections associated with the nine climate models, one Student’s t-test was performed in each grid cell over the nine percentage change values to test for a significant difference of their mean from zero (H0: percentage change is equal to zero, H1: percentage change is different from zero). Statistical significance (p) was set at the 5% level.
Although the compartment model allows studying the temporal dynamics of the Aedes species, it requires an a priori knowledge of the potential mosquito presence, such as the dynamics given by species distribution models. Hence, to produce realistic contemporary mosquito density maps, we first built spatial species distribution for the two species in the region. Such distributions consisted of spatial predictions of probability of occurrence of each species, computed from species distribution models (SDMs) based on occurrence data from Kraemer et al.20 and topography and land cover data (see Figure S3A for such species distributions). Details about the SDM are provided in the “Species Distribution Modeling” section. These maps were then used to weight the contemporary simulations of species spatial densities given by the mosquito dynamical model. These weighted densities maps were then linearly scaled to vary between 0 and 1 to produce a simple and interpretable index of density.
It has to be noted that future percentage changes relied only on climate variations and on the process-based model. Indeed, the SDM weighting was only based on topography and current contemporary land cover data (no future projections of land cover were used); its effect thus did not appear in the final ratio describing density percentage change.
Species Distribution Modeling
Geographical distribution of both species was modeled in the study area based on occurrence data and land-cover and topography maps. Occurrence data for both species were obtained from Kraemer et al.20 and subsampled over the two countries of the region with the largest available data, Thailand and Vietnam, resulting in a data set of 102 occurrences for Ae. albopictus and 725 occurrences for Ae. aegypti (Figure S3A). For both species, the same number of pseudo-absences was randomly sampled in both countries. Land cover data from the FROM-GLC initiative (Finer Resolution Observation and Monitoring – Global Land Cover)47 and topography data from the Shuttle Radar Topography Mission (SRTM)48 were used as independent variables. Land cover data consisted in a classification of land surface in eight categories (cropland, forest, grassland, shrubland, wetland, water, impervious surface, and bare land) at a spatial resolution of 10 m. Because occurrence data was available on a 0.01-degree grid, land cover data were converted to this scale by computing proportions of each land cover category in each grid cell. Topography data was available at a 90 m spatial resolution but converted to 0.01° by computing elevation and slope averages and variability of slope in each grid cell.
An ensemble modeling approach was used to model presence/absence as dependent on land cover and topography. Ensemble modeling was performed using the biomod2 R package (version 4.0.3; R Development Core Team), considering a combination of generalized linear modeling (GLM), gradient boosting modeling (GBM), classification tree analysis (CTA), artificial neural network (ANN), flexible discriminant analysis (FDA) and random forest (RF) to compute spatial predictions of probability of observation. Individual model quality was assessed through a 5-fold cross-validation process and area under the receiver operating characteristic curve (ROC AUC) quality metrics. All models with AUC values <0.85 were discarded (Table S1); next, the predictions of the remaining models were averaged with weights proportional to their AUC to compute a single prediction at a 0.01° resolution over the entire study area (Figure S3A). The resulting distribution maps were then converted to match the 10-minutes-of-arc grid used to compute simulations from the process-based climate-driven model. Distribution maps were thus used to weight the contemporary simulations of species spatial densities given by this process-based model, as previously described.
Process-Based Model Sensitivity to Climate Variables
A posteriori sensitivity of modeled species to temperature and precipitation was estimated as follows: 10,000 combinations of 100 constant values of temperature (between 10° and 40°C), and 100 constant values of precipitation (between 0 and 20mm/d) were used to simulate the evolution of mosquito density over 30 d. Mosquito density values at the end of these 30 d were then mapped as a function of temperature and precipitation, allowing a mean view of the modeled vector densities sensitivity to both temperature and precipitation.
Validation of Contemporary Temporal Dynamics
Entomological survey data from Phnom Penh and Vientiane were used to estimate the seasonal dynamics of Ae. albopictus and Ae. aegypti densities. In Phnom Penh, surveys consisted in the sampling of 40 pagodas in the city (Figure S4), using 5 ovitraps per pagoda picked up every 8 wk between March 2019 and February 2020, resulting in 1,033 sampling points. Ovitraps without filter paper (black buckets)49 were used to collect immature stages (larvae), which were then emerged in laboratory for species identification using morphological characters.50 For each pagoda and each month, the proportion of emerged adults among collected larvae was used to extrapolate a number of captured larvae of each species. Extrapolated numbers of captured larvae were then standardized by pagoda (divided by the pagoda maximum) to account for differences in species abundance level between pagodas. In Lao PDR, surveys consisted in the sampling of five villages from the Vientiane province, using two to four BG sentinel traps per village picked up every 2 wk between May 2016 and March 2020 (Figure S4), resulting in 440 sampling points. Collected mosquitoes were identified using morphological characters.50
Total numbers of captures of Ae. albopictus and Ae. aegypti, standardized by the corresponding number of traps, were computed in both cities and compared to standardized seasonal dynamics of densities simulated in these cities based on the local contemporary climate. Simulated densities were standardized using a multiplicative coefficient so that simulation and recorded curves had the same AUC.
Results
Model outputs based on contemporary seasonal variations of temperature and precipitations were compared to seasonal variations of density recorded from field surveys in Phnom Penh and Vientiane (Figure 1). In Phnom Penh, entomological field surveys showed seasonal dynamics of Aedes densities consisting in two peaks of density in June and October for Ae. aegypti and a consistent rising trend from March to October, followed by a rapid decrease for Ae. albopictus. In Vientiane, entomological surveys showed a less complex, one-peak symmetric seasonal dynamics of densities of both species. The model seasonal predictions captured the dynamics observed in both cities and for both species: the single-peak seasonal dynamics of both species in Vientiane and the more complex two-peak seasonal dynamics of Ae. aegypti and symmetric single-peak seasonal dynamics of Ae. albopictus in Phnom Penh. Precipitation was clearly the main driver of these dynamics, because it presented the same two-peak asymmetrical seasonality in Phnom Penh and one-peak symmetrical seasonality in Vientiane (Figure S5).
Figure 1. Comparison of temporal dynamics of simulated and recorded Ae. albopictus and Ae. aegypti adult female densities in Phnom Penh, Cambodia, and Vientiane, Lao PDR. Blue dotted lines correspond to monthly standardized captures of Ae. aegypti (top panels) and Ae. albopictus (bottom panels) from entomological surveys in Phnom Penh (left panels) and Vientiane (right panels). Red lines without dots represent monthly standardized simulated densities of both species, extracted from grid cells corresponding to Phnom Penh and Vientiane locations.
Figure 1 is a set of four line graphs. On the left, two graphs titled Phnom Penh (Cambodia), plotting extrapolated standardized number of larvae captured, ranging from 0.2 to 0.8 in increments of 0.2 and 0.0 to 0.6 in increments of 0.2 (left y-axis) and standardized simulated larvae density, ranging from 0 to 0.8, in increments of 0.2 and 0.0 to 0.6 in increments of 0.2 (right y-axis) across months from January to December (x-axis) for Aedes aegypti and Aedes albopictus. On the right, two graphs titled Vientiane (Lao People’s Democratic Republic), plotting standardized number of adults captured, ranging from 0.0 to 0.6 in increments of 0.2 and 0.0 to 0.8 in increments of 0.2 (left y-axis) and standardized simulated adult density, ranging from 0 to 0.6, in increments of 0.2 and 0.0 to 0.8 in increments of 0.2 (right y-axis) across months from January to December (x-axis) for Aedes aegypti and Aedes albopictus.
The model was then used to compute seasonal variations of Ae. albopictus and Ae. aegypti densities at continental scale in Southeast Asia, using the gridded contemporary observations and climate model future projections of temperature and precipitation (see Figures S6−S9). In the following, summer and winter refer to warmest and coldest trimesters in the region, corresponding respectively to April-May-June and December-January-February.
During summer months, Ae. aegypti showed elevated densities in lowlands and coastal areas (Figure 2), a spatial structure driven by its association to urban habitat (Figures S3B−C). This pattern contrasted with wintertime when Ae. aegypti densities were much lower over the domain due to lower temperature and precipitation (Figure S6).
Figure 2. Modeled Ae. albopictus and Ae. aegypti adult female densities in Southeast Asia, based on contemporary climate and future projections. Panels A describe the spatial variations of modeled Aedes adult female density (top: Ae. albopictus; bottom: Ae. aegypti) for summer (April, May, June) and winter months (December, January, February), based on topography and land cover and on contemporary climate data input. Panels B describe projections of modeled Aedes adult female density in summer and winter months at the middle (2041–2060) and end of the century (2081–2100), based on future climate projections associated with the SSP1 2.6 and SSP3 7.0 emission scenarios. Percentage changes correspond to ratios of projected modeled density over contemporary modeled densities, averaged over the nine simulations associated with the nine climate models. Positive (red) and negative (blue) percentage changes correspond respectively to higher and lower predicted mosquito density in comparison with the contemporary situation. Purple grid cells correspond to percentage changes greater than a threshold of 130%, mostly corresponding to areas of null to low contemporary adult female density. Diagonal hashing represent grid cells where percentage changes associated with the nine climate models are significantly different from zero (Student’s t-test, p <0.05).
Figure 2A is a set of four Southeast maps that depicts Aedes densities under contemporary climate. The first two maps depict Aedes aegypti in the months of April, May, June, December, January, and February. A scale depicting adult density index ranges from 0.0 to 1.0 in increments of 0.2. The last two maps depict Aedes albopictus in the months of April, May, June, December, January, and February. A scale depicting adult density index ranges from 0.0 to 1.0 in increments of 0.2. Figure 2B is a set of sixteen Southeast maps, depicting Aedes density changes under climate projections from the years 2041 to 2060 and 2081 to 2100. On the top, there are eight maps, of which four maps depict shared socio-economic pathway 1 (2.6) and four maps depict shared socio-economic pathway 3 (7.0). A scale depicting density percentage changes ranges from negative 100 to negative 50 percent in increments of 50, negative 50 to 20 percent in increments of 30, negative 20 to 10 percent in increments of 10, 10 to 30 percent in increments of 20, 30 to 60 percent in increments of 30, and 60 to 130 percent in increments of 70. At the bottom, there are eight maps, of which four maps depict shared socio-economic pathway 1 (2.6) and four maps depict shared socio-economic pathway 3 (7.0). A scale depicting density percentage changes ranges from negative 100 to negative 50 percent in increments of 50, negative 50 to 20 percent in increments of 30, negative 20 to 10 percent in increments of 10, 10 to 30 percent in increments of 20, 30 to 60 percent in increments of 30, and 60 to 130 percent in increments of 70.
For both SSP1 2.6 and SSP3 7.0 future climates, our model predicted that the Ae. aegypti densities would mostly increase in comparison with contemporary levels, except for a slight decrease in central Myanmar, northern Thailand, and Cambodia between April and May following SSP3 7.0 at the end of the century (Figure 2; Figure S9). Summer density evolution showed higher increases in the northern part of the domain, whereas winter densities showed stronger increases in the southern part of the domain, with increases reaching up to 100% for SSP3 7.0 and still 60% for SSP1 2.6 at the end of the century (Figure 2). These seasonal density increases were mostly driven by future temperature increases (Figure S10). Relative increases in densities in both winter and summer accentuated later in the century, most notably in the business-as-usual scenario SSP3 7.0, except in northern Myanmar during the winter months, where SSP1 2.6 resulted in higher densities than SSP3 7.0 (Figure 2).
Contemporary densities of Ae. albopictus were globally similar to Ae. Aegypti, with maxima densities occurring during summer−autumn regionally (Figure 2; Figure S8) due to higher temperature and precipitation during these months. As for future Ae. aegypti evolutions, Ae. albopictus showed an increase in densities during the winter months in all climate scenarios (Figure 2; Figure S9), reaching +50% in 2041−2060 and up to +100% in 2081−2100 for the business-as-usual scenario (SSP3 7.0). In contrast to the trends of Ae. aegypti during summer months, Ae. albopictus showed marked dipoles of increasing and decreasing densities in the northern and southern parts of the continent, respectively (Figure 2; Figure S9), for all periods and climate scenarios. This contrasted pattern accentuated along the century (Figure S11) and with increased greenhouse gas emissions corresponding to decreased mitigation scenarios (Figure S12). The dipole accentuation corresponded to topographic constraints, with increased densities in highlands and decreased densities in lowlands (Figure S13).
The sensitivity of the process-based model to temperature and precipitation was estimated through 10,000 simulations of population dynamics across a range of temperature and precipitation values (Figure 3). Higher precipitation resulted in higher adult female densities for both species (Figures 3A,C). However, the model displayed an optimal temperature above which increasing temperatures resulted in lower adult female densities. This thermal optimum was much warmer for Ae. aegypti (33°C; Figure 3A) than for Ae. albopictus (29°C; Figure 3C). In the range of temperature and precipitation values used here to evaluate the sensitivity of the model, future changes in temperature were expected to be of greater importance on model outputs than future changes in precipitation. Indeed, temperature increases at the end of the century following the SSP3 7.0 scenario were shown to be relatively uniform within the domain and to reach as much as 4.7°C (Figure S7), whereas changes in precipitation were usually projected to be weak and within −1 to +1mm/d, only reaching −3.5 OR +3.5mm/d in localized areas (e.g., south Myanmar in September; Figure S7). During summer months in southern Vietnam, rising temperatures predicted at the end of the century along the SSP3 7.0 scenario were thus expected to bring Ae. aegypti and Ae. albopictus closer to their thermal optimums (Figure 3A,C), resulting in an increase in densities (Figures 3B,D). In contrast, in central Myanmar, where contemporary temperatures are higher, a rise was expected to exceed thermal optimum of both species (Figures 3A,C), resulting in a decrease of densities (Figures 3B,D). In Thailand, rising temperatures were shown to have a contrasting effect on the two species, bringing Ae. aegypti closer to its thermal optimum (Figure 3A), resulting in an increase in densities (Figure 3B) and exceeding the optimum of Ae. albopictus (Figure 3C), which in turn led to a decrease in densities (Figure 3D).
Figure 3. Process-based model sensitivity to temperature and precipitation. Process-based model simulations were performed over various constant values of temperature and precipitations. Model outputs were extracted after 30 d of simulation and plotted against corresponding temperature and precipitations (Panels A and C). White dots and arrows represent evolution of temperature and precipitations from contemporary to future projections (2081–2100/SSP3 7.0) at a specific month and location, displayed on the panels B and D. Note: SSP, Shared Socioeconomic Pathways.
Figures 3A and 3C are graphs, plotting temperature (degree Celsius) ranging from 15 to 40 in increments of 5 (y-axis) across simulated density (color scale) according to precipitations millimeter per day ranging from 0 to 15 in increments of 5 (x-axis) for Aedes aegypti and Aedes albopictus. A scale depicts the simulated adult density after 30 days ranges from 0 to 1 in increments of 0.5. Figures 3B and 3D are Southeast Asia maps, depicting the Aedes density percentage change from contemporary to future projections in the month of May for Aedes aegypti and Aedes albopictus. A scale depicting density percentage changes ranges from negative 100 to negative 50 percent in increments of 50, negative 50 to 20 percent in increments of 30, negative 20 to 10 percent in increments of 10, 10 to 30 percent in increments of 20, 30 to 60 percent in increments of 30, 60 to 130 percent in increments of 70.
Finally, Figure 4 summarizes the evolution of both Aedes species for the next century as a function of the four climate scenarios and for the two seasons in the highlands and lowlands (regions that were shown to show coherent evolutions). Because future changes in precipitation were shown to be globally null in both highlands and lowlands (Figure S14), future precipitation had overall weak influences on the species’ future evolution. By contrast, future changes in temperature were always positive and similar in both highlands and lowlands (Figure S14). The differing response of both species in highlands and lowlands during summer months, with a weaker rise in Ae. aegypti densities in lowlands in comparison with highlands (+5% and +33% at the end of the century for the SSP3 7.0) and a decrease in Ae. albopictus in lowlands (−24%; Figure 4), can be explained by warmer contemporary temperatures in lowlands. Indeed, a future rise in temperature in already warm lowlands is likely to result in a) a small increase in Ae. aegypti densities due to contemporary densities already close to their thermal optimum, and b) a decrease of Ae. albopictus densities due to excessive heat bringing the species beyond its thermal optimum. In highlands during summer months and in the entire domain during winter months, rising temperatures resulted in an increase in densities of both species because their present-day background climate was well below the excessive heat thresholds. Indeed, at the end of the century following the SSP3 7.0, Ae. aegypti and Ae. albopictus densities increased by 15% and 21%, respectively, in the entire domain during winter months.
Figure 4. Evolution of projected Aedes density across the 21st century following four SSP scenarios. Grid cells were split in two groups considering their elevation (Highlands: elevation >600m, Lowlands: elevation >600m). For each species, scenario, period and month, points and bars represent respectively mean and standard deviation of percentage change values associated to all climate models and all grid cells. Note: SSP, Shared Socioeconomic Pathways.
Figure 4 is a set of 8 error bar graphs. On the left, the four graphs are titled April, May, June, plotting density percentage change in highlands, ranging from negative 50 to 50 percent in increments of 50 and density percentage change in lowlands, ranging from negative 50 to 50 percent in increments of 50, each for aedes aegypti and aedes albopictus (y-axis) across years, ranging as 2021 to 2040, 2041 to 2060, 2061 to 2080, and 2081 to 2100 (x-axis) for shared socio-economic pathway 1 (2.6), shared socio-economic pathway 2 (4.5), shared socio-economic pathway 3 (7.0) and shared socio-economic pathway 5 (8.5). On the right, the four graphs are titled December, January, February, plotting density percentage change in highlands, ranging from negative 50 to 50 percent in increments of 50 and density percentage change in lowlands, ranging from negative 50 to 50 percent in increments of 50, each for aedes aegypti and aedes albopictus (y-axis) across years, ranging as 2021 to 2040, 2041 to 2060, 2061 to 2080, and 2081 to 2100 (x-axis) for shared socio-economic pathway 1 (2.6), shared socio-economic pathway 2 (4.5), shared socio-economic pathway 3 (7.0) and shared socio-economic pathway 5 (8.5).
The potential effects of climate mitigation measures was investigated by computing the relative change in adult density associated with a transition from SSP3 7.0 to SSP1 2.6 at the end of the century. Over the entire domain, mitigation measures would result in an average yearly decrease in Ae. aegypti densities of 5% and an increase in Ae. albopictus densities of 7%. Such mitigation measures would even result in a global increase of Ae. albopictus densities of 28% during summer months, because colder temperatures would bring the species closer to its thermal optimum (Figure S15C and D). For Ae. aegypti during summer months, these mitigation measures would result in an average increase in densities of 2%, although with some local variations, e.g., increases of ∼20%−40% in central Thailand and Myanmar and decreases of about 10%−20% in Vietnam and in the north of the domain (Figure S15B).
Discussion
Main Results
A process-based model was used here to simulate climate-dependent seasonal population dynamics of two human arbovirus vectors, Ae. aegypti and Ae. albopictus, in Southeast Asia, for present and future climates. Using climate model projections of temperature and precipitation enabled us to produce predictions of vector densities up to the end of the 21st century in four climate scenarios. Results suggest that densities of Ae. aegypti mosquitoes, which are currently the main vectors of dengue fever in Southeast Asia, will increase in most parts of the continent during the 21st century in all scenarios. Even strict implementation of the Paris Agreement’s greenhouse gas emissions reductions [here, the SSP1 2.6, Intergovernmental Panel on Climate Change (IPCC) report]51,52 would not mitigate this rise in mosquito densities. Indeed, following the SSP1 2.6 pathway instead of the current SSP3 7.0 would result in a very limited decrease in Ae. aegypti densities and even a slight increase in Ae. albopictus densities.
Given the current climate model projections, the projected changes of Aedes densities are more likely to be driven by a future change of temperature than a change in precipitation, which do not show clear evolutions between climate models. Indeed, although climate models systematically predict a rise of temperatures in all Southeast Asia, they do not agree in predicting the direction of the change in precipitation in most parts of the continent, especially during summer months (Figure S7). Given the agreement of the modeled temperature projections, and even with such precipitation uncertainties in state-of-the-art model CMIP6 outputs, mosquito densities modeled from all separate climate projections agreed in the direction of the future change of Aedes species, with an overall increase of Ae. aegypti densities and dipoles of increased and decreased Ae. albopictus densities in highlands and lowlands, respectively.
Implications of Rising Temperatures
Health implications.
Because Ae. aegypti is the main vector of dengue fever in Southeast Asia, rising temperatures on the continent, along the current pathway where the SSP1 2.6 seems unlikely and the SSP2 4.5/SSP3 7.0 seem more likely,53 thus pose a real threat of increased dengue risk. Indeed, although relationships between vector indices and dengue risk are difficult to establish,11 Aedes vector density has been shown to correlate positively with dengue occurrence.12–15 An increase of vector density, added to a lengthening of the season of high density (Figure S16) due to a lengthening of warmer seasons, are thus likely to result in higher incidence, higher numbers of outbreaks, and a lengthening of the epidemic season. Rising temperatures may also amplify this pattern through other mechanisms, such as an increase of Aedes feeding rate (up to a threshold of 36°C)54,55 or a reduction of the extrinsic incubation period, as it has been evidenced for dengue fever.54,56,57 Yet, the effects of rising temperatures on mosquitoes’ densities may also be mitigated, because mosquitoes’ life span may shorten with higher temperatures,35,37,55 reducing the risk of disease transmission between two hosts.58,59 Ryan et al. incorporated these complex biological factors into a global transmission model and provided future projections associated with various climate scenarios.60 Their results in Southeast Asia align with ours, because they predicted an increase in disease transmission through Ae. aegypti as well as a reduction of transmission through Ae. albopictus. The same conclusions were reached in our study and in Ryan et al. about the effect of greenhouse gas emission mitigation measures, which were expected to attenuate transmission through Ae. aegypti but increase transmission through Ae. albopictus. Using various covariates including climate variables, Messina et al. built a statistical model of dengue occurrence at global scale to provide future predictions of climate change consequences on dengue risk.61 Their conclusions, however, contrast with ours in Southeast Asia, where they predict no real change in dengue risk in the future. These comparisons highlight the importance of considering models of different natures to reach scientific consensus and make reliable predictions about the health effects of climate change.
Ecology implications.
Rising temperatures over the continent are likely to benefit Ae. aegypti, because the process-based model displays a high optimal temperature of 33.3°C (Figure 3). In contrast, rising temperatures are not expected to benefit Ae. albopictus in all of Southeast Asia, because the model displays a colder optimal temperature for this species at ∼29.1°C (Figure 3), which is likely to be exceeded in most parts of Southeast Asia during summer months at the end of the century, according to all climate models. This factor resulted in the model predicting a decrease of Ae. albopictus densities everywhere except in the northern parts of Southeast Asia (Figure 2).
Regions where the model predicts rising Ae. albopictus densities match with topography maps and correspond to regions of higher altitude (Figure S13), where rising temperatures may get mosquitoes closer to their optimal temperature. This change could result in an expansion of the niche of the species toward higher altitudes, as has been projected in northern and southern America, Africa,19,62 and Asia.63
An important consideration is that those overall optimal temperatures from the model were driven by the underlying temperature-dependent functions used to describe development and mortality rates of each life stage, based on the literature. For instance, the transition rate from larvae to pupae shows a lower optimal temperature for Ae. albopictus (28°C) than for Ae. aegypti (31°C; Figure S17). This factor, as well as considerations of climate variable uncertainties, naturally lead to the discussion on the potential limitations of this study.
Limitations.
Outputs from the model provide only indices of the relative density of Aedes mosquitoes, but not quantitative evolutions of number of adults per surface area. To be able to provide such information, one needs to rely on the quantification of the density of breeding sites, which is not available at this spatial scale from observations. Although relative indices of adult densities from the present modeling could provide valuable information on seasonal variability and future evolution, we advocate for a major effort in observations of breeding site adult densities on regional scales.
The two species were considered separately in our modeling framework, with no interaction between them. Yet, they can often share the same breeding sites and interspecific competition can occur among aquatic stages.64,65 For instance, both species can share the same breeding sites in urban environments (e.g., used tires).66 Competition between Aedes species was not implemented in our framework because it depends on very complex mechanisms that could not be accurately described at this spatial scale, such as the diversity of breeding sites,67 the nature of available nutrients,65 and the history of species spread.68 Our predictions about future changes in densities of each species should thus be interpreted as their separate response to changes in their climate suitability. Interspecific competition has the potential to increase or decrease our predictions for each species, for instance, in areas where they use similar breeding sites, such as urban areas.
Seasonal variations of densities were assessed using climate data at a monthly scale. This process did not allow us to account for the effects of extreme climate events occurring at a submonthly scale and responsible for extreme values of temperature or precipitation. For instance, extreme rainfall events can result in a drop in egg, larva, and pupa densities due to the washing of breeding sites, yet such mechanism could not be accounted for here due to the smoothing of monthly climate data. As pointed out in the IPCC reports,51,52 such extreme events are forecast to increase in the future and will need to be considered in future studies.
Considering the objective of asserting the impact of climate over Aedes densities, only rain-filled breeding sites were considered in the modeling framework. The presence of human-filled breeding sites (e.g., water storage containers, flower vases) typically found in urban settings where the SDM model predicts highest probabilities of occurrence (Figure S3B) could attenuate the effect of rainfall over density variations. Yet, we believe this is of negligible impact because urban settings also present numerous rain-filled artificial breeding sites, such as buckets, tires, and plastic bottles.69–71
Although seasonal variability of mosquito density was assessed through the process-based model, its baseline spatial variability was estimated with statistical SDMs. SDMs predicted higher probability of presence in urban areas (high proportion of impervious surface land cover type) for both species. This pattern could be due to some extent to an observational bias in the occurrence database, with populated areas potentially more likely to be surveyed than rural areas. However, the potential observational bias is only pertinent when examining contemporary density maps but has no consequences for the conclusions about the future relative evolution of densities, which did not depend on SDM outputs. Still, SDM outputs could be pertinent to inform the future evolution of densities, for instance by using predictions of future urban expansion such as provided by Huang et al.72
Several sources of uncertainty can, however, compromise future projections. First, the process-based model does not apprehend adaptive abilities of the species, potentially responsible for variations of biological responses to varying temperature, both on spatial scales with potentially differing genetic strains between regions73–75 and on evolutionary scale, with species likely to adapt to changing environmental conditions. Such a mechanism is particularly relevant for Aedes species, considering the role of adaptation to cold temperatures in the recent invasion of Ae. albopictus in temperate regions, such as Europe and northern America.76,77
Another source of uncertainty comes from the climate model data used as inputs in the process-based model. In our study, particular focus was given to projections corresponding to the SSP3 7.0, because it is considered as one likely scenario if current greenhouse gas emissions are not curbed drastically to reach the levels in the Paris Agreements.53 Still, uncertainty remains about the projections from climate models following this scenario, especially for precipitation projections at the regional scales.78 To take into account such uncertainties and capture their magnitude, nine different climate models were considered in our study. Although differing in their magnitude, outputs from computations based on all climate models converged in predicting the global trends of Aedes densities—i.e., a regional increase of Ae. aegypti densities in the continent for all scenarios and a contrast between an increase of Ae. albopictus densities in highlands and a decrease in lowlands.
A main source of uncertainty in density projections remained in the parameters and weather-dependent functions used to describe the mortality and transitions rates between life stages of mosquito cycles. An interesting consideration is that a similar process-based compartment model developed by Liu-Helmersson et al.25 using different functions and parameters than the present study reached opposite conclusions in some parts of Southeast Asia for projections of Ae. aegypti densities. The roots of such discrepancies were explored here. They probably lie in important quantitative differences in temperature-dependent transition functions between the Liu-Helmersson et al. study and the present study, such as transition rates from larvae to pupae and from pupae to adult (Figure S18). Indeed, a sharp decline in transition rates used by Liu-Helmersson et al.25 above 20°–22°C is consistent with their predicted decrease of adult densities in Southeast Asia lowlands, where rising temperatures are more likely to exceed the optimal temperature threshold of their model.
Such divergence in functions and outputs reveals the importance of proper parameter estimates in process-based models. Process-based models are based on solid scientific knowledge of the species biology, but confidence in such models’ predictions can only come from a confidence in the underlying functions used to describe species’ response to given meteorological variations. Unfortunately, there are too few observational-based studies to choose such parametrization with confidence, and these may also be spatially dependent. Given the paucity of observations, researchers rely on the current literature to implement such climate-dependent parametrization, and we advocate for more observational/lab studies to estimate how the mosquito life cycles depend on meteorological variables at regional scales across the world.
Validation
Because weather variables show great variability at the seasonal scale, the ability of our model to describe adult density response to these variables was attested to through comparison with seasonal dynamics of mosquito densities measured from field survey. Validation of spatial dynamics was not feasible here due to the inherent difficulty of gathering spatially consistent mosquito density data. Indeed, most methods used to survey mosquito abundance or density involve the use of traps to capture eggs and adults, which can be subject to spatially inconsistent sources of bias. Indeed, adult traps can only capture an unknown and variable proportion of adults in a location, whereas egg traps can be biased by local availability of breeding sites (a higher number of surrounding breeding sites dilutes the number of adults that will lay eggs in the trap).79 Mark–release–recapture methods are able to approximate consistently the abundance of mosquitoes in different locations80–83 but are very costly and time-consuming.
Field data from Phnom Penh and Vientiane were used here to estimate only seasonal variations of Aedes densities. In Phnom Penh, records covered only a single year of sampling and were therefore not likely to be fully representative of the seasonal variations of densities in the location. Still, seasonal dynamics of densities, although contrasted between the two species, appeared to both match with outputs from the process-based model.
Further Uses of Modeled Mosquito Densities for Vectorborne Diseases
Model outputs describing vector risk can be used in different ways to investigate how it translates into disease risk. Such process-based model outputs have indeed been used as input in dengue mechanistic transmission models.84 Temporally variable indices of adult density, for instance those modeled from historical observed gridded climate data (e.g., from the Worldclim data set), could also be used in statistical models as an independent covariate to build better disease predictive models when long-term data from density surveys are not available. Outputs from large-scale process-based models of Ae. aegypti density have indeed been shown to correlate with local dengue fever caseload.85 Yet, causal relationships between vector density and disease risk are not completely apprehended for diseases like dengue fever11,86 as a result of a lack of long-term reliable longitudinal entomological survey data.11,13 In such contexts, process-based model outputs can be useful as proxies to investigate empirical relationships between vector density and recorded disease dynamics.
Conclusion
This work brings new insights about the benefits and limitations of process-based approaches for the modeling of mosquito densities and provides a new tool for investigating the climate dependency of mosquito-borne diseases dynamics. Most important, these findings carry further evidence that climate changes that are human in origin will impact ecosystems and, in turn, public health by highlighting the threat of a rise in densities of a potent vector for numerous infectious diseases in Southeast Asia for the coming 21st century. We show that reductions of greenhouse gas emissions following strictly the Paris Agreements, although unlikely at the date of the present study, would have a limited impact on this increase in vector risk.
Supplementary Material
Click here for additional data file.
Click here for additional data file.
Acknowledgments
This work was funded by the Agence Française de Développement (AFD), in the framework of the ECOMORE II project (http://ecomore.org/).
Model outputs are available for download at https://doi.org/10.23708/NYX0NV.
==== Refs
References
1. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, et al. 2013. The global distribution and burden of dengue. Nature 496 (7446 ):504–507, PMID: , 10.1038/nature12060.23563266
2. Petersen LR, Jamieson DJ, Powers AM, Honein MA. 2016. Zika virus. N Engl J Med 374 (16 ):1552–1563, PMID: , 10.1056/NEJMra1602113.27028561
3. Caglioti C, Lalle E, Castilletti C, Carletti F, Capobianchi MR, Bordi L, et al. 2013. Chikungunya virus infection: an overview. New Microbiol 36 (3 ):211–227, PMID: .23912863
4. LaBeaud AD, Bashir F, King CH. 2011. Measuring the burden of arboviral diseases: the spectrum of morbidity and mortality from four prevalent infections. Popul Health Metr 9 (1 ):1, PMID: .21219615
5. WHO (World Health Organization). 2019. Dengue and Severe Dengue. https://www.who.int/news-room/fact-sheets/detail/dengue-and-severe-dengue [accessed 1 March 2021].
6. Morrison AC, Zielinski-Gutierrez E, Scott TW, Rosenberg R. 2008. Defining challenges and proposing solutions for control of the virus vector Aedes aegypti. PLoS Med 5 (3 ):e68, PMID: , 10.1371/journal.pmed.0050068.18351798
7. Ooi E-E, Goh K-T, Gubler DJ. 2006. Dengue prevention and 35 years of vector control in Singapore. Emerg Infect Dis 12 (6 ):887–893, PMID: , 10.3201/eid1206.051210.16707042
8. Vega-Rúa A, Lourenço-de-Oliveira R, Mousson L, Vazeille M, Fuchs S, Yébakima A, et al. 2015. Chikungunya virus transmission potential by local aedes mosquitoes in the Americas and Europe. PLoS Negl Trop Dis 9 (5 ):e0003780, PMID: , 10.1371/journal.pntd.0003780.25993633
9. Rodriguez-Morales AJ. 2015. Zika: the new arbovirus threat for Latin America. J Infect Dev Ctries 9 (6 ):684–685, PMID: , 10.3855/jidc.7230.26142684
10. Kramer LD, Ciota AT. 2015. Dissecting vectorial capacity for mosquito-borne viruses. Curr Opin Virol 15 :112–118, PMID: , 10.1016/j.coviro.2015.10.003.26569343
11. Bowman LR, Runge-Ranzinger S, McCall PJ. 2014. Assessing the relationship between vector indices and dengue transmission: a systematic review of the evidence. PLoS Negl Trop Dis 8 (5 ):e2848, PMID: , 10.1371/journal.pntd.0002848.24810901
12. Barrera R, Amador M, MacKay AJ. 2011. Population dynamics of Aedes aegypti and dengue as influenced by weather and human behavior in San Juan, Puerto Rico. PLoS Negl Trop Dis 5 (12 ):e1378, PMID: , 10.1371/journal.pntd.0001378.22206021
13. Cromwell EA, Stoddard ST, Barker CM, Van Rie A, Messer WB, Meshnick SR, et al. 2017. The relationship between entomological indicators of Aedes aegypti abundance and dengue virus infection. PLoS Negl Trop Dis 11 (3 ):e0005429, PMID: , 10.1371/journal.pntd.0005429.28333938
14. Chadee DD. 2009. Dengue cases and Aedes aegypti indices in Trinidad, West Indies. Acta Trop 112 (2 ):174–180, PMID: , 10.1016/j.actatropica.2009.07.017.19632189
15. Rodriguez-Figueroa L, Rigau-Perez JG, Suarez EL, Reiter P. 1995. Risk factors for dengue infection during an outbreak in Yanes, Puerto Rico in 1991. Am J Trop Med Hyg 52 (6 ):496–502, PMID: , 10.4269/ajtmh.1995.52.496.7611553
16. Sedda L, Taylor BM, Eiras AE, Marques JT, Dillon RJ. 2020. Using the intrinsic growth rate of the mosquito population improves spatio-temporal dengue risk estimation. Acta Trop 208 :105519, PMID: , 10.1016/j.actatropica.2020.105519.32389450
17. Marinho RA, Beserra EB, Bezerra-Gusmão MA, Porto V. D S, Olinda RA, Dos Santos CAC, et al. 2016. Effects of temperature on the life cycle, expansion, and dispersion of Aedes aegypti (Diptera: Culicidae) in three cities in Paraiba, Brazil. J Vector Ecol 41 (1 ):1–10, PMID: , 10.1111/jvec.12187.27232118
18. Kweka EJ, Baraka V, Mathias L, Mwang’onde B, Baraka G, Lyaruu L, et al. 2018. Ecology of Aedes Mosquitoes, The Major Vectors of Arboviruses in Human Population. In: Dengue Fever: A Resilient Threat in the Face of Innovation. Falcón-Lezama JA, Betancourt-Cravioto M, Tapia-Conyer R, eds. 10.5772/intechopen.81439.
19. Kraemer MUG, Reiner RC, Brady OJ, Messina JP, Gilbert M, Pigott DM, et al. 2019. Past and future spread of the arbovirus vectors Aedes aegypti and Aedes albopictus. Nat Microbiol 4 (5 ):854–863, PMID: , 10.1038/s41564-019-0376-y.30833735
20. Kraemer MUG, Sinka ME, Duda KA, Mylne A, Shearer FM, Brady OJ, et al. 2015. The global compendium of Aedes aegypti and Ae. albopictus occurrence. Sci Data 2 :150035, PMID: , 10.1038/sdata.2015.35.26175912
21. Ducheyne E, Tran Minh NN, Haddad N, Bryssinckx W, Buliva E, Simard F, et al. 2018. Current and future distribution of Aedes aegypti and Aedes albopictus (Diptera: Culicidae) in WHO Eastern Mediterranean region. Int J Health Geogr 17 (1 ):4, PMID: , 10.1186/s12942-018-0125-0.29444675
22. Ding F, Fu J, Jiang D, Hao M, Lin G. 2018. Mapping the spatial distribution of Aedes aegypti and Aedes albopictus. Acta Trop 178 :155–162, PMID: , 10.1016/j.actatropica.2017.11.020.29191515
23. Focks DA, Haile DG, Daniels E, Mount GA. 1993. Dynamic life table model for Aedes aegypti (Diptera: Culicidae): analysis of the literature and model development. J Med Entomol 30 (6 ):1003–1017, PMID: , 10.1093/jmedent/30.6.1003.8271242
24. Hopp MJ, Foley JA. 2001. Global-scale relationships between climate and the dengue fever vector, Aedes aegypti. Clim Change 48 (2/3 ):441–463, 10.1023/A:1010717502442.
25. Liu-Helmersson J, Brännström Å, Sewe MO, Semenza JC, Rocklöv J. 2019. Estimating past, present, and future trends in the global distribution and abundance of the arbovirus vector Aedes aegypti under climate change scenarios. Front Public Health 7 :148, PMID: , 10.3389/fpubh.2019.00148.31249824
26. Cattarino L, Rodriguez-Barraquer I, Imai N, Cummings DAT, Ferguson NM. 2020. Mapping global variation in dengue transmission intensity. Sci Transl Med 12 (528 ), 10.1126/scitranslmed.aax4144.
27. Lugito NPH. 2017. Trends of dengue disease epidemiology. Virology (Auckl) 8 :1178122X17695836, PMID: , 10.1177/1178122X17695836.28579763
28. Boyer S, Lopes S, Prasetyo D, Hustedt J, Sarady AS, Doum D, et al. 2018. Resistance of Aedes aegypti (Diptera: Culicidae) populations to deltamethrin, permethrin, and temephos in Cambodia. Asia Pac J Public Health 30 (2 ):158–166, PMID: , 10.1177/1010539517753876.29502428
29. Marcombe S, Fustec B, Cattel J, Chonephetsarath S, Thammavong P, Phommavanh N, et al. 2019. Distribution of insecticide resistance and mechanisms involved in the arbovirus vector Aedes aegypti in Laos and implication for vector control. PLOS Negl Trop Dis 13 (12 ):e0007852, PMID: , 10.1371/journal.pntd.0007852.31830027
30. Calvez E, Pommelet V, Somlor S, Pompon J, Viengphouthong S, Bounmany P, et al. 2020. Trends of the dengue serotype-4 circulation with epidemiological, phylogenetic, and entomological insights in Lao PDR between 2015 and 2019. Pathogens 9 (9 ):728, 10.3390/pathogens9090728.32899416
31. Eyring V, Bony S, Meehl GA, Senior CA, Stevens B, Stouffer RJ, et al. 2016. Overview of the coupled model intercomparison project phase 6 (CMIP6) experimental design and organization. Geosci Model Dev 9 (5 ):1937–1958, 10.5194/gmd-9-1937-2016.
32. Cailly P, Tran A, Balenghien T, L’Ambert G, Toty C, Ezanno P, et al. 2012. A climate-driven abundance model to assess mosquito control strategies. Ecological Modelling 227 :7–17, 10.1016/j.ecolmodel.2011.10.027.
33. Tran A, L’Ambert G, Lacour G, Benoît R, Demarchi M, Cros M, et al. 2013. A rainfall- and temperature-driven abundance model for Aedes albopictus populations. Int J Environ Res Public Health 10 (5 ):1698–1719, PMID: , 10.3390/ijerph10051698.23624579
34. Tran A, Mangeas M, Demarchi M, Roux E, Degenne P, Haramboure M, et al. 2020. Complementarity of empirical and process-based approaches to modelling mosquito population dynamics with Aedes albopictus as an example—application to the development of an operational mapping tool of vector populations. PLoS One 15 (1 ):e0227407, PMID: , 10.1371/journal.pone.0227407.31951601
35. Delatte H, Gimonneau G, Triboire A, Fontenille D. 2009. Influence of temperature on immature development, survival, longevity, fecundity, and gonotrophic cycles of Aedes albopictus, vector of chikungunya and dengue in the Indian Ocean. J Med Entomol 46 (1 ):33–41, PMID: , 10.1603/033.046.0105.19198515
36. Dieng H, Rahman GMS, Abu Hassan A, Che Salmah MR, Satho T, Miake F, et al. 2012. The effects of simulated rainfall on immature population dynamics of Aedes albopictus and female oviposition. Int J Biometeorol 56 (1 ):113–120, PMID: , 10.1007/s00484-011-0402-0.21267602
37. Magori K, Legros M, Puente ME, Focks DA, Scott TW, Lloyd AL, et al. 2009. Skeeter buster: a stochastic, spatially explicit modeling tool for studying Aedes aegypti population replacement and population suppression strategies. PLoS Negl Trop Dis 3 (9 ):e508, PMID: , 10.1371/journal.pntd.0000508.19721700
38. Xin X, Wu T, Shi X, Zhang F, Li J, Chu M, et al. 2019. BCC BCC-CSM2MR model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.3035.
39. Voldoire A. 2019. CNRM-CERFACS CNRM-CM6-1 model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.4197.
40. Voldoire A. 2019. CNRM-CERFACS CNRM-ESM2-1 model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.4199.
41. Swart NC, Cole JNS, Kharin VV, Lazare M, Scinocca JF, Gillett NP, et al. 2019. CCCma CanESM5 model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.3690.
42. John JG, Blanton C, McHugh C, Radhakrishnan A, Rand K, Vahlenkamp H, et al. 2018. NOAA-GFDL GFDL-ESM4 model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.8691.
43. Boucher O, Denvil S, Levavasseur G, Cozic A, Caubel A, Foujols M-A, et al. 2019. IPSL IPSL-CM6A-LR model output prepared for CMIP6 ScenarioMIP ssp370. Earth System Grid Federation. 10.22033/ESGF/CMIP6.5265.
44. Tachiiri K, Abe M, Hajima T, Arakawa O, Suzuki T, Komuro Y, et al. 2019. MIROC MIROC-ES2L model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.5751.
45. Shiogama H, Abe M, Tatebe H. 2019. MIROC MIROC6 model output prepared for CMIP6 ScenarioMIP ssp370. Version 20200310. Earth System Grid Federation. 10.22033/ESGF/CMIP6.5752.
46. Yukimoto S, Koshiro T, Kawai H, Oshima N, Yoshida K, Urakawa S, et al. 2019. MRI MRI-ESM2.0 model output prepared for CMIP6 ScenarioMIP ssp370. Version 20203010. Earth System Grid Federation. 10.22033/ESGF/CMIP6.6915.
47. Gong P, Liu H, Zhang M, Li C, Wang J, Huang H, et al. 2019. Stable classification with limited sample: transferring a 30-m resolution sample set collected in 2015 to mapping 10-m resolution global land cover in 2017. Science Bulletin 64 (6 ):370–373, 10.1016/j.scib.2019.03.002.
48. Jarvis A, Reuter H, Nelson A, Guevara E. 2008. Hole-filled SRTM for the globe version 3, from the CGIAR-CSI SRTM 90m database. https://srtm.csi.cgiar.org/srtmdata/ [accessed 10 July 2020].
49. Maquart PO, Fontenille D, Boyer S. 2021. Recent and massive invasion of aedes (Stegomyia) albopictus (Skuse, 1894) in Phnom Penh, Cambodia. Parasit Vectors 14 (1 ):113, PMID: , 10.1186/s13071-021-04633-5.33602318
50. Rattanarithikul R, Harbach RE, Harrison BA, Panthusiri P, Coleman RE, Richardson JH. 2010. Illustrated keys to the mosquitoes of Thailand. VI. Tribe Aedini. Southeast Asian J Trop Med Public Health 41 (suppl 1 ):1–225, PMID: .20629439
51. IPCC (Intergovernmental Panel on Climate Change). 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. https://www.ipcc.ch/report/2019-refinement-to-the-2006-ipcc-guidelines-for-national-greenhouse-gas-inventories/ [accessed 1 March 2021].
52. The PhilPapers Foundation. Climate Change and Land: An IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems. Shukla PR, Skeg J, Calvo Buendia E, Masson-Delmotte V, Pörtner H-O, Roberts DC, et al., eds. https://philpapers.org/rec/SHUCCA-2.52 [accessed 1 March 2021].
53. UNEP (United Nations Environmental Programme). 2021. Emissions Gap Report 2021: The Heat Is On A World of Climate Promises Not Yet Delivered. https://wedocs.unep.org/20.500.11822/36990 [accessed 1 March 2021].
54. Morin CW, Comrie AC, Ernst K. 2013. Climate and dengue transmission: evidence and implications. Environ Health Perspect 121 (11–12 ):1264–1272, PMID: , 10.1289/ehp.1306556.24058050
55. Yang HM, Macoris MLG, Galvani KC, Andrighetti MTM, Wanderley DMV. 2009. Assessing the effects of temperature on the population of Aedes aegypti, the vector of dengue. Epidemiol Infect 137 (8 ):1188–1202, PMID: , 10.1017/S0950268809002040.19192322
56. Watts DM, Burke DS, Harrison BA, Whitmire RE, Nisalak A. 1987. Effect of temperature on the vector efficiency of Aedes aegypti for dengue 2 virus. Am J Trop Med Hyg 36 (1 ):143–152, PMID: , 10.4269/ajtmh.1987.36.143.3812879
57. Brady OJ, Golding N, Pigott DM, Kraemer MUG, Messina JP, Reiner RC, et al. 2014. Global temperature constraints on Aedes aegypti and Ae. albopictus persistence and competence for dengue virus transmission. Parasit Vectors 7 :338, PMID: , 10.1186/1756-3305-7-338.25052008
58. Mordecai EA, Caldwell JM, Grossman MK, Lippi CA, Johnson LR, Neira M, et al. 2019. Thermal biology of mosquito-borne disease. Ecol Lett 22 (10 ):1690–1708, PMID: , 10.1111/ele.13335.31286630
59. Smith DL, Battle KE, Hay SI, Barker CM, Scott TW, McKenzie FE, et al. 2012. Ross, Macdonald, and a theory for the dynamics and control of mosquito-transmitted pathogens. PLoS Pathog 8 (4 ):e1002588, PMID: , 10.1371/journal.ppat.1002588.22496640
60. Ryan SJ, Carlson CJ, Mordecai EA, Johnson LR. 2019. Global expansion and redistribution of aedes-borne virus transmission risk with climate change. PLoS Negl Trop Dis 13 (3 ):e0007213, PMID: , 10.1371/journal.pntd.0007213.30921321
61. Messina JP, Brady OJ, Golding N, Kraemer MUG, Wint GRW, Ray SE, et al. 2019. The current and future global distribution and population at risk of dengue. Nat Microbiol 4 (9 ):1508–1515, PMID: , 10.1038/s41564-019-0476-8.31182801
62. Equihua M, Ibáñez-Bernal S, Benítez G, Estrada-Contreras I, Sandoval-Ruiz CA, Mendoza-Palmero FS, et al. 2017. Establishment of Aedes aegypti (L.) in mountainous regions in Mexico: increasing number of population at risk of mosquito-borne disease and future climate conditions. Acta Trop 166 :316–327, PMID: , 10.1016/j.actatropica.2016.11.014.27863974
63. Acharya B, Cao C, Xu M, Khanal L, Naeem S, Pandit S, et al. 2018. Present and future of dengue fever in Nepal: mapping climatic suitability by ecological niche model. Int J Environ Res Public Health 15 (2 ):187, 10.3390/ijerph15020187.29360797
64. Juliano SA, Lounibos LP, O’Meara GF. 2004. A field test for competitive effects of Aedes albopictus on A. aegypti in South Florida: differences between sites of coexistence and exclusion? Oecologia 139 (4 ):583–593, PMID: , 10.1007/s00442-004-1532-4.15024640
65. Murrell EG, Juliano SA. 2008. Detritus type alters the outcome of interspecific competition between Aedes aegypti and Aedes albopictus (Diptera: Culicidae). J Med Entomol 45 (3 ):375–383, PMID: , 10.1093/jmedent/45.3.375.18533429
66. Abílio AP, Abudasse G, Kampango A, Candrinho B, Sitoi S, Luciano J, et al. 2018. Distribution and breeding sites of Aedes aegypti and Aedes albopictus in 32 urban/peri-urban districts of Mozambique: implication for assessing the risk of arbovirus outbreaks. PLoS Negl Trop Dis 12 (9 ):e0006692, PMID: , 10.1371/journal.pntd.0006692.30208017
67. Chareonviriyaphap T, Akratanakul P, Nettanomsak S, Huntamai S. 2003. Larval habitats and distribution patterns of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse), in Thailand. Southeast Asian J Trop Med Public Health 34 (3 ):529–535, PMID: .15115122
68. Cunze S, Kochmann J, Koch LK, Klimpel S. 2018. Niche conservatism of Aedes albopictus and Aedes aegypti - two mosquito species with different invasion histories. Sci Rep 8 (1 ):7733, PMID: , 10.1038/s41598-018-26092-2.29769652
69. Arunachalam N, Tana S, Espino F, Kittayapong P, Abeyewickreme W, Wai KT, et al. 2010. Eco-bio-social determinants of dengue vector breeding: a multicountry study in urban and periurban Asia. Bull World Health Organ 88 (3 ):173–184, PMID: , 10.2471/BLT.09.067892.20428384
70. Edillo FE, Roble ND, Ii NDO. 2012. The key breeding sites by pupal survey for dengue mosquito vectors, Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse), in Guba, Cebu City, Philippines. Southeast Asian J Trop Med Public Health 43 (6 ):1365–1374, PMID: .23413699
71. Rao BB, Harikumar P, Jayakrishnan T, George B. 2011. Characteristics of Aedes (Stegomyia) albopictus Skuse (Diptera: Culicidae) breeding sites. Southeast Asian J Trop Med Public Health 42 (5 ):1077–1082, PMID: .22299432
72. Huang K, Li X, Liu X, Seto KC. 2019. Projecting global urban land expansion and heat island intensification through 2050. Environ Res Lett 14 (11 ):114037, 10.1088/1748-9326/ab4b71.
73. Tippelt L, Werner D, Kampen H. 2019. Tolerance of three Aedes albopictus strains (Diptera: Culicidae) from different geographical origins towards winter temperatures under field conditions in northern Germany. PLoS One 14 (7 ):e0219553, PMID: , 10.1371/journal.pone.0219553.31310645
74. Tippelt L, Werner D, Kampen H. 2020. Low temperature tolerance of three Aedes albopictus strains (Diptera: Culicidae) under constant and fluctuating temperature scenarios. Parasit Vectors 13 (1 ):587, PMID: , 10.1186/s13071-020-04386-7.33225979
75. Kamimurai K, Matsuse IT, Takahashi H, Komukai J, Fukuda T, Suzuki K, et al. 2002. Effect of temperature on the development of Aedes aegypti and Aedes albopictus. Med Entomol Zool 53 (1 ):53–58, 10.7601/mez.53.53_1.
76. Sherpa S, Blum MGB, Després L. 2019. Cold adaptation in the Asian tiger mosquito’s native range precedes its invasion success in temperate regions. Evolution 73 (9 ):1793–1808, PMID: , 10.1111/evo.13801.31313825
77. Goubert C, Henri H, Minard G, Valiente Moro C, Mavingui P, Vieira C, et al. 2017. High-throughput sequencing of transposable element insertions suggests adaptive evolution of the invasive Asian tiger mosquito towards temperate environments. Mol Ecol 26 (15 ):3968–3981, PMID: , 10.1111/mec.14184.28517033
78. Intergovernmental Panel on Climate Change. 2021. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press.
79. Focks DA, UNDP (United Nations Development Programme)/World Bank/WHO Special Programme for Research and Training in Tropical Diseases. 2004. A Review of Entomological Sampling Methods and Indicators for Dengue Vectors, https://apps.who.int/iris/handle/10665/68575 [accessed 1 March 2021].
80. Maciel-de-Freitas R, Eiras ÁE, Lourenço-de-Oliveira R. 2008. Calculating the survival rate and estimated population density of gravid Aedes aegypti (Diptera, Culicidae) in Rio de Janeiro. Cad Saúde Pública 24 (12 ):2747–2754, 10.1590/S0102-311X2008001200003.19082265
81. Mercer DR, Marie J, Bossin H, Faaruia M, Tetuanui A, Sang MC, et al. 2012. Estimation of population size and dispersal of Aedes polynesiensis on Toamaro motu, French Polynesia. J Med Entomol 49 (5 ):971–980, PMID: , 10.1603/me11234.23025176
82. Neira M, Lacroix R, Cáceres L, Kaiser PE, Young J, Pineda L, et al. 2014. Estimation of Aedes aegypti (Diptera: Culicidae) population size and adult male survival in an urban area in Panama. Mem Inst Oswaldo Cruz 109 (7 ):879–886, PMID: , 10.1590/0074-0276140136.25410991
83. Villela DAM, Codeço CT, Figueiredo F, Garcia GA, Maciel-de-Freitas R, Struchiner CJ, et al. 2015. A Bayesian hierarchical model for estimation of abundance and spatial density of Aedes aegypti. PLoS One 10 (4 ):e0123794, PMID: , 10.1371/journal.pone.0123794.25906323
84. Focks DA, Daniels E, Haile DG, Keesling JE. 1995. A simulation model of the epidemiology of urban dengue fever: literature analysis, model development, preliminary validation, and samples of simulation results. Am J Trop Med Hyg 53 (5 ):489–506, PMID: , 10.4269/ajtmh.1995.53.489.7485707
85. Hopp M, Foley J. 2003. Worldwide fluctuations in dengue fever cases related to climate variability. Clim Res 25 :85–94, 10.3354/cr025085.
86. Scott TW, Morrison A. 2004. Chapter 14: Aedes aegypti density and the risk of dengue-virus transmission. In: Ecological Aspects for Application of Genetically Modified Mosquitoes, vol. 2. Takken W, Scott TW, eds. Cham, Switzerland: Springer Frontis Series, 187–206.
| 36473499 | PMC9726451 | NO-CC CODE | 2022-12-09 23:25:58 | no | Environ Health Perspect. 2022 Dec 6; 130(12):127002 | utf-8 | Environ Health Perspect | 2,022 | 10.1289/EHP11068 | oa_other |
==== Front
REL
sprel
RELC Journal
0033-6882
1745-526X
SAGE Publications Sage UK: London, England
10.1177/00336882221137944
10.1177_00336882221137944
Editorial
Editorial: What is at the Heart of Teaching and Learning in a Post-pandemic Recovery World?
https://orcid.org/0000-0003-0136-2532
Yeo Marie Alina SEAMEO RELC, Singapore
Marie Alina Yeo, SEAMEO RELC, 30 Orange Grove Road, Singapore, 258352, Singapore. Email: [email protected]
12 2022
12 2022
12 2022
53 3 485489
© The Author(s) 2022
2022
SAGE Publications
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
typesetterts19
==== Body
pmcAt this time last year, just as we were transitioning from a COVID-19 pandemic to a COVID-19 endemic world, I was writing the editorial for the December 2021 issue of the RELC Journal. I celebrated the agility, resilience, and professionalism of teachers around the region who did their best to provide learning opportunities during periods of lockdown. I also raised concerns about learning loss and the increasing digital divide, as well as the blurring of boundaries between teachers’ personal and professional lives.
Thankfully, most countries have now entered a post-pandemic recovery phase with the resumption of face-to-face teaching or at least some form of hybrid teaching. Not surprisingly, the digital transformation that occurred over the past two years has continued. As many institutions and teachers witnessed first-hand the power of technology in engaging learners and providing useful data to inform teaching improvement, we are keen to continue to use tools and technologies.
As language teachers, teacher educators, and researchers, we may be wondering and worrying about this emphasis on technology, asking basic questions such as Can my school afford new technology? How can I find out about the latest tools? How do I choose from the multitude of tools? How do I learn to use these tools effectively? We may even be asking important pedagogical questions: How can I use this tool to improve teaching and learning? How will I know if the technology I use has made a difference to my students?
We can take heart by remembering that it is less about technology and more about the learners and their learning. At the recent BETTS Asia 2022 Conference in Bangkok, Thailand, this was the overwhelming message – that humans, not technology – must be at the heart of teaching and learning. At the end of the day, language teaching and language teacher education are not about technology or how it is used. It is about why we use it, for whom we use it, and the extent to which technology use can enhance or worsen our lives. As early as 2003, Castro (cited in Benito-Osorio et al., 2013: 275) wrote, “Constructivist learning cannot be achieved by the mere introduction of ICT, but must be understood as a type of education whose success largely depends on the humans behind it.”
As we emerge from the pandemic, there has been more emphasis on the human aspects of online learning, such as mental health (Alibudbud, 2021), teacher self-efficacy (Cataudella et al., 2021; Ma et al., 2021), compassion fatigue (Yang, 2021), and the need for teachers to develop socio-emotional competence (Hadar et al., 2020). For this reason, the kind of research and publication that we do, focusing on the impact of teaching approaches, methods, and materials on learners and their learning and the development of learner and teacher attitudes and identity, remains relevant pre-, during, and post-pandemic. It is with this in mind that I introduce the December 2022 issue of the RELC Journal.
In This Issue
Even as we approach some semblance of normality in teaching and travel, there seems to be no let-up in the number of submissions we have received. We, therefore, have yet another bumper issue containing 27 original manuscripts, all of which offer useful findings that will help to improve our understanding and practice of language education and language teacher professional development. I have summarized the key findings and urge readers to look more closely at the articles that are of personal interest and relevance.
The first six articles, which examine various aspects of teaching speaking and listening, found that: for Japanese English as a foreign language (EFL) learners at the intermediate level, an output-based shadowing procedure with explicit instruction and feedback improved phonemic discrimination abilities (Hamada, 2022);
native English speaking teachers (NESTs) may not necessarily be ideal English speakers for students to imitate (Kong and Kang, 2022);
some task features such as their nature and purpose, repetition, familiarity, and ease of task topics influence successful engagement (Aubrey et al., 2022);
learners’ planning processes and subsequent task performances differed significantly depending on their note-taking strategies, the interpersonal dynamics between pairs, and their second language (L2) proficiency (Leeming et al., 2022);
teacher educators’ first language (L1) use may intentionally or incidentally shape the attitudes of student teachers (Werang and Harrington, 2022);
input enhancement, rhythmic priming, and perception-based instruction may be useful in areas other than traditional grammar and pronunciation instruction (Lee, 2022).
The next four papers address various aspects of teacher reading, vocabulary development, and writing, offering these key findings: in making sense of visual stimuli, teachers engaged in four integrated categories of comprehension processes, namely: anticipation or preparation, sampling, deepening, and regulation (Bautista and Gutierrez, 2022);
the New Concept English (NCE) textbook series used in China provided opportunities for learning mid-frequency vocabulary (Yang and Coxhead, 2022);
while both self-assessment and indirect feedback helped learners develop self-regulation, self-assessment reduced maladaptive behavior more than indirect feedback (Vasu et al., 2022); and
a teacher's beliefs about the appropriateness of a particular approach may not be reflected in actual instructional practice (Jeon, 2022).
The final research articles concern teacher education. The key ideas from these are the following: by collaborating with language teachers, content teachers’ attitudes towards English medium instruction (EMI) teaching became more positive (Lu, 2022);
experience of engagement with multiliteracies’ pedagogies throughout their professional preparation provided teachers with a source of professional knowledge and created opportunities to change their pre-conceptions about English language teaching (Maia, 2022);
teachers navigated their Action Research identity construction across the four stages: Plan, Act, Observe, and Reflect (Nazari, 2022); and
Teacher educators utilized a variety of individual and community “funds” of knowledge in order to enhance their understanding of English language teaching and evidence-based practice (Banegas, 2022).
These articles are based on primary research undertaken by teachers and teacher educators from a wide range of countries, such as Japan, Malaysia, Indonesia, China, Taiwan, Brazil, and Argentina. They address gaps in our knowledge about particular areas in TESOL, offer strong theoretical background, employ rigorous methods of investigation, and provide new findings that are discussed with a focus on implications for practice.
Alongside these research articles, the RELC Journal features “Innovations in Practice” reports. These shorter articles serve to provide emerging scholars with an opportunity to share the findings of action research undertaken to address a specific classroom challenge. The three innovations featured are as follows: a Responsive Multimodal Oral Presentation Pedagogy (RMO2P) designed to systematically scaffold learning through the integration of TED videos with Web 2.0, collaborative learning, and teacher feedback (Sze Seau and Azman, 2022);
the integration into and implementation of online peer review on oral presentations in an undergraduate English literature curriculum (Ho, 2022); and
the use of a concordance software program, AntConc, to teach grammar in International English Language Testing System (IELTS) writing classes (Pham 2020).
These reports provide an easy-to-read and practical source of information for teachers. They include step-by-step information about why and how the innovation was introduced, an evaluation of its effectiveness, and suggestions for teachers who may want to try the same technique in their own classrooms.
Following these reports is a thematic review entitled Research of Language Teacher Identity: Status Quo and Future Directions (Sang 2022). The article summarizes theoretical conceptualizations of language teacher identity, examines the status quo of contemporaneous language teacher identity research, discusses the socialization process in language teacher identity development, and suggests further research directions. A highlight of this issue is an interview with Professor Li Wei, the Director and Dean of the Institute of Education, conducted by Professor Lawrence Zhang. The issue ends with books reviews on topics including ways of “being Chinese,” evaluating L2 vocabulary and grammar instruction, developing expertise through experience, boredom in foreign language classrooms, and language proficiency testing in Asia. Finally, the technology reviews evaluate three technology tools, namely, Clips, Grammarly, and Kahoot!, from a second language acquisition (SLA) lens. Taken together, these research articles, Innovation in Practice reports, thematic review, one Conversations with Experts interview, book reviews, and technology reviews provide some novel insights with a strong emphasis on how the findings can be applied in practice.
As I end this editorial, I would like to return to the question: What is at the heart of teaching and learning in a post-pandemic recovery world? Each of us might answer this differently, but for us at the SEAMEO Regional Language Centre (RELC) in Singapore, at the heart of what we do is our mission, which is to develop language teacher education in the region and promote international cooperation among language professionals. We would not be able to achieve our mission without the support of our editors at SAGE, guest editors, reviewers, and of course, our readers, so I would like to thank you for joining us to serve the needs and hopes of the TESOL community as we collectively move into a post-pandemic future.
ORCID iD: Marie Alina Yeo https://orcid.org/0000-0003-0136-2532
==== Refs
References
Alibudbud R (2021) On online learning and mental health during the COVID-19 pandemic: Perspectives from the Philippines. Asian Journal of Psychiatry 66 : 102867.34600400
Aubrey S King J Almukhaild H (2022) Language learner engagement during speaking tasks: A longitudinal study. RELC Journal 53 (3 ): 519–533.
Banegas DL (2022) Teacher educators’ funds of knowledge for the preparation of future teachers. RELC Journal 53 (3 ): 686–702.
Bautista JC Gutierrez MR (2022) A grounded theory on the comprehension processing of teachers as ESL readers of multimodal still visuals. RELC Journal 53 (3 ): 582–596.
Benito-Osorio D Peris-Ortiz M Armengot CR , et al . (2013) Web 5.0: The future of emotional competences in higher education. Global Business Perspectives 1 (3 ): 274–287.
Cataudella S Carta SM Mascia ML , et al. (2021) Teaching in times of the COVID-19 pandemic: A pilot study on teachers’ self-esteem and self-efficacy in an Italian sample. International Journal of Environmental Research and Public Health 18 (15 ): 8211.34360503
Hadar LL Ergas O Alpert B , et al . (2020) Rethinking teacher education in a VUCA world: student teachers’ social-emotional competencies during the COVID-19 crisis. European Journal of Teacher Education 43 (4 ): 573–586.
Hamada Y (2022) Developing a New Shadowing Procedure for Japanese EFL Learners. RELC Journal 53 (3 ): 490–504.
Ho E (2022) Online peer review of oral presentations. RELC Journal 53 (3 ): 712–722.
Jeon H (2022) Exploring the nature of teaching for transfer in EAP: A case study. RELC Journal 53 (3 ): 627–641.
Kong ML Kang HI (2022) Identity and accents: Do students really want to speak like native speakers of English? RELC Journal 53 (3 ): 505–518.
Lee BJ (2022) Enhancing listening comprehension through kinesthetic rhythm training. RELC Journal 53 (3 ): 567–581.
Leeming P Aubrey S Lambert C (2022) Collaborative pre-task planning processes and second-language task performance. RELC Journal 53 (3 ): 534–550.
Lu YH (2022) A case study of EMI teachers’ professional development: The impact of interdisciplinary teacher collaboration. RELC Journal 53 (3 ): 642–656.
Ma K Chutiyami M Zhang Y , et al . (2021) Online teaching self-efficacy during COVID-19: Changes, its associated factors and moderators. Education and Information Technologies 26 (6 ): 6675–6697.33723481
Maia AADM (2022) English language teacher education and the multiliteracies pedagogy: Constructing complex professional knowledge and identities. RELC Journal 53 (3 ): 657–671.
Nazari M (2022) Plan, Act, Observe, Reflect, Identity: Exploring teacher identity construction across the stages of Action Research. RELC Journal 53 (3 ): 672–685.
Pham QHP (2020) A corpus-based approach to grammar instruction in IELTS writing classes. RELC Journal 53 (3 ): 723–730.
Sang Y (2022) Research of language teacher identity: status quo and future directions. RELC Journal 53 (3 ): 731–738.
Sze Seau L Azman H (2022) Introducing a responsive multimodal oral presentation pedagogy: Integrating TED videos with Web 2.0, collaborative learning and teacher feedback. RELC Journal 53 (3 ): 703–711.
Vasu KAP Mei Fung Y Nimehchisalem V , et al . (2022) Self-regulated learning development in undergraduate ESL writing classrooms: Teacher feedback versus self-assessment. RELC Journal 53 (3 ): 612–626.
Werang EA Harrington M (2022) Avoiders and embracers: Attitudes towards L1 use by Indonesian EFL student teachers. RELC Journal 53 (3 ): 551–566.
Yang C (2021) Online teaching self-efficacy, social–emotional learning (SEL) competencies, and compassion fatigue among educators during the COVID-19 pandemic. School Psychology Review 50 (4 ): 505–518.
Yang L Coxhead A (2022) A corpus-based study of vocabulary in the new concept English textbook series. RELC Journal 53 (3 ): 597–611.
| 0 | PMC9726633 | NO-CC CODE | 2022-12-14 23:52:22 | no | 2022 Dec; 53(3):485-489 | utf-8 | null | null | null | oa_other |
==== Front
Affilia
Affilia
AFF
spaff
Affilia
0886-1099
1552-3020
SAGE Publications Sage CA: Los Angeles, CA
10.1177/08861099221137058
10.1177_08861099221137058
Original Article
“All I was Thinking About was Shattered”: Women's Experiences Transitioning Out of Anti-Trafficking Shelters During the COVID-19 Lockdown in Uganda
Namy Sophie 1
Namakula Sylvia 2
Nabachwa Agnes Grace 2
Ollerhead Madeleine 1
Tsai Laura Cordisco 3
Kemitare Jean 4
https://orcid.org/0000-0003-3543-8339
Bolton Kelly 5
https://orcid.org/0000-0002-4508-8822
Nkwanzi Violet 5
Carlson Catherine 5
1 Healing and Resilience after Trauma (HaRT), Asturias, Spain
2 Healing and Resilience after Trauma (HaRT), Kampala, Uganda
3 Carr Center for Human Rights Policy, 33574 Harvard University John F Kennedy School of Government , Cambridge, Massachusetts, USA
4 Urgent Action Fund- Africa, Kampala, Uganda
5 School of Social Work, 8063 The University of Alabama , Tuscaloosa, Alabama, USA
* Sophie Namy, Healing and Resilience after Trauma (HaRT), Asturias, Spain. Email: [email protected]
4 12 2022
4 12 2022
08861099221137058© The Author(s) 2022
2022
SAGE Publications
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
Human trafficking is an egregious violation of fundamental human rights and a global challenge. The long-term harms to survivors’ physical, psychological and social wellbeing are profound and well documented, and yet there are few studies exploring how to best promote resilience and holistic healing. This is especially true within shelter programs (where the majority of anti-trafficking services are provided) and during the transition out of residential shelter care, which is often a sensitive and challenging process. The current study begins to address this gap by centering the lived experiences of six women residing in a trafficking-specific shelter in Uganda as they unexpectedly transitioned back to their home communities due to the COVID-19 lockdown. We explore this pivotal moment in participants’ post-trafficking journey, focusing on how these women described and interpreted their rapidly changing life circumstances—including leaving the shelter, adjusting back to the community setting, and simultaneously navigating the uncertainties of a global pandemic. Four core themes emerged from the analysis: economic insecurities as a cross-cutting hardship; intensification of emotional and physical symptoms; social disruptions; and sources of hope and resilience. By centering their personal stories of struggle and strength, we hope to elevate survivors’ own accounts and draw on their insights to identify actionable considerations for future programming.
human trafficking
community reintegration
mental health
qualitative research
COVID-19
Urgent Action Fund- Africa University of Alabama https://doi.org/10.13039/100011531 Small Grants Program edited-statecorrected-proof
typesetterts19
==== Body
pmcBackground
Human trafficking is an egregious violation of fundamental human rights and a global challenge. While it is difficult to accurately assess prevalence given its illegality, available data estimates that 40.3 million people worldwide are trafficked each year for sexual exploitation, forced labor, forced marriage, slavery, and other forms of human trafficking (UN WOMEN, 2020). People of all genders can be trafficked, however, women and girls comprise the majority of those affected, most often for sexual exploitation. Women and girls belonging to marginalized social groups are most at risk, reflecting how systems of patriarchal power and intersecting oppressions lie at the root of human trafficking (Caretta, 2015; Hu, 2019; Lockyer, 2020).
The long-term harms of human trafficking on survivors’ physical, psychological and social well-being are profound (Doherty & Morley, 2016; Zimmerman et al., 2008). Mental health consequences include post-traumatic stress disorder (PTSD), anxiety disorders, depression, low self-esteem, flashbacks, suicidality, and dissociation (Oram et al., 2012; Zimmerman et al., 2008). Other serious consequences include social stigma and isolation, loss of income, housing and food insecurity, abusive relationships, and physical health problems (Cordisco Tsai, 2017; Doherty & Morley, 2016; Gerassi & Nichols, 2017).
More recent literature suggests that the COVID-19 pandemic amplifies both the risk factors for human trafficking and its harmful sequela. Due to COVID-19 public health measures, survivors may experience barriers in accessing healthcare services, increased financial stress/loss of livelihoods, disruptions in support programs, grief over the loss of loved ones, among other challenges—compromising their health and wellbeing and increasing the risk of re-victimization (Todres & Diaz, 2021). Moreover, transportation restrictions, quarantines, and lockdowns can trigger memories of traumatic events experienced while being trafficked, such as forced isolation and a lack of basic mobility (Cordisco Tsai et al., 2021; Jimenez et al., 2021).
While a growing literature base documents the risk factors and consequences of human trafficking, far less is understood regarding how best to promote survivors’ resilience and holistic healing (Doherty & Morley, 2016; Salami et al., 2018). This is especially true within shelter programs where quality of services varies considerably. Initially, shelters operated mainly as emergency establishments to house individuals “rescued” by law enforcement during raids in areas suspected of human trafficking. However, the use of shelters has broadened substantially with both private non-governmental and public facilities operating around the world. These range from temporary shelters to provide basic needs, long term residential facilities, foster care programs, transitional housing, and more. Currently the majority of services for survivors of human trafficking are centralized within shelter programs, despite concerns around consistency, insufficient freedom in care, limited client participation, and more (Cordisco Tsai et al., 2022; Hacker et al., 2015).
Moreover few studies have examined the necessary conditions for healthy and successful transitions out of shelter care, a critical process that can be challenging for numerous reasons. For example, trafficking-specific shelters are often isolated from the community, meaning that survivors may live for years inside shelters without programs that promote social inclusion, facilitate ongoing connection with their families, or support in preparing for life outside the shelter (Dutta, 2017; Cordisco Tsai et al., 2020). Survivors have reported several difficulties when leaving shelter settings—especially when transitions are abrupt—including a sense of abandonment, conflict and disconnection when reconnecting with family, a lack of social support, difficulties completing education or obtaining employment, and experiences of violence and stigmatization in the community (Cordisco Tsai et al., 2020; Dutta, 2017).
Survivors, practitioners, and researchers have all expressed the need to strengthen community reintegration services and better promote the conditions for sustained well-being at this critical junction in a survivor's post-trafficking experience (Dutta, 2016; Surtees, 2017). Yet designing stronger programs requires learning from those most affected, and few studies have provided space for survivors’ to narrate and draw meaning from their experiences (Le, 2016; Lockyer, 2020). In particular, the process of community reintegration after leaving residential shelter care—where many survivors live and receive support for an extended period—has received scant attention in the literature (Cordisco Tsai et al., 2020).
The current study aims to address these gaps by exploring the lived experiences of six women as they unexpectedly transitioned from a trafficking-specific shelter to their home communities due to the COVID-19 lockdown in Uganda. We focus on how these women interpreted and drew meaning from their rapidly changing life circumstances—leaving the shelter, adjusting back to the community setting, and simultaneously navigating the uncertainties of a global pandemic. By centering their personal stories of struggle and resilience, we hope to elevate survivors’ own accounts and draw on their insights to identify considerations for future programming. Further, survivors’ reflections on the nuanced ways in which their experiences intersect with the broader context of COVID-19 deepens understanding of the potential exacerbating effects of the pandemic for women already experiencing systemic vulnerabilities.
Context and Methods
Human Trafficking in Uganda
Peer-reviewed literature on human trafficking in Uganda remains limited, with no official estimation of prevalence. However, existing reports indicate that Uganda is both a source and a destination country for human trafficking and that traffickers exploit both domestic and foreign victims. Ugandan government records indicate that 666 victims were identified in 2020, of which 575 were female, 497 were transnational, and 222 were children (U.S. Department of State, 2021). The government prosecuted 202 cases in 2020, the majority (140) were sex trafficking cases, 54 were labor cases, and 8 were unknown (U.S. Department of State, 2021).
As is common around the world, the true number of trafficking cases and individuals affected is likely much higher than official counts. For example, in Northern Uganda, the Lord's Resistance Army (LRA), a rebel group that fought the government for over 20 years, abducted at least 20,000 boys and girls and recruited them into its ranks as sex slaves, cooks, combatants, and other forms of servitude (Kelly et al., 2016). According to the U.S. Department of State (2021), children and young women from economically vulnerable families are most at risk, with reports of children as young as seven exploited for forced labor in agriculture, fishing, forestry, cattle herding, mining, stone quarrying, brick making, carpentry, steel manufacturing, street vending, bars, restaurants, and domestic service. Most recently, the increase in labor externalization of domestic workers from Uganda to Asia and Middle Eastern countries in search of economic opportunities has increased women and girls’ risk of being exploited into forced labor, sex trafficking, and organ harvesting (U.S. Department of State, 2021). Other structural drivers are prevalent, including patriarchal gender norms that condone violence against women in some circumstances and reinforce male sexual entitlement (Namy et al., 2017).
While the government of Uganda has made recent strides in recognizing and addressing human trafficking as a major human rights issue in the country, mental health services for survivors remain limited (Namy et al., 2021). Non-governmental organizations (NGOs) provide the vast majority of available services to trafficking victims and survivors, including shelter care, psycho-social counseling, medical treatment, family tracing, resettlement support, and vocational training (U.S. Department of State, 2017).
Setting and Participants
All participants in this study were residing in a female-only residential shelter in Kampala, Uganda prior to the COVID-19 pandemic. The shelter is run by an anti-trafficking NGO that provides holistic programming for survivors of human trafficking, predominantly for sexual exploitation. Services at the shelter include psycho-social support, individual case management, education/vocational training, and extracurricular activities.1 The shelter allows women who were living with children at the time of enrollment into their programs, or gives birth after already being enrolled, to bring/keep their children with them at the shelter. If children were living with another guardian at the time of enrollment, the NGO provides medical and educational support for the children, but does not encourage women to bring their children to the shelter.
On March 25, 2020, Uganda entered a nation-wide lockdown immediately following the country's first confirmed case of COVID-19. The mitigation measures were swift and widespread, including a ban on the use of all public and private transport, the closure of all non-food businesses, and a national curfew. The country maintained one of the most far reaching policy responses globally, extending the lockdown several times and not fully opening the economy for nearly 2 years. Recognizing the uncertainty of the situation, the NGO gave all clients 18 years or older the option to return to their home communities before “locking down” the shelter. The six women in this study opted to leave the shelter. While the transition home was rapid and unexpected (prompted by the lockdown), the NGO provided financial support for transport home, and ongoing services such as remote case management support and an emergency cash transfer of 120,000 UGX/month (approximately 35 USD equivalent). The shelter staff continued to follow up with the six participants and they were enrolled into the NGO's Community-based Care program for 2 years, where they continued to receive psycho-social services (counseling), training in income generating activities and were provided with start-up capital.
At the time of the interviews, the women in this study were all living in their home communities in Uganda: two in rural settings and four in peri-urban settings (peri-urban areas are typically understood as rapidly evolving transitional zones situated between cities and rural areas). The socio-demographic profile of the six women is presented in Table 1, including a summary of their trauma histories (these data were collected in February 2020, just before COVID-19 was declared a global pandemic). As is common with many survivors of human trafficking, most of the women in our study have experienced multiple traumatic events, including rape, torture, lack of shelter, and witnessing violent death. Broadly speaking, in comparison to NGO clients who continued to reside in the shelter, the women who opted to leave were older, more likely to be in a relationship, and more likely to have children (Carlson et al., under review). The desire to reunite with family members during undertain times, especially young children who had previously been living with a guardian, was the most common reason shared for going home.
Table 1. Participant Profile.
Socio-demographic characteristics n %
Age
20–24 years 3 50
25–29 years 3 50
Partner status
Single 3 50
In a partnership 2 33
Married 0 0
Widowed 1 17
Number of children
0 2 33
1 3 50
2 or more 1 17
Educational attainment
Primary or less 3 50
Secondary or tertiary 3 50
Duration of time living at the shelter
Less than 3 months 0 0
3 to 6 months 2 33
More than 6 months 4 67
Trauma history
Ever experienced sexual violence 4 67
Ever experienced torture 1 17
Ever witnessed a violent death 4 67
Ever experienced lack of shelter 4 67
Data Collection
The data analyzed for this study is part of a larger research project evaluating Move with HaRT, a mind-body mental health intervention to promote healing and resilience for women and girls who have experienced trafficking and other forms of violence (Namy et al., 2021). The larger study enrolled 20 participants (14 of whom continued to reside in the shelter during the study) and involved six rounds of survey data and two rounds of in-depth interviews (IDIs). To address the research focus of this manuscript, we include only one round of IDIs with the six women who left the shelter due to the COVID-19 pandemic.
The interviews (n = 6) were carried out in June 2020 using mobile phones, since all non-essential travel was prohibited during the national lockdown (see Namakula et al., 2021 for specific steps taken to maintain a trauma-informed approach using mobile phones). Each IDI lasted for approximately 1 hour and was conducted in either English or Luganda (depending on the participant's preference). The June 2020 interviews explored the following topics: the underlying motivations for leaving the shelter, participants’ experiences in their home communities during lockdown, and perceived changes in physical, emotional, and social wellbeing since the start of the COVID-19 pandemic. Findings draw on the interview data as well as observations captured during daily debriefs (for example, interviewers commented on any perceived changes in participants’ wellbeing and safety since their previous interaction).
Approval for the overall research project was obtained from University of Alabama (19-10-2964) and the Uganda National Council of Science and Technology (SS420ES), with an approved amendment to use remote-based data collection after the start of the lockdown. Interviews were conducted by two research team members (SN and AGN) from Uganda, both with extensive prior experience facilitating interviews on sensitive topics such as violence, HIV, and mental health. Several steps were taken to enhance emotional and physical safety throughout; interviewers established rapport, listened to nonverbal cues, and proactively checked-in on participants’ wellbeing during interviews. Daily debriefs with the research team allowed for troubleshooting emergent issues and processing emotional responses of interviewers—an important step for minimizing the risk of secondary trauma (see Billing et al., 2022 for further details of steps taken to promote researcher wellbeing in this study). All participants were offered a voluntary referral to their case manager and mandatory referrals were made in the event of suicide risk.
Data Analysis
Audio recordings of the interviewers were transcribed verbatim and simultaneously translated into English (as needed) by the researcher who facilitated the interview. We used an interpretative phenomenological approach (IPA), characterized as “participant-oriented” and frequently used to “investigate and interpret the ‘lived experiences’ of people who have experienced similar (common) phenomenon” (Alase, 2017, p. 11). Consistent with IPA, the interview guides posed general questions that provided space for individuals to interpret and draw meaning from their own experiences. For example, the guide included questions that explored any perceived changes in physical, emotional, and social wellbeing since transitioning out of the shelter (which coincided with the COVID-19 lockdown).
The analytic process was iterative, and utilized several common strategies for IPA (Smith & Shinebourne, 2012) including (1) Codebook development: an initial codebook was created based on observations and preliminary themes identified during team debriefs. Subsequently, two co-authors immersed in the data made revisions and integrated inductive codes; (2) Coding: line-by-line coding in Atlas.ti (Version 8 Windows), with double coding for the first three transcripts to enhance consistent interpretation of the final codebook; (3) Analysis outputs: To help identify and refine themes within and across participants, we created a “participant wellbeing matrix” that captured key dimensions of physical, emotional, and social wellbeing for each woman in the study. In addition, six “code summaries” were developed that included detailed commentary of extracted data for all conceptually similar codes; (4) Final theme identification: All analysis outputs were reviewed and jointly discussed by the research team (CC, SN, SN, AGN) to select themes that best reflected the commonalities and unique aspects related to the transition out of shelter care during the COVID-19 lockdown, and participant-identified dimensions of physical, psychological and social wellbeing.
Findings
Overall, participants shared numerous challenges to their emotional, physical, social, and financial wellbeing as a result of their abrupt transition from the shelters to community-based living. They also described how their trafficking and trauma histories were intensified by this transition during the pandemic, dramatically altering their lives and future aspirations. Our analysis identified four core themes related to how these women described and interpreted their rapidly changing life circumstances, each described separately below. Note that all six women have been assigned a pseudonym to protect confidentiality.
Economic Insecurities as a Cross-Cutting Hardship
All six women acknowledged having experienced profound economic hardship after transitioning home, often discussing their financial situation several times in the same interview and linking this to feeling a lack of safety and wellbeing. Even with a monthly cash transfer provided by the NGO, many still struggled to meet basic daily needs—a stark contrast from life at the shelter. As Barbra explained, “they [NGO staff] cared about my life and were mindful of everything that concerned my life, whether I have eaten food, whether I am feeling well—and they would try everything possible to make sure that I am stress free.” Later in the interview, she speculated on whether she would have survived during the pandemic without this financial support:My life would have been so bad. I cannot lie to you. If it was not for [the NGO] our lives would have been so miserable. The money we are getting is used to cater for the basic needs not luxuries, so I do not even know how we would have survived. At this moment you cannot get support from people, everyone is crying and talking about COVID, so then who would have come to our rescue? (Barbra, 26 years).
The cash transfer also enabled several of the women to provide for their household during the lockdown, some having to do this for the first time in their adult lives. For some women who had been experiencing the heavy weight of economic expectations given their “time abroad,”2 they interpreted their newfound ability to be a source of stability and material support in their household as meaningful and validating:You might even laugh at me. The brightest part [since being home] is I am so happy, only that I cannot go out and say it out loud. I had never owned 100,000 UGX [approximately $28 USD] in my life! By the time I went out of the country, the maximum that I had owned was 50,000 UGX [approximately $14 USD]. I left [Uganda] hoping that I was also going to hold some good money, but that never happened. (Barbra, 26 years)
For others, the realities of multiple family members and limited (or nonexistent) options for paid work meant that the cash transfer was simply not enough to secure basic needs for their household, and sometimes personal necessities, including healthcare, were overlooked. Sarah put it succinctly, “I have never been fine [since leaving the shelter]. I have been in the hospital, if not this week then the next week. I am sick and the problem was money.” Similarly, when asked about the “hardest part” of the lockdown so far, Amira recounted:The hardest part was sickness. The time when I fell sick, yet I did not have proper treatment… I have ulcers, so this time I fell sick and I did not have enough money to go for treatment, during that time [the NGO] had not yet started giving us some money, so it was so challenging for me. And during that time my daughter also fell sick and she also needed treatment. (Amira, 26 years)
Some women attributed financial hardships to their emotional state. For example, Grace explained how she felt food insecurity compromised her fundamental sense of safety:Overall, I haven‘t been very safe. I have had a lot of challenges in my daily life. What I explained to you earlier, you can‘t be safe if you don‘t have food, you can‘t be safe if your health is not good, because all that has not being going on well for me. […] The biggest challenge I have found, is that you may want something or to do something, but you cannot get it or do it. For example food and healthcare, and where you stay … the situation has been bad, and still is. (Grace, 22 years)
Later in the same interview she circled back to her financial situation and pondered how it had shaped (and constrained) her sense of self and aspirations for the future:I am filled with sadness and a lot of pain in my heart. I can‘t do anything now because I have no money. And I have no one to help me now. I reflect about the reason why I had gone abroad which was to come back and start up life, but I instead got problems. I sit down time after time and wonder what I can do now. I can‘t do anything. If I had some money, I would start working and be able to better my life without any hardships, that‘s what I keep asking myself. That is it. (Grace, 22 years)
Intensification of Emotional and Physical Symptoms
Pronounced emotional and physical distress emerged as another shared experience, which participants frequently linked to COVID as a direct cause or an exacerbating factor. For instance, when asked to reflect on their emotional wellbeing and what they thought was most affecting it, several participants emphasized how the pandemic had “disorganized” their plans and “shattered” their aspirations that had been taking shape at the shelter. A sense of having “failed” and feeling powerlessness to move forward because of COVID was pervasive across the interviews. As observed in the interview with Hope, this experience was at times interpreted as a sharp departure from the sense of relative wellbeing she had previously experienced in the shelter:My life was good, I had started feeling joy. We had hopes and each of us had programs, which were going well [before COVID] … but ever since that disease broke out, things changed so much, and we had to change priorities. So, I am here as if I am stuck. I am like a failure. It‘s even hard for me to explain it more. … All that I was thinking about was shattered. I no longer do the things I used to. I had different plans [before COVID] and hoped that by now I would be at my own pace, setting up my own things, but everything got stuck. That is the challenge I have right now. (Hope, 29 years)
Christine shared that the most pronounced hardship was being unproductive. For her, this was associated with a loss of agency—feeling that she was “just seated like a baby”:The hardest part of it [lockdown], I would say, to me is that it has put me down. Put me down. I expected another thing, now I am down. And also sometimes, you know for me, I am not used to sitting without doing anything. But it has made me to be bored which is not good for me … It is not good, I made a loss for the year. It means I did a loss, because I was unable to do the things that I was supposed to do. I am just seated like a baby as if I have nothing. (Christine, 20 years)
Barbra's analysis was unique, in that she situated the desperation she was experiencing in a global perspective. To her, it was the scope of the crisis and magnitude of collective suffering that eclipsed the personal hardships she had previously been able to endure:At my age, I had never witnesses trying events that befall the entire country. I would say that in my life COVID is the first pandemic that has affected the entire world. We have been used to personal problems like being hungry, suffering over this and that but life would still continue. So we would plan and have dreams amidst these challenges, but then suddenly something happens that affects the entire world and disorganizes all your life dreams. (Barbra, 26 years)
Only one of the women explicitly mentioned being afraid of contracting COVID-19. However, nearly everyone suffered a continuation (or intensification) of physical health conditions that had begun before the lockdown. Amira suffered from ulcers—commonly associated with stress—and had to discontinue treatment after leaving the shelter when she was unable to access care. Sarah had been raped, and could not sleep because of chronic worry about her HIV status and a persistent leg injury also sustained during her trafficking experience. She also described generalized weakness and body pains—symptoms experienced by several other women in the study:Okay like now for me, I am thinking I have not yet known my HIV status, what if I am HIV positive, how will I appear? Regarding that one, I think a lot. Because I was just raped from abroad and, when they made the first test, they never told me the results. And the second thing, me, I feel weak. I don‘t work, I don‘t do anything. I just stay inside the house. I feel general body weakness and back pain. I also think, ‘when will my leg be healed’? That one I also think about it. (Amira, 26 years)
Similar to Amira, Grace drew explicit connections between her physical health symptoms, emotional difficulties and past traumas. She had previously been hospitalized for fainting while at the shelter, and overall was in poor physical health. During the interview she speculated on how her experiences abroad—and inability to share them with others—was causing physical pain (headaches) as well as emotional distress and other symptoms commonly used to describe depression in the Ugandan context (“lots of thoughts,” “hating the world,” etc.):Sometimes I would have nothing giving me headache, but I had a lot of thoughts and was uncomfortable where I was [my state of mind]. I can say it this way—inside, you feel afraid. You fear, but I couldn‘t give a clear explanation of what was bothering me to myself, or even to anyone who asked to know. Other times, I would get things that bother me in my heart, and I feel like I hate this world. And I would detest it so much that I ask, God why? But I would never understand it at all. Sometimes I try to avoid thinking about things that break my heart, which hurt me or bother me … You keep it at heart. There is no one to tell … Sometimes, you leave everything to yourself and say, God knows. When you can, cry. You cry alone without anyone noticing. You decide to cry to get off what is on the heart. (Grace, 22 years)
Social Disruptions
The women described how leaving the shelter during the lockdown caused both positive and negative social interactions, ranging from the elation of reconnecting with a child, to outright stigma in other situations. Many of the women had enjoyed close relationships with the staff and their housemates at the shelter, using familial language (“auntie,” “mother,” “sister,” etc.) and referring to the constant companionship as an important source of support. The abrupt rupture of this social network was yet another important change that accompanied the transition home, triggering feelings of loss and loneliness for several women. For instance, Hope identified her emerging friendships at the shelter as a source of healing prior to the lockdown, and surmised that her current isolation caused her to “back slide:”The truth is, I had started feeling well [before COVID] because I had started getting friends in the places I used to visit during our training, and learning from people‘s experiences. I was feeling good, however COVID had negatively changed everything. It has made me back slide [so much] that I even lost connection with some constructive people I had met. (Hope, 29 years)
During the interviews, all the women were asked about their decision to leave the shelter. Five participants explained that the primary motivation was to reunite with their young children or family members after the sustained separation while abroad and living at the shelter. Anxiety over these familial separations was also driven by uncertainties around COVID-19. For instance, Barbra recalled being afraid that she would receive a phone call at the shelter that her daughter had contracted the virus and died. Once back in their home communities, several participants experienced a positive (re)connection with family members. When asked if there was anything that had helped them to feel better, two participants--Christine and Amira—both emphasized the joy of living in the same household with their young daughters:I am now living with my daughter which I have always wanted, so I take on any opportunity to be with her. I feel safe and happy when I am with her. She is also happy when I am around. She is happy that mama is staying at home. She has changed, she has grown and now recognizes me as her mama. (Christine, 20 years)
I enjoy having time with my baby, like when I am done with my work I sit and relax and have time with my baby and we play, because whenever I am busy with work I don‘t have time for her. So after work, that is the time we have together to play and laugh. During that time I feel so good, just like any other parent. (Amira, 26 years)
These positive social interactions were juxtaposed with several examples of interpersonal conflict, often linked to circumstances surrounding their trafficking experiences and the patriarchal general norms existing in the community. Hope's situation was particularly challenging. She felt her family no longer cared for her and recounted how they often mocked her for not having money despite having gone abroad. She also described strong distrust following an incident involving her children. When Hope confronted a male family member about this, he threatened her directly with violence:Relatives can even sell off your things or property, so I don‘t trust them completely. So with trusting I have limits, I may trust some people or even not trust others because some people you trust may end up hurting you, like the ones you trusted before, so I have a fear. When I was abroad, I was working and would buy things for my children, but when I came back, and now that I was back in the village, I would find even the people staying with my children using the things for themselves not for my children … I asked him [male family member] that ‘why would you do this to my things, yet children are even sleeping on the floor.?’ … He added that ‘I am freely advising you to stop saying to me words about that. If you ever talk to me about these items, I would beat you up and dump you in a swamp.’ (Hope, 29 years)
Grace spoke in detail about the joy of bonding with her during the lockdown, who she had been missing. However, she also described a profound distrust of others, and felt that her inability to connect was linked to the inexplicable anger she had experienced since the pandemic:I usually get a lot of challenges, because these days I have a lot of anger ever since COVID started. Since we left the shelter and came back home, I have a lot of anger these days. I don‘t even know where it comes from. If someone does anything to me, I get so angry, though I may not show it. I feel so so angry at that person. That‘s my problem. (Grace, 22 years)
Outright hostility from the community was not uncommon—and most of the women related at least one example of community accusations or stigma linked to being trafficked. Sarah described these hardships in detail, as well as the emotional toll she experienced as a result:I have been stressed with other people, I feel uncomfortable. Some other people look at me, they know I have been in [name of shelter] and even others never knew that I went abroad, now they are just coming to know. I would become uncomfortable. Others say that ‘she went abroad and came back with nothing.’ Like things are being spoken by my family members, they share with other people and they start laughing at me … Sometimes when I am inside the house and I see people look at me and they laugh, and even talk about me, I fear … I feel so small and uncomfortable. I feel bad about myself when they talk about me, even if it is not true. (Sarah, 25 years)
Sources of Hope and Resilience
Despite these extremely difficult circumstances, all participants demonstrated their capacity for resilience and hope at some point during the interviews. Sarah used a portion of her cash transfer to buy a goat, and expressed a sense of pride in her investment and its potential to secure future income. Barbra found work planting crops with her mother, and reflected on how this helped her to keep busy and gave her a sense of purpose. And several other participants described personal practices that helped them better navigate COVID-related uncertainties, such as prayer, cooking, “quiet time” and positive self-talk about the future. Sarah described keeping a hopeful outlook largely rooted in her faith:I am a gift of God. I managed to come back [home], and up to now, I am still alive, which means I have a future ahead and there is something God has planned for me. It is just that the time has not yet reached, but there is something for me. That is what I always tell my mind. I think about the disease, I think about people working but me I am not, people getting money, but I don‘t. But I tell myself God gave me a chance to meet an organization [the NGO]. I don‘t know what they are planning, but I see there is future for me. That is what I always tell my mind when I am thinking. (Sarah, 25 years)
When asked to consider how they were able to sustain hope in spite of such trying circumstances, nearly all the women alluded to the relevance of emotional support, either from family members, a trusted friend, their NGO case manager (who provided remote services during the lockdown) as well as the researchers in this study. The ability to be validated and seen by someone—without judgment—emerged as particularly salient, as noted by Sarah (referring to her case manager) and Hope (referring to a member of the research team).When I have a problem and I share it, like, with a friend that I trust--not family, I get some confidence, I cool down. Like last week I was totally down. I contacted my case manager she called me back, but I was just crying on the phone, explaining what happened. But before the call ended, I was somehow better. I like sharing my problems and or secrets with other people. (Sarah, 25 years)
What has made me feel good, it‘s you people [research team]. Whenever I hear from you I feel hopeful. I didn‘t know that I would ever meet such people. It makes me stronger. When I get your call, I feel energized … If someone can call and ask how I am doing, I get hope. (Hope, 29 years)
Discussion
To our knowledge, this is the first in-depth qualitative study to explore the lived experiences of survivors of human trafficking as they transitioned out of shelter care during the COVID crisis. Overwhelmingly this was a time of immense hardship, magnified by these women's histories of being trafficked and their rapidly changing life circumstances. At the same time, reconnecting with family brought moments of joy, and participants’ stories highlighted compelling examples of courage and resilience.
Consistent with our methodological approach, findings are not intended to be broadly representative of any one group or demographic. Rather our aim is to elevate the experiences of six individual women and how they draw meaning from navigating several pivotal life events simultaneously: transitioning out of shelter care, readjusting to life in the community, and experiencing the COVID-19 pandemic and nation-wide lockdown. By centering women's own interpretation and perspectives, these findings address a gap in survivor-generated knowledge which is essential for informing programs and policies that speak to the lived realities of those most affected.
Overall our findings broadly align with existing literature from various disciplines. First, trauma-focused scholarship indicates that persons with past trauma histories are at heightened risk for experiencing maladaptive reactions to new traumas (Fossion et al., 2015). For survivors of human trafficking in particular, the COVID-19 lockdown may be experienced as a trauma in-and-of-itself (Lee et al., 2020), given similarities between quarantine/forced isolation and the experience of being trafficked. The profound suffering while in lockdown described by the women in our study supports this interpretation. Further, several studies have described how trauma—and sexual violence in particular—can become “embodied,” causing sustained impacts on a person's physical body, physiology, and associated emotional states (Fields et al., 2020; Herman, 1992; Van der Kolk, 2014). In our research, participants’ descriptions of their overall wellbeing aligns with this understanding, particularly their own analysis of the interconnections between physical ailments and emotional distress linked to their trafficking experiences.
Secondly, feminist literature frequently emphasizes the salience of women's economic independence to bolster various aspects of their agency and autonomy (Kabeer, 2008) —for example, to exit abusive relationships, claim decision making power in the household and resist other inequitable gender norms. Research with survivors of human trafficking has similarly emphasized the importance of sustainable employment for self-worth and belonging as a critical component of recovery and protection against revictimization (Cordisco Tsai, 2017; Hacker et al., 2015). For instance, a qualitative study with survivors of human trafficking in Nepal found that women who returned to their home communities without money and were unable to pay off debts experienced social stigma and rejection from family members (Simkhada, 2008)—a finding that resonates with the experiences shared by some of the participants in our study.
Finally, while relatively few studies focus on the transition out of shelter care, available literature supports our findings, including the likelihood of experiencing community accusations, strained relationships with family members after extended separations, distress over the abrupt loss of direct support provided in the shelter environment, mental health challenges, and difficulty reestablishing a sense of purpose and meaning (Cordisco Tsai et al., 2020; Le, 2016;Surtees, 2017).
Our study makes an additional contribution to the current knowledge base by exploring how survivors’ experiences intersected with the COVID-19 crisis in Uganda. Participants reflected on the nuanced ways in which the pandemic exacerbated existing difficulties, and—at times—created new challenges, for instance, intensifying physical symptoms given the difficulty in accessing care, contributing to a feeling of unproductivity and “disrupting” personal plans for the future. It is also possible that the strict lockdown policy in Uganda circumscribed the potential for positive shifts highlighted in other research, such as a “heightened sense of freedom and self-determination” described by survivors in the Philippines upon leaving the shelter (Cordisco Tsai et al., 2020, p. 12).
Several limitations should be noted. As with all in-depth qualitative research, we do not expect these findings to be generalizable beyond study participants, though their experiences may offer insight into patterns and dynamics relevant to other survivors in similar circumstances. In addition, while the qualitative researchers established strong rapport and had previously engaged with participants in-person (prior to lockdown), the remote nature of these interviews may have limited interpretation and understanding based on non-verbal cues. While most of the interviews were conducted in Luganda, data were transcribed and translated into English for analysis, and some linguistic and cultural nuances may not have been captured.
Taken as a whole, the experiences of the women participating in this research underscore the importance of creating feminist, trauma-informed programs to support reintegration and adjustment to life in the community. Such approaches must acknowledge (and attend to) the structural conditions that too often impede healing for women survivors, including patriarchal norms and behaviors that contribute to a lack of meaningful work opportunities, social stigma, and repeated experiences of systemic trauma(s). These same issues lie at the root of women and girls’ vulnerability to trafficking in the first place (Cameron et al., 2020; Gacinya, 2020). As many activists, scholars and practitioners have argued, the prevention—and gradual elimination—of human trafficking requires decisive and urgent action to dismantle these structural inequalities.
Further, drawing on what we learned from the women in our study, we identify four actionable directions to better support holistic healing and wellbeing for survivors as they transition out of shelter care—an experience that affects nearly all survivors who have engaged with the residential shelter system: Ensure viable livelihood opportunities and expand the definition of safety for survivors. “Exiting” human trafficking is not a one-off event, and once home women will be immersed in the same contexts that made them vulnerable to being trafficked initially (Cordisco Tsai, 2017). It is imperative for survivors to access stable employment, in sectors that are prioritized by women themselves (Richardson et al., 2009), and that economic support given to survivors is coupled with financial training. This can mitigate the risk of re-trafficking and, as discussed by the women in our research, the ability to provide for their families is intricately bound to a sense of purpose and hope for securing a better future.
Create structures for survivors to maintain core social connections including with family (while living at the shelter) and with staff (after transitioning home). Continuity in positive relationships emerged as a source of resilience in our study—whereas, for some women, the experience of family betrayals and/or outright conflict was re-traumatizing. Other anti-trafficking researchers and activists have argued for the importance of avoiding abrupt social disruptions during community reintegration, particularly in light of the profound breach of trust many survivors experience when trafficked (Cordisco Tsai et al., 2020; Le, 2016; Surtees, 2017).
Address and mitigate challenging social dynamics likely to emerge, including financial expectations (e.g., becoming “rich” since living abroad), accusations, and/or stigma directed at survivors. As others have argued, doing so may require that shelters conduct outreach (and/or counseling) to family members and community leaders, and facilitate collaboration between families and clients in planning for their transition home (Cordisco Tsai et al., 2020). In addition, programming could consider creative ways to prepare women and equip them with practical skills to better navigate these difficulties (e.g., role playing scenarios, discussion groups, etc.), as well as a plan for crisis management should the situation become overwhelming.
Integrate specific programming to promote physical and emotional healing, with a particular focus on the “embodied imprint” of systemic trauma. Increasing evidence suggests that yoga and other contemplative activities can address somatic and mental health symptoms following sexual violence (Fields et al., 2020) and initiatives such as the Feminist Republic (Urgent Action Fund-Africa, 2021) explicitly links somatic and movement-based practices to their conceptualization of healing justice for marginalized communities. Providing such activities within the shelter setting could help establish a stronger foundation for sustained resilience as survivors transition out of shelter care.
To support long-term wellbeing for survivors, anti-trafficking programming must move beyond reducing immediate harm to addressing the structural impediments to healing and promoting women's agency, social support, and resilience (Clawson et al., 2007). Developing such programs requires exploring, listening to, and prioritizing the experiences of survivors—particularly during pivotal moments such as the transition out of shelter care back to life in the community.
Author Biographies
Sophie Namy is a feminist practitioner and researcher who works to prevent violence against women and support holistic healing for survivors.
Sylvia Namakula is a social behavioral researcher working in the areas of mental health, human trafficking, violence against women, HIV/AIDS, and reproductive health.
Agnes Grace Nabachwa is a researcher and practitioner focused on promoting women's rights and preventing all forms of violence against women and children.
Madeleine Ollerhead is a gender and conflict expert with varied experience working in research and influencing for feminist non-profit organisations.
Laura Cordisco Tsai is a social work scholar and practitioner in the areas of human trafficking and gender based violence, specializing in the development of empowering and trauma-informed services for survivors.
Jean Kemitare is a feminist activist who co-created the GBV Prevention Network and has authored a range of methodologies to strengthen individual and organizational capacity to address violence against women.
Kelly Bolton is currently pursuing her Ph.D. Candidate with a focus on child welfare, trauma and neurobiology, and trauma-informed mental health treatment.
Violet Nkwanzi is a social worker and Ph.D. Candidate with a focus on mental health of survivors of human trafficking and young people in resource-constrained settings. Catherine Carlson is a social work researcher who focuses on the intersection of violence and mental health, and the implementation of evidence-based interventions in low-resource settings
Author's Note: We express our immense gratitude to the women who participated in this research for their courage and sharing their experiences. We also thank the NGO for supporting this study and providing referral and case management support.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Urgent Action Fund- Africa, University of Alabama, (Small Grants Program).
ORCID iDs: Kelly Bolton https://orcid.org/0000-0003-3543-8339
Violet Nkwanzi https://orcid.org/0000-0002-4508-8822
1. Further details of the NGO and shelter program are intentionally omitted to protect participants’ confidentiality and safety.
2. Most of the women in this research were trafficked outside Uganda and frequently used the term “abroad” to refer to their trafficking experiences.
==== Refs
References
Alase A. (2017). The interpretative phenomenological analysis (IPA): A guide to a good qualitative research approach. International Journal of Education and Literacy Studies, 5 (2 ), 9–19. 10.7575/aiac.ijels.v.5n.2p.9
Billing L. Carlson C. Namakula S. Namy S. Nabachwa A. Gevers A. (2022). Researching with HaRT: promoting researcher wellbeing through self and collective care. Sexual Violence Research Initiative.
Cameron E. C. Cunningham F. J. Hemingway S. L. Tschida S. L. Jacquin K. M. (2020). Indicators of gender inequality and violence against women predict number of reported human trafficking legal cases across countries. Journal of Human Trafficking, 1–15. 10.1080/23322705.2020.1852000 32190715
Caretta M. A. (2015). Casa rut: A multilevel analysis of a “good practice” in the social assistance of sexually trafficked Nigerian women. Affilia: Journal of Women and Social Work, 30 (4 ), 546–559. 10.1177/0886109915572846
Clawson H. J. Salomon A. Goldbatt Grace L. (2007). Treating the hidden wounds: Trauma treatment and mental health recovery for victims of human trafficking. U.S. Department of Health and Human Services.
Cordisco Tsai L. (2017). Family financial roles assumed by sex trafficking survivors upon community re-entry: Findings from a financial diaries study in the Philippines. Journal of Human Behavior in the Social Environment, 27 (4 ), 334–345. 10.1080/10911359.2017.1288193
Cordisco Tsai L. Eleccion J. Panda A. (2021). Impact of the COVID-19 pandemic on survivors of human trafficking in the Philippines. Journal of Modern Slavery, 6 (2 ), 231–246. 10.22150/jms/MIGO6786
Cordisco Tsai L. Lim V. Nhanh C. (2020). I feel like we are people who have never known each other before: The experiences of survivors of human trafficking and sexual exploitation transitioning from shelters to life in the community. Forum: Qualitative Social Research, 21 (1 ), Article 16. 10.17169/fqs-21.1.3259
Cordisco Tsai L. Lim V. Nhanh C. (2022). Shelter-based services for survivors of human trafficking in Cambodia: Experiences and perspectives of survivors. Qualitative Social Work, 21 (3 ), 523–541. 10.1177/14733250211010901
Doherty S. Morley R. (2016). Promoting psychological recovery in victims of human trafficking. In Malloch M. Rigby P. (Eds.), Human trafficking: The complexities of exploitation (pp. 121–135). Edinburgh University Press.
Dutta S. (2016). Institutional care in India: Investigating processes for social reintegration. Children and Youth Services Review, 66 , 144–153. 10.1016/j.childyouth.2016.05.010
Dutta S. (2017). Experiences of young Indian girls transiting out of residential care homes. Asian Social Work and Policy Review, 11 , 16–29. 10.1111/aswp.12107
Fields A. Namy S. Dartnall E. (2020). SVRI Knowledge Exchange: Body-focused mental health approaches with survivors of sexual violence. Sexual Violence Research Initiative.
Fossion P. Leys C. Kempenaers C. Braun S. Verbanck P. Linkowski P. (2015). Beware of multiple traumas in PTSD assessment: The role of reactivation mechanism in intrusive and hyper-arousal symptoms. Aging & Mental Health, 19 (3 ), 258–263. 10.1080/13607863.2014.924901 24927132
Gacinya J. (2020). “Gender inequality as the determinant of human trafficking in Rwanda”. Sexuality, Gender & Policy, 3 (1 ), 70–84. 10.1002/sgp2.12018
Gerassi L. B. Nichols A. J. (2017). Sex trafficking and commercial sexual exploitation: Prevention, advocacy, and trauma-informed practice. Springer Publishing Company.
Hacker D. Levine-Fraiman Y. Halili I. (2015). Ungendering and regendering shelters for survivors of human trafficking. Social Inclusion, 3 (1 ), 35–51. 10.17645/si.v3i1.173
Herman J. (1992). Trauma and recovery. Basic Books.
Hu R. (2019). Examining social service providers’ representation of trafficking victims: A. Feminist postcolonial lens. Affilia: Journal of Women and Social Work, 34 (4 ), 421–438. 10.1177/0886109919868832
Jimenez E. Bravo-Balsa L. Brotherton V. Dang M. Gardner A. Gul M. Lucas B. Such L. Wright N. (2021). Risks and Impacts of COVID-19 for Modern Slavery Survivors in the UK and the USA: Rapid Evidence Review. Rights Lab.
Kabeer N. (2008). Paid work, women‘s empowerment and gender justice: critical pathways of social change. Pathways of Women’s Empowerment working paper 3. Institute of Development Studies.
Kelly J. T. Branham L. Decker M. R. (2016). Abducted children and youth in Lord’s resistance army in Northeastern Democratic Republic of the Congo (DRC): Mechanisms of indoctrination and control. Conflict and Health, 10 (1 ), 1–11. 10.1186/s13031-016-0078-5 26865857
Le P. D. (2016). Reconstructing a sense of self: Trauma and coping among returned women survivors of human trafficking in Vietnam. Qualitative Health Research, 27 (4 ), 509–519. 10.1177/1049732316646157 27206456
Lee S. A. Mathis A. A. Jobe M. C. Pappalardo E. A. (2020). Clinically significant fear and anxiety of COVID-19: A psychometric examination of the coronavirus anxiety scale. Psychiatry Research, 290 (113112 ), 1–7. 10.1016/j.psychres.2020.113112
Lockyer S. (2020). Beyond inclusion: Survivor-leader voice in anti-human trafficking organizations. Journal of Human Trafficking, 8 (2 ), 135–156. 10.1080/23322705.2020.1756122
Namakula S. Nabachwa A. Namy S. Carlson C. Nkwanzi V. Ng L. Galaway K. (2021). Trauma-informed phone interviews: Learning from the COVID-19 quarantine. Best Practices in Mental Health, 17 (1 ), 18–27.
Namy S. Carlson C. Morgan K. Nkwanzi V. Neese J. (2021). Healing and resilience after trauma (HaRT) yoga: Programming with survivors of human trafficking in Uganda. Journal of Social Work Practice, 36 (1 ), 87–100. 10.1080/02650533.2021.1934819
Namy S. Carlson C. O’Hara K. Nakuti J. Bukuluki P. Lwanyaaga J. Namakula S. Nanyunja B. Wainberg M. L. Naker D. Michau L. (2017). Towards a feminist understanding of intersecting violence against women and children in the family. Social Science & Medicine, 184 , 40–48. 10.1016/j.socscimed.2017.04.042 28501019
Oram S. Stöckl H. Busza J. Howard L. M. Zimmerman C. (2012). Prevalence and risk of violence and the physical, mental, and sexual health problems associated with human trafficking: Systematic review. PLoS Medicine, 9 (5 ), e1001224. 10.1371/journal.pmed.1001224 22666182
Richardson D. Poudel M. Laurie N. (2009). Sexual trafficking in Nepal: Constructing citizenship and livelihoods. Gender, Place and Culture, 16 (3 ), 259–278. 10.1080/09663690902836300
Salami T. Gordon M. Coverdale J. Nguyen P. T. (2018). What therapies are favored in the treatment of the psychological sequelae of trauma in human trafficking victims? Journal of Psychiatric Practice, 24 (2 ), 87–96. 10.1097/PRA.0000000000000288 29509178
Simkhada P. (2008). Life histories and survival strategies amongst sexually trafficked girls in Nepal. Children & Society, 22 (3 ), 235–248. 10.1111/j.1099-0860.2008.00154.x
Smith J. A. Shinebourne P. (2012). Interpretative phenomenological analysis. In Cooper H. Camic P. M. Long D. L. Panter A. T. Rindskopf D. Sher K. J. (Eds.), APA Handbook of research methods in psychology, volume 2: Research designs: Quantitative, qualitative, neuropsychological, and biological (pp. 73–82). American Psychological Association.
Surtees R. (2017). Moving on. Family and community reintegration among Indonesian trafficking victims. Nexus Institute.
Todres J. Diaz A. (2021). COVID-19 and human trafficking-the amplified impact on vulnerable populations. JAMA Pediatrics, 175 (2 ), 123–124. 10.1001/jamapediatrics.2020.3610 32955557
United Nations Women (UN Women) (2020). COVID-19 and enabling violence against women and girls. [online: brief]. Available at: https://www.unwomen.org/- /media/headquarters/attachments/sections/library/publications/2020/issue-brief-covid-19- and-ending-violence-against-women-and-girls-en.pdf?la = en&vs = 5006 (Accessed 10 February 2021).
Urgent Action Fund Africa (2021). Fostering Cultures of Care. [online: brief]. Available at: https://www.uaf-africa.org/fostering-cultures-of-care/ (Accessed 5 August 2021).
U. S. Department of State (June 2017). “Trafficking in persons report”, available at: https://www.state.gov/trafficking-in-persons-report/ (accessed 12 March 2021).
U. S. Department of State (June 2021). “Trafficking in persons report”, available at: https://www.state.gov/reports/2021-trafficking-in-persons-report/uganda/ (accessed 3 July 2022).
Van der Kolk B. A. (2014). The body keeps the score: brain, mind and body in the healing of trauma. Viking Press.
Zimmerman C. Hossain M. Yun K. Gajdadziev V. Guzun N. Tchomarova M. Watts C. (2008). The health of trafficked women: A survey of women entering post trafficking services in Europe. American Journal of Public Health, 98 (1 ), 55–59. 10.2105/AJPH.2006.108357 18048781
| 0 | PMC9726634 | NO-CC CODE | 2022-12-08 23:18:17 | no | Affilia. 2022 Dec 4;:08861099221137058 | utf-8 | Affilia | 2,022 | 10.1177/08861099221137058 | oa_other |
==== Front
Qual Soc Work
Qual Soc Work
spqsw
QSW
Qualitative Social Work
1473-3250
1741-3117
SAGE Publications Sage UK: London, England
10.1177_14733250221144050
10.1177/14733250221144050
Main Paper
“Band-Aids on Bullet Holes”: Experiences of pediatric hospital social workers after 1 year of the COVID-19 pandemic
https://orcid.org/0000-0003-0937-8299
McKenna John L
Gender Multispecialty Service, 1862 Boston Children’s Hospital , Boston, MA, USA
Center for Gender Surgery, 1862 Boston Children’s Hospital , Boston, MA, USA
Ross Abigail M
139220 Fordham Graduate School of Social Service , New York, NY, USA
Social Work Department, 1862 Boston Children’s Hospital , Boston, MA, USA
https://orcid.org/0000-0002-4982-1323
Boskey Elizabeth R
Center for Gender Surgery, 1862 Boston Children’s Hospital , Boston, MA, USA
Social Work Department, 1862 Boston Children’s Hospital , Boston, MA, USA
John L McKenna, Gender Multispecialty Service and Center for Gender Surgery, Boston Children’s Hospital, 300 Longwood Ave, Boston, MA 02115-5724, USA. Email: [email protected]
5 12 2022
5 12 2022
14733250221144050© The Author(s) 2022
2022
SAGE Publications
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
COVID-19 has continued to bring devastation to children and families, even 1 year into the pandemic. The rise of the Black Lives Matter movement has also led to renewed attention to systemic racism in the United States and awareness of how the pandemic has further exacerbated health inequities that disproportionately affect communities of color. Pediatric hospital social workers have played a key role since the beginning of the pandemic in responding to the resulting behavioral health crisis and helping to address social disparities. There is a need to understand how the roles and experiences of pediatric social workers have evolved during the first year of the pandemic. In this qualitative study, a series of practice-setting based focus groups were conducted with social workers to capture (a) what has changed or stayed the same since the beginning of COVID-19, (b) thoughts and experiences on diversity, equity, and inclusion with particular attention to race and racism, and (c) perspectives about the long-term implications of COVID-19 on the profession of social work. The Framework Method was used to analyze data, from which six superordinate themes emerged: burnout/coping; the impact of patient acuity; awareness of racial inequity in patient care; awareness of social determinants of health; social worker inclusion in hospital decision-making and policy reform; and grief/racial inequity. An overview of what has stayed the same, what has changed, and what the future may hold for pediatric hospital social workers is reviewed.
pediatrics
hospital social work
COVID-19
healthcare
interprofessional practice
edited-statecorrected-proof
typesetterts10
==== Body
pmcIntroduction
On 11 March 2020, the World Health Organization declared a global pandemic due to the worldwide spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), commonly known as COVID-19 (Liu et al., 2020). Government and healthcare systems across the globe have struggled immensely to contain the spread of the virus (The Lancet Editorial Board, 2020) and, to date, over 80 million cases having been reported in the United States alone (Centers for Disease Control, 2020). One year since the emergence of COVID-19 in the United States has brought about multiple waves of infection, locally prescribed social distancing and self-isolation practices, and changes in safety protocols occurring in response to the rise of new variants (e.g., Reicher and Drury, 2021).
It is well-documented that the COVID-19 pandemic has led to a global mental health crisis for both the general population and front-line health care workers (e.g., Galli et al., 2020). The pandemic has also magnified disparities in health and access to healthcare that particularly impact communities of color (Devakumar et al., 2020). Although not directly related to the pandemic, during this period the murders of George Floyd, Breonna Taylor, among many others, led to the rise of the #BlackLivesMatter movement and brought renewed focus to the problem of police brutality and the pervasiveness of racism as a public health issue that endangers the lives of ethnically and racially marginalized individuals (Johnson-Agbakwu et al., 2020). In the United States, healthcare workers, and especially social workers, have witnessed the intersection of the COVID-19 pandemic and upsurge in racism that has contributed to the unprecedented demand for mental health services (e.g., Kaslow et al., 2020).
In response to what has become known as “America’s mental health crisis” (American Hospital Association, 2022) and other health disparities exacerbated by the COVID-19 pandemic, social workers have stepped forward to support patients’ increasing socioemotional and economic needs, help patients gain access to necessary technology for safe communication, and simultaneously offer ongoing emotional support to colleagues (Cook et al., 2020; Walter-McCabe, 2020). Social workers serving youth populations have also been tasked with responding to the widespread detrimental emotional, social, and academic effects that school closings have had on child wellbeing, as well as the ways in which the pandemic has amplified housing and food insecurity that disproportionately affects communities of color (e.g., Daftary et al., 2021). Essential child protective services have also been impacted by the COVID-19 pandemic as home visits took place virtually, which was accompanied by both new challenges and new opportunities (Ferguson et al., 2022). Although the COVID-19 pandemic has significantly affected social workers and the services provided (e.g., Miller and Cassar, 2021), there is evidence to suggest that the challenges that pediatric social workers face are unique (Ross et al., 2021).
During the early phases of the pandemic, our team examined the impact of the pandemic and associated public health precautions on pediatric social workers providing care in a hospital setting. We conducted virtual focus groups with social workers employed across various settings in a large urban, pediatric hospital to better understand the evolution of social workers’ responsibilities and roles within integrated teams in response to the pandemic. Multiple themes emerged from analyses that elucidated (a) the impact the pandemic has had on the practice of social work and social workers themselves, (b) institutional assistance and obstacles to effective social work across hospital settings, and (c) perspectives on how social work will respond to future pandemic recovery efforts (Ross et al., 2021). Additional analyses about the essential worker designation revealed that many social workers agree their work is essential regardless of whether services are provided in-person or virtually, and that it is important that other healthcare professionals acknowledge social workers as essential team members through the same recognition, compensation, and support offered to healthcare workers of other professions (Schneider et al., 2022). Although salient themes related to the need to more adequately address health disparities exacerbated by the pandemic did emerge, it was not possible to assess how the rise in racist rhetoric and violence had impacted the lives and work of pediatric social workers during the acute phase of COVID-19 due to the timing of the focus groups (Spring 2020). As such, the present study had three primary aims: (1) to examine what had and had not changed for pediatric social workers in a hospital setting 1-year into the COVID-19 pandemic; (2) to directly analyze social workers’ thoughts and experiences on diversity, equity, and inclusion with particular attention to race and racism; and (3) to understand social worker perspectives about the long-term implications of the COVID-19 on the profession of social work.
Methods
All participants were recruited from the Social Work Department of a large, urban, quaternary pediatric hospital located in the northeastern United States. The Social Work Department is comprised of over 200 Masters’ level licensed clinical social workers who employed across more than 50 clinics and programs. Within the hospital enterprise, social workers practice across a wide range of settings, including emergency, inpatient, outpatient, primary care, specialty programs spanning both inpatient and outpatient locations.
We intentionally employed the same recruitment methodology used in previous research with this population (Ross et al., 2021; Schneider et al., 2022), wherein participants were recruited via an internal departmental email listserv with the permission of department administrators. Consenting participants were asked to enroll themselves in a focus group designated by their primary practice setting or role: Leadership/Administration, Specialty Program, Outpatient/Primary Care, Inpatient, or Emergency.
All focus groups were conducted asynchronously using Focus Group It, a qualitative research cloud-based focus group software platform previously employed in similar research studies (Ross et al., 2021; Schneider et al., 2022). Asynchronous focus group participation enabled participants to type responses to various research questions and engage in dialogue with other participants via posting responses to their answers, and logging in and out of the platform over the time period the focus group was operational (Gamarel et al., 2021). A unique advantage of this data collection method is that it allowed participants to respond to questions and comment on others’ responses when it was most convenient for them as opposed to having to attend a live virtual focus group. This method also preserved participant anonymity, as no video component was employed, and participants could choose the name under which their replies were posted.
Similar to the first round of focus groups conducted in April 2020, 12 discussion topics were designed to elicit perspectives on social work roles, responsibilities, and practice 1 year into the pandemic, the intersection of race and racism with COVID-19, and challenges and future directions for the social work field were released over a 21-day period (May 17th and 4 June 2021). New questions were released every 3 days and participants could access their designated group at any time. Reminders were sent to all participants when new questions were released. Each group remained open for the duration of the 21-day period to give participants time to complete their responses and dialogue with others asynchronously.
Data analysis
A team of three researchers, all doctorally trained in psychology or social work, analyzed focus group data using the Framework Method, an approach to thematic analysis that is widely used among interdisciplinary teams in the conduct of health-related research (Ritchie and Spencer, 2002; Ritchie et al., 2013). A core feature which differentiates the Framework Method from other qualitative approaches is the development of a matrix that enables researchers to systematically analyze data by both participants and themes. This allows for consideration of how themes occur across participants while being able to also consider a particular participant’s experience (Gale et al., 2013). Consistent with well-established framework analytic procedures (see Gale et al., 2013), all members of the research team initially read each focus group transcript in full to familiarize themselves with the data. After familiarization, team members used line-by-line open coding to classify data from each focus group to facilitate systematic comparisons with other parts of the dataset. Upon completion of initial line-by-line coding, the coding team met regularly to develop and refine a working analytical framework. All data from focus groups were subsequently indexed by all team members using the working analytical framework and charted into the framework matrix. To interpret the charted data and generate themes, the research team met weekly to interrogate data categories through comparisons between and within both individual participants and practice setting groups. Findings are presented using a narrative analytic framework situated within Bronfenbrenner’s (1979) ecological systems model.
Results
Thirty-four social workers participated in the second round of focus groups, representing three emergency room social workers, eleven specialty program-based social workers, eleven outpatient or primary care social workers, six inpatient social workers, and three social workers from leadership or administration. Bronfenbrenner’s (1979) ecological systems model was employed as the primary analytic framework, with six superordinate themes emerging from qualitative analyses. Themes spanned four primary categories, with three corresponding to individual-, professional/hospital-, and societal-level factors, and one spanning all ecological systems model levels (see Figure 1). Themes captured (a) experiences of burnout and coping among social workers 1 year into the COVID-19 pandemic, (b) the impact of patients’ behavioral acuity on social workers, (c) renewed awareness of racial inequity in patient care, (d) increased awareness of social determinants of health (SDOH), (e) the need for social workers to have more involvement in hospital decision making and policy reform, and (f) the long-lasting implications of grief and racial inequity. Subthemes also emerged as part of the qualitative analysis. In the sections that follow, themes and subthemes are presented within the four categories noted above.Figure 1. Qualitative themes.
Individual-level experiences
Burnout and coping
One year into the COVID-19 pandemic, social workers reported feeling exhausted, overworked, and, at times, demoralized. Multiple participants emphasized how challenging it was to be a social worker during this time because they were tasked with not only providing support to patients and their families, but also co-workers and hospital staff who were struggling with the trauma and overwhelm of working in a pediatric healthcare center during a global pandemic. One participant commented:“It feels hard to be a social worker right now. The overall morale on my unit is very low and staff has been relying on me for support more than ever. I feel burnt out from providing intensive support to patient families as well as to staff.” [Inpatient social worker A]
Several participants also commented on how the onset of pandemic fatigue (i.e., the exhaustion that comes with repeated safety precautions and demotivation to keep up such practices; WHO, 2020) had contributed to feelings of burnout and negatively impacted their work. This appeared to be particularly notable among social workers in inpatient and emergency department settings, as well as for those social workers located in specialty program or outpatient clinics who were required to deliver in-person care:“The idea of pandemic fatigue is real and it really set in for me back in January. [...] I think I was mostly on autopilot and able to come and do what was needed. When we entered the new year of 2021 I felt like I had hit a brick wall head on and even the most menial of tasks took herculean effort.” [Outpatient/primary care social worker A]
Across all groups, participants shared that the development and practice of new coping skills were, at times, an effective combatant to pandemic fatigue as well as mounting work and personal distress. For example, one participant reported that they felt as though their coping abilities significantly improved in order to meet the demands on social workers:“I think that overall my coping capacity increased, mostly out of necessity. [...] I think my repertoire of coping skills is significantly bigger than it was a year and a half ago.” [Specialty social worker A]
Two subthemes emerged from analyses that significantly exacerbated experiences of burnout. These themes captured: (a) the incongruence between expectations placed on social workers and the insufficient resources available, and (b) the impact of trauma and racism on burnout.
It isn’t enough
Many of the social workers who participated in this study expressed ongoing frustration and stress from being expected to “fix” complex, systematic issues that children and their families faced during the COVID-19 pandemic but not being provided with the resources to do so. One participant commented on how such expectations from medical teams precipitated changes in job requirements, and further contributed to burnout:“The level of stress as a social worker has significantly increased since the start of the pandemic as the psychosocial needs of the patients and families continue to outweigh the resources available. The expectations of the medical team for social work to ‘fix it’ has at times felt overwhelming, and I have not felt empowered to manage their expectations or set boundaries.” [Outpatient/primary care social worker B]
For social workers who voiced these experiences, they generally appeared to be more closely correlated with interprofessional team dynamics and functioning, or lack thereof, as opposed to circumstances associated with work location or practice setting.
The impact of trauma and racism
Participants also commented on how the broader context of systemic racism and ongoing violence against communities of color had significantly impacted social workers’ mental health and experiences of burnout from clinical care. For example, a participant wrote about how the intersection of a global pandemic, rising racial tension, and resulting violence had been stressful and draining and had negatively impacted their own emotional wellbeing:“I think the pandemic and societal stressors such as racial tensions and violence have had a significant impact on my own mental health. As Social Workers, we are present to support our patients and families, who during the past year and a half have also been under significant stress due to the pandemic and racial tensions and violence. It has been hard to not feel impacted by carrying the weight of helping patients and families, while also experiencing a significant amount of stress and anxiety. It was rough for a long time, experiencing anxiety, sleep disturbances, and symptoms of depression.” [Inpatient social worker B]
In addition, many participants shared that they experienced guilt and depression for not having the time or resources to devote to directly helping communities of color and creating change at mezzo- and macro-levels. Although numerous participants reported prioritizing health equity and racial inclusion in their roles as social workers, they also questioned the level of impact they could have given the enormity of systemic racism and injustice.
Professional and hospital-level factors
Behavioral health crisis in the context of the great resignation
As a result of considerable burnout among social workers and limited institutional support, many individuals resigned within the first year of the COVID-19 pandemic. What is often referred to as “The Great Resignation” has undoubtedly contributed to the scarcity of mental health resources amid an ongoing behavioral health crisis for children and their families. Participants in the present study noted how staffing shortages and long wait times had negatively impacted the type of acute care patients and their families receive:“The number of inpatient beds is not enough to meet the demand. [...] For example, a child who was engaging in, or thinking about engaging in, non-suicidal self-injury might be treated on an outpatient basis. Now, because outpatient resources are scarce, we are hospitalizing those kids who would normally be treated in less restrictive environments.” [Emergency department social worker A]
Across all groups, participants also reported that this scarcity of resources for patients and families had resulted in more behavioral health acuity of patients and increasing caseloads. One participant noted these factors had decreased morale and increased stress among social workers:“I feel most supported by the floor RNs and other I/P [inpatient] Social Workers, and I think that both groups feel quite exhausted and morale remains low in both groups. One common theme is not feeling supported with staff ratios to patients or acuity.” [Inpatient social worker B]
Mental health needs of new patients
An important subtheme to emerge from analyses captured how the negative emotional impact of the COVID-19 pandemic caused many children and families to access behavioral health services for the first time, which added additional demand to an already stressed system. For example, one participant commented on how children/families who had never previously needed or engaged in mental health support required more time, education, and resources from social workers:“The main difference in the work has been that more time/attention is spent with parents/families regarding their child’s need for psych treatment. Many families are experiencing mental health crises for the first time, and as such require more psychoeducation and management.” [Emergency department social worker B]
Working with patients and families new to mental health crises is considerably more time-consuming than working with families who are well known to an emergency department or agency, and/or have experience navigating behavioral and mental health services. This extra effort and time, in combination with higher volumes of patients needing to be seen, further increased stress and may have led social workers to resign from their positions.
Societal-level factors
Racial inequity
The behavioral health crisis fueled by the COVID-19 pandemic also highlighted enormous, longstanding, and macro-level healthcare inequities that are rooted in systemic racism. Multiple social workers in the present study reported becoming significantly more aware of how medical systems fail children and families of color, particularly Black children and families. One participant discussed how structural racism has negatively impacted the care families of color have received and how past traumatic experiences with healthcare systems has understandably led these families to be wary of healthcare professionals and doubtful they will be well taken care of:“I have been able to see firsthand how a crisis like the pandemic painfully exposes the systemic racism that many of our clients experience on a daily basis. This year has also taught me how broken our public health care system is and the extreme vulnerability this presents to so many Americans, especially families of color. I have observed how access to COVID related care is vastly unequal and the lack trust people of color have in the medical system to this day, as a result of past racist abuse, is yet another embarrassing fact of white supremacy culture.” [Outpatient/primary care social worker B]
Many participants wrote about knowing that racism is a major problem that affects healthcare long before the COVID-19 pandemic, but that social work as a profession needs to prioritize direct action and advocacy in combating institutional racism. Social workers in this study commented on how their past silence on issues of racism have been harmful, and how essential it is for social workers self-reflect and take an active, anti-racist stance in their work:“[...] the pandemic has brought all of the inequities in our hospital, community, the culture, and country at large to light. If you are black, brown or a POC [person of color] your experience is vastly different to those who are white. If anything, the pandemic has revealed how little I had been doing as social worker to focus on these issues and lift marginalized voices. My previous efforts were not enough. [...] Our profession needs to acknowledge our racist roots and do more to improve the lives of all people.” [Outpatient/primary care social worker A]
Another participant offered a striking simile of how different communities are impacted by the pandemic based on privilege and access:“Inequities that existed before the pandemic were exacerbated during the pandemic, evident in the glaring disparities across races. A colleague said to me that the pandemic is like we are all in the same storm, but different size boats.” [Inpatient social worker C]
Greater awareness of social determinants of health
Across all groups, participants noted that greater awareness SDOH is a critical component in addressing social inequities that place children and families of color at a social and medical disadvantage in healthcare. Many participants shared that attention to SDOH within the context of the COVID-19 pandemic was important for better understanding how to respond in health emergencies in underserved communities. For example, one participant emphasized how the health crisis has disproportionately and negatively impacted SDOH for communities of color:“The glaring systemic racism that was evident in the number of cases in minority communities, lack of testing available, lack of vaccinations when they initially became available, length of lines at food pantries, limited amount of employment opportunities and struggle between childcare and maintaining employment.” [Inpatient social worker D]
To help address such inequities, another participant shared their belief that social workers should be part of each hospital department in order to adequately respond to institutional inequities that contributed to poor SDOH for families during the pandemic:“Access to social work uniformly across the hospital will have to be addressed in order to address the mental health concerns post pandemic, as well as address access inequities, and ultimately, the inequities of our systems that result in housing instability and food insecurity, and access to health care. If we’re truly going to address social determinants of health, then we need to increase advocacy supports in all areas for all of the above.” [Leadership/administration social worker A]
Across participants and groups, there was agreement that an increased awareness of the inequities in access to healthcare, food, shelter, and reliable employment was essential for helping the most vulnerable and at-risk community members. As exemplified above, many respondents shared that social workers must do their part to help enhance SDOH for children and their families during the COVID-19 pandemic response.
Need for advocacy and policy change
Another societal-level theme to emerge from analyses was that social workers in our sample felt strongly that advocacy within healthcare systems was needed to address the mounting mental health crisis exacerbated by the COVID-19 pandemic. Many participants described how the current approaches and responses to this crisis have been temporary, short-term solutions that do not address longstanding and problematic underlying inadequacies and deficiencies. One participant stated:“It feels like putting band-aids on bullet holes because of the degree to which the weaknesses in our healthcare and welfare systems have been exacerbated by the pandemic.” [Specialty social worker B].
Some participants also underscored the strong need for a concentrated effort by the social work profession to advocate for change at the hospital-, state-, and policy-levels to help address the growing mental health challenges, grief, and trauma that have existed for families and were magnified by the COVID-19 pandemic:“Our profession has a key role in advocacy and policy. Our voice in the areas of police reform, racism, LGBTQ+, social inequalities is essential! We need to be vocal in how some communities have been disproportionately impacted by COVID.” [Specialty social worker C]
Two subthemes emerged that captured how social workers can better advocate for systemic change and contribute to policy reform. These themes captured the need for social workers to (a) be included at the many different levels of a hospital healthcare system, and (b) be part of specialized disaster response teams.
Social work inclusion at all levels
In order for social workers to be able to advocate for system-wide change and contribute to efforts to “fix” the complex problems they are tasked with by other professionals, participants in this study stated that social workers must work at all levels of a hospital healthcare system. This included social worker’s holding higher administrative positions so that social worker voices could be appropriately amplified when emergency response and hospital-wide policy decisions were being made. One participant (Specialty social worker D) commented that social workers becoming included in more positions of power might also unify social workers who are otherwise siloed onto specific specialty teams by stating: “The profession should be doing all levels of practice and working to tear down the walls that divide one social worker/team from another.” Another participant shared their belief that if social workers had initially been listened to at the start of the COVID-19 pandemic, many challenges and the mental health crisis could have been diminished:“[...] I think social workers need to be more involved in the planning and oversight of pandemic response. [...] I suspect that certain aspects of this disaster could have been mitigated, if not avoided, if the response had been planned through a social work lens. I think that clinical social workers would have been able to give some really valuable insight into how macro policies would realistically play out at the micro level.” [Specialty social worker A]
Needs for a social worker response team
Some participants in the study shared the perspective that hospital systems would greatly benefit from a team of social workers that are prepared to respond to emergencies related to future disasters or pandemics. One participant shared how they thought such a specialized group could be of assistance:“I strongly believe that the creation of a specialized unit of social workers trained in psychological first aid should be incorporated into the emergency management strategic planning. Whether the next disaster is man-made, biological, or natural there should be team of social workers ready to be deployed to serve our patients as part of an emergency response. I think social workers are a forgotten, vital, and underutilized resource in emergency management and mitigation.” [Specialty social worker E]
It is clear that social workers felt a need for representation at higher levels of management in order to take a more active role in important decision making when responding to emotional and behavioral crises that come from traumatic world events.
Looking towards the future
Grief and social inequity
The final theme to emerge from participants’ responses related to the long-lasting implications that the COVID-19 pandemic will have on mental health and our understanding of social inequity. Numerous respondents shared that they anticipated that grief and a sense of loss would remain long after the pandemic had passed, especially for minoritized communities.“I think the pandemic has exposed the weak points in our society, which we will continue to deal with for many years to come. Even if the pandemic is winding down, the problems we have as a result will take much more time to address.” [Emergency department social worker C]
However, they also expressed that in communal hardship comes the possibility for reflection, collaboration, and action to better support those who are most vulnerable and underserved. As shared by one participant:“The impact that COVID had on our minority communities is overwhelming, and the collective grief is so significant - the loss of time, the loss of lives, the loss of livelihood, the loss of stability, the loss of feeling safe. With that, though, people have been brought to our knees, and opportunities have existed for more compassionate listening, self-awareness, and learning.” [Inpatient social worker E]
In general, social workers who participated in this study felt they would be tasked with helping to manage the socioemotional difficulties that have arisen from the pandemic for years to come and that they have the skills that are needed to help children and their families manage these difficult and traumatic experiences.
Discussion
To our knowledge, this is the first study to capture experiences and thoughts of pediatric hospital social workers 1 year into the COVID-19 pandemic. Results of this present study are important for understanding and contextualizing what has remained the same and what has changed for this population of social workers since the acute phase of the pandemic, as well as the implications the COVID-19 pandemic has for the future of social work.
What has remained the same?
When comparing themes that emerged in these focus groups to those in our initial series of focus groups at the start of the pandemic, many persisted. For example, COVID-19 has continued to impact social work as a profession and social workers themselves. Many participants of this study explained that the combination of role expansion in a hospital setting to meet the needs of children and families, in combination with caring for colleagues and the expectation that social workers should fix large, systemic challenges without the necessary resources, has led to chronic experiences of stress. During the acute phase of the pandemic, social workers felt overwhelmed and powerless, and it appears that these experiences have persisted and evolved into feelings of burnout and pandemic fatigue (Ross et al., 2021). Other researchers have also documented the impact of burnout on social workers (e.g., Peinado and Anderson, 2020) and other essential healthcare workers, as well as the negative effects burnout can have on mental health and completion of professional responsibilities (Sharifi et al., 2021).
Similar to experiences participants shared at the beginning of the pandemic (Ross et al., 2021), some social workers continued to feel as though other members of multidisciplinary teams do not fully understand social workers’ roles in the context of the pandemic. In the initial phase of this research, results suggested that a joint effort between social work as a profession educating other disciplines, and institutions working to shift organizational cultures to value social work, could lead to increased awareness of the role and value of social workers (Ross et al., 2021). Social worker’s perspectives captured in the current study expanded upon this recommendation by advocating for social workers to be included at all institutional levels to better inform other disciplines of services and expertise offered. It is likely that this representation will lead to a better sense of community among social workers and increase morale, which has been linked to improved job satisfaction and decrease in turnover (e.g., Martin, 1981).
Another recommendation that has remained consistent since the beginning of the pandemic has been the call for a social worker preparedness response team (Ross et al., 2021). Findings from the current study highlight social workers’ continuing belief that the organization of an emergency/disaster response team of social work professionals could be tremendously helpful, based on first-hand observations of healthcare shortcomings from the earlier stages of the COVID-19 pandemic. The utilization and deployment of such social work teams have been found to offer essential support to patients and other hospital healthcare systems (e.g., Yu et al., 2020) and thus it is likely such benefits would be applicable in the pediatric setting.
What has changed?
Findings from the current study have also highlighted multiple changes and evolutions in social worker’s experiences after enduring 1 year of the COVID-19 pandemic. A new but expected evolution of chronic burnout was found to contribute to worsening mental health, which multiple participants discussed during the focus group. Specifically, group members shared that the stress of the pandemic, coupled with social stressors such as racial tension and violence, had deleterious effects on their own mental health. This is, unfortunately, a common outcome of frontline healthcare workers during the pandemic (e.g., Santarone et al., 2020). Another noteworthy change from the beginning of the pandemic was that many social workers, and other healthcare professionals, resigned from their positions after the incredible surge in case management and patient complexity, and the toll such demands had taken on emotional and physical wellbeing. The resignations of healthcare professionals across the world underscore how difficult and traumatic working in a hospital setting during the pandemic has been (e.g., Jiskrova, 2022). It is important to note, however, that many of the social workers who participated in this study also described how the nature of social work during a global pandemic has led to increased coping capacity and the development of new self-care strategies. Although increased institutional economic and socioemotional support remains essential so that coping skills are not inherently necessary in the profession, this ability to adapt is just one of many examples of social worker resiliency during COVID-19 (see Cook et al., 2020; Tosone, 2020).
Another important change from reported experiences during the acute phase of the pandemic (Ross et al., 2021) was social workers’ increased awareness of racial violence, longstanding healthcare inequities that have failed children and families of color, and how the pandemic has disproportionately and negatively impacted social determinants of health for communities of color. Participants shared that these deeper reflections and enhanced awareness were spurred by the murders of George Floyd and Breonna Taylor—among many others, the rise of the Black Lives Matter movement, and calls to reckon with systemic racism as a profession. Many social workers have called for their colleagues to acknowledge their own history of complicity and participate in the dismantling racist systems that perpetuate social injustices against Black people in America (e.g., McCoy, 2020; Ioakimidis and Maglajlic, 2020). Themes from the present study indicated that pediatric social workers in hospital settings share in this perspective and agree it is an ethical obligation of social work as a profession to pursue social change, particularly on behalf of vulnerable and oppressed community members. However, multiple group participants shared that they were not able to devote as much of their time and emotional resources to this critical social justice issue as they would like because of the unrelenting demands on the healthcare systems and consequential rapid role expansion of social workers. Although the COVID-19 pandemic has brought about enhanced awareness of racism, some social workers have not yet been able to behaviorally implement the changes that align with their anti-racist and social justice values.
Where do we go from here?
Looking towards the future, many pediatric hospital social workers voiced beliefs that approaches and responses to the mental health crisis exacerbated by the COVID-19 pandemic—even 1 year later—are both insufficient and often only temporary, short-term solutions that do not offer adequate, sustainable solutions to underlying deficiencies and disparities of the healthcare system. One participant offered a striking visual that compared the current efforts of social workers and institutional pandemic responses to placing band-aids on bullet holes. In other words, this participant was making the case that quick fixes do not make for effective, long-term solutions, in the same way that band-aids are not an effective way to address a bullet wound. Many participants discussed how the need for advocacy and change at hospital-, state-, and policy-levels that directly target major and complex challenges are crucial for progress. Examples included addressing chronic staffing shortages, including social work at higher levels of administration, developing a social worker crisis response team in preparation for future disasters, and protecting social workers’ time so they can properly advocate for system-wide change and engage in actively anti-racist initiatives that amplify voices of communities of color and move the profession towards social justice. Scholars have emphasized the critical importance of social workers being agents of change and advocates of social justice and human rights (e.g., Finn, 2020), which underscores the necessity of social workers being provided the time and resources to accomplish such initiatives (see McCoy, 2020). Devotion of resources and expertise to address these systemic issues will likely also aid in the management of the chronic grief and negative economic and socioemotional devastation that will surely continue to impact our most vulnerable communities long after the pandemic has ended.
Limitations
There are several major limitations in our study. The first is the lack of demographic information, which was not collected due to low response rates to the demographic survey in the initial wave (Ross et al., 2021). The second limitation is that participants were recruited exclusively from one large, urban pediatric hospital in New England and so all themes may not be generalizable to hospital social workers in other contexts or geographic locations. It may be helpful for this study to be replicated with social workers serving adult populations or providing services in other areas of the United States. Finally, the respondents may be a non-representative subset of hospital social workers in the institution.
Conclusion
The current study is one of the first to explore how the experiences of pediatric hospital social workers have changed over the course of the COVID-19 pandemic. Identified themes represented individual-, professional/hospital-, and societal-level factors, as well as implications for the future. Key shifts from perspectives shared during the initial acute phase of the pandemic include experiences of the ways in which chronic stress has evolved into burnout and a need for improved coping skills, the impact trauma and racism has had on social workers’ own mental health, the impact of the “great resignation” on social work as a profession, and an increased awareness of racial inequity and social determinants of health. Looking towards the future, social workers acknowledge the need for social justice advocacy to best support the needs of the communities of color who have been disproportionately impacted by the pandemic, as well as the necessary resources to address their anticipated, long-term grief. Finally, this research shows that social workers continue to recognize and underscore the need for additional institutional support to not only meet the demands of the current mental health crisis, but to address complex, longstanding structures that promote health disparities in the healthcare system—especially in the area of mental health. The COVID-19 pandemic has raised awareness of a mental health crisis that was already affecting many under-resourced communities, particularly communities of color. This study suggests the social work workforce recognizes their role in advocating for change.
ORCID iDs
John L McKenna https://orcid.org/0000-0003-0937-8299
Elizabeth R Boskey https://orcid.org/0000-0002-4982-1323
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
==== Refs
References
American Hospital Association (2022) Statement of the American Hospital Association to the Committee on Ways and Means of the United States House of Representatives: “America’s Mental Health Crisis”. Washington, DC: American Hospital Association.
Bronfenbrenner U (1979) The Ecology of Human Development: Experiments by Nature and Design. Cambridge, MA: Harvard University Press.
Centers for Disease Control (2020) COVID Data Tracker Weekly Review: April 15th 2022 . Washington, DC: U.S Department of Health and Human Services. https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidview/index.html
Cook LL Zschomler D Biggart L , et al. (2020) The team as a secure base revisited: remote working and resilience among child and family social workers during COVID-19. Journal of Children’s Services 14 (4 ): 259–266.
Daftary A Sugrue EP Gustman BD , et al. (2021) Pivoting during a pandemic: school social work practice with families during COVID-19. Children & Schools 43 (2 ): 71–78.
Devakumar D Shannon G Bhopal SS , et al. (2020) Racism and discrimination in COVID-19 responses. Lancet 395 (10231 ): 1194.
Ferguson H Kelly L Pink S (2022) Social work and child protection for a post-pandemic world: the re-making of practice during COVID-19 and its renewal beyond it. Journal of Social Work Practice 36 (1 ): 5–24.
Finn JL (2020) Just Practice: A Social Justice Approach to Social Work. Oxford: Oxford University Press.
Gale NK Heath G Cameron E , et al. (2013) Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Medical Research Methodology 13 (1 ): 1–8.23297754
Galli F Pozzi G Ruggiero F , et al. (2020) A systematic review and provisional meta-nalysis on psychopathologic burden on health care workers of coronavirus outbreaks. Frontiers in Psychiatry 11 : 568664.33192692
Gamarel KE Stephenson R Hightow-Weidman L (2021) Technology-driven methodologies to collect qualitative data among youth to inform HIV prevention and care interventions. Mhealth 7 (34 ).
Ioakimidis V Maglajlic RA (2020) Black lives matter, biopolitics and the social work response. The British Journal of Social Work 50 (6 ): 1647–1651.
Johnson-Agbakwu CE Ali NS Oxford CM , et al. (2020) Racism, COVID-19, and health inequity in the USA: a call to action. Journal of Racial and Ethnic Health Disparities 9 (52–58 ): 1–7.
Jiskrova GK (2022) Impact of COVID-19 pandemic on the workforce: from psychological distress to the great resignation. Journal of Epidemiology and Community Health 2022 : 1–2.
Kaslow NJ Friis-Healy EA Cattie JE , et al. (2020) Flattening the emotional distress curve: a behavioral health pandemic response strategy for COVID-19. American Psychologist 75 (7 ): 875–886.32538638
The Lancet Editorial Board (2020) Reviving the US CDC. Lancet 395 (10236 ): 1521.32416772
Liu YC Kuo RL Shih SR (2020) COVID-19: The first documented coronavirus pandemic in history. Biomedical Journal 43 (4 ): 328–333.32387617
Martin PY (1981) Multiple constituencies, dominant societal values, and the human service administrator: Implications for delivery. Administration in Social Work 4 (2 ): 15–27.
McCoy H (2020) Black Lives matter, and yes, you are racist: the parallelism of the twentieth and twenty-first centuries. Child and Adolescent Social Work Journal 37 (5 ): 463–475.32836724
Miller JJ Cassar J (2021) Self-care among healthcare social workers: the impact of COVID-19. Social Work in Health Care 60 (1 ): 30–48.33550956
Peinado M Anderson KN (2020) Reducing social worker burnout during COVID-19. International Social Work 63 (6 ): 757–760.
Reicher S Drury J (2021) Pandemic fatigue? How adherence to Covid-19 regulations has been misrepresented and why it matters. BMJ: British Medical Journal 372 : n137.33461963
Ritchie J Lewis J Nicholls CM , et al. (2013) Qualitative Research Practice: A Guide for Social Science Students and Researchers. Thousand Oaks, CA: Sage Publications.
Ritchie J Spencer L (2002) The Qualitative Research Companion. Thousand Oaks, CA: Sage Publications.
Ross AM Schneider S Muneton-Castano YF , et al. (2021) You never stop being a social worker:” Experiences of pediatric hospital social workers during the acute phase of the COVID-19 pandemic. Social Work in Health Care 60 (1 ): 8–29.33657982
Santarone K McKenney M Elkbuli A (2020) Preserving mental health and resilience in frontline healthcare workers during COVID-19. The American Journal of Emergency Medicine 38 (7 ): 1530.32336584
Schneider SE Ross AM Boskey ER (2022) ‘We are essential:’ Pediatric health care social workers’ perspectives on being designated essential workers during the COVID-19 pandemic. Social Work in Health Care 61 (1 ): 36–51.35138996
Sharifi M Asadi-Pooya AA Mousavi-Roknabadi RS (2021) Burnout among healthcare providers of COVID-19; a systematic review of epidemiology and recommendations. Annals of Agricultural and Environmental Medicine: AAEM 9 (1 ): 1–17.
Tosone C (2020) Shared Trauma, Shared Resilience During a Pandemic: Social Work in The Time of COVID-19. Springer Nature.
Walter-McCabe HA (2020) Coronavirus pandemic calls for an immediate social work response. Social Work in Public Health 35 (3 ): 69–72.32286936
World Health Organization (2020) Pandemic Fatigue: Reinvigorating the Public to Prevent COVID-19, Regional WHO Office for Europe. WHO. https://apps.who.int/iris/bitstream/handle/10665/335820/WHO-EURO-2020-1160-40906-55390-eng.pdf
Yu Z Tan W Niu L (2021) The experiences of the good companions response team the, during the COVID-19 pandemic in Wuhan, China: a multi-professional team led by social workers. Asia Pacific Journal of Social Work and Development 31 (1–2 ): 132–138.
| 0 | PMC9726635 | NO-CC CODE | 2022-12-08 23:18:17 | no | Qual Soc Work. 2022 Dec 5;:14733250221144050 | utf-8 | Qual Soc Work | 2,022 | 10.1177/14733250221144050 | oa_other |
==== Front
J Biol Rhythms
J Biol Rhythms
JBR
spjbr
Journal of Biological Rhythms
0748-7304
1552-4531
SAGE Publications Sage CA: Los Angeles, CA
10.1177/07487304221124661
10.1177_07487304221124661
Letter
Time of Day of Vaccination Does Not Associate With SARS–CoV–2 Antibody Titer Following First Dose of mRNA COVID–19 Vaccine
https://orcid.org/0000-0003-3928-0735
Yamanaka Yujiro *†1
Yokota Isao ‡
Yasumoto Atsushi §
Morishita Eriko ‖
Horiuchi Hisanori ¶
* Laboratory of Life and Health Sciences, Graduate School of Education and Faculty of Education, Hokkaido University, Sapporo, Japan
† Research and Education Center for Brain Science, Hokkaido University, Sapporo, Japan
‡ Department of Biostatistics, Graduate School of Medicine, Hokkaido University, Sapporo, Japan
§ Division of Laboratory and Transfusion Medicine, Hokkaido University Hospital, Sapporo, Japan
‖ Department of Hematology, Kanazawa University Hospital, Kanazawa, Japan
¶ Department of Molecular and Cellular Biology, Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan
1 Yujiro Yamanaka, Laboratory of Life and Health Sciences, Graduate School of Education and Faculty of Education, Hokkaido University, North-11, West-7, Kita-Ku, Sapporo 060-0811, Japan; e-mail: [email protected].
12 2022
12 2022
12 2022
37 6 700706
© 2022 The Author(s)
2022
SAGE Publications
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
The immune system exhibits circadian rhythms, and its response to viral infection is influenced by the circadian clock system. Previous studies have reported associations between the time of day of vaccination against COVID-19 and production of anti-SARS-CoV-2 antibody titer. We examined the effect of vaccination time of day on anti-SARS-CoV-2 antibody titer after the first dose of vaccination with the mRNA-1273 (Moderna) COVID-19 vaccine in an adult population. A total of 332 Japanese adults participated in the present study. All participants were not infected with SARS-CoV-2 and had already received the first dose of mRNA-1273 2 to 4 weeks prior to participating in the study. The participants were asked to provide basic demographic characteristics (age, sex, medical history, allergy, medication, and mean sleep duration), the number of days after the first dose of vaccination, and the time of day of vaccination. Blood was collected from the participants, and SARS-CoV-2 antibody titers were measured. Ordinary least square regression was used for assessing the relationship between basic demographic characteristics, number of days after vaccination, time of day of vaccination, and the log10-transformed normalized antibody titer. The least square mean of antibody titers was not associated with the vaccination time and sleep durations. The least square means of antibody titers was associated with age; the antibody titers decreased in people aged 50 to 59 years and 60 to 64 years. The present findings demonstrate that the vaccination time with mRNA-1273 was not associated with the SARS-CoV-2 antibody titer in an adult population, suggesting that these results do not support restricting vaccination to a particular time of day. The present findings may be useful in optimizing SARS-CoV-2 vaccination strategies.
SARS-CoV-2
antibody titer
mRNA-1273
time of day
general adults
Japan Agency for Medical Research and Development https://doi.org/10.13039/100009619 JP20ek0210154 typesetterts1
==== Body
pmcThe circadian clock system generates and integrates daily rhythms in many physiological and behavioral functions. Immune function is also influenced by the circadian clock system (Scheiermann et al., 2013; Wyse et al., 2021). The primary outcome of vaccine administration is the production of an antibody for the disease targeted by the vaccine. Regarding the effect of vaccination on antibody production, an increase in antibody titer due to inactivated influenza vaccine was higher in older individuals but not tested in younger individuals receiving morning vaccination than in those receiving afternoon vaccination (Kurupati et al., 2017; Long et al., 2016a, 2016b; Phillips et al., 2008). Adjusting the time of day of vaccine administration could be a simple yet effective strategy to enhance antibody production, which helps protect against viral infection. Circadian rhythms have been increasingly accepted as an important determinant of immune response. Many of immune parameters such as cytokine levels, antigen presentation, and lymphocyte proliferation show circadian variation both in mice and humans (Arjona and Sarkar, 2006; Hornung et al., 2002; Petrovsky and Harrison, 1998). In addition, the severity of viral infection is dependent on the time of day of encountering the pathogen (Borrmann et al., 2021; Sengupta et al., 2019). The rhythm of T-cell counts and its functions are dependent on glucocorticoid rhythm which is under the control of circadian pacemaker in the brain (Shimba et al., 2018; Dimitrov et al., 2009). The circadian clock controls the magnitude of response of T cell to antigen presentation, and the circadian clock within T cell is required for the circadian regulation (Nobis et al., 2019). At the molecular clock level, mice loss of Bmal1 led to be elevated viral burden and more severe lung pathology in mice infected at specific time of day compared with the wild-type mice (Ehlers et al., 2018). Bmal1 deletion in either club cells or alveolar cells resulted in increased morbidity and inflammation as well as mortality after H1N1 influenza type A virus, respectively (Zhuang et al., 2019; Issah et al., 2021). Thus, circadian clock system and circadian variation of various immune functions may contribute to time of day modulation of vaccine’s efficacy.
The first official case of coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was reported in December 2019. It has since spread rapidly and evolved into a pandemic. To protect against COVID-19, the COVID-19 vaccines have been quickly developed in a short period of time. Regarding the time of day of vaccination on the antibody response to the COVID-19 vaccine, 2 recent reports (Wang et al., 2022; Zhang et al., 2021) have found that the time of day of administration differentially affects the SARS-CoV-2 antibody titer after the first- and second-dose vaccination in health care workers. The first study demonstrated a stronger immune response to morning vaccination with an inactivated COVID-19 vaccine (Sinovac Covid-19 vaccine), which resulted in a higher antibody level (Zhang et al., 2021). Another study demonstrated the opposite conclusion, with higher antibody levels detected when the SARS-CoV-2 mRNA vaccine BNT162b2 (Pfizer-BioNTech) was administered in the afternoon rather than in the morning (Wang et al., 2022). From these 2 reports in health care workers, it has been revealed that the antibody titer response following the COVID-19 vaccine was also influenced by vaccine type, participant age, sex, and number of days post vaccination. However, there are currently no reports investigating the effect of time of day of administration on the immune response to mRNA-1273 (Moderna) in adult population with no experience of night and rotating shift work. In the present study, we examined whether the time of day of vaccination affects the SARS-CoV-2 antibody titer after the first dose of mRNA-1273 in the Japanese adult population.
Materials and Methods
Ethics Statement
This study was approved by the Ethical Review Board for Life Science and Medical Research, Hokkaido University Hospital (approval no. 021-0012) and was performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants after they received an explanation of the study.
Study Design and Recruitment
This study was performed between 11 and 27 August 2021. Participants who received the first dose of mRNA-1273 2 to 4 weeks prior were recruited using a mailing list at Hokkaido University. In the present study, the participants received the COVID-19 vaccine at the Hokkaido University as the COVID-19 in-house vaccination program. In the University’s in-house vaccination program, a total of 22,000 university members (18 years of age or older) including students, faculty, staff members, and other approved personnel of Hokkaido University and the Otaru University of Commerce were available to receive the COVID-19 vaccine at the Hokkaido University Gym from 17 July to 12 September 2021. The vaccination program was only conducted on weekends, and the participants were required to make a reservation when they prefer to get the vaccine. All health care workers had already received the COVID-19 vaccine at the start of the in-house vaccination program. Therefore, all the participants were considered neither shift workers nor health care workers. Considering the feasibility of this study, we set the sample size to 500. We did not conduct power analysis before collecting samples. A total of 419 participants were enrolled in the study, and 332 of the 419 participants completed demographic questions, including age, sex, date and time of vaccination, past medical history, medication use (antiplatelet drugs, anticoagulant drugs, antihypertensive drugs, diabetes treatment drugs, and hyperlipidemia drugs), allergic history, and mean sleep duration. To measure the level of antibody titers of the SARS-CoV-2 antibody, blood samples were collected at the Hokkaido University Hospital. Blood samples were separated into serum and stored at −80 °C until use. Serum antibody titers against the SARS-CoV-2 neutralizing antibody were assayed using a chemiluminescence enzyme immunoassay and HISCL SARS-CoV-2 S-IgG reagent (Sysmex Corporation, Kobe, Japan). The titer range of detection was 200 to 20,000 BAU/mL; titers under the lower limit of detection were considered 20 BAU/mL. We also measured antibody to SAR-CoV-2 N-IgG (Sysmex Corporation, Kobe, Japan) to identify individuals who have previously been infected by the SARS-CoV-2 virus. Individuals previously infected with COVID-19 were excluded from the analysis. The cut off value of the antibody titer of N-IgG to identify the individuals previously infected was set at 10 BAU/mL.
Statistical Analysis
A general linear regression model was used for assessing the relationship of common logarithm (log10) transformed SARS-CoV-2 antibody titers with days after vaccination and time of day of vaccination adjusting sex, age, allergy, medication, and sleep duration (5-6 h, 7 h, and 8-9 h). The time of day of vaccination were categorized into 2 groups of morning and afternoon group. JMP Pro 16 software (SAS Institute Japan) was used for all the statistical analyses. A confidence level was set to 95%.
Results
Table 1 summarizes the association between the demographic characteristics (sex, ages, allergy, medication, chronic diseases, sleep duration, and time of day of vaccination) and log10 transformed antibody titers in the study participants (n = 332).
Table 1. Demographic characteristics of participants and mean antibody titers.
SARS-CoV-2 Antibody Titer (Log10 BAU/mL)
Morning Afternoon
Characteristics n M SD n M SD
Sex
Male 74 2.39 0.43 74 2.50 0.36
Female 86 2.43 0.30 98 2.46 0.33
Ages
20-29 years 69 2.49 0.29 72 2.47 0.31
30-39 years 22 2.34 0.26 28 2.53 0.34
40-49 years 37 2.47 0.36 36 2.49 0.37
50-59 years 27 2.30 0.34 25 2.40 0.44
60-64 years 5 1.76 0.90 11 2.47 0.28
Allergy
Yes 46 2.42 0.32 55 2.52 0.40
No 114 2.41 0.39 117 2.45 0.32
Medication
Yes 26 2.36 0.35 31 2.41 0.38
No 134 2.42 0.37 141 2.49 0.34
Chronic diseases
Yes 31 2.34 0.55 31 2.48 0.37
No 129 2.43 0.31 141 2.47 0.34
Sleep duration
5-6 h 74 2.37 0.42 70 2.50 0.38
7 h 67 2.46 0.30 73 2.43 0.32
8-9 h 19 2.40 0.33 29 2.51 0.35
The response of SARS-CoV-2 antibody titers to mRNA-1273 has been reported to increase 2 to 4 weeks after the first dose of vaccination (Widge et al., 2021). In the present study, the effect of post vaccination days on the antibody titers was analyzed by conveniently dividing into 3 groups based on the days after vaccination (13-20, 21-27, and 28-45 days after the first dose of vaccination). Table 2 shows the mean and standard deviation of log10 transformed antibody titers in 13 to 20, 21 to 27, and 28 to 45 days. The antibody titers were similar between the 3 categories of the day after vaccination in 13 to 20, 21 to 27, and 28 to 45 days group, respectively.
Table 2. Mean antibody titers association between the days after vaccinations and time of day of vaccination.
Days After Vaccination SARS-CoV-2 Antibody Titer (Log10 BAU/mL)
Morning Afternoon
n M SD n M SD
13-20 days 37 2.40 0.32 27 2.44 0.40
21-27 days 73 2.47 0.31 97 2.48 0.31
28-45 days 50 2.33 0.46 48 2.47 0.39
Table 3 summarizes the association between antibody titer and several factors using a general linear regression analysis. Antibody titers were decreased in participants aged 50 to 59 and 60 to 64 years independent of time of day of vaccination (Figure 1a, Table 3). There were no obvious differences for antibody titers between morning and afternoon groups. From a general linear regression model, a significant association was not found between the time of day of vaccination and antibody titers (Figure 1b, Table 3).
Table 3. The association between antibody titer and several factors using a general linear regression analysis.
Factors SARS-CoV-2 AB (Log10 BAU/mL)
LS-mean Difference [95% CI]
Sex
Male 2.39 ref
Female 2.39 –0.004 [2.30, 2.46]
Age
20-29 years 2.47 Ref
30-39 years 2.44 –0.034 [2.28, 2.52]
40-49 years 2.47 –0.001 [2.36, 2.57]
50-59 years 2.34 –0.128 [2.09, 2.33]
60-64 years 2.24 –0.233 [1.81, 2.19]
Allergy
Yes 2.41 Ref
No 2.37 –0.043 [2.24, 2.41]
Medication
Yes 2.35 Ref
No 2.43 0.074 [2.36, 2.64]
Chronic diseases
Yes 2.41 Ref
No 2.37 –0.432 [2.19, 2.46]
Sleep duration
5-6 h 2.39 Ref
7 h 2.38 –0.008 [2.29, 2.46]
8-9 h 2.39 –0.001 [2.27, 2.51]
Days after vaccination
13-20 days 2.39 Ref
21-27 days 2.42 0.039 [2.36, 2.57]
28-45 days 2.36 –0.026 [2.22, 2.45]
Time of day
Morning 2.36 Ref
Afternoon 2.42 0.064 [2.41, 2.56]
Abbreviations: CI = confidence interval; LS-mean = least squares mean, which indicates the estimated mean in each factor by the general linear regression model.
Figure 1. Associations between the least square means of SARS-CoV-2 antibody titers, participant’s ages, and time of day of vaccination. Association between the least square means of log10 transformed SARS-CoV-2 antibody titers between the participant’s ages (a). *p < 0.05 vs. 20 to 29 years. †p < 0.05 vs. 40 to 49 years. Student’s t-test. Association between the least square means of antibody titers and time of day of vaccination in the morning and afternoon (b).
Discussion
The present study examined whether the vaccination time of day alters the SARS-CoV-2 antibody titer after the first dose of the mRNA-1273 vaccine in a sample from the Japanese adult population. A general linear regression analysis revealed no significant association between the time of day of vaccination and the SARS-CoV-2 antibody titer (Figure 1b, Table 3). The antibody titers were decreased in participants aged 50 to 59 years and 60 to 64 years. The decrease in antibody titer in older age groups were consistent with the previous studies (Anastassopoulou et al., 2022; Grupel et al., 2021; Müller et al., 2022).
In the case of BNT162b2, vaccination in the evening (1500-2059 h) was associated with a stronger immune response after vaccination, which resulted in an increase in antibody level in 16- to 29-year-old male and female patients and 30- to 39-year-old male health care workers (Wang et al., 2022). Furthermore, no influence of vaccination time on the antibody response in health care workers who received the ChAdOx1 adenovirus vaccine (AstraZeneca adenovirus vaccine) was demonstrated (Wang et al., 2022). In contrast to the immune response to BNT162b2, the first and second doses of inactivated COVID-19 vaccine in the morning (0900-1100 h) lead to a stronger immune response and increase in antibody titer as compared with the evening vaccination (1500-1700 h) at 28 and 56 days after the vaccination (Zhang et al., 2021). The present results showed no significant association between the time of vaccination and antibody titer (Table 3, Figure 1b), in contrast with the above recent reports (Wang et al., 2022; Zhang et al., 2021). Taken together, the present and recent findings suggest that the effect of the time of day of vaccination on the host immune response may depend on various factors, such as the type of vaccine (mRNA, inactivated, and adenoviral vaccines), sex, age, and number of days after vaccination (Wang et al., 2022). Furthermore, the differential effect of the vaccination time on the host immune response between BNT162b2 and mRNA-1273 might be explained by the large difference in the vaccine dose. BNT162b2 contained 30 μg (0.3 mL) of mRNA, whereas mRNA-1273 contained 100 μg (0.5 mL) of mRNA, which is more than 3 times the amount of mRNA compared with BNT162b2. A recent comparative study in health care workers revealed that higher antibody titers were observed in participants vaccinated with 2 doses of mRNA-1272 than in those vaccinated with BNT162b2 mRNA (Steensels et al., 2021; Khoury et al., 2021). Thus, the higher mRNA content in the mRNA-1273 vaccine compared with the BNT162b2 vaccine might explain the difference in effects regarding the time of day of vaccination between the mRNA-1273 and BNT162b2 mRNA vaccines. Another aspect to be considered is the association between participant race and antibody titer against the COVID-19 vaccine, which is not yet fully understood.
The present study is the first to demonstrate the influence of the time of day of vaccination on the antibody titer in Japanese adults who have received the first dose of mRNA-1273. However, our study has a few limitations which must be considered when interpreting these results in the broader context of translational opportunities within the circadian context. The time of day of vaccination was dependent on the time of day when the participants visited the vaccination site, so that the present study was not a randomized controlled trial. We did not examine other factors, such as participants’ physical activity (Chastin et al., 2021), circadian rhythm (Sengupta et al., 2021), and natural immune system (Netea et al., 2022), which have been associated with variations in the magnitude of response of the host immune system to viral infection. These limitations reduce our ability to conclusively rule out the possibility of circadian influence on antibody response humans. Future studies would also benefit from including other surrogates of immune response to vaccination, in addition to antibody titers. In addition, our data did not examine the effect of the vaccination time on the antibody titer after the second dose of vaccination, which induces a marked increase in the antibody titer (Steensels et al., 2021; Tré-Hardy et al., 2021). Furthermore, our data contained intra- and inter-individual differences in the host immune system. Further studies will be needed to carry out serial blood sample collections or cross-over studies that administer morning and evening vaccinations to the same individuals after performing the sample size estimation by a power analysis. It would be also needed to evaluate the effect of time of day of vaccination not only on the antibody titers but also other immune response markers (Zhang et al., 2021). While the current study does not support restricting vaccination to certain times of day, given the limitations associated with the study size and the non-randomized design and considering the large body of literature supporting clock-gated immune responses, we believe vaccination strategies should include time-of-day variable both in the vaccination and antibody testing steps. Larger randomized studies are needed to answer this question conclusively.
In conclusion, the present study provides evidence that the vaccination time with mRNA-1273 was not associated with the SARS-CoV-2 antibody titer in adult population, suggesting that these results do not support restricting vaccination to a particular time of day. The present findings may be useful in optimizing SARS-CoV-2 vaccination strategies.
This work was supported by the Agency for Medical Research and Development (AMED) under grant number JP20ek0210154.
The author(s) have no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
ORCID iD: Yujiro Yamanaka https://orcid.org/0000-0003-3928-0735
==== Refs
References
Anastassopoulou C Antoni D Manoussopoulos Y Stefanou P Argyropoulou S Vrioni G Tsakris A (2022) Age and sex associations of SARS-CoV-2 antibody responses post BNT162b2 vaccination in healthcare workers: A mixed effects model across two vaccination periods. PLoS ONE 17 :e0266958.35486622
Arjona A Sarkar DK (2006) The circadian gene mPer2 regulates the daily rhythm of IFN-gamma. J Interferon Cytokine Res 26 :645-649.16978068
Borrmann H McKeating JA Zhuang X (2021) The circadian clock and viral infections. J Biol Rhythms 36 :9-22.33161818
Chastin SFM Abaraogu U Bourgois JG Dall PM Darnborough J Duncan E Dumortier J Pavón DJ McParland J Roberts NJ , et al . (2021) Effects of regular physical activity on the immune system, vaccination and risk of community-acquired infectious disease in the general population: Systematic review and meta-analysis. Sports Med 51 :1673-1686.33877614
Dimitrov S Benedict C Heutling D Westermann J Born J Lange T (2009) Cortisol and epinephrine control opposing circadian rhythms in T cell subsets. Blood 113 :5134-5143.19293427
Ehlers A Xie W Agapov E Brown S Steinberg D Tidwell R Sajol G Schutz R Weaver R Yu H , et al . (2018) BMAL1 links the circadian clock to viral airway pathology and asthma phenotypes. Mucosal Immunol 11 :97-111.28401936
Grupel D Gazit S Schreiber L Nadler V Wolf T Lazar R Supino-Rosin L Perez G Peretz A Ben Tov A , et al . (2021) Kinetics of SARS-CoV-2 anti-S IgG after BNT162b2 vaccination. Vaccine 39 :5337-5340.34393018
Hornung V Rothenfusser S Britsch S Krug A Jahrsdörfer B Giese T Endres S Hartmann G (2002) Quantitative expression of toll-like receptor 1-10 mRNA in cellular subsets of human peripheral blood mononuclear cells and sensitivity to CpG oligodeoxynucleotides. J Immunol 168 :4531-4537.11970999
Issah Y Naik A Tang SY Forrest K Brooks TG Lahens N Theken KN Mermigos M Sehgal A Worthen GS , et al . (2021) Loss of circadian protection against influenza infection in adult mice exposed to hyperoxia as neonates. Elife 10 :e61241.33650487
Khoury DS Cromer D Reynaldi A Schlub TE Wheatley AK Juno JA Subbarao K Kent SJ Triccas JA Davenport MP (2021) Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection. Nat Med 27 :1205-1211.34002089
Kurupati RK Kossenkoff A Kannan S Haut LH Doyle S Yin X Schmader KE Liu Q Showe L Ertl HCJ (2017) The effect of timing of influenza vaccination and sample collection on antibody titers and responses in the aged. Vaccine 35 :3700-3708.28583307
Long JE Drayson MT Taylor AE Toellner KM Lord JM Phillips AC (2016a) Corrigendum to “Morning vaccination enhances antibody response over afternoon vaccination: A cluster-randomised trial” [Vaccine 34 (2016) 2679-2685]. Vaccine 34 :4842.27543455
Long JE Drayson MT Taylor AE Toellner KM Lord JM Phillips AC (2016b) Morning vaccination enhances antibody response over afternoon vaccination: A cluster-randomised trial [published correction appears in Vaccine]. Vaccine 34 :2679-2685.27129425
Müller L Kannenberg J Biemann R Hönemann M Ackermann G Jassoy C (2022) Comparison of the measured values of quantitative SARS-CoV-2 spike antibody assays. J Clin Virol 155 :105269.36029637
Netea MG Domínguez-Andrés J van de Veerdonk FL van Crevel R Pulendran B van der Meer JWM (2022) Natural resistance against infections: Focus on COVID-19. Trends Immunol 43 :106-116.34924297
Nobis CC Dubeau Laramée G Kervezee L Maurice De Sousa D Labrecque N Cermakian N (2019) The circadian clock of CD8 T cells modulates their early response to vaccination and the rhythmicity of related signaling pathways. Proc Natl Acad Sci U S A 116 :20077-20086.31527231
Petrovsky N Harrison LC (1998) The chronobiology of human cytokine production. Int Rev Immunol 16 :635-649.9646180
Phillips AC Gallagher S Carroll D Drayson M (2008) Preliminary evidence that morning vaccination is associated with an enhanced antibody response in men. Psychophysiology 45 :663-666.18346041
Scheiermann C Kunisaki Y Frenette PS (2013) Circadian control of the immune system. Nat Rev Immunol 13 :190-198.23391992
Sengupta S Brooks TG Grant GR FitzGerald GA (2021) Accounting for time: Circadian rhythms in the time of COVID-19. J Biol Rhythms 36 :4-8.32875944
Sengupta S Tang SY Devine JC Anderson ST Nayak S Zhang SL Valenzuela A Fisher DG Grant GR López CB , et al . (2019) Circadian control of lung inflammation in influenza infection. Nat Commun 10 :4107.31511530
Shimba A Cui G Tani-Ichi S Ogawa M Abe S Okazaki F Kitano S Miyachi H Yamada H Hara T , et al . (2018) Glucocorticoids drive diurnal oscillations in T cell distribution and responses by inducing interleukin-7 receptor and CXCR4. Immunity 48 :286-298.e6.29396162
Steensels D Pierlet N Penders J Mesotten D Heylen L (2021) Comparison of SARS-CoV-2 antibody response following vaccination with BNT162b2 and mRNA-1273. JAMA 326 :1533-1535.34459863
Tré-Hardy M Cupaiolo R Wilmet A Antoine-Moussiaux T Della Vecchia A Horeanga A Papleux E Vekemans M Beukinga I Blairon L (2021) Immunogenicity of mRNA-1273 COVID vaccine after 6 months surveillance in health care workers: A third dose is necessary. J Infect 83 :559-564.34437927
Wang W Balfe P Eyre DW Lumley SF O’Donnell D Warren F Crook DW Jeffery K Matthews PC Klerman EB , et al . (2022) Time of day of vaccination affects SARS-CoV-2 antibody responses in an observational study of health care workers. J Biol Rhythms 37 :124-129.34866459
Widge AT Rouphael NG Jackson LA Anderson EJ Roberts PC Makhene M Chappell JD Denison MR Stevens LJ Pruijssers AJ , et al . (2021) mRNA-1273 Study Group. Durability of responses after SARS-CoV-2 mRNA-1273 vaccination. N Engl J Med 384 :80-82.33270381
Wyse C O’Malley G Coogan AN McConkey S Smith DJ (2021) Seasonal and daytime variation in multiple immune parameters in humans: Evidence from 329,261 participants of the UK Biobank cohort. iScience 24 :102255.33817568
Zhang H Liu Y Liu D Zeng Q Li L Zhou Q Li M Mei J Yang N Mo S , et al . (2021) Time of day influences immune response to an inactivated vaccine against SARS-CoV-2. Cell Res 31 :1215-1217.34341489
Zhuang X Magri A Hill M Lai AG Kumar A Rambhatla SB Donald CL Lopez-Clavijo AF Rudge S Pinnick K , et al . (2019) The circadian clock components BMAL1 and REV-ERBα regulate flavivirus replication. Nat Commun 10 :377.30670689
| 36154515 | PMC9726636 | NO-CC CODE | 2022-12-08 23:18:17 | no | J Biol Rhythms. 2022 Dec; 37(6):700-706 | utf-8 | J Biol Rhythms | 2,022 | 10.1177/07487304221124661 | oa_other |
==== Front
Work Occup
Work Occup
WOX
spwox
Work and Occupations
0730-8884
1552-8464
SAGE Publications Sage CA: Los Angeles, CA
10.1177/07308884221143063
10.1177_07308884221143063
Original Research Article
Precarious Employment and Well-Being: Insights from the COVID-19 Pandemic
https://orcid.org/0000-0001-5416-8499
Mai Quan D. 1
Song Lijun 2
https://orcid.org/0000-0003-1374-0131
Donnelly Rachel 2
1 Rutgers University, New Brunswick, NJ, USA
2 5718 Vanderbilt University , Nashville, TN, USA
Quan D. Mai, Department of Sociology, 26 Nichol Ave., New Brunswick, NJ 08901, USA. Email: [email protected]
4 12 2022
4 12 2022
07308884221143063© The Author(s) 2022
2022
SAGE Publications
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
While precarious employment is not a new concept, it has been brought to the center of scholarly and public discourse worldwide by the unprecedented COVID-19 pandemic. This essay delineates how precarious employment shapes well-being and situates that relationship in the context of the COVID-19 pandemic. The essay also provides an overview of how the nine articles boldly investigate how these two layers of global risk—precarious employment and the pandemic—interact to shape individuals’ well-being. In addition to advancing theoretical and empirical knowledge by analyzing timely data from diverse sources and populations, these articles call for more efforts on worker protection reforms and government financial support.
precarious employment
well-being
COVID-19 pandemic
edited-statecorrected-proof
typesetterts19
==== Body
pmcIn recent decades, a series of political, economic, technological, and cultural factors jointly propelled the rise of precarious employment in both the developed and developing world. Following Kalleberg (2009), we define precarious work as employment arrangements that are uncertain, unpredictable, and risky for workers. Against a backdrop of the shift from the Keynesian model of state intervention to neoliberal economic policies, scholars have attributed the proliferation of precarious work to a variety of factors such as the weakening of labor unions, the financialization of the economy, the digitalization of the workforce, and the rise of the discourse of individualism and personal responsibility (Kalleberg 2009, Kalleberg and Vallas 2017, Mai 2018, Morgan and Nelligan 2018, Vallas and Cummins 2015, Vallas and Schor 2020, Western and Rosenfeld 2011). These structural shifts have allowed employers to maintain a flexible workforce, increase the use of nonstandard workers, benefit from relaxed employment standards, and have great discretion in hiring and firing. The emboldening of business and capital corresponds directly to the weakening of working class power. Workers are exposed to more low-paid and short-term jobs, bear more risks that were previously assumed by governments or businesses, and have less access to collective representation. The advent of the precariat as a social class (Standing 2011) parallels the emergence of the risk society (Beck 2004), in which uncertainty and insecurity become core features of the lived experiences of millions of workers across the globe (Bauman 2013, Hacker 2019).
The surge in precarity and its repercussions have inspired a broad stream of scholarly work, policy discussion, and social activism. Earlier scholarship probing the ramifications of precarious work focused on employment-related elements such as pay, benefits, and job satisfaction (Aletraris 2010, Kalleberg et al. 2000, Ko and Yeh 2013). The literature on the impact of precarious employment has since expanded beyond strictly employment-related outcomes to analyze how nonstandard work influences different aspects of workers’ “precarious lives” (Kalleberg 2018). Emerging scholarship in this line of inquiry has documented how precarious employment generates significant consequences for various measures of well-being such as physical and mental health (Benach et al. 2016, Donnelly 2022, Donnelly et al. Forthcoming, Gevaert et al. 2021, Glavin et al. 2021, Macmillan and Shanahan 2021, Mai et al. 2019), prospects for career advancement and social mobility (Ayala-Hurtado 2022, Mai 2021, Pedulla 2016), ability to balance work and family life (Choper et al. 2022, Lim 2017, Rao 2017, Schneider et al. 2019), and sense of identity (Vallas and Christin 2018). This area of scholarly inquiry continues to flourish and is becoming a key body of literature in sociology, social psychology, industrial relations, management, and health sciences. In this special issue, we consider well-being to encompass mental and physical health, socioeconomic status and mobility, work-life balance, and other factors that contribute to quality of life.
Since late 2019, the COVID-19 pandemic has wreaked havoc on the employment experiences of billions of employees around the world, threatening their well-being and harming their livelihoods. We argue that the pandemic has changed the experience of work precarity for many workers in different ways. First, some precarious workers suddenly became even more vulnerable. For example, employees’ already insecure working circumstances in restaurants, pubs, and movie theaters became more uncertain because of sudden layoffs and benefit cuts. As a result of their contracts not being renewed, many self-employed workers, independent contractors, gig workers, and freelancers faced further disruption to their income streams. It is also likely that many other workers in positions with limited autonomy and more uncertainty experienced heightened vulnerability during this time. Second, a new class of precarious workers emerged in the form of essential frontline service personnel who had frequent close interactions with customers, often without adequate safety measures. The “essential worker” language emerged largely during the pandemic and includes employees in groceries, retail stores, transportation, healthcare, nursing homes, and delivery services. Workers in essential frontline roles have been especially vulnerable, not only to the COVID-19 virus but also to overworking, lack of schedule control, and minimal benefits. Third, the pandemic placed employees in “good” occupations in a vulnerable position. Millions of highly skilled and highly paid full-time workers suddenly experienced a sense of precarity after being temporarily laid off or compelled to work for fewer hours, often for an undetermined period.
Given the widespread social and economic upheaval associated with the pandemic, it is critical to explore how the changes in working conditions and the general uncertainties have influenced precarious workers’ well-being, exacerbated existing dimensions of social stratification, and possibly inspired public policy debates and interventions. Several questions are worth asking at this juncture, for example: How do employment-related stress and pandemic-induced uncertainty combine to affect workers’ health? How do variations in implementation of workplace safety protocols impact workers’ physical and mental well-being? What are the roles of management and unions in shaping workers’ response to employment uncertainty caused by the pandemic? How do workers mobilize individual and family resources to cope with the risk of employment precarity and illness? How does the relationship between work and well-being vary cross-nationally and across various dimensions of social stratification (race/ethnicity, gender, sexuality, class, disability status, etc.)? Despite some notable studies, there is a dire need for innovative theoretical and empirical visions on how the pandemic influenced the well-being of precarious workers. This gap in the existing literature is understandable and, in some ways, expected because the pandemic is a relatively recent phenomenon. At the time of writing, we are still in the midst of a public health crisis with rising cases in many parts of the world.
This special issue's primary intention is to push the field forward by filling the gap in research on how the pandemic changed the contours of the relationship between precarious work and well-being. In the remainder of this introductory essay, we provide a brief overview of the state of the literature on the association between precarious work and well-being. We subsequently describe how the pandemic might have complicated that relationship. We then discuss in detail how the innovative articles published in this two-part special issue offer distinctive perspectives on the interplay between work, well-being, and the pandemic.
Precarious Employment and Well-Being
The scholarship on the consequences of precarious work has generated substantial evidence that this mode of employment affects workers’ well-being. In a very comprehensive and forward-looking article, Benach et al. (2016) set the agenda for research on precarious work until 2025. They described several mechanisms through which precarious work might lead to adverse health outcomes and poor life quality. The authors explained that precarious workers experience higher exposure to poor working conditions, psychosocial stressors associated with low work autonomy and powerlessness, and various issues related to managing professional and personal lives. The authors shrewdly called for more research exploring a “detailed understanding of the pathways and mechanisms” (p.233) linking precarious work and health. Since then, scholars have generated further evidence on how these three mechanisms operate to shape the linkage between precarious work and well-being. The following section offers a few examples of such recent evidence. We focus on research published since 2016.
Social marginalization can operate as a key mechanism linking precarious work to poor health. Macmillan and Shanahan (2021) argued that precarious employment undermines self-efficacy, reduces social integration, and decreases social capital more broadly. Precariously employed workers are predicted to lack work autonomy, have a low capacity for social integration at work, and have fragmented social networks beyond the workplace. These factors jointly undermine their mental health. Relatedly, Glavin et al. (2021) demonstrated how freelancing work through an app-based service company fostered a sense of alienation and powerlessness. The high level of algorithmic control embedded in platforms limits worker interaction with employers and customers, obfuscates information about job availability, and intensifies competition through various forms of customer evaluation and gamification. These strategies constrain workers autonomy and generate a sense of isolation, which are detrimental to workers’ mental well-being.
In addition to social marginalization as a mechanism, the uncertainty embedded in precarious work also has important implications for workers’ ability to plan for their personal lives, particularly starting a family and having children. Lim (2017) showed that precarious employment arrangements are “bad jobs” for marriage: all indicators of precarious work decrease the odds of first marriage by up to 40% for men, while having part-time employment delays entry into first marriage for women. The inherent uncertainty in employment makes planning for romantic relationships and potential marriage prospects challenging (Rao 2017). Relatedly, unpredictable schedules—a common feature of nonstandard employment—pose significant challenges to workers’ ability to organize their lives. Relying on data from the Shift Project—an innovative dataset on retail and food service workers—scholars have demonstrated that on-call shifts, shift timing changes, work hour volatility, and short advance notice make workers struggle to arrange childcare and increase work-life conflict (Harknett et al. 2022, Luhr et al. 2022). Given the benefits of marriage for health (Umberson and Karas Montez 2010) and the detriment of work-life conflict (Bianchi and Milkie 2010), social disruption could be a key mechanism linking precarious work to mental and physical health.
Intimately linked with difficulties in managing personal lives is financial hardship. It is hardly surprising that, relative to their full-time counterparts, precarious workers generally receive lower income and higher levels of income volatility. Material hardship associated with precarious employment has implications for workers’ asset accumulation, housing quality, and access to healthcare. These factors in turn erode health and quality of life. Recent studies have added more evidence to document the dire financial picture associated with precarious employment. Schneider and Harknett (2021) documented that workers with schedule uncertainty are more likely to experience hunger, as well as overall hardship including residential, medical, and utility hardship. Using in-depth interviews with freelancing creative workers, Butler and Stoyanova Russell (2018) found that comedians regularly suppress their financially induced anxiety to keep a constant flow of gigs and to maintain relationships with producers even when the pay rate is low and the promptness of payment is questionable. Ferrante et al. (2019) found that for Italian men precarious work impacts mental health primarily through financial strain.
Beyond their personal lives, precarious employment might also threaten workers’ socioeconomic well-being by hindering the progression of their professional careers. Whether histories of nonstandard employment operate as “stepping-stone” or “dead-end” constitutes a significant debate. In their meta-analysis of studies that investigated the impact of temporary job on subsequent labor market performances, Filomena and Picchio (2022) reported that out of 78 observations from 64 articles, 45% provide evidence in favor of the dead-end hypothesis, while 23% report ambiguous or mixed findings. The authors cautiously noted that research in this literature typically compares careers of workers who experienced precarity to careers of the ones who did not using non-randomized observational data. In recent years, studies using experimentally designed randomization also shed light on this topic. Pedulla (2016) that part-time employment history is as scarring for male workers as a year of unemployment. Mai (2021) reported that a freelancing work history reduces workers’ odds of being invited to a job interview for a full-time position by about 30 percent. Altogether, this evidence illustrates the cost in terms of subsequent employment prospects as precarious workers navigate out of precarity to embrace organizational careers.
In sum, existing work shows that precarious employment affects workers’ well-being by exposing them to hazardous working conditions, threatening their sense of self and social relations, obscuring their ability to plan for their personal lives, increasing their financial hardship, and diminishing their career prospects. In the following section, we explore how the pandemic has complicated the relationship between work and well-being.
Work and Well-Being During the Pandemic
The COVID-19 pandemic affected the landscape of work for almost all workers. At the start of the pandemic, unemployment in the United States surged to 14.7% in April 2020 (Bureau of Labor Statistics 2020) as large sectors of the economy shut down. Even among workers who remained employed, wage loss was endemic. Evidence suggests that 60% of workers who retained their jobs experienced a wage cut or wage freeze in the first months of the pandemic (Cajner et al. 2020). In addition to changes in employment status and wages, the nature of work changed considerably. For example, a new category of worker—essential worker—emerged to describe workers critical to the functioning of society and who continued to work in person (CISA 2020). On the other hand, many workers who had the ability to telework began working remotely; about 40% of working-age Americans worked from home at the start of the pandemic (Pew Research Center, 2020). Overall, the COVID-19 pandemic created a unique landscape of work and many of these changes have endured beyond the initial months of the crisis.
Although most workers have been impacted by the COVID-19 pandemic in some way, the impact of the pandemic could be particularly profound among precariously employed workers. For instance, workers in nonstandard employment arrangements typically have less power, less security, and fewer protections than workers in standard employment relationships (Benach et al. 2016, Hacker 2019). As such, the pandemic likely added uncertainty to already precarious working conditions. Moreover, this heightened vulnerability has occurred at a time when many countries have spent decades rolling back social safety nets. Precarious workers, then, may be a vulnerable population with concerning trajectories of well-being during and after the pandemic.
This special issue focuses on experiences related to work and employment, especially among precariously employed workers, as a key determinant of well-being during the pandemic. Indeed, mental health and well-being worsened considerably, during the COVID-19 pandemic such that rates of depression and anxiety in the United States were approximately four times higher in April–June 2020 compared to a similar period in 2019 (Czeisler et al. 2020, Ettman et al. 2020). Scholars speculated that the unparalleled transformation of work resulted in numerous challenges that likely diminished well-being. Indeed, recent research shows that income loss and financial insecurity increased the risk of depression and anxiety during the pandemic (e.g., Donnelly and Farina (2021); Zheng et al. (2021)). Among workers who remained employed, new challenges emerged that could undermine mental health. Essential workers, for example, experienced more mental health concerns than non-essential workers (e.g., Bell et al. (2021)Mayer et al. (2022)) perhaps due to the increased risk of COVID-19 infection at work, greater exposure to overwork, and minimal benefits. While this initial research suggests that work-related changes have undermined mental health and well-being during the pandemic, much remains to be known about the many aspects of employment, especially precarious employment, that could affect well-being during the pandemic.
The possible mechanisms linking precarious work to well-being during the COVID-19 pandemic are likely numerous and multifaceted. One possibility is that the well-documented mechanisms outlined above (i.e., psychosocial stressors, difficulty managing work and nonwork life, financial hardship) would still be relevant but may, in fact, be more salient during this time. For example, financial hardship among the precariously employed could be more consequential if workers had to miss work due to COVID-19 exposure and illness within their household and/or had additional expenses related to protective equipment or food for children when schools were closed. Another possibility is that new mechanisms linking precarious work and well-being emerged during the pandemic. For instance, precarious workers may have less ability to protect themselves and their family members from COVID-19 exposure at work and/or have less power to advocate for a safe working environment. Together, these mechanisms could diminish well-being among workers. However, existing studies have not yet tested specific mechanisms linking precarious work and well-being during the pandemic.
An important consideration is that work experiences have been marked by inequity during the pandemic. Racially minoritized adults, women, and adults with less education were more likely to lose their jobs in the initial months of the pandemic (Kochhar 2021, Moen et al. 2020). Moreover, many workers, primarily women, were forced to leave the labor force or reduce their work hours in the absence of childcare and in-person schooling (Collins et al. 2021a, Collins et al. 2021b). Compared to non-essential workers, essential workers are more likely to be racially minoritized adults with lower educational attainment and wages (Blau et al. 2021). Thus, while the pandemic has affected most workers, the burden of adverse work experiences likely falls largely on marginalized populations.
Inequities in work and stressful work-related experiences have dire consequences for well-being. Indeed, prior research posits that inequities in work, stress, and adverse experiences contribute to inequities in mental health and well-being (see Pearlin et al. (2005) Notably, inequities in work-related experiences by gender, race/ethnicity, and socioeconomic status have occurred alongside other pandemic-related inequities such as higher rates of COVID-19 cases and mortality and more instances of discrimination and harassment among racially minoritized individuals (Andrasfay and Goldman 2021, Garcia et al. 2021, Laurencin and Walker 2020). The unequal landscape of work and other stressors during the pandemic could fuel inequities in mental health and well-being—a possibility that must be explored.
Given the changes in work occurring alongside a rise in depression, anxiety, and other mental health concerns, it is necessary to advance scholarship on work and well-being during the COVID-19 pandemic. While the literature on work and well-being is robust and spans countless disciplines, the pandemic offers an opportunity to test and refine existing theories during a unique historical period and to shed light on the experiences of marginalized workers. Thus, in this special issue, we aim to understand linkages between work and well-being across the world during the pandemic in ways that can inform existing theories and future research.
Insights on Two Layers of Global Risk: Precarious Employment in the COVID-19 Crisis
This two-part special issue centers on the coexistence of two layers of global risk: the risk of precarious employment in the face of the COVID-19 pandemic. This special issue grew out of a virtual international conference, “Precarious Employment and Well-Being During the COVID-19 Pandemic,” which was co-sponsored by Rutgers and Vanderbilt Universities and held on January 21, 2022. The issue organizes nine papers thematically into two parts. Part I comprises four quantitative articles related to “Market Conditions, Employment Quality, and Workplace Politics” and appears in the February 2023 issue of this journal. Part II includes three quantitative and two mixed-method papers tackling the theme of “The Experience and Perception of Employment Precarity” and is scheduled for publication in the May 2023 issue of Work and Occupations. Part I begins with Alon's challenge to the conceptualization and operationalization of precarious work (Alon Forthcoming), followed by two articles on job quality (Reynolds and Kincaid Forthcoming, Rho et al. Forthcoming) and one article on workplace politics (Woods et al. Forthcoming). As Alon points out, while being useful and popular for its simultaneous consideration of employment instability and employment-contingent outcomes, a comprehensive definition of precarious work limits research to only employed individuals. This limitation can create two issues (sample selection bias and truncated heterogeneity of employment instability), which overlook labor force dynamics and lead to biased estimates of precarious work and its consequences. Considering that recessions aggravate these two issues, Alon identifies the pandemic as a unique opportunity for the appraisal of the two issues via the comparison of two market conditions (pre-COVID-19 near full employment era and COVID-19 recession era) among the entire working-age population. Her analysis of multiple years of cross-sectional survey data in Israel demonstrates the presence of these two issues. Employment instability is higher in the COVID-19 recession era than the pre-COVID-19 era. During the COVID-19 recession era, employment instability is negatively associated with the probability of employment, and employment status is in turn negatively associated with economic insecurity at all levels of employment instability, whereas these patterns do not appear during the pre-COVID-19 era. Alon ends her article with a call for a conceptual shift from precarious work to a more inclusive construct, precarious employment, and more effective estimates of “the true scope of employment precarity.”
Precarious work comes in diverse forms. One of the relatively new but increasingly prevalent forms is gig work mediated by digital platform technologies. Theoretically speaking, gig work has both advantages and disadvantages and scholars debate whether its benefits can offset its drawbacks. The COVID-19 crisis adds another puzzling layer to this debate. Competing perspectives are possible on whether the pandemic pushes gig workers toward or away from gig work and whether increased gig working hours increase financial returns. Reynolds and Kincaid examine these competing perspectives with their focus on one specific type of gig work, microtask work. As their analysis of panel data from U.S. workers on Amazon's Mechanical Turk platform shows, one third of existing workers increased their microtask hours, especially for those who lost household income or wage/salary hours. However, supporting the platform-dependence and precarity perspectives, increased working hours generated little financial returns, especially for those without other sources of income. Reynolds and Kincaid echo the warning that “gig work is not a viable substitute for the social safety net” and the necessity for worker protection reforms.
One major characteristic of precarious work is job insecurity. Job insecurity is harmful and unarguably exacerbated by the pandemic. It is unclear, however, how job insecurity is associated with workers’ response during the COVID-19 recession and how workplace factors affect that association. Rho and colleagues address these research gaps and center their focus on workers’ voice and three workplace factors (unions, managers, and employment arrangements). Results from a sample of Illinois and Michigan workers do not support the job insecurity as a stressor perspective such that job insecurity is not associated with voice. Despite their positive associations with voice, unions and managerial receptiveness do not serve as a stress buffer. That is, insecure workers remain less likely to exercise voice than secure workers as confidence in organized labor or perception of managerial receptiveness increases. Consistent with the less investment perspective, insecure nonstandard workers are less likely to speak up than their secure counterparts, whereas this pattern does not appear among standard workers.
Precarious work and the pandemic do not exist in a political vacuum. Instead, they co-exist in a world with intense political polarization. Among others, issues on science and COVID-19 are highly politically polarized. Woods, Schneider, and Harknett investigate the unanswered question of how political polarization shapes the relationship between COVID-19 workplace safety measures and individual workers’ well-being. They use an underexamined measure of political polarization (support for political leaders) and analyze a sample of U.S. frontline service sector workers, who are more proximate to the risks of COVID-19 in their workplaces. They find that Biden voters react more to workplace risks than Trump voters. The positive associations between COVID-19 workplace safety measures and workers’ feeling of safety and mental well-being apply only to Biden voters.
In Part II, five articles center on the experiences and perception of employment precarity during the pandemic. Three quantitative studies show stressful dark sides of precarity during the pandemic (Brown and Ciciurkaite Forthcoming, Grace Forthcoming, Wu Forthcoming) whereas two mixed-method papers suggest possible temporary bright sides(Ravenelle and Kowalski Forthcoming, Schieman et al. Forthcoming). As for the harmful dark sides, some populations bear multiple sources of stress and their high stress levels are further exacerbated by the pandemic. Brown and Ciciurkaite explore simultaneously three sources of stress (employment precarity, disability, and discrimination) and their main and interactive effects on depressive symptoms. They use community survey data in the United States and measure employment precarity as one single factor based on ten items, which include employment challenges and three aspects of uncertainty (contract, development, and income). Consistent with the minority stress framework and the stress process model, each source of stress is positively associated with depressive symptoms, and the dual stressful effects of employment precarity and discrimination are stronger for those with disabilities. Brown and Ciciurkaite encourage more policy and research efforts on the marginalized disabled working population.
Different from Brown and Ciciurkaite, who treat employment precarity as one single construct, Wu emphasizes the distinction between its objective and subjective dimensions. Employing survey data on workers from 27 E.U. member states, Wu captures the objective dimension of precarity as contractual instability and the subjective dimension using job insecurity and emotional precariousness. She also measures an additional stressor, the COVID-19 risk at work. In line with the stress process model, overall employment precarity is negatively associated with mental and subjective well-being. After it is decomposed into its two dimensions, only the subjective dimension plays a significant role. Between the two subjective indicators, emotional precariousness exerts a stronger impact. Contrary to the stress process model, the COVID-19 risk at work is protective of well-being outcomes net of indicators of employment precarity. Wu speculates possible explanations such as increased social integration.
According to the stress process model, stress is a process involving stress proliferation (from primary to secondary stressors) and coping resources. Grace applies this model fully to the relationship between a primary stressor (job displacement due to COVID-19) and mental well-being, using national survey data on U.S. adults. Furlough or job loss due to the pandemic (versus stable employment and job loss due to other reasons) is positively associated with depressive symptoms and anger partly via three pathways: two secondary stressors (financial strains and anticipatory stress about economic security) and coping resources. Among the three pathways, anticipatory worry about economic security is most salient. Grace urges more research on future-oriented stressors and the enforcement of worker protection policies.
The two mixed-method articles illustrate how the unique pandemic context manifests possible temporary bright sides through changing perceptions of employment precarity and time. In contrast to the stress process model, Schieman and colleagues propose the forced vacation perspective. This perspective argues that being temporarily laid off (versus continuous employment) early into the pandemic is buffered by the countervailing conditions of the earlier pandemic context and leads to a temporary, unforeseen, and forced vacation. They employ data from a national longitudinal survey of working adults and in-depth interviews in Canada. Their quantitative findings support the forced vacation perspective such that the association between temporary job disruption and psychological distress is initially negative and then becomes null. Their qualitative results suggest three possible explanations: temporary exemption from work stress, reduced internal attribution, and mitigated financial strain. Schieman and colleagues do warn us about the temporary quality of vacations and the generalizability of their findings. Temporary job disruption can become harmful as it turns chronic or in societies with fewer government supports.
The Great Resignation, as described by Ravenelle and Kowalski, exemplifies the dramatic impacts of the pandemic on the labor market. Its causes remain under heated debate. Among others, the work passion ideology explanation argues that work passion can inspire workers to leave their jobs and actively pursue more fulfilling and lucrative careers. Going beyond prior work that limited this ideology primarily to creative professionals, Ravenelle and Kowalski demonstrate its adoption among low-wage precarious workers during the pandemic. In their mixed-method panel study of precarious and gig-based service workers in the New York metropolitan area, they document these workers’ self-reflection process from the pandemic's earlier period to its second surge. Under the condition of relative financial security generated by unprecedented government financial assistance, workers shift their experiences of time from “spending time” instrumentally for money to “investing time” in themselves for personal growth. They change their conceptualization of time from a “use-mindset” to an “investment-mindset.” They move their goal beyond employability to prioritize work passion. With the ending of government financial supports, Ravenelle and Kowalski are concerned about the continuity and realization of the work passion logic and emphasize the need for future long-term research.
While precarious work is as old as capitalism, it is brought to the center of scholarly and public discourse worldwide by the unprecedented COVID-19 pandemic. It is an unfortunate but unique and urgent research opportunity for us to understand how these two layers of global risk—precarious work and the pandemic—interplay with each other to shape people's well-being. The nine cutting-edge articles in this special issue boldly pursue this research opportunity. As a whole, they employ unique timely data—quantitative and/or qualitative—from diverse sources, populations, and regions or societies. They advance our theoretical and empirical knowledge. In brief, as they suggest, we should adopt the more inclusive concept of precarious employment, compare the costs and benefits of different forms of precarious work, listen to precarious workers’ voice within different workplaces, analyze the role of politics and political polarization, pay more attention to marginalized working populations facing multiple sources of stress, distinguish the objective and subjective dimensions of employment precarity, lay out the stress proliferation process, and identify the temporary bright or beneficial consequences. The policy implications are clear and consistent. More efforts are needed for worker protection reforms and government financial supports.
Acknowledgments
The authors are grateful for support from Daniel B. Cornfield, the journal's editor-in-chief, whose insights were vital in the production of the special issue. The authors also thank participants in the virtual mini-conference entitled Precarious Employment and Well-Being During the COVID-19 Pandemic: A Special Issue and Mini-Conference, co-hosted by Rutgers and Vanderbilt Universities on January 21, 2022. We also appreciate the valuable support from the Assistant Guest Editors, Darwin Baluran and Rachel Zajdel.
Author Biographies
Quan D. Mai is an Assistant Professor of Sociology at Rutgers University. His research explores the interplay of work, race, and space in shaping patterns of social stratification. His work has appeared in American Journal of Sociology, Social Forces, and Social Science & Medicine.
Lijun Song is an Associate Professor of Sociology at Vanderbilt University. Her major research interests include social networks, medical sociology, and social stratification. Her work has appeared in Social Forces, Journal of Health and Social Behavior, Society and Mental Health, Social Psychology Quarterly, Social Science and Medicine, and Social Networks.
Rachel Donnelly is Assistant Professor of Sociology at Vanderbilt University. Her research focuses on determinants of health across the life course, with an emphasis on stress, work, and family. With support from the National Institutes of Health, she examines how stressors and state-level policies jointly shape inequities in mental health.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article
ORCID iDs: Quan D. Mai https://orcid.org/0000-0001-5416-8499
Rachel Donnelly https://orcid.org/0000-0003-1374-0131
==== Refs
References
Aletraris L. (2010). How satisfied are they and why? A study of job satisfaction, job rewards, gender and temporary agency workers in Australia. Human Relations, 63 (8 ), 1129–1155. 10.1177/0018726709354131
Alon S. (Forthcoming). The measurement of precarious work and market conditions: Insights from the COVID-19 disruption on sample selection. Work and Occupations, 0 (0 ), 10.1177/07308884221127636
Andrasfay T. Goldman N. (2021). Reductions in 2020 us life expectancy due to COVID-19 and the disproportionate impact on the black and latino populations. Proceedings of the National Academy of Sciences, 118 (5 ), e2014746118. 10.1073/pnas.2014746118
Ayala-Hurtado E. (2022). Narrative continuity/rupture: Projected professional futures amid pervasive employment precarity. Work and Occupations, 49 (1 ), 45–78. 10.1177/07308884211028277
Bauman Z. (2013). Liquid Modernity. John Wiley & Sons.
Beck U. (2004). Ulrich Beck: A Critical Introduction to Risk Society. JSTOR.
Bell C. Williman J. Beaglehole B. Stanley J. Jenkins M. Gendall P. Rapsey C. Every-Palmer S. (2021). Challenges facing essential workers: A cross-sectional survey of the subjective mental health and well-being of New Zealand healthcare and ‘other’ essential workers during the COVID-19 lockdown. BMJ open, 11 (7 ), http://dx.doi.org/10.1136/bmjopen-2020-048107
Benach J. Vives A. Tarafa G. Delclos C. Muntaner C. (2016). What should we know about precarious employment and health in 2025? Framing the agenda for the next decade of research. International Journal of Epidemiology, 45 (1 ), 232–238. 10.1093/ije/dyv342 26744486
Bianchi S. M. Milkie M. A. (2010). Work and family research in the first decade of the 21st century. Journal of Marriage and Family, 72 (3 ), 705–725. 10.1111/j.1741-3737.2010.00726.x
Blau F. D. Koebe J. Meyerhofer P. A. (2021). Who are the essential and frontline workers? Business Economics, 56 (3 ), 168–178. 10.1057/s11369-021-00230-7 34253931
Brown R. L. Ciciurkaite G. (Forthcoming). Precarious employment during the COVID-19 pandemic, disability-related discrimination, and mental health. Work and Occupations, 0 (0 ), 07308884221129839. 10.1177/07308884221129839
Bureau of Labor Statistics (2020). Unemployment rate rises to record high 14.7 percent in April 2020. The Economics Daily.
Butler N. Stoyanova Russell D. (2018). No funny business: Precarious work and emotional labour in stand-up comedy. Human Relations, 71 (12 ), 1666–1686. 10.1177/0018726718758880
Cajner T. Crane L. D. Decker R. A. Grigsby J. Hamins-Puertolas A. Hurst E. Kurz C. Yildirmaz A. (2020). “The Us Labor Market During the Beginning of the Pandemic Recession.” Vol.: National Bureau of Economic Research.
Choper J. Schneider D. Harknett K. (2022). Uncertain time: Precarious schedules and job turnover in the us service sector. ILR Review, 75 (5 ), 1099–1132. 10.1177/00197939211048484
CISA (2020). “Identifying Critical Infrastructure During Covid-19.”.
Collins C. Landivar L. C. Ruppanner L. Scarborough W. J. (2021a). COVID-19 and the gender gap in work hours. Gender, Work & Organization, 28 (S1) , 101–112. 10.1111/gwao.12506 32837019
Collins C. Ruppanner L. Landivar L. C. Scarborough W. J. (2021b). The gendered consequences of a weak infrastructure of care: School reopening plans and Parents’ employment during the COVID-19 pandemic. Gender & Society, 35 (2 ), 180–193. 10.1177/08912432211001300
Czeisler M. É. Lane R. I. Petrosky E. Wiley J. F. Christensen A. Njai R. Weaver M. D. Robbins R. Facer-Childs E. R. Barger L. K. (2020). Mental health, substance use, and suicidal ideation during the COVID-19 pandemic—United States, June 24–30, 2020. Morbidity and Mortality Weekly Report, 69 (32 ), 1049. 10.15585/mmwr.mm6932a1 32790653
Donnelly R. (2022). Precarious work in midlife: Long-term implications for the health and mortality of women and men. Journal of Health and Social Behavior, 63 (1 ), 142–158. 10.1177/00221465211055090 34794348
Donnelly R. Farina M. P. (2021). How do state policies shape experiences of household income shocks and mental health during the COVID-19 pandemic? Social Science & Medicine, 269 , 113557. 10.1016/j.socscimed.2020.113557 33308909
Donnelly R. Zajdel R. Farina M. P. (Forthcoming). Inequality in household job insecurity and mental health: Changes during the COVID-19 pandemic. Work and Occupations, 0 (0 ), https://doi.org/10.1177/07308884221123255
Ettman C. K. Abdalla S. M. Cohen G. H. Sampson L. Vivier P. M. Galea S. (2020). Prevalence of depression symptoms in us adults before and during the COVID-19 pandemic. JAMA network Open, 3 (9 ), e2019686–e86. 10.1001/jamanetworkopen.2020.19686 32876685
Ferrante G. Fasanelli F. Gigantesco A. Ferracin E. Contoli B. Costa G. Gargiulo L. Marra M. Masocco M. Minardi V. Violani C. Zengarini N. d’Errico A. Ricceri F. (2019). Is the association between precarious employment and mental health mediated by economic difficulties in males? Results from two Italian studies. BMC Public Health, 19 (1 ), 869. 10.1186/s12889-019-7243-x 31269944
Filomena M. Picchio M. (2022). Are temporary jobs stepping stones or dead ends? A systematic review of the literature. International Journal of Manpower, 43 (9 ), 60–74. 10.1108/IJM-02-2022-0064
Garcia M. A. Homan P. A. García C. Brown T. H. (2021). The color of COVID-19: Structural racism and the disproportionate impact of the pandemic on older black and Latinx adults. The Journals of Gerontology: Series B, 76 (3 ), e75–e80. 10.1093/geronb/gbaa114
Gevaert J. Van Aerden K. De Moortel D. Vanroelen C. (2021). Employment quality as a health determinant: Empirical evidence for the waged and self-employed. Work and Occupations, 48 (2 ), 146–183. 10.1177/0730888420946436
Glavin P. Bierman A. Schieman S. (2021). Über-Alienated: Powerless and alone in the gig economy. Work and Occupations, 48 (4 ), 399–431. 10.1177/07308884211024711
Grace M. K. (Forthcoming). The contributions of social stressors and coping resources to psychological distress among those who experienced furlough or job loss due to COVID-19. Work and Occupations, 0 (0 ), 10.1177/07308884221123325
Hacker J. S. (2019). The Great Risk Shift: The New Economic Insecurity and the Decline of the American Dream. Oxford University press.
Harknett K. Schneider D. Luhr S. (2022). Who cares if parents have unpredictable work schedules?: Just-in-time work schedules and child care arrangements. Social Problems, 69 (1 ), 164–183. 10.1093/socpro/spaa020
Kalleberg A. (2009). Precarious work, insecure workers: Employment relations in transition. American Sociological Review, 74 (1 ), 1–22. 10.1177/000312240907400101
Kalleberg A. (2018). Precarious Lives: Job Insecurity and Well-Being in Rich Democracies. Polity Press.
Kalleberg A. Reskin B. Hudson K. (2000). Bad jobs in America: Standard and nonstandard employment relations and job quality in the United States. American Sociological Review, 65(2), 256–278. 10.2307/2657440
Kalleberg A. Vallas S. (2017). Probing precarious work: Theory, research, and politics. In Kalleberg A. Vallas S. (Eds.), Precarious work, vol. 31, research in the sociology of work (pp. 1–30). Emerald Publishing Limited.
Ko J.-J. R. Yeh Y.-J. Y. (2013). Worker satisfaction following employment restructuring: Effects of nonstandard workers and downsizing on job satisfaction in Taiwan. Social Indicators Research, 110 (2 ), 453–467. 10.1007/s11205-011-9937-z
Kochhar R. (2021). “Hispanic Women, Immigrants, Young Adults, Those with Less Education Hit Hardest by Covid-19 Job Losses. Pew Research Center. June 9, 2020.”
Laurencin C. T. Walker J. M. (2020). A pandemic on a pandemic: Racism and COVID-19 in blacks. Cell Systems, 11 (1 ), 9–10. 10.1016/j.cels.2020.07.002 32702320
Lim S. (2017). “Bad jobs” for marriage: Precarious work and the transition to first marriage. Research in the Sociology of Work, 31 , 399–427. 10.1108/S0277-283320170000031015
Luhr S. Schneider D. Harknett K. (2022). Parenting without predictability: Precarious schedules, parental strain, and work-life conflict. RSF: The Russell Sage Foundation Journal of the Social Sciences, 8 (5 ), 24–44. 10.7758/rsf.2022.8.5.02
Macmillan R. Shanahan M. J. (2021). Why precarious work is bad for health: Social marginality as key mechanisms in a multi-national context. Social Forces, 100 (2 ), 821–851. 10.1093/sf/soab006
Mai Q. (2018). Precarious work in Europe: Assessing cross-national differences and institutional determinants of work precarity in 32 European countries. Research in the Sociology of Work, 31 , 273–306. 10.1108/S0277-283320170000031009
Mai Q. Jacobs A. Schieman S. (2019). Precarious sleep? Nonstandard work, gender, and sleep disturbance in 31 European countries. Social Science & Medicine, 237 , 112424. 10.1016/j.socscimed.2019.112424 31400590
Mai Q. D. (2021). Unclear signals, uncertain prospects: The labor market consequences of freelancing in the new economy. Social Forces, 99 (3 ), 895–920. https://doi.org/10.1093/sf/soaa043
Mayer B. Arora M. Helm S. Barnett M. (2022). Essential but ill-prepared: How the COVID-19 pandemic affects the mental health of the grocery store workforce. Public Health Reports, 137 (1 ), 120–127. 10.1177/00333549211045817 34524904
Moen P. Pedtke J. H. Flood S. (2020). “Disparate disruptions: Intersectional COVID-19 employment effects by age, gender, education, and race/ethnicity. Work, Aging and Retirement, 6 (4 ), 207–228. 10.1093/workar/waaa013 33214905
Morgan G. Nelligan P. (2018). The Creativity Hoax: Precarious Work in the Gig Economy. Anthem Press.
Pearlin L. I. Schieman S. Fazio E. M. Meersman S. C. (2005). Stress, health, and the life course: Some conceptual perspectives. Journal of Health and Social Behavior, 46 (2 ), 205–219. 10.1177/002214650504600206 16028458
Pedulla D. (2016). Penalized or protected? The consequences of non-standard and mismatched employment histories. American Sociological Review, 81 (2 ), 262–289. 10.1177/0003122416630982 27182069
Pew Research Center (2020). “Most Americans Say Coronavirus Outbreak Has Impacted Their Lives.” March 2020 report.
Rao A. H. (2017). “You Don’t Dare Plan Much”: Contract work and personal life for international early-career professionals. Research in the Sociology of Work, 31 , 429–453.
Ravenelle A. J. Kowalski K. C. (Forthcoming). “It’s not like chasing channel:” spending time, investing in the self, and pandemic epiphanies. Work and Occupations, 0 (0 ), 10.1177/07308884221125246
Reynolds J. Kincaid R. (Forthcoming). Gig work and the pandemic: Looking for good pay from bad jobs during the COVID-19 crisis. Work and Occupations, 0 (0 ), 10.1177/07308884221128511
Rho H. J. Riordan C. Ibsen C. L. Lamare J. R. Tapia M. (Forthcoming). Do workers speak up when feeling job insecure? Examining workers’ response to precarity during the COVID-19 pandemic. Work and Occupations, 0(0).
Schieman S. Mai Q. Badawy P. Kang R. W. (Forthcoming). A forced vacation? The stress of being temporarily laid off during a pandemic. Work and Occupations, 0(0).
Schneider D. Harknett K. (2021). Hard times: Routine schedule unpredictability and material hardship among service sector workers. Social Forces, 99 (4 ), 1682–1709. 10.1093/sf/soaa079
Schneider D. Harknett K. Stimpson M. (2019). Job quality and the educational gradient in entry into marriage and cohabitation. Demography, 56 (2 ), 451–476. 10.1007/s13524-018-0749-5 30617947
Standing G. (2011). The Precariat: The New Dangerous Class. A&C Black.
Umberson D. Karas Montez J. (2010). Social relationships and health: A flashpoint for health policy. Journal of Health and Social Behavior, 51 (1_suppl ), S54–S66. 10.1177/0022146510383501 20943583
Vallas S. Cummins E. (2015). Personal branding and identity norms in the popular business press: Enterprise culture in an age of precarity. Organizational Studies, 36 (3 ), 293–319. 10.1177/0170840614563741
Vallas S. Schor J. B. (2020). What do platforms do? Understanding the gig economy. Annual Review of Sociology, 46 (1 ), 273–294. 10.1146/annurev-soc-121919-054857
Vallas S. P. Christin A. (2018). Work and identity in an era of precarious employment: How workers respond to “personal branding” discourse. Work and Occupations, 45 (1 ), 3–37. 10.1177/0730888417735662
Western B. Rosenfeld J. (2011). Unions, norms, and the rise in us wage inequality. American Sociological Review, 76 (4 ), 513–537. 10.1177/0003122411414817
Woods T. Schneider D. Harknett K. (Forthcoming). The politics of prevention: Polarization in how workplace COVID-19 safety practices shaped the well-being of frontline service sector workers. Work and Occupations, 0 (0 ), 10.1177/07308884221125821.
Wu Q. (Forthcoming). Employment precarity, COVID-19 risk, and workers’ well-being during the pandemic in Europe. Work and Occupations, 0 (0 ), 10.1177/07308884221126415.
Zheng J. Morstead T. Sin N. Klaiber P. Umberson D. Kamble S. DeLongis A. (2021). Psychological distress in North America during COVID-19: The role of pandemic-related stressors. Social Science & Medicine, 270 , 113687. 10.1016/j.socscimed.2021.113687 33465600
| 0 | PMC9726637 | NO-CC CODE | 2022-12-08 23:18:17 | no | Work Occup. 2022 Dec 4;:07308884221143063 | utf-8 | Work Occup | 2,022 | 10.1177/07308884221143063 | oa_other |
==== Front
J Biol Rhythms
J Biol Rhythms
JBR
spjbr
Journal of Biological Rhythms
0748-7304
1552-4531
SAGE Publications Sage CA: Los Angeles, CA
36124632
10.1177/07487304221123455
10.1177_07487304221123455
Original Articles
A Naturalistic Actigraphic Assessment of Changes in Adolescent Sleep, Light Exposure, and Activity Before and During COVID-19
Rynders Corey A. *†
Bowen Anne E. ‡
Cooper Emily §
Brinton John T. §
Higgins Janine §
Nadeau Kristen J. §
Wright Kenneth P. Jr. *‖
https://orcid.org/0000-0003-4755-8151
Simon Stacey L. §1
* Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, Colorado
† Department of Kinesiology, School of Education and Human Development, University of Virginia, Charlottesville, Virginia
‡ Children’s Hospital Colorado, Aurora, Colorado
§ Department of Pediatrics, University of Colorado Anschutz Medical Campus, Aurora, Colorado
‖ University of Colorado Boulder, Boulder, Colorado
1 Stacey L. Simon, Department of Pediatrics, University of Colorado Anschutz Medical Campus, 13123 E 16th Avenue, Box B395, Aurora, CO 80045, USA; e-mail: [email protected].
12 2022
12 2022
12 2022
37 6 690699
© 2022 The Author(s)
2022
SAGE Publications
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
The majority of high school–aged adolescents obtain less than the recommended amount of sleep per night, in part because of imposed early school start times. Utilizing a naturalistic design, the present study evaluated changes in objective measurements of sleep, light, and physical activity before (baseline) and during the first wave of the COVID-19 pandemic (during COVID-19) in a group of US adolescents. Sixteen adolescents (aged 15.9 ± 1.2 years, 68.8% female) wore an actigraphy monitor for 7 consecutive days during an in-person week of school before the pandemic (October 2018-February 2020) and again during the pandemic when instruction was performed virtually (May 2020). Delayed weekday sleep onset times of 1.66 ± 1.33 h (p < 0.001) and increased sleep duration of 1 ± 0.87 h (p < 0.001) were observed during COVID-19 compared with baseline. Average lux was significantly higher during COVID-19 compared with baseline (p < 0.001). Weekday physical activity parameters were not altered during COVID-19 compared with baseline, except for a delay in the midpoint of the least active 5 h (p value = 0.044). This analysis provides insight into how introducing flexibility into the traditional school schedule might influence sleep in adolescents.
COVID-19
school start times
sleep
adolescents
circadian rhythms
National Center for Advancing Translational Sciences https://doi.org/10.13039/100006108 UL1 TR002535 National Institute of Diabetes and Digestive and Kidney Diseases https://doi.org/10.13039/100000062 K01 DK113063 National Institute of Diabetes and Digestive and Kidney Diseases https://doi.org/10.13039/100000062 K23 DK117021 typesetterts1
==== Body
pmcA biologically induced slowed build-up of homeostatic sleep drive (sleep pressure) across the day and a physiological delay in circadian rhythms leads to late sleep onset in adolescence (Crowley et al., 2018). Combined with today’s academic and psychosocial demands, imposed early school start times, and daily profile of electronics use and light exposure, nearly 80% of high school–age adolescents obtain less than the recommended 8 to 10 h sleep per night (Paruthi et al., 2016; Crowley et al., 2018; Centers for Disease Control and Prevention, 2019). Insufficient sleep and circadian misalignment are associated with numerous negative outcomes, including mental health problems, obesity and dysregulated metabolism, learning difficulties, and poor academic achievement (Crowley et al., 2018).
Adolescents were profoundly affected in all aspects of daily life by the school closures and stay-at-home orders implemented in the United States in spring 2020 to mitigate the spread of COVID-19. Since the onset of these health measures, evidence is emerging of subsequent changes in daily routines and lifestyle activities in youth (Bates et al., 2020). Adolescents showed patterns of delayed bed and wake times, longer sleep duration, and less daytime sleepiness during COVID-19 per subjective self- and parent reports (Gruber et al., 2020; Becker et al., 2021; Bruni et al., 2021; Lavigne-Cerván et al., 2021; Illingworth et al., 2022). However, these studies are limited to primarily retrospective reports prior to COVID-19 and lack of objective assessment.
A later sleep-wake schedule may be more in line with adolescents’ circadian rhythms (Crowley et al., 2018). Indeed, multiple studies have confirmed the benefits of later high school start times (allowing for later wake time) on adolescent sleep and well-being (Owens et al., 2017; Meltzer et al., 2021b; Biller et al., 2022). Yet, going to bed and sleeping in later limits opportunities for participation in physical activity and for obtaining morning light exposure which serves to synchronize and entrain circadian rhythms (Youngstedt et al., 2016; Bates et al., 2020). Furthermore, high levels of light exposure from electronics, particularly late at night, may further delay sleep timing and the circadian system (Wams et al., 2017; Hisler et al., 2020). Lack of daily structure, increased electronics use, and spending less time outside were all associated with an irregular sleep-wake schedule and greater delay in self-reported sleep times during COVID-19 in adolescents (Amran, 2022). To our knowledge, despite the relationship and importance to sleep, light and activity levels have not been systematically evaluated in adolescents during COVID-19.
Utilizing a naturalistic design, we evaluated changes in the objective measurement of sleep, light, and activity before and during the COVID-19 pandemic in adolescents. Without the restrictions of early start times and the structure of traditional schooling, we hypothesized that adolescents would have longer, later, and less variable sleep, increased and later light exposure, and decreased physical activity during COVID-19 compared with prior.
Materials and Methods
Participants from a study examining insulin sensitivity and sleep in adolescents (ClinicalTrials.gov: NCT03500458) were invited to participate in an additional study week during COVID-19 stay-at-home orders. Inclusion criteria for the primary study included habitually sedentary (<3 h of reported physical activity per week) high school students 14 to 19 years of age with < 7 h sleep on school nights without a diagnosis of a sleep disorder or regular use of medications affecting sleep. The study was approved by the Colorado Multiple Institutional Review Board, and participants and guardians who previously consented for the primary study provided verbal consent/assent for the optional COVID-19 week. Participants completed 7 consecutive days of at-home monitoring at two time points: the baseline week took place prior to COVID-19, between October 2018 and February 2020, and the COVID-19 week occurred in May 2020. Participants were required to be attending traditional, in-person high school prior to COVID-19, and all study participation took place during the academic year while school was in session. Efforts were made to avoid data collection during daylight saving weeks and over school breaks; one study participant participated at the end of daylight savings at baseline. Actigraphy devices were delivered to families and returned by mail. During both study weeks, participants were asked to maintain their current, typical schedule.
Outcome Measures
Participants wore a Spectrum Plus actigraphy monitor (Philips Respironics, Bend, OR) on their nondominant wrist for 7 consecutive days. Participants were asked to press the event marker button on the watch at the time that they attempted sleep and again upon awakening, and concurrent sleep diaries were completed daily to facilitate scoring. Data were scored using proprietary software (Actiware Version 6, Philips Respironics, Pittsburgh, PA) and standard scoring rules (Ancoli-Israel et al., 2015).
Sleep-Wake
The following variables were derived separately for all recording days and averaged over weekdays (Sunday-Thursday nights) and weekends (Friday and Saturday nights): sleep onset, sleep offset, sleep midpoint, total sleep duration (difference between sleep onset and offset minus wake after sleep onset), and sleep efficiency. In addition, social jetlag (difference between weekend and weekday sleep midpoints; Mathew et al., 2019) and sleep regularity (intraindividual standard deviations of sleep onset, offset, midpoint, and duration) were calculated.
Light
The Actiwatch Spectrum calculates white light illuminance (lux) on a minute-by-minute basis by integrating the input from separate red-, green-, and blue-colored light sensors. The light data were summarized as average lux values over 24 h, during the average estimated waking window, and during the period 2 h before average bedtime, and as a percentage of time at > 1000 lux and < 100 lux, indicative of likely exposure to outdoor and indoor light, respectively (Bhandary et al., 2021).
Activity
The Actiwatch Spectrum estimates physical activity levels using accelerometer counts, with higher values indicating more activity. Changes in physical activity patterns over 24 h were evaluated from the accelerometer count data using the nonparametric methods first described by Witting et al. (1990) and more recently reviewed by Gonçalves et al. (2015). Briefly, these methods attempt to estimate the stability and variability of the 24 h rest-activity cycle across days of measurement. The ‘nparACT’ package (Blume et al., 2016) for R Core Team (R Development Core Team, 2020) was used to derive the following variables:
Inter-daily Stability (IS): An estimate of stability of the 24 h rest-activity cycle across days, where a value of 1 = perfect stability.
Intra-daily Variability (IV): An estimate of the fragmentation of the 24 h rest-activity cycle where a value of 0 = a perfect sine wave with no fragmentation and 2 = no apparent pattern in the 24 h rest-activity cycle.
The L5 indicates the period of time with the lowest 5 h of activity (in accelerometer counts).
The M10 indicates the period of time with the highest 10 h of activity (in accelerometer counts).
The relative amplitude (RA) is the ratio of the M10 and L5 average accelerometer counts. Higher values indicate more robust 24 h rest-activity patterns.
Questionnaires
Participants self-reported demographic information at baseline. Chronotype was assessed at baseline with the Morningness-eveningness Scale for Children (MESC), a validated 10-item self-report multiple-choice measure (Carskadon et al., 1993). Scores range from 10 to 42 with higher scores indicating more morning preference and lower scores indicating more evening preference. Using cut-off points based on the 25th to 75th percentiles, individuals with scores of 10 to 23 were categorized as evening-type, individuals with scores of 24 to 27 were categorized as intermediate-type, and individuals with scores of 28 to 40 were categorized as morning-type (Díaz-Morales and Gutiérrez Sorroche, 2008). At the COVID-19 assessment, participants responded to questions derived for the purpose of the current study asking them to report if they spent more, less, or the same amount of time engaged in behaviors including electronics/technology, physical activity, time outside, social interactions, and schoolwork. They also completed the PROMIS Anxiety and Depression short form measures at both baseline and during COVID-19 to assess anxiety and depression symptoms over the past week (Irwin et al., 2012). Raw scores are summed and converted to T-scores, with scores of 55 and below described as “within normal limits,” 55 to 60 “mild,” 60 to 70 “moderate,” and >70 “severe” (Kaat et al., 2019).
Analyses
All analyses were stratified by weekday and weekend. Comparisons of each outcome (sleep, light, and activity) from baseline to the period during COVID-19 outcome were evaluated using paired sample t tests or Wilcoxon signed-rank test. Shapiro-Wilk tests were used to determine the normality of outcomes and those identified as non-normal (Shapiro-Wilk p value < 0.05) were compared using the nonparametric Wilcoxon signed-rank test. Data are presented as mean ± standard deviation (SD) regardless of the test used. To account for multiple comparisons, p values were adjusted using the Holm–Bonferroni method. The p values were adjusted within each type of outcome (sleep, light, and activity) for weekdays and weekends separately.
We hypothesized that the MESC score (evening, intermediate, or morning chronotype) might explain any observed changes in sleep given that the participants may have had more flexibility in their daily schedules during the COVID-19 week. To assess this, linear mixed models were fit to predict sleep onset, sleep offset, sleep duration, and sleep efficiency based on the day of the week, study time point (baseline, during COVID-19), and chronotype. In addition, interactions were included for night-by-study time point, and study time point-by-chronotype, with this time-by-chronotype interaction being of particular interest in addressing our hypothesis. The linear mixed models accounted for within-subject correlations using an AR(1) covariance structure and a random intercept by subject. All models were fit separately for weekdays and weekends. The estimated difference between baseline and COVID-19 was calculated from the linear mixed models for each chronotype at each study time point. For each model, a Type 3 Test of Fixed Effects for the time-by-chronotype interaction was used to directly address if the chronotype explains the observed changes in sleep between study time points. The p values from the Type 3 Tests of Fixed Effects were adjusted for multiple comparisons using the Holm–Bonferroni method. The p values were adjusted for weekdays and weekends separately.
Linear mixed models were fit using SAS Software (SAS Institute Inc., Cary, NC, USA), while all other analyses were conducted in the R computing environment (R Development Core Team, 2020). Results with an adjusted p value < 0.05 were considered statistically significant.
Sensitivity Analyses
This study included one subject who had a set school start time during COVID-19. A sensitivity analysis was performed to assess if the inclusion of this subject altered the results of the statistical analyses. In addition, one subject’s baseline data were collected during time change at the end of daylight savings. A separate sensitivity analysis was performed to assess the impact of the inclusion of this subject in analyses. For each of these sensitivity analyses, all statistical analyses were performed with the exclusion of the associated subject, and results were assessed to determine if the results changed statistical or clinical significance when the subject was excluded. The exclusion of these participants did not change the statistical results appreciably; thus, both participants were retained for subsequent analyses.
Results
Sixteen participants completed procedures at both baseline and COVID-19 follow-up weeks. All participants had ≥ 4 weekdays and ≥ 2 weekend days of actigraphy data at both baseline and during COVID-19, with the exception of one participant that did not have any baseline weekend date. This participant was excluded from all weekend analyses except for linear mixed models which allow for missing data. Participants were on average 15.9 ± 1.2 years old at baseline, 68.8% female, 87.5% White, and 25.0% Hispanic/Latino. Fifty percent of participants were classified as evening chronotype, while 18.8% and 31.3% were classified as intermediate and morning chronotype, respectively. During the COVID-19 week, all participants reported participation in online learning due to in-person school closures. Prior to COVID-19, reported school start time ranged from 0730 to 0830 h. Only 1 participant (6.3%) had a set start time for online learning during COVID-19, which was reportedly consistent with their baseline school start time (0830 h); the remaining participants reported that learning was conducted according to their own schedule. Participants reported doing their schoolwork primarily in the afternoon (1200 -1700 h; 47%) during COVID-19, while most participants completed schoolwork in the evening (after 1800 h; 53%) at baseline. The majority of participants (67%) reported spending less time on schoolwork during COVID-19 compared with baseline.
Sixty percent of participants endorsed spending less time in social interactions with peers, 53% spent more time outside, and approximately half the sample reported spending more time engaged in physical activity during COVID-19 compared with baseline. Time spent using technology and electronics increased for 93% of participants during COVID-19 compared with baseline. Ratings of anxiety and depression symptoms remained on average in the normal to mild range at both baseline (mean PROMIS depression t-scores = 53.3 ± 10.9; mean PROMIS anxiety t scores = 50.5 ± 11.1) and COVID-19 (mean PROMIS depression t scores = 54.6 ± 9.8; mean PROMIS anxiety t scores = 46.9 ± 11.4) with no significant change between timepoints (all p values > 0.05). No participants reported COVID-19-related symptoms or a prior diagnosis of COVID-19 while participating in the study.
Change in Sleep
Actigraphy-estimated sleep measures at baseline and during COVID-19 are presented in Table 1 and Figure 1. Participants obtained on average 1 more hour of sleep per night on weekdays during COVID-19 compared with baseline (p < 0.001), while weekend sleep duration during COVID-19 was not significantly different from baseline (p = 0.268). At baseline, 10.6% of all recorded sleep episodes were greater than 8 h in duration, compared with 24.8% of all recorded sleep episodes during the COVID-19 week (χ2 test p = 0.008). Bedtime, waketime, and the midpoint of sleep were each significantly delayed on both weekdays and weekends during COVID-19 compared with baseline (all p values < 0.05). The largest delay was observed in weekday waketime which occurred nearly 3 h later during COVID-19 than baseline. Sleep efficiency decreased modestly by −3.75% (p < 0.001) during the weekdays but was not significantly changed during the weekends (p = 0.268). Social jetlag (defined here as the difference in weekday and weekend sleep midpoints; Mathew et al., 2019) was not statistically different between time points (p = 0.058). Similarly, sleep regularity was also not significantly different during COVID-19 compared with baseline (all p values > 0.05; see Supplemental Table 1).
Table 1. Weekday and weekend sleep parameters assessed with wrist actigraphy in N = 16 teenagers before the COVID-19 pandemic and during the initial stay-at-home and safer-at-home phase of the pandemic in Colorado. Data are shown as mean ± SD.
Baseline During COVID-19 Δ Adjusted p value
Weekday recordings
Time of sleep onset 00:01 ± 00:59 01:41 ± 01:26 1.66 ± 1.13 <0.001
Time of sleep offset 06:30 ± 00:35 09:23 ± 01:19 2.88 ± 0.94 <0.001
Time of sleep midpoint 03:16 ± 00:45 05:32 ± 01:18 2.27 ± 0.9 <0.001
Sleep duration (h) 5.93 ± 0.41 6.93 ± 0.86 1 ± 0.87 <0.001
Sleep efficiency (%) 87.9 ± 4.46 84.15 ± 3.51 -3.75 ± 3.1 <0.001
Social jetlag (h)* 0.92 ± 1.5 0.46 ± 0.89 -0.46 ± 1.81 0.058
Weekend recordings
Time of sleep onset 00:27 ± 01:08 02:07 ± 01:52 1.66 ± 1.79 0.01
Time of sleep offset 08:30 ± 01:12 09:40 ± 01:17 1.15 ± 1.21 0.01
Time of sleep midpoint 04:28 ± 00:59 05:53 ± 01:25 1.43 ± 1.38 0.007
Sleep duration (h) 7.25 ± 1.11 6.76 ± 1.36 -0.49 ± 1.2 0.268
Sleep efficiency (%) 84.54 ± 4.51 82.33 ± 5.65 -2.21 ± 6.12 0.268
Data are shown as mean ± standard deviation. Change represents difference in activity variables from baseline to during COVID-19 and p values correspond to paired-sample t-tests or Wilcoxon signed-rank tests (indicated by *), adjusted for multiple comparisons using the Holm–Bonferroni method.
Bold values indicate p < 0.05.
Figure 1. Sleep timing (a) and duration (b) at baseline and during the COVID-19 pandemic on weekdays and weekends. Sleep onset and offset were significantly delayed on both weekdays and weekends during the COVID-19 week compared with baseline. Sleep duration on weekdays significantly increased during COVID-19 compared with baseline, while weekday sleep duration was not significantly different between time points.
Change in Sleep by Chronotype
Estimated changes in sleep onset, sleep offset, sleep duration, and sleep efficiency from baseline to during COVID-19 by chronotype from linear mixed models are presented in Supplemental Table 2. Although the estimated sleep outcome changed between baseline and COVID-19 for each chronotype, none of the time-by-chronotype interactions were statistically significant (all p values > 0.05), indicating that MESC chronotype did not significantly impact the observed changes in these sleep outcomes from baseline to during COVID-19.
Change in Light
Light variables are presented in Table 2 and Figure 2. Weekday average lux over 24 h and while awake were significantly higher during COVID-19 compared with baseline (p = 0.001 for both outcomes), with participants more than doubling average lux levels during COVID-19. Weekday average light exposure in the 2 h prior to bedtime (i.e., evening light) was also significantly higher during COVID-19 compared with baseline (p value = .007). In addition, the percentage of time spent at > 1000 lux was significantly higher and the percentage of time at < 100 lux was significantly lower during COVID-19 compared with baseline (p values = 0.007 for both outcomes).
Table 2. Weekday and weekend light levels were assessed with wrist actigraphy in N = 16 teenagers before the COVID-19 pandemic and during the initial stay-at-home and safer-at-home phase of the pandemic in Colorado.
Baseline During COVID-19 Δ Adjusted p value
Weekday recordings
24 h average, lux* 85.17 ± 59.53 234.29 ± 175.35 149.12 ± 173.1 0.001
Waking average, lux* 109.48 ± 76.44 343.48 ± 303.83 234 ± 295.28 0.001
Average 2 h prior to sleep, lux 86.02 ± 64.69 237.9 ± 167.97 151.88 ± 168.97 0.007
Percent of time > 1000 lux* 1.37 ± 1.26 7.37 ± 7.56 5.99 ± 7.51 0.007
Percent of time < 100 lux 83.95 ± 10.16 72.58 ± 12.25 –11.37 ± 11.89 0.007
Weekend recordings
24 h average, lux 87.59 ± 60.8 240.25 ± 179.82 152.66 ± 178.57 0.015
Waking average, lux* 112.4 ± 78.2 352.41 ± 312.32 240.01 ± 304.63 0.003
Average 2 h prior to sleep, lux 77.51 ± 58.4 248.28 ± 198.01 170.77 ± 208.13 0.015
Percent of time > 1000 lux* 1.43 ± 1.28 7.63 ± 7.75 6.2 ± 7.73 0.013
Percent of time < 100 lux* 83.69 ± 10.46 72.36 ± 12.65 –11.33 ± 12.31 0.007
Data are shown as mean ± standard deviation. Change represents the difference in the light variables from baseline to during COVID-19 and p values correspond to paired-sample t tests or Wilcoxon signed-rank tests (indicated by *), adjusted for multiple comparisons using the Holm–Bonferroni method.
Bold values indicate p < 0.05.
Figure 2. Weekday light (ln lux; a) and daily accelerometer counts (b) at baseline and during the COVID-19 pandemic. The standard deviation in the average local time of sleep onset and offset are shown by the black and red horizontal bars. In addition, for accelerometry, horizontal lines are included to indicate the standard deviation in the local time of the lowest level of activity over a 5-h period (L5) and the highest level of activity over a 10-h period (M10). The gray background indicates the range (minimum and maximum) of the observed light exposures or accelerometer counts. Color version of the figure is available online.
Change in Activity
Physical activity variables are presented in Table 3 and Figure 2. The weekday midpoint of the least active 5 h was significantly delayed by approximately 2 h during COVID-19 compared with baseline (p value = 0.044). There were no other significant differences in the actigraphy-estimated activity outcomes between baseline and the COVID-19 week for weekdays or weekends.
Table 3. Weekday and weekend physical activity variables assessed with wrist actigraphy in N = 16 teenagers before the COVID-19 pandemic and during the initial stay-at-home and safer-at-home phase of the pandemic in Colorado.
Baseline During COVID-19 Δ Adjusted p value
Weekday recordings
Interdaily stability 0.62 ± 0.1 0.51 ± 0.15 –0.11 ± 0.19 0.147
Intradaily variability 0.82 ± 0.24 0.75 ± 0.22 –0.07 ± 0.18 0.388
Relative amplitude* 0.93 ± 0.05 0.84 ± 0.14 –0.08 ± 0.15 0.147
Avg. counts, least active 5 h (L5)* 11.66 ± 6.78 23.7 ± 19.95 12.04 ± 20.33 0.147
Avg. counts, most active 10 h (M10)* 317.56 ± 81.28 307.31 ± 125.3 –10.25 ± 103.86 0.807
Mid-point time of the L5* 0610 ± 0744 h 0823 ± 0732 h 0213 ± 0902 h 0.044
Mid-point time of the M10 1447 ± 0242 h 1502 ± 0220 h 0016 ± 0229 h 0.807
Weekend recordings
Interdaily stability 0.67 ± 0.14 0.76 ± 0.12 0.09 ± 0.2 0.488
Intradaily variability 0.57 ± 0.15 0.64 ± 0.25 0.07 ± 0.29 0.671
Relative amplitude* 0.94 ± 0.03 0.92 ± 0.05 –0.02 ± 0.04 0.500
Avg. counts, least active 5 h (L5) 9.81 ± 4.01 13.84 ± 8.9 4.03 ± 6.29 0.158
Avg. counts, most active 10 h (M10)* 359.89 ± 145.28 361.39 ± 153.35 1.5 ± 155.77 0.762
Mid-point time of the L5* 0515 ± 0553 h 0523 ± 0142 h 0008 ± 0648 h 0.106
Mid-point time of the M10* 1509 ± 0213 h 1626 ± 0141 h 0117 ± 0250 h 0.488
Data are shown as mean ± standard deviation. Change represents the difference in activity variables from baseline to during COVID-19 and p values correspond to paired-sample t tests or Wilcoxon signed-rank tests (indicated by *), adjusted for multiple comparisons using the Holm–Bonferroni method.
Bold values indicate p < 0.05.
Discussion
The current analysis of changes in objectively assessed sleep, light, and activity from before and during the first wave of the COVID-19 pandemic found that adolescents delayed sleep times, increased sleep duration, and received more light exposure during COVID-19 compared with baseline. With the majority of participants free from the constraints of an early school start time, adolescents obtained longer sleep duration during the COVID-19 pandemic, notably by sleeping later in the mornings on school days. This equates to an additional 5 h of sleep over the course of a school week. Sleep efficiency decreased modestly on weekdays during COVID-19, but the clinical significance of this change is unclear. Moreover, bedtimes and waketimes shifted later on both weekdays and weekends, with the most significant delay in weekday waketime of ~3 h. These findings are similar to naturalistic studies of U.S. and Brazilian adolescents that self-reported a delay in bed and waketimes of 1-1.5 and 1.5 to 2 h, respectively, and increased sleep duration during COVID-19 compared with assessments that took place prior to COVID-19 (Becker et al., 2021; Genta et al., 2021). In a sample of college students assessed via sleep log before and during COVID-19, weekday time in bed increased, weekday sleep timing delayed, and social jetlag was reduced during COVID-19 (Wright et al., 2020). In contrast, social jetlag did not significantly change in our sample.
One potential contributor to these changes in sleep includes greater flexibility of schedules associated with online learning and lack of commute to school resulting in greater opportunity for more sleep. Indeed, a cross-sectional study of 6- to 12-grade students in the United States found that instruction type was significantly associated with timing and duration of sleep during COVID-19 such that bed and waketimes were latest and sleep opportunity was longer on nights when students had online/asynchronous learning compared with online/synchronous, and in-person schooling (Meltzer et al., 2021a). This is consistent with studies completed prior to the pandemic that have shown that delaying high school start times by 50-70 min allows students to obtain ~40-45 min more weeknight sleep compared with high school students before the change in start times (Meltzer et al., 2021b) and compared with students at schools that maintained an early school start time of 0730 h (Widome et al., 2020).
Although sleep duration increased and the percentage of adolescents that obtained more than 0800 h of sleep more than doubled during COVID-19 in our sample, approximately 75% of adolescents still obtained insufficient sleep during COVID-19. These findings highlight that adolescent sleep is modifiable but that additional strategies beyond flexibility in schedule may be needed to improve sleep health and specifically sleep duration to recommended levels for age. While there was no significant effect of chronotype on the observed changes in sleep variables in the current study, other studies suggest strategies to improve sleep health individualized depending on chronotype and other characteristics may be needed (Gradisar et al., 2011; Blake et al., 2019).
In the current study, adolescents had greater light exposure during COVID-19 over the 24-h period, while awake, and in the 2 h prior to sleep, as well as increased the percentage of time > 1000 lux, which may reflect greater exposure to sunlight or electronics. These findings should be considered within the limitations of the current data such that there was a longer daylight period during the COVID-19 data collection period in May compared to participants who completed their baseline assessment in the fall or winter. Notably, only roughly half of the participants reported spending more time outside, while nearly all participants endorsed spending more time using electronics during COVID-19 compared with baseline. In a sample of youth attending a morning or afternoon-shift school schedule, evening electronics use was correlated with later bedtimes and shorter time in bed regardless of school start time (Arrona-Palacios, 2017). Thus, future studies should include measurement of daily electronics usage to evaluate its role in sleep timing. In a study conducted prior to the pandemic, actigraphy-measured light exposure and dim light melatonin onset (DLMO) were compared between 14 young adults with delayed sleep and 14 matched controls and found that the delayed sleep group had relatively greater exposure to white and blue light 2 h after DLMO, a circadian time with maximal phase-delay effect (Van der Maren et al., 2018). A limitation of the current study is that we were not able to evaluate melatonin; consideration of the circadian phase in adolescents is important for future studies.
We did not find a significant change in objective activity parameters from baseline to COVID-19, except for a delay in the weekday midpoint of the least active 5 h. This may be due to the similarly observed delay in sleep times. These findings are in contrast to a study of school-age children in Israel who wore Actiwatch and demonstrated a significant decrease in time engaged in moderate-to-vigorous physical activity during COVID-19 compared with before COVID-19 (Guo et al., 2021). This difference may be in part due to age, as physical activity decreases throughout adolescence, primarily replaced by sedentary activity (Kandola et al., 2020). Moreover, inclusion criteria for the current parent study required participants to have a typical low level of physical activity (< 3 h of reported physical activity per week), although this is broadly representative of the adolescent population (Guthold et al., 2020). Notably, participants in the current sample were evenly split in their subjective report of engaging in more and less physical activity during COVID-19 compared with baseline.
The current findings should be considered within the limitations of the study, including a small sample size. Our sample included adolescents that obtained insufficient sleep during the academic year prior to COVID-19 and thus may not be generalizable to all youth. Potentially important variables, such as daily patterns of electronics usage, specific virtual school schedules, individual and family stressors, and other factors that may have impacted sleep behaviors were not assessed in the present study. Using the Actiwatch to derive estimates of physical activity is not ideal and may have missed changes in sedentary behaviors which may be more accurately measured by other devices. Similarly, the Actiwatch has been found to underestimate light levels compared with gold standard measures (Howell et al., 2021). In addition, the validity of actigraphy devices in estimating certain sleep parameters is inconsistent, with the devices tending to underestimate onset latency in adolescents (Meltzer et al., 2015). Finally, intraindividual standard deviations were used to estimate sleep regularity, but future analysis using the Sleep Regularity Index is recommended (Phillips et al., 2017). However, strengths of the study include objective measurement of sleep, activity, and light both before and during COVID-19. Although the study was performed in the early months of the pandemic when restrictions to mitigate the spread of disease and safety concerns limited research activities, the current analysis is a significant contribution to the existing COVID-19 literature that primarily relies on subjective measurement and retrospective comparisons.
Unintended effects of the lifestyle changes during the COVID-19 pandemic, such as the shift to online learning, may have provided adolescents the opportunity to obtain longer and later sleep, although the majority of adolescents still obtained insufficient sleep during COVID-19. These findings may inform parents and policy makers as they consider shifting high schools schedules to start no earlier than 0830 h, in line with recommendations (Adolescent Sleep Working Group et al., 2014; Ziporyn et al., 2022). Continued efforts by clinicians to promote healthy sleep habits and inform adolescents of the potential benefits of obtaining healthy sleep remain important. Finally, additional research to better understand the impact of these changed sleep behaviors on daytime functioning, academic performance, and health outcomes is particularly urgent as schools return to in-person learning and plan for future academic years.
Supplemental Material
sj-docx-1-jbr-10.1177_07487304221123455 – Supplemental material for A Naturalistic Actigraphic Assessment of Changes in Adolescent Sleep, Light Exposure, and Activity Before and During COVID-19
Click here for additional data file.
Supplemental material, sj-docx-1-jbr-10.1177_07487304221123455 for A Naturalistic Actigraphic Assessment of Changes in Adolescent Sleep, Light Exposure, and Activity Before and During COVID-19 by Corey A. Rynders, Anne E. Bowen, Emily Cooper, John T. Brinton, Janine Higgins, Kristen J. Nadeau, Kenneth P. Wright Jr. and Stacey L. Simon in Journal of Biological Rhythms
This study was supported by funding from K23 DK117021 (S.L.S.), K01 DK113063 (C.A.R.), UL1 TR002535 (Colorado CTSA).
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: K.P.W. reports during the conduct of the study being a consultant to/and/or receiving personal fees from Circadian Therapeutics, Inc., Circadian Biotherapies, Inc., Philips, Inc. outside the submitted work. All other authors declare that there are no conflicts of interest.
ORCID iD: Stacey L. Simon https://orcid.org/0000-0003-4755-8151
Supplementary material is available for this article online.
==== Refs
References
Adolescent Sleep Working Group, Committee on Adolescence, and Council on School Health (2014) School start times for adolescents. Pediatrics 134 :642-649.25156998
Amran MS (2022) Psychosocial risk factors associated with mental health of adolescents amidst the COVID-19 pandemic outbreak. Int J Soc Psychiatry 68 :6-8.33158391
Ancoli-Israel S Martin JL Blackwell T Buenaver L Liu L Meltzer LJ Sadeh A Spira AP Taylor DJ (2015) The SBSM guide to actigraphy monitoring: clinical and research applications. Behav Sleep Med 13 :S4-S38.26273913
Arrona-Palacios A (2017) High and low use of electronic media during nighttime before going to sleep: a comparative study between adolescents attending a morning or afternoon school shift. J Adolesc 61 :152-163.29111446
Bates LC Zieff G Stanford K Moore JB Kerr ZY Hanson ED Barone Gibbs B Kline CE Stoner L (2020) COVID-19 impact on behaviors across the 24-hour day in children and adolescents: physical activity, sedentary behavior, and sleep. Children (Basel) 7 :138.32947805
Becker SP Dvorsky MR Breaux R Cusick CN Taylor KP Langberg JM (2021) Prospective examination of adolescent sleep patterns and behaviors before and during COVID-19. Sleep 44:zsab054.
Bhandary SK Dhakal R Sanghavi V Verkicharla PK (2021) Ambient light level varies with different locations and environmental conditions: potential to impact myopia. Plos One 16 :e0254027.34234353
Biller AM Molenda C Zerbini G Roenneberg T Winnebeck EC (2022) Sleep improvements on days with later school starts persist after 1 year in a flexible start system. Sci Rep 12 :2787.
Blake MJ Latham MD Blake LM Allen NB (2019) Adolescent-sleep-intervention research: current state and future directions. Curr Dir Psychol Sci 28 :475-482.
Blume C Santhi N Schabus M (2016) nparACT’ package for R: a free software tool for the non-parametric analysis of actigraphy data. MethodsX 3 :430-435.27294030
Bruni O Malorgio E Doria M Finotti E Spruyt K Melegari MG Villa MP Ferri R (2021) Changes in sleep patterns and disturbances in children and adolescents in Italy during the Covid-19 outbreak. Sleep Med 91 :166-174.33618965
Carskadon MA Vieira C Acebo C (1993) Association between puberty and delayed phase preference. Sleep 16 :258-262.8506460
Centers for Disease Control and Prevention (2019) Youth risk behavior survey data. https://www.cdc.gov/healthyyouth/data/yrbs/index.htm
Crowley SJ Wolfson AR Tarokh L Carskadon MA (2018) An update on adolescent sleep: new evidence informing the perfect storm model. J Adolesc 67 :55-65.29908393
Díaz-Morales JF Gutiérrez Sorroche M (2008) Morningness-eveningness in adolescents. Span J Psychol 11 :201-206.18630661
Genta FD Rodrigues Neto GB Sunfeld JPV Porto JF Xavier AD Moreno CRC Lorenzi-Filho G Genta PR (2021) COVID-19 pandemic impact on sleep habits, chronotype, and health-related quality of life among high school students: a longitudinal study. J Clin Sleep Med 17 :1371-1377.33666168
Gonçalves BS Adamowicz T Louzada FM Moreno CR Araujo JF (2015) A fresh look at the use of nonparametric analysis in actimetry. Sleep Med Rev 20 :84-91.25065908
Gradisar M Dohnt H Gardner G Paine S Starkey K Menne A Slater A Wright H Hudson JL Weaver E , et al . (2011) A randomized controlled trial of cognitive-behavior therapy plus bright light therapy for adolescent delayed sleep phase disorder. Sleep 34 :1671-1680.22131604
Gruber R Saha S Somerville G Boursier J Wise MS (2020) The impact of COVID-19 related school shutdown on sleep in adolescents: a natural experiment. Sleep Med 76 :33-35.33070000
Guo YF Liao MQ Cai WL Yu XX Li SN Ke XY Tan SX Luo ZY Cui YF Wang Q , et al . (2021) Physical activity, screen exposure and sleep among students during the pandemic of COVID-19. Sci Rep 11 :8529.33879822
Guthold R Stevens GA Riley LM Bull FC (2020) Global trends in insufficient physical activity among adolescents: a pooled analysis of 298 population-based surveys with 1·6 million participants. Lancet Child Adolesc Health 4 :23-35.31761562
Hisler GC Hasler BP Franzen PL Clark DB Twenge JM (2020) Screen media use and sleep disturbance symptom severity in children. Sleep Health 6 :731-742.32861729
Howell CM McCullough SJ Doyle L Murphy MH Saunders KJ (2021) Reliability and validity of the Actiwatch and Clouclip for measuring illumination in real-world conditions. Ophthalmic Physiol Opt 41 :1048-1059.34387902
Illingworth G Mansfield KL Espie CA Fazel M Waite F (2022) Sleep in the time of COVID-19: findings from 17000 school-aged children and adolescents in the UK during the first national lockdown. Sleep Adv 3 :zpab021.35128401
Irwin DE Gross HE Stucky BD Thissen D DeWitt EM Lai JS Amtmann D Khastou L Varni JW DeWalt DA (2012) Development of six PROMIS pediatrics proxy-report item banks. Health Qual Life Outcomes 10 :22.22357192
Kaat AJ Kallen MA Nowinski CJ Sterling SA Westbrook SR Peters JT (2019) PROMIS® Pediatric Depressive Symptoms as a Harmonized Score Metric. J Pediatr Psychol 45 :271-280.
Kandola A Lewis G Osborn DPJ Stubbs B Hayes JF (2020) Depressive symptoms and objectively measured physical activity and sedentary behaviour throughout adolescence: a prospective cohort study. Lancet Psychiat 7 :262-271.
Lavigne-Cerván R Costa-López B Juárez-Ruiz de Mier R Real-Fernández M Sánchez-Muñoz de León M Navarro-Soria I (2021) Consequences of COVID-19 confinement on anxiety, sleep and executive functions of children and adolescents in Spain. Front Psychol 12 :565516.33664690
Mathew GM Li X Hale L Chang AM (2019) Sleep duration and social jetlag are independently associated with anxious symptoms in adolescents. Chronobiol Int 36 :461-469.30786775
Meltzer LJ Saletin JM Honaker SM Owens JA Seixas A Wahlstrom KL Wolfson AR Wong P Carskadon MA (2021a) COVID-19 instructional approaches (in-person, online, hybrid), school start times, and sleep in over 5,000 U.S. adolescents. Sleep 44 :zsab180.34401922
Meltzer LJ Wahlstrom KL Plog AE Strand MJ (2021b) Changing school start times: impact on sleep in primary and secondary school students. Sleep 44 :zsab048.33855446
Meltzer LJ Walsh CM Peightal AA (2015) Comparison of actigraphy immobility rules with polysomnographic sleep onset latency in children and adolescents. Sleep Breath 19 :1415-1423.25687438
Owens JA Dearth-Wesley T Herman AN Oakes JM Whitaker RC (2017) A quasi-experimental study of the impact of school start time changes on adolescent sleep. Sleep Health 3 :437-443.29157637
Paruthi S Brooks LJ D’Ambrosio C Hall WA Kotagal S Lloyd RM Malow BA Maski K Nichols C Quan SF , et al . (2016) Recommended amount of sleep for pediatric populations: a consensus statement of the American Academy of Sleep Medicine. J Clin Sleep Med 12 :785-786.27250809
Phillips AJK Clerx WM O’Brien CS Sano A Barger LK Picard RW Lockley SW Klerman EB Czeisler CA (2017) Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Sci Rep 7 :3216.28607474
R Development Core Team (2020) R: A Language and Environment for Statistical Computing. Vienna (Austria): R Foundation for Statistical Computing.
Van der Maren S Moderie C Duclos C Paquet J Daneault V Dumont M (2018) Daily profiles of light exposure and evening use of light-emitting devices in young adults complaining of a delayed sleep schedule. J Biol Rhythms 33 :192-202.29463186
Wams EJ Woelders T Marring I van Rosmalen L Beersma DGM Gordijn MCM Hut RA (2017) Linking light exposure and subsequent sleep: a field polysomnography study in humans. Sleep 40 :zsx165.29040758
Widome R Berger AT Iber C Wahlstrom K Laska MN Kilian G Redline S Erickson DJ (2020) Association of delaying school start time with sleep duration, timing, and quality among adolescents. JAMA Pediatr 174 :697-704.32338727
Witting W Kwa IH Eikelenboom P Mirmiran M Swaab DF (1990) Alterations in the circadian rest-activity rhythm in aging and Alzheimer’s disease. Biol Psychiatry 27 :563-572.2322616
Wright KP Jr Linton SK Withrow D Casiraghi L Lanza SM Iglesia H Vetter C Depner CM (2020) Sleep in university students prior to and during COVID-19 Stay-at-home orders. Curr Biol 30 :R797-R798.32693068
Youngstedt SD Kline CE Elliott JA Zielinski MR Devlin TM Moore TA (2016) Circadian phase-shifting effects of bright light, exercise, and bright light + exercise. J Circadian Rhythms 14 :2.27103935
Ziporyn TD Owens JA Wahlstrom KL Wolfson AR Troxel WM Saletin JM Rubens SL Pelayo R Payne PA Hale L , et al . (2022) Adolescent sleep health and school start times: setting the research agenda for California and beyond. A research summit summary. Sleep Health 8 :11-22.34991996
| 36124632 | PMC9726638 | NO-CC CODE | 2022-12-08 23:18:17 | no | J Biol Rhythms. 2022 Dec; 37(6):690-699 | utf-8 | J Biol Rhythms | 2,022 | 10.1177/07487304221123455 | oa_other |
==== Front
Nefrologia (Engl Ed)
Nefrologia (Engl Ed)
Nefrologia
2013-2514
Sociedad Española de Nefrología. Published by Elsevier España, S.L.U.
S2013-2514(22)00149-3
10.1016/j.nefroe.2022.11.023
Short Original
Increase in peritoneal dialysis-related hospitalization rates after telemedicine implementation during the COVID-19 pandemic
Aumento de las tasas de hospitalización relacionadas con la diálisis peritoneal después de la implementación de telemedicina durante la pandemia de la COVID-19Nerbass Fabiana Baggio a*
Vodianitskaia Raíssa Martins b
Ferreira Helen Caroline c
Sevignani Gabriela ac
Vieira Marcos Alexandre ac
Calice-Silva Viviane abc
a Fundação Pró-Rim, Rua Alexandre Dumas, 50, CEP: 89227-630 Joinville, SC, Brazil
b Escola de Medicina, Universidade da Região de Joinville – UNIVILLE, Rua Paulo Malschitzki – Zona Industrial Norte, CEP: 89219-710 Joinville, SC, Brazil
c Centro de Tratamento de Doenças Renais, Rua Xavier Arp, 15, CEP: 89227-630 Joinville, SC, Brazil
⁎ Corresponding author.
7 12 2022
7 12 2022
18 6 2021
31 10 2021
© 2021 Sociedad Española de Nefrología. Published by Elsevier España, S.L.U.
2021
Sociedad Española de Nefrología
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction and objectives
To minimize our peritoneal dialysis (PD) population exposure to coronavirus disease (COVID-19), in April 2020 we developed and implemented a telemedicine program. In this investigation, we aimed to compare the hospitalization rates and metabolic disorders in patients undergoing PD 6 months before and after the COVID-19 pandemic and telemedicine implementation.
Materials and methods
This single-center retrospective analysis included all active prevalent patients undergoing PD from April 2020. Dialysis records were reviewed to obtain clinical, demographic, laboratory, appointment, and hospitalization data. We compared hospitalization rates (total, non-PD-related, and PD-related), hospitalization-associated factors, and metabolic disorders (hemoglobin, serum potassium, and serum phosphate) 6 months before and after the pandemic.
Results
Our sample comprised 103 participants. During the pre-pandemic and post-pandemic periods, there were 13 and 27 hospital admissions, respectively. The total hospitalization incident rate ratio (IRR) was 2.48 (95% confidence interval [CI], 1.29–4.75). PD-related hospitalizations increased from 3 to 15 episodes (IRR = 7.25 [95% CI, 2.11–24.78]). In the pre-pandemic period, the educational level was lower in participants hospitalised due to PD-related issues than in participants not hospitalised. In the post-pandemic period, only sex distribution differed between patients not hospitalised and those hospitalised due to non-PD-related issues. Only serum potassium levels changed significantly in the post-pandemic period (4.79 ± 0.48 vs. 4.93 ± 0.54 mg/dL; P < 0.01).
Conclusion
This study showed a significant increase in hospitalization rates after the COVID-19 pandemic period and telemedicine implementation, mainly due to PD-related infectious causes. Strategies to improve distance monitoring assistance are needed for the PD population.
Introducción y objetivos
Para minimizar la exposición de nuestra población de diálisis peritoneal (DP) a la enfermedad por coronavirus (COVID-19), en abril del 2020 desarrollamos e implementamos un programa de telemedicina. En esta investigación, nuestro objetivo fue comparar las tasas de hospitalización y los trastornos metabólicos en pacientes sometidos a DP 6 meses antes y después de la pandemia de COVID-19 y la implementación de la telemedicina.
Materiales y métodos
Este análisis retrospectivo de un solo centro incluyó a todos los pacientes prevalentes activos sometidos a DP desde abril del 2020. Se revisaron los registros de diálisis para obtener datos clínicos, demográficos, de laboratorio, de citas y de hospitalización. Comparamos las tasas de hospitalización (total, no relacionada con la DP y relacionada con la DP), los factores asociados a la hospitalización y los trastornos metabólicos (hemoglobina, potasio sérico y fosfato sérico) 6 meses antes y después de la pandemia.
Resultados
Nuestra muestra fue compuesta por 103 participantes. Durante los períodos prepandémico y pospandémico, hubo 13 y 27 ingresos hospitalarios, respectivamente. La razón de la tasa de incidentes de hospitalización (TIR) total fue de 2,48 (intervalo de confianza [IC] del 95%, 1,29-4,75). Las hospitalizaciones relacionadas con la DP aumentaron de 3 a 15 episodios (TIR = 7,25 [IC del 95%, 2,11-24,78]). En el período prepandémico, el nivel educativo fue más bajo en los participantes hospitalizados debido a problemas relacionados con la DP que en los participantes no hospitalizados. En el período posterior a la pandemia, solo la distribución por sexo difirió entre los pacientes no hospitalizados y los hospitalizados debido a problemas no relacionados con la DP. Solo los niveles de potasio sérico cambiaron significativamente en el período pospandémico (4.79 ± 0.48 frente a 4.93 ± 0.54 mg/dL; P < 0.01).
Conclusión
Este estudio mostró un aumento significativo en las tasas de hospitalización después del período pandémico de COVID-19 y la implementación de la telemedicina, principalmente debido a causas infecciosas relacionadas con la DP. Se necesitan estrategias para mejorar la asistencia de monitoreo a distancia para la población con DP.
Keywords
COVID-19
Hospitalization
Peritoneal dialysis
Telemedicine
Palabras clave
COVID-19
Hospitalización
Diálisis peritoneal
Telemedicina
==== Body
pmcIntroduction
In March 2020, the government of Santa Catarina State, southern Brazil, declared an state of emergency to prevent and combat coronavirus disease (COVID-19).1 Simultaneously, the Federal Board of Medicine authorized telemedicine for distance monitoring,2 and the International Society of Peritoneal Dialysis stated that ‘people on peritoneal dialysis (PD) should stay at home. Hospital visits should be minimized only for urgent indications. Consultations should otherwise be conducted by telemedicine’.3
Following the authorities’ recommendations to minimize the PD population's exposure to COVID-19, who are more likely to develop severe disease4 our dialysis center immediately developed and implemented a telemedicine program.5 Therefore, our healthcare team started monitoring our patients undergoing PD via telephone calls; prior to this, all patients were consulted via face-to-face appointments every month.
Before the pandemic, many PD centers worldwide had already implemented telemedicine for monitoring patients undergoing PD. Most teleconference tools have shown successful results in terms of patient satisfaction, time-saving, economic advantages, and PD outcomes.6, 7, 8, 9, 10 Videoconference is the preferred tool for telemedicine consultations for patients undergoing PD. In addition to allowing a closer interaction between the healthcare personnel and patients, it enables a visual inspection of the catheter exit site and effluent bag for signs of peritonitis.11 However, it demands a good quality and stable internet connection, adequate hardware and software access, and training skills for patients and workers. These requirements are challenging for low- and middle-income countries even in a non-pandemic scenario and inviable in an emergency when migration for distance monitoring needs to be fast. During the pandemic, patient's behavior in seeking healthcare assistance may also be impacted due to the fear of contamination in healthcare sites; the quality of hospital assistance may also be impacted in this troublesome period.
To evaluate the impact of the pandemic and telemedicine implementation on health outcomes, in this single-center retrospective analysis, we aimed to compare the hospitalization rates and metabolic disorders 6 months before and after the COVID-19 pandemic in patients undergoing PD.
Methods
This retrospective study included participants from a single center located in southern Brazil. The 6-month pre-pandemic period was considered from September 2019 to February 2020 and the 6-month post-pandemic period was considered from April to September 2020.
Patients monitoring
Before the pandemic, all patients (N = 117) were monitored via presential PD clinic visits by a multidisciplinary team. They had monthly appointments with nephrologists and renal nurses. Whenever necessary, they were referred to psychologists, dietitians, and social workers. As we previously described,5 in April 2020, the nursing team at our center started contacting patients every month via telephone calls before the scheduled appointment with nephrologists.
Distance monitoring was accomplished whenever they reported being clinically well, their laboratory tests within normal values, and they provided telemedicine consent. The medical team used a telephone call as a telemedicine tool because most patients were elderly and had deprived social conditions; hence, videoconference was not feasible for everyone. Telephone calls were done monthly, according to pre-scheduled date and time, by the nurse and nephrologist from a specific number linked to the dialysis unit. Calls were recorded after patients’ verbal authorization was requested at the begging of the call.
During these calls, nurses and nephrologists applied a minimum questionnaire with questions to address the main aspects of the clinical evaluation, such as bowel functioning, stool aspects, urine output, blood pressure values, weight, and glycemic control. In addition, other aspects related to the dialysis were evaluated: mean daily ultrafiltration; the presence of signs and symptoms of hypervolemia such as edema, shortness of breath, cough; catheter exit site aspect and catheter and cycler performance.
Blood samples were also collected monthly in satellite laboratory facilities to whom the requisition was sent, and patients could go directly to the closest unit to have their blood draw. Lab results were analyzed before the doctor's calls, and the modifications in treatment and medications were done during the calls. During COVID-19, no essential measurements such as KT/V, peritoneal equilibrium test and others, were post-pone to avoid the need for the patients to come to the facility and expose themselves to infection, which was recommended by the International Society of Peritoneal Dialysis (ISPD)3 and by the Brazilian Society of Nephrology.12
Participants
We included adult patients who had started PD treatment at least 4 months before the isolation period (March 2020) and were active in the program in April 2020.
Data collection
Dialysis records were reviewed to obtain clinical, demographic, laboratory, appointment, and hospitalization data. We considered hypervolemia, catheter mechanical issues, and technique-related infections such as peritonitis and tunnel infection as PD-related hospitalization causes. In our center, all peritonitis cases begin the treatment hospitalized, regardless of its severity and independent of germ type. After culture results, according to clinical evaluation and control cell counts patients can be discharged to complete the treatment at home. All other causes were considered non-PD related. For demographic comparisons between participants according to hospitalization status, patients with more than one hospitalization event that included a PD-related, were allocated to the PD-related hospitalization group (one patient pre-pandemic and one patient post-pandemic). For metabolic results, we calculated the mean hemoglobin, serum phosphate, and potassium levels from all available results. The percentage of participants with the mean values within the target was calculated.
Data analysis
We calculated the incidence rate (IR) for total hospitalizations and PD-related hospitalization episodes in both periods to obtain the incidence rate ratio (IRR) and 95% confidence interval (95% CI). We also calculated the total mortality IR and IRR for all patients who underwent PD and were followed up at our center during both periods.
Statistical analysis was performed using SPSS software version 21.0 for Windows (SPSS, Inc., Chicago, IL, USA). Results are expressed as mean and standard deviation or median and interquartile range according to variable distribution. For group comparisons, we used Student's t-test for variables with normal distribution or the Mann–Whitney U test for skewed variables. The chi-square test and Fisher's exact tests were used for categorical variables, and P values < 0.05 were considered statistically significant.
Results
One hundred and three patients undergoing PD were included in this analysis, and their main characteristics are presented in Table 1 . Most patients were female, Caucasian, and had a caretaker. Only four patients were on continuous ambulatory peritoneal dialysis (CAPD), and almost half of the patients had diabetes.Table 1 Demographical characteristics of the participants (N = 103).
Table 1Age (years) 56.9 ± 16.2
Female N (%) 57 (55.3)
Caucasian N (%) 77 (74.8)
Education (years at school) 8 (5–11)
Caretaker (yes) N (%) 62 (60)
Diabetes mellitus N (%) 47 (45.6)
APD N (%) 99 (96.1)
PD vintage (month) 16 (8–27)
APD: automated peritoneal dialysis; PD: peritoneal dialysis.
During the 6-month post-pandemic period, there were a total of 539 medical appointments, 444 (82%) via telemedicine and 95 (18%) face-to-face visits. All patients had at least one telemedicine appointment, and 47 (46%) did not have a face-to-face visit during the follow-up period. We did not find any statistical difference regarding the demographical characteristics of patients followed exclusively by telemedicine compared to others.
Fig. 1 shows hospitalization episodes. Before the pandemic, there were 13 hospital admissions in 12 patients (1 patient was admitted twice), resulting in an IR of 21 per 1000 patient-months. During the post-pandemic period, there were 27 admissions (4 patients were admitted due to COVID-19) in 23 patients (2 patients were admitted twice and 2 patients were admitted three times), resulting in an IR of 52 per 1000 patient-months. The IRR for total hospitalization was 2.48 (95% CI, 1.29–4.75).Fig. 1 Hospitalization events 6-month pre- and 6-month post the pandemic.
The total number of PD-related hospitalizations increased from 3 (IR = 4 per 1000 patient-month) to 15 (IR = 29 per 1000 patient-month), representing 23% to 55% of the total hospitalizations, respectively. The IRR for PD-related hospitalization was 7.25 (95% CI, 2.11–24.78).
We also calculated the total mortality rates for all patients undergoing PD who were followed up at our center. The total mortality rate was 13.2/1000 patient-months and 17.1/1000 patient-months before and after the pandemic, resulting in an IRR of 1.29 (95% CI, 0.62–2.64). The main pre-pandemic mortality causes were: sepsis (n = 5); multiple organ failure (n = 2); respiratory failure (n = 2); others/unknown (n = 4). Post-pandemic were: COVID-19 (n = 4), cardiovascular (n = 3); metabolic (n = 3); sepsis (n = 2); others/unknown (n = 3).
We analyzed the participants’ demographic characteristics according to hospitalization status before and after the pandemic (Table 2 ). Compared to patients not hospitalised, patients hospitalised due to PD-related issues had a lower education level during the pre-pandemic period. In the post-pandemic period, only sex distribution differed between patients not hospitalised and those hospitalised due non-PD-related issues.Table 2 Comparison of participants demographical characteristics according to hospitalization status before and after the pandemic (N = 103).
Table 2 Pre-pandemic hospitalization Post-pandemic hospitalization
Variables No hospitalization Non-PD-related PD-related No hospitalization Non-PD-related PD-related
(N = 92) (N = 9) (N = 3) (N = 80) (N = 10) (N = 13)
Age (years) 56.4 ± 16.2 59.5 ± 18.3 67.7 ± 7.6 57.1 ± 16.2 58.1 ± 15.7 55.2 ± 17.8
Gender
Female N (%) 52 (56.5) 2 (22) 3 (100) 45 (56) 2 (20)* 10 (77)
Male N (%) 40 (43.5) 7 (78) 0 (0) 35 (44) 8 (80) 3 (23)
Caucasian N (%) 70 (76) 6 (66) 1 (36) 59 (74) 8 (80) 10 (77)
Education (years) 8 (5–11) 8 (5–11) 5 (5–5)* 8 (5–11) 8 (5–11) 8 (5–11)
Caretaker (yes) N (%) 55 (60) 4 (50) 3 (100) 48 (60) 5 (50) 9 (61)
Diabetes mellitus N (%) 41 (44.6) 3 (37.5) 3 (100) 37 (46) 4 (40) 6 (46)
PD vintage (month) 15 (8–26) 23.5 (18–46) 24 (19–49) 16 (9–27) 20 (5–25) 23 (7–33)
PD: peritoneal dialysis.
*P < 0.05.
The mean laboratory results were compared between both periods. Only serum potassium levels increased significantly in the post-pandemic period compared to those in the pre-pandemic period (Table 3 ). The percentage of participants with mean results within the target decreased 12, 6 and 1 percentage points for serum potassium, hemoglobin, and serum respectively.Table 3 Comparison of laboratory results before and after the pandemic (N = 103).
Table 3 Pre-pandemic Post-pandemic P
Hemoglobin (g/dL) 10.6 ± 1.1 10.5 ± 1.1 0.08
10–12 g/dL N (%) 67 (65) 60 (59)
Serum phosphate (mg/dL) 6.3 ± 1.8 6.2 ± 1.8 0.15
3.5–5.5 mg/dL N (%) 43 (42) 42 (41)
Serum potassium (mEq/L) 4.79 ± 0.48 4.93 ± 0.54 0.006
3.5–5.5 mEq/L N (%) 97 (94) 84 (82)
Discussion
In this retrospective analysis, we found a significant increase in the total hospitalization rate due to PD-related causes and a slightly higher prevalence of metabolic disorders after the start of the pandemic and telemedicine implementation.
In active patients included in our PD program, the total hospitalization rate before the pandemic more than doubled after the pandemic (IRR = 2.48 [95% CI, 1.29–4.75]) and PD-related causes increased more than seven times (IRR = 7.25 [95% CI, 2.11–24.78]). Among the PD-related hospitalizations, infectious causes, which were absent in the pre-pandemic period, were the most prevalent in the post-pandemic period (9 of 15 [60%], eight patients were hospitalised due to peritonitis). We raised possible causes for this finding: lack of proper training for telemedicine transition for the healthcare team and patients; difficulties in selecting patients that could benefit from face-to-face visits; the impossibility of catheter exit-site visual inspection via the telemedicine tool used (telephone call); and patients’ delay in seeking medical evaluation at the first signs of infection due to fear of exposure to COVID-19.
Our literature search found only one study that implemented a similar telemedicine protocol and reported pre- and post-pandemic results. The investigation was conducted in PD centers in the Dominican Republic and included 946 patients. Contrary to our results, the authors did not find a difference in the hospitalization or peritonitis rates during the first 3 months before and after the pandemic.13 Years before the pandemic, Gallar et al. in 2007, reported a shorter duration of hospitalization in a group of patients undergoing PD who were followed up via hybrid monitoring (face-to-face appointments or telemedicine by teleconference every other month) than in a group of patients undergoing PD who were followed up via face-to-face appointments.10
We found few significant differences in patient characteristics according to the hospitalization status. Compared to patients not hospitalised, those hospitalised for a PD-related cause had a lower educational level only in the pre-pandemic period. There was a lower prevalence of women in the non-PD-related hospitalization group than in the no hospitalization group in the post-pandemic period.
No significant differences were found in the mean hemoglobin and serum phosphate levels between the two periods; however, serum potassium levels increased in the post-pandemic period than in the pre-pandemic period, and the percentage of patients with inadequate examination results increased from 6% (before the pandemic) to 18% (after the pandemic). Hyperkalaemia was the most prevalent disturb, since only 1 patient presented hypokalaemia in both periods, however, these differences do not seem to be clinically relevant. In patients undergoing PD, risk factors for hyperkalaemia have been poorly studied and have been associated with anuria.14 Based on our clinical experience, we believe this increment might reflect a possible decrease in dialysis prescription and dietary adherence.
Our results suggest that the healthcare model implemented for distance monitoring of our patients undergoing PD was not effective in evaluating signals not complained by the patients during the consultation, which could be reflected in clinical decompensations. This might have been implicated in the observed increase in PD-related hospitalization rates during the pandemic. As previously shown, when well-managed, distance monitoring can deliver quality care for this population with some extra advantages such as time-saving and lower cost compared to the current routine6, 8 In addition to routine consultations, technological solutions for remote daily monitoring of home-based automated peritoneal dialysis (APD) treatments are also available, making it feasible to detect problems early and correct inadequate dialysis delivery. Using this tool, researchers found a significant reduction in hospitalization rate and days after propensity score matching with APD group without this monitoring.15 Currently, this technology is not available for most developing countries due to cost issues.
Our study has some limitations. First, because of the observational retrospective design, no conclusions can be drawn regarding causality. In addition, the clinical context of our single-center population and the telemedicine model implemented in a pandemic scenario needs to be considered when extrapolating results to other populations. To the best of our knowledge, this is the first study to compare hospitalization rates and metabolic disorders 6 months before and after the pandemic. In addition, the outcome analysis of the same patients undergoing PD in both periods increased the power of the exposure effect.
In conclusion, in our PD population, we found a significant increase in hospitalization rates after the COVID-19 pandemic and telemedicine implementation mainly due to PD-related infectious causes. Strategies to improve telemedicine assistance with proper training and tools are needed for the PD population.
Conflict of interest
All authors declare that there is no conflict of interest.
==== Refs
References
1 Estado De Santa Catarina. Decreto no 515, de 17 de março de 2020. Published 2020. http://www.saude.sc.gov.br/coronavirus/arquivos/decreto_515_17_03_20.pdf [accessed 18.3.20].
2 Conselho Federal de Medicina. Ofício CFM No 1756/2020. Published 2020. http://portal.cfm.org.br/images/PDF/2020_oficio_telemedicina.pdf [accessed 20.3.20].
3 International Society of Peritoneal Dialysis. ISPD: Strategies regarding COVID-19 in PD patients. Published online 2020:1-4. http://ispd.org/wp-content/uploads/ISPD-PD-management-in-COVID-19_ENG.pdf [accessed 25.3.21].
4 World Health Organization. https://www.who.int/health-topics/coronavirus#tab=tab_1.
5 Deboni L.M. Neermann E.M.V. Calice-Silva V. Hanauer M.A. Moreira A. Ambrósio A. Development and implementation of telehealth for peritoneal dialysis and kidney transplant patients monitoring during the COVID-19 pandemic Braz J Nephrol 2020 1 7 10.1590/2175-8239-jbn-2020-0137 Published online November 30
6 Gallar P. Gutierrez M. Ortea O. Rodriguez I. Oliet A. Herrero J. Usefulness of telemedicine in the follow-up of peritoneal dialysis patients Nefrologia 26 2006 365 371 16892826
7 Dey V. Jones A. Spalding E.M. Telehealth: acceptability, clinical interventions and quality of life in peritoneal dialysis SAGE Open Med 4 2016 10.1177/2050312116670188 205031211667018
8 Magnus M. Sikka N. Cherian T. Lew S.Q. Satisfaction and improvements in peritoneal dialysis outcomes associated with telehealth Appl Clin Inform 8 2017 214 225 10.4338/ACI-2016-09-RA-0154 28246673
9 Milan Manani S. Rosner M.H. Virzì G.M. Giuliani A. Berti S. Crepaldi C. Longitudinal experience with remote monitoring for automated peritoneal dialysis patients Nephron 142 2019 1 9 10.1159/000496182 30699410
10 Gallar P. Vigil A. Rodriguez I. Ortega O. Gutierrez M. Hurtado J. Two-year experience with telemedicine in the follow-up of patients in home peritoneal dialysis J Telemed Telecare 13 2007 288 292 10.1258/135763307781644906 17785025
11 Scarpioni R. Manini A. Chiappini P. Remote patient monitoring in peritoneal dialysis helps reduce risk of hospitalization during Covid-19 pandemic J Nephrol 33 2020 1123 1124 10.1007/s40620-020-00822-0 32780306
12 Calice-Silva V. Cabral A.S. Bucharles S. Moura-Neto J.A. Figueiredo A.E. Franco R.P. Good practices recommendations from the Brazilian Society of Nephrology to Peritoneal Dialysis Services related to the new coronavirus (Covid-19) epidemic Braz J Nephrol 42 Suppl. 1 2020 18 21 10.1590/2175-8239-jbn-2020-s106
13 Polanco E. Aquey M. Collado J. Campos E. Guzman J. Cuevas-Budhart M.A. A COVID-19 pandemic-specific, structured care process for peritoneal dialysis patients facilitated by telemedicine: therapy continuity, prevention, and complications management Ther Apher Dial 2021 10.1111/1744-9987.13635 Published online
14 Goncalves F.A. de Jesus J.S. Cordeiro L. Piraciaba M.C.T. de Araujo L.K.R.P. Steller Wagner Martins C. Hypokalemia and hyperkalemia in patients on peritoneal dialysis: incidence and associated factors Int Urol Nephrol 52 2020 393 398 10.1007/s11255-020-02385-2 32016907
15 Sanabria M. Buitrago G. Lindholm B. Vesga J. Nilsson L. Yang D. Remote patient monitoring program in automated peritoneal dialysis: impact on hospitalizations Perit Dial Int 2019 1 7 10.3747/pdi.2018.00287 2015
| 36494279 | PMC9726640 | NO-CC CODE | 2022-12-08 23:18:17 | no | Nefrologia (Engl Ed). 2022 Dec 7; doi: 10.1016/j.nefroe.2022.11.023 | utf-8 | Nefrologia (Engl Ed) | 2,022 | 10.1016/j.nefroe.2022.11.023 | oa_other |
==== Front
Br J Anaesth
Br J Anaesth
BJA: British Journal of Anaesthesia
0007-0912
1471-6771
British Journal of Anaesthesia. Published by Elsevier Ltd.
S0007-0912(22)00638-9
10.1016/j.bja.2022.10.043
Book Review
Bhatia Aneeta
Chen James
Huang Jiapeng ∗
Louisville, KY, USA
∗ Corresponding author.
7 12 2022
7 12 2022
Vasu T. Managing Long COVID Syndrome2022TFM Publishing Limited300Price £25.00. ISBN 978191375520122 9 2022
24 10 2022
© 2022 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.
2022
British Journal of Anaesthesia
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmcThe global coronavirus disease 2019 (COVID-19) pandemic, an infectious disease caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has affected >617 million people worldwide.1 Despite the rapid identification of the genetic code of this RNA virus and the monumental speed at which drugs and vaccines were developed and introduced, the disease spread rapidly as more contagious variants of the virus continue to emerge. Long COVID syndrome is an illness encompassing a wide range of new, returning, or ongoing health problems that people experience at least 4 weeks after being infected with SARS-CoV-2. Estimates of the proportion of people who experience long COVID syndrome vary: 13.3% at 1 month or longer after infection; 2.5% at 3 months or longer, based on self reporting; and >30% at 6 months among patients who were hospitalised.2 Multisystem organ damage from the initial infection might lead to long-term symptoms that, if persistent, are diagnosed as long COVID syndrome. Long COVID syndrome can cause devastating physical, neuropsychological, and emotional disabilities with major global socioeconomic burden.2
Many aspects of long COVID syndrome and the underlying mechanisms remain poorly understood. To provide insights into the pathophysiology, epidemiology, diagnosis, and management of long COVID syndrome and to create awareness of this unique syndrome, Thanthullu Vasu, a clinician and consultant in pain management at the University Hospital of Leicester (UK) has authored this timely publication. Managing Long COVID Syndrome is a concise yet comprehensive clinical book on a relatively new, complex, and debilitating global disease of potential interest to clinicians, students, and patients. In this 310-page book, the author uses up to date peer-reviewed journals and scientific publications to discuss the subject in 10 sections comprising 52 short chapters. In the first two sections, the author leads readers through the identification of the disease, its aetiology, epidemiology, and clinical manifestations. Causative mechanisms of the lingering and debilitating multifactorial symptoms of exertional malaise, shortness of breath, fatigue, myalgia, and cognitive decline with brain fog are extensively discussed. Of particular interest in this section is the chapter on the psychological trauma that persists long after recovery, specifically in those with post-intensive care syndrome. The residual anxiety, depression from grief and loss, unresolved pain, and sleep disturbance can also be deeply disturbing and debilitating. Given the high incidence of SARS-CoV-2 infection worldwide, the potential burden of long COVID syndrome is concerning. Therefore, the author purposefully presents guidelines from different health organisations and countries to offer global perspectives.
The next few sections cover management of long COVID syndrome, including a special chapter on pain management with complimentary and novel pain therapies. Sections on pain management and long COVID syndrome in children are particularly interesting to anaesthesiologists as they may encounter these patients in both the pain clinic and operating rooms. Of note, 8% of children experience long COVID syndrome at 3–6 months after initial infection, and children are also more sensitive to the social and psychological effects of lockdown and isolation.
Managing Long COVID Syndrome is an easy-to-read book providing basic and advanced information relating to long COVID and offering strategies with medical solutions to manage many long COVID symptoms. Each chapter has bulleted key points and well organised tables to enable readers to grasp important concepts easily. It raises awareness amongst clinicians and scientists, and highlights the need for continued research into the underlying pathophysiology. The entire book is well organised into sections covering virology, pathophysiology, classification, presenting features, diagnosis, management, physical, psychological consequences, and ongoing research, along with global policies and guidelines. In addition, it leaves readers appropriately cognisant of the overall socioeconomic impact of long COVID syndrome.
Despite being an overall informative read, the book does have its drawbacks. Realising there are more flexible resources, such as journal review articles and online resources discussing long COVID syndrome, the literature for which is still rapidly evolving and for which no consensus exists regarding diagnosis and management, the value of a textbook would be significantly enhanced by the provision of an online or dynamically updated version. The book lacks scientific and visual details of the causative metabolic pathways leading to neurological, psychological, and cognitive decline that would interest clinicians partially because of a limitation of available scientific knowledge. It lacks patient stories to keep the reader interested and emotionally involved. The book fails to cover the disease process and elevated anxiety and depression in pregnant persons.3 , 4 The colour combination used in the key points summary, charts, and tables could be more appealing. Finally, this book appears to focus on burden of disease and management strategies as measured and treated in high-income countries, not in countries with limited resources.
In conclusion, the few negatives cannot compete with the strength of the content of this book and the education it offers. It is a comprehensive resource covering a vast variety of mental and physical health issues from long COVID syndrome that affects our society today. Managing Long COVID Syndrome is highly recommended as being amongst the few publications on a highly contagious disease with lasting consequences. For many with limited knowledge or experience with long COVID syndrome, this book brings forth awareness and empathy for confused, misunderstood, often isolated, and unsupported patients. While scientific evidence accumulates, the implications of long COVID syndrome are potentially far reaching for patients and clinicians around the world. The book partially addresses an otherwise unmet need and is potentially useful to both healthcare professionals and policy makers.
Declaration of interest
The authors declare that they have no conflicts of interest.
==== Refs
References
1 Johns Hopkins University & Medicine Coronavirus Resource Center. Available from: https://coronavirus.jhu.edu/map.html (accessed 1 October 2022).
2 Centers for Disease Control and Prevention. Long COVID or post-COVID conditions. Available from: https://www.cdc.gov/coronavirus/2019-ncov/long-term-effects/index.html (accessed 1 October 2022).
3 Lebel C. MacKinnon A. Bagshawe M. Elevated depression and anxiety symptoms among pregnant individuals during the COVID-19 pandemic J Affect Disord 277 2020 5 13 32777604
4 Khamees R.E. Taha O.T. Ali T.Y.M. Anxiety and depression during pregnancy in the era of COVID-19 J Perinat Med 49 2021 674 677 34062628
| 0 | PMC9726641 | NO-CC CODE | 2022-12-08 23:18:17 | no | Br J Anaesth. 2022 Dec 7; doi: 10.1016/j.bja.2022.10.043 | utf-8 | Br J Anaesth | 2,022 | 10.1016/j.bja.2022.10.043 | oa_other |
==== Front
Am J Obstet Gynecol MFM
Am J Obstet Gynecol MFM
American Journal of Obstetrics & Gynecology Mfm
2589-9333
Elsevier Inc.
S2589-9333(22)00261-0
10.1016/j.ajogmf.2022.100831
100831
Original Research
Assessing the impact of telehealth implementation on postpartum outcomes for Black birthing people
Kumar Natasha R. MD ⁎
Arias Maria Paula MDMSCE
Leitner Kirstin MD
Wang Eileen MD
Clement Elizabeth G. MD
Hamm Rebecca Feldman MD, MSCE
Department of Obstetrics and Gynecology, University of Pennsylvania Health Systems, Philadelphia, Pennsylvania
⁎ Corresponding Author: Natasha R. Kumar, MD; 3400 Spruce Street, Philadelphia PA 19104, Phone: (717) 712-8582
7 12 2022
7 12 2022
1008318 7 2022
10 11 2022
5 12 2022
© 2022 Elsevier Inc. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
The COVID-19 pandemic led to rapid uptake of telemedicine services, which have been shown to be potentially cost-saving and of comparable quality to in-person care for certain populations. However, there are some concerns regarding feasibility of implementation for marginalized populations and the impact of widespread implementation of these services on health disparities has not been well studied.
Objective
To assess the impact of implementation of telehealth for postpartum care during the COVID-19 pandemic on racial disparities in visit attendance and completion of postpartum care goals.
Study Design
In this retrospective cohort study at a single tertiary care center, we compared differences in outcomes between all Black and non-Black patients who had scheduled postpartum visits (PPV) prior to and following implementation of telehealth for postpartum care. Our primary outcome was PPV attendance. Secondary outcomes included postpartum depression screening, contraception selection, breastfeeding status, completion of postpartum 2-hour glucose tolerance test (GTT) and cardiology follow-up for hypertensive disorders of pregnancy. In multivariable analysis, interaction terms evaluated the differential impact of telehealth implementation by race.
Results
Of the 1,579 patients meeting inclusion criteria (pre-implementation: 780, post-implementation: 799), 63% self-identified as Black. In the pre-implementation period, Black patients were less likely to attend a postpartum visit than non-Black patients (Black: 63.9% vs non-Black: 88.7%, aOR= 0.48, 95% CI: 0.29-0.79). In the post-implementation period, there was no difference in postpartum visit attendance by race (Black: 79.1% vs non-Black: 88.6%, aOR= 0.74, 95% CI: 0.45-1.21). Significant differences across race in postpartum depression screening during the pre-implementation period also became non-significant in the post-implementation period. Implementation of telehealth for postpartum care significantly reduced racial disparities in postpartum visit attendance (interaction p=0.005).
Conclusion
Implementation of telehealth for postpartum care during the COVID-19 pandemic is associated with decreased racial disparities in postpartum visit attendance.
Keywords
Implementation science
Racial disparities
Postpartum care
Telemedicine
COVID-19 pandemic
==== Body
pmcINTRODUCTION
The United States has one of the highest pregnancy-related mortality rates among high-income countries. More than half of pregnancy-related deaths occur postpartum and state-level data show that three in five maternal deaths may be preventable.1 The racial disparities in pregnancy-related mortality in the United States are deeply troubling; between 2007 and 2016, there were 40.8 pregnancy-related deaths per 100,000 live births among non-Hispanic Black mothers compared with 12.7 pregnancy-related deaths per 100,000 live births among non-Hispanic White mothers.2 Increasing attention has been given to improving and expanding postpartum care, including state-level legislation extending postpartum Medicaid coverage,3-5 publication of professional guidelines and checklists for the postpartum period,6, 7 and development of innovative postpartum healthcare delivery tools.8
It is important to think critically about racial disparities in health outcomes or engagement with care. Holding pregnant patients responsible for poor reproductive or neonatal outcomes “ignores the circumstances, environments and situations in which each woman seeks to maintain health, to become pregnant, and to safely give birth to children.”9 Structural racism, or the “ways in which societies foster racial discrimination through mutually reinforcing systems. . . [which] in turn reinforce discriminatory beliefs, values, and distribution of resources” has impacted the provision of obstetric care.10 A contemporary example of structural racism is the relationship between the mutually reinforcing systems of residential segregation and policing; communities that were subject to “redlining,” or the historic segregation and disinvestment through racially motivated policies dictating housing loan provision now experience increased rates of fatal encounters with police in present day.11, 12 For pregnant patients specifically, historical injustices enacted on marginalized communities not only have reduced trust and engagement with healthcare providers and their recommendations,13, 14 but also have systemically enforced resource deprivation such as lack of transportation, stable housing, and adequate nutrition. These issues directly impact engagement with postpartum care and interpregnancy health, as populations that lack stable housing, transportation or food resources may prioritize these needs over healthcare-related demands or may distrust healthcare providers due to potential relationships with family regulatory systems (i.e. the U.S. child welfare system) that may act in response to significant resource deprivation, among other issues. Reinventing healthcare delivery systems to provide flexibility for patients in how and when they engage with care may mitigate and eliminate these disparities, which is an essential part of work to improve poor obstetric outcomes in the United States.
Prior to the COVID-19 pandemic, limited data demonstrated that perinatal telemedicine can provide health outcomes comparable with in-person care for diabetes, hypertension, and perinatal depression.15 Preliminary data also demonstrated that telemedicine could potentially be cost-saving when utilized on a large scale.16 The COVID-19 pandemic then led to rapid development of healthcare infrastructure and institution of modified reimbursement structures that support the widespread implementation of telemedicine. However, it remains unclear whether telehealth models will address disparities in perinatal care, as feasibility of these services can be limited by access to technology and availability of childcare support.17 On one hand, telehealth directly addresses concerns regarding lack of transportation that limit access to care for marginalized populations, as demonstrated by data from telemedicine programs for antenatal care for rural populations.18 At our own institution, remote blood pressure monitoring programs have also been shown to reduce racial disparities in postpartum follow-up for hypertensive disorders of pregnancy.19 On the other hand, these populations may not have access to the technology required to facilitate telehealth, such as stable internet access or cell phones. A retrospective cohort of patients in New York City during the early portion of the COVID-19 pandemic demonstrated that patients with public insurance were significantly less likely to utilize telehealth for prenatal care than patients with private insurance.20 Given the paucity of data on impact of telehealth implementation on postpartum care, particularly for marginalized populations, we sought to investigate how implementation of telehealth for postpartum care impacted racial disparities in postpartum visit attendance and completion of postpartum care goals during the beginning of the COVID-19 pandemic.
MATERIALS AND METHODS
This is a retrospective cohort study at an urban tertiary care center with two clinical sites serving a diverse patient population comparing a 3-month period prior to implementation of telehealth for postpartum care (March 16-June 16, 2019) to the identical calendar months following implementation of telehealth in 2020. This center delivers over 4,000 patients annually. Patients were included in this analysis if they had a scheduled postpartum visit (PPV) of any modality at an outpatient obstetric clinic affiliated with this tertiary care center between 21 and 56 days postpartum. There were no exclusion criteria.
In response to the COVID-19 pandemic, our institution transitioned the majority of postpartum visits to telehealth, either via video and audio or audio only, starting on March 16, 2020. Outpatient obstetric clinics recommended telehealth PPVs as the first-line modality for visits to limit COVID-19 transmission, while also offering limited in-person PPVs for specific clinical needs such as incision checks for wound complications, Pap smear completion, and initiation of long-acting reversible contraception (LARC) for patients who had made this desire clear prior to discharge from delivery admission. Telehealth visits were scheduled via telephone or video by office staff, between 2-6 weeks postpartum depending on each patient's clinical needs. Telehealth visits were preferentially performed via video telehealth platform requiring internet (or cellular data) access; if patients had issues with the video platform, then the visit was switched to audio only via telephone. In-person follow up visits after PPVs were scheduled for colposcopy completion and for LARC initiation (if this desire became evident during their PPV). Because of infrastructure development preceding the pandemic, implementation of widespread telemedicine at the institutional level was smooth, with minimal interruptions in service operations.
Our primary outcome was PPV attendance, defined as completion of the postpartum visit regardless of modality (in-person or telehealth). Secondary outcomes included postpartum depression screening (defined as completion of an EDPS, PHQ-2 or PHQ-9), contraception choice, feeding plan at time of postpartum visit and, if applicable, follow-up for completion of 2-hour GTT (for patients with gestational diabetes mellitus), Pap smear and colposcopy completion (for patients with cervical dysplasia), and cardiology follow-up visit (for patients with preeclampsia with severe features) by 3 months postpartum. Secondary outcomes that were documented at the postpartum visit, i.e. postpartum depression screening, feeding plan at time of postpartum visit, and contraception, were only abstracted if patients completed a postpartum visit. Other secondary outcomes, namely 2-hour GTT, Pap smear and colposcopy, and cardiology follow-up, were only abstracted when completion was clinically indicated based on institutional guidelines.
To minimize potential confounding, we opted to compare differences across racial groups rather than within the Black population across time. In this way, we compare groups that are subject to similar environmental factors during a rapidly evolving pandemic. We subsequently used interaction term analysis to assess whether there was a reduction in disparities due to telehealth implementation.
Data was collected by research staff on demographics, PPV attendance, and postpartum outcomes through electronic health record review. Race/ethnicity were self-reported within the electronic medical record (EMR). Bivariate analyses were performed using Wilcoxon rank-sum tests or t-tests for continuous variables and chi-square tests for categorical variables. Multivariate comparison and interaction term analysis were performed for our primary outcome as the remainder of the data on secondary outcomes was exploratory in nature. Multivariable logistic regressions comparing the odds of the primary outcome in the pre- and post-implementation groups for Black and non-Black patients included all variables with p<0.20 in the bivariate analyses for association with both exposure and outcome. Backwards stepwise regression was used to create parsimonious models using p>0.20 for elimination. Interaction analyses were also performed to assess the impact of telehealth implementation on differences in postpartum outcomes by race. Covariates initially evaluated in the multivariable models included age (continuous variable), education (categorical variable defined as high school education or less vs more than high school education), parity (categorical variable defined as nulliparous vs multiparous), insurance (categorical variable defined as privately insured vs publicly insured), scant prenatal care (defined as <5 prenatal visits), gestational diabetes, chronic hypertension, and hypertensive disorders of pregnancy. After stepwise backwards regression, education, parity, insurance and scant prenatal care remained in the models for the pre-implementation group and the interaction analysis, while only insurance, parity and scant prenatal care remained for the post-implementation group.
All analyses were performed using Stata version 17 (StataCorp LLC, College Station, TX). This study was approved by the Institutional Review Board with waiver of informed consent.
RESULTS
Of the 1,579 subjects meeting inclusion criteria, 780 had a PPV scheduled in the pre-telehealth period and 799 in the post-telehealth period. In our post-implementation group, 317 patients had audio telehealth visits, 174 patients had video telehealth visits, and 156 patients had in-person visits (data not shown; 15 missing values). 63% of patients self-identified as Black (n=996). The majority of our non-Black patients self-identified as White (26% of overall study population). 16 patients (1% of overall study population) self-identified as Latinx.
Demographic and clinical characteristics of study groups are detailed in Table 1 . In both pre- and post-implementation groups, Black patients were younger and were more likely to have a high school education or less, to be publicly insured, to be multiparous, and to receive scant prenatal care (Table 1). They were also more likely to have chronic hypertension or hypertensive disorders of pregnancy (Table 1).Table 1 Demographic and Clinical Characteristics of Patients Receiving Postpartum Care Prior to and Following Implementation of Telehealth
Table 1: Pre-Intervention (n=780)b Post-Intervention (n=799)c
Characteristicsa Black
(n=513) Non-Blackd
(n=267) P value Black
(n=483) Non-Blacke
(n=316) P value
Maternal age 27.9 (24.0-32.6) 32.7 (29.8-36.1) <0.001 28.0 (23.5-33.4) 32.3 (29.6-35.1) <0.001
Education
≤HS
>HS
379 (76.1%)
119 (23.9%)
64 (24.2%)
201 (75.6%) <0.001
385 (80.7%)
92 (19.3%)
74 (23.9%)
236 (76.1%) <0.001
Insurance
Private
Public
119 (24.3%)
370 (75.6%)
195 (75.6%)
63 (24.4%) <0.001
121 (25.6%)
347 (74.2%)
239 (76.4%)
74 (23.6%) <0.001
Nulliparous 175 (34.1%) 141 (52.8%) <0.001 177 (36.7%) 182 (57.6%) <0.001
Scant prenatal care
(<5 visits) 70 (13.8%) 14 (5.3%) <0.001 80 (16.8%) 17 (5.5%) <0.001
Gestational diabetes 25 (4.9%) 20 (7.5%) 0.14 34 (7.0%) 25 (7.9%) 0.65
Pregestational diabetes 12 (2.5%) 7 (2.8%) 0.76 16 (3.6%) 7 (2.4%) 0.38
Chronic hypertension 55 (10.7%) 7 (2.6%) <0.001 54 (11.2%) 13 (4.1%) <0.001
Hypertensive disorder of pregnancy 189 (36.8%) 64 (24.0%) <0.001 157 (32.5%) 73 (23.1%) 0.004
Mental health disorder 122 (23.8%) 59 (22.1%) 0.60 110 (22.8%) 64 (20.3%) 0.40
a Median (Q1-Q3) for maternal age, n(%) for remainder of characteristics
b Missing values for pre-intervention group: education, n=763; insurance, n=747; scant prenatal care, n=773; pregestational diabetes, n=735
c Missing values for post-intervention group: education, n=787; insurance, n=781; scant prenatal care, n=786; pregestational diabetes, n=740
d White, n=188 (24.0%); Asian, n=65 (8.3%); Latinx, n=7 (1.0%); Native American, n=9 (1.2%).
e White, n=227 (28.4%); Asian, n=72 (9.0%); Latinx n=9 (1.1%); Native American, n=5 (0.6%)
In both unadjusted and adjusted analyses, Black patients were less likely to attend a postpartum visit than non-Black patients prior to implementation of telehealth (63.9% vs. 88.7%, p=<0.001; adjusted odds ratio [aOR]= 0.48, 95% confidence interval [CI]: 0.29-0.79). After implementation of telehealth, Black patients were still less likely than non-Black patients to attend a postpartum visit on bivariate analysis (79.1% vs. 88.6%, p<0.001), but after controlling for insurance, education, parity and scant prenatal care, there was no longer a significant difference in postpartum visit attendance by race (aOR= 0.74, 95% CI: 0.45-1.21). There was a statistically significant differential impact of implementation of telehealth on postpartum visit attendance by race (interaction p = 0.005).
Unadjusted analyses for secondary outcomes are also shown in Table 2 . Prior to implementation of telehealth, Black patients were significantly less likely than non-Black patients to receive a postpartum depression screen, complete a Pap smear, breastfeed their infants, and attend cardiology follow-up visits prior to implementation of telehealth. In addition, prior to implementation of telehealth, Black patients were more likely than non-Black patients to obtain long-acting reversible contraception (LARC) or present to the emergency department for care postpartum. Following telehealth implementation, differences by race were no longer significant for postpartum depression screening, Pap smear completion, cardiology follow-up visit attendance and LARC initiation. Black patients remained less likely to breastfeed and more likely to visit the emergency department for postpartum care after implementation of telehealth.Table 2 Bivariate Analysis of Postpartum Milestone Completion for Patients Receiving Postpartum Care Prior To and Following Implementation of Telehealth
Table 2: Pre-Intervention (n=780)a Post-Intervention (n=799)b
Black Non-Black P value Black Non-Black P value
Postpartum visit attendance 328/513 (63.9%) 237/267 (88.7%) <0.001 382/483 (79.1%) 280/316 (88.6%) <0.001
Depression screeninga 169/328 (51.5%) 199/237 (84.0%) <0.001 324/382 (84.8%) 247/280 (88.2%) 0.21
Any breastfeedinga 221/327 (67.6%) 199/237 (84.0%) <0.001 243/382 (63.6%) 230/278 (82.7%) <0.001
Glucose screeningb 6/25 (24.0%) 6/20 (30.0%) 0.65 7/34 (20.6%) 8/25 (32.0%) 0.32
Cardiology follow-upb 19/43 (44.2%) 10/13 (76.9%) 0.04 26/43 (60.5%) 10/16 (62.5%) 0.89
ED presentations 47/513 (9.2%) 12/267 (4.5 %) 0.02 34/483 (7.0%) 10/316 (3.2%) 0.02
Hospital readmissions 36/513 (7.0%) 19/267 (7.1%) 0.96 46/483 (9.5%) 16/316 (5.1%) 0.02
Any contraceptiona
LARC or BTL 197/231 (85.3%)
89/193 (46.1%) 191/224 (85.3%)
64/180 (35.6%) 1.00
0.04 213/253 (84.2%)
70/206 (34.0%) 238/261 (91.2%)
61/229 (26.6%) 0.02
0.10
Pap completionb 49/86 (57.0%) 41/47 (87.2%) <0.001 31/74 (41.9%) 21/39 (53.9%) 0.23
Colposcopy completionb 1/7 (14.3%) 10/40 (25.0%) 0.54 3/8 (37.5%) 12/27 (44.4%) 0.72
a These outcomes were collected for all patients who completed a postpartum visit.
b These outcomes were collected for all eligible patients: Pap, n=246; colposcopy, n=82; glucose screening, n=104; cardiology follow-up, n=115
Adjusted analyses are demonstrated in Table 3 . After adjusting for education, parity, insurance status and scant prenatal care, Black patients were significantly less likely than non-Black patients to complete postpartum depression screening (aOR=0.48, 95% CI: 0.29-0.80) prior to implementation of telehealth. Following implementation of telehealth, differences in postpartum depression screening by race were no longer significant (aOR=0.72, 95% CI: 0.43-1.24).Table 3 Multivariable Analysis Comparing Odds of Accomplishing Postpartum Milestones between Black and non-Black patients
Table 3: Pre-Implementation Groupa Post-Implementation Groupb
Postpartum visit attendance 0.48 (0.29-0.79) 0.74 (0.45-1.21)
Depression screening 0.48 (0.29-0.80) 0.72 (0.43-1.24)
Any contraception
LARC/BTL 1.38 (0.72-2.66)
0.90 (0.52-1.55) 0.57 (0.30-1.06)
1.62 (1.01-2.61)
Pap completion 0.33 (0.10-1.04) 0.99 (0.36-2.68)
Colposcopy completion 0.78 (0.03-18.37)c 8.08 (0.59-110.7)
Any breastfeeding 0.69 (0.41-1.15) 0.58 (0.31-0.89)
Glucose screening 1.34 (0.26-6.91)c 0.55 (0.14-2.18)c
Cardiology follow-up 0.57 (0.09-3.57) 0.52 (0.13-2.15)
ED presentations 1.06 (0.49-2.28) 2.15 (0.93-4.99)
Hospital readmissions 0.89 (0.43-1.89) 1.92 (0.98-3.77)
a Regressions for the pre-implementation group controlled for insurance, education, parity and scant prenatal care based on backwards stepwise selection.
b Regressions for post-implementation group controlled for insurance, parity and scant prenatal care based on backwards stepwise selection.
c Scant prenatal care was not included in multivariable logistic regression due to collinearity.
DISCUSSION
Principal Findings
In this retrospective cohort study, implementation of telehealth in postpartum care was associated with amelioration of disparities in postpartum visit attendance between Black and non-Black patients. While our data on secondary outcomes is exploratory, the trends in postpartum depression screening indicate there may be potential for reduction of disparities in these domains with telehealth implementation as well.
Results in Context
Existing data on telehealth provision of postpartum care has overall been reassuring,21 demonstrating equivalent or sometimes improved quality of care relative to in-person care. Offering lactation consulting services via telehealth has been associated with increased success and maintenance of exclusive breastfeeding.22 A systematic review also demonstrated that telehealth interventions improve obstetrical outcomes, including breastfeeding status and continuation of oral and injectable contraception.23 Remote blood pressure monitoring has led to increased compliance, retention, and patient satisfaction.24-26
Our data showed a reduction in racial disparities in postpartum visit attendance with telehealth implementation. Existing literature showed mixed effects of telehealth implementation on disparities. On one hand, studies have demonstrated improvement of postpartum outcomes such as breastfeeding and contraception initiation with implementation of telehealth, independent of race.23 A limited number of studies have examined the impact of telehealth implementation on racial disparities in postpartum outcomes such as hypertension follow-up, demonstrating improvement with incorporation of telehealth.27 On the other hand, several studies show decreased uptake of telehealth implementation by Black and Latinx patients,28, 29 including retrospective cohort demonstrating decreased engagement with prenatal care by publicly insured patients after implementation of telehealth.20 It is unsurprising that the literature examining the impact of telehealth implementation on marginalized populations does not demonstrate consistent impact across interventions, as marginalized communities vary widely in their needs and preferences. While our initial findings are promising, ongoing research – likely requiring qualitative work in addition to quantitative analysis – will be required to assess impact of implementation. Additional questions regarding patients’ experience with care provided via telehealth also must be addressed in future research; for example, patients without adequate childcare support may be able to attend a telehealth postpartum visit, thus satisfying a quantitative metric assessing access to care, but may have a very different experience of that care than they would with in-person care because they are caring for their children while attending the appointment.
Clinical and Research Implications
Our exploratory data on other postpartum outcomes showed a mixed effect of telehealth implementation on racial disparities. The trend in postpartum depression screening indicates potential for reduction of disparities in this domain as differences in the pre-implementation period became non-significant in the post-implementation period. While our study did not assess differences across time within racial groups, the absolute rates of Pap completion and LARC initiation may have declined for both Black and non-Black patients from pre- to post-implementation, raising concerns about the impact of telehealth implementation on achievement of postpartum care goals that require in-person examination by providers; however, this effect was not observed for postpartum colposcopy. Future studies examining the impact of telehealth implementation on health disparities should focus on patient experiences and quality of care for outcomes that ultimately required in-person care.
Strengths and Limitations
Our study has several strengths as well as some limitations. Our data includes a sizable cohort of Black patients. Standardized electronic medical records between inpatient and outpatient settings at our institution allowed for consistent clinical documentation and extensive individual chart review facilitated collection of detailed patient level data. The comparison of Black to non-Black patients, rather than White patients, supports an equity-focused framework supported by reproductive justice advocates and health equity researchers by not establishing the White population as an aspirational norm.30 Limitations of this study include its retrospective nature. Our assessment of secondary postpartum outcomes was limited by the data available in the electronic medical record. Our analysis cannot account for the overall impact of the COVID-19 pandemic, which led to rapid implementation and de-implementation of evolving practices for our post-implementation group. Because the post-implementation study period was at the start of the pandemic, our findings may not be generalizable to current practice. Further research should explore whether the findings of this paper remain applicable to current practices in telehealth. Given the difficulty in distinguishing the impact of telehealth implementation from other less definitive effects of the COVID-19 pandemic, the impact of telehealth upon racial disparities in postpartum outcomes may need to be reevaluated outside of the pandemic. We are also unable to independently report on ethnicities such as the Latinx population due to the limitations of our sample size. Characterizing care in other marginalized communities should be a focus of future research.
Conclusions
As the COVID-19 pandemic has led to the widespread adoption of telehealth models of prenatal and postpartum care, it is critical to evaluate the impact of these interventions on access to and quality of care for marginalized populations. Our study indicates that telehealth implementation may ameliorate disparities in postpartum care for Black patients. Encouraging ongoing support for innovative delivery mechanisms such as telehealth may lead to more equitable obstetric care during the postpartum period.
FINANCIAL SUPPORT: The authors did not receive financial support for this work.
ACKNOWLEDGEMENTS: The authors have no acknowledgements.
Condensation: Implementation of telehealth for postpartum care during the COVID-19 pandemic was associated with a significant improvement in racial disparities in postpartum visit attendance.
AJOG at a Glance
A. This study was conducted to assess the impact of implementation of telemedicine upon racial disparities in postpartum care.
B. In this retrospective cohort, disparities in postpartum visit attendance for Black patients resolved after implementation of telehealth. While our data on other postpartum care milestones are exploratory, the trend in postpartum depression screening indicates there may be potential for reduction or resolution of disparities in this domain with telehealth implementation.
C. Telemedicine had previously been shown to be of comparable quality and potentially cost-saving in select populations. This study shows that telemedicine could potentially ameliorate racial disparities in certain postpartum outcomes.
Blinded Conflict of Interest Statement: The authors have no conflicts of interest to disclose.
Reference
1. Petersen EE, Davis NL, Goodman D, et al. Vital signs: pregnancy-related deaths, United States, 2011–2015, and strategies for prevention, 13 states, 2013–2017. Morbidity and Mortality Weekly Report 2019;68:423.
2. Hoyert DL, Miniño AM. Maternal mortality in the United States: changes in coding, publication, and data release, 2018. 2020.
3. Kumar NR, Borders A, Simon MA. Postpartum Medicaid extension to address racial inequity in maternal mortality: American Public Health Association, 2021 (vol 111).
4. Ranji U, Gomez I, Salganicoff A. Expanding postpartum Medicaid coverage. Women's Health Policy Issue Brief 2019.
5. Eckert E. Preserving the momentum to extend postpartum Medicaid coverage. Women's Health Issues 2020;30:401-04.
6. McKinney J, Keyser L, Clinton S, Pagliano C. ACOG Committee Opinion No. 736: optimizing postpartum care. Obstetrics & Gynecology 2018;132:784-85.
7. Gibson KS, Hameed AB. Society for Maternal-Fetal Medicine Special Statement: Checklist for postpartum discharge of women with hypertensive disorders. American Journal of Obstetrics & Gynecology 2020;223:B18-B21.
8. Hirshberg A, Downes K, Srinivas S. Comparing standard office-based follow-up with text-based remote monitoring in the management of postpartum hypertension: a randomised clinical trial. BMJ quality & safety 2018;27:871-77.
9. Scott KA, Britton L, McLemore MR. The ethics of perinatal care for black women: dismantling the structural racism in “mother blame” narratives. The Journal of perinatal & neonatal nursing 2019;33:108-15.
10. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and interventions. The Lancet 2017;389:1453-63.
11. Mitchell J, Chihaya GK. Tract level associations between historical residential redlining and contemporary fatal encounters with police. Social Science & Medicine 2022;302:114989.
12. Bloch S, Phillips SA. Mapping and making gangland: A legacy of redlining and enjoining gang neighbourhoods in Los Angeles. Urban Studies 2021;59:750-70.
13. Quinn SC. Belief in AIDS as a form of genocide: implications for HIV prevention programs for African Americans. Journal of Health Education 1997;28:S-6-S-12.
14. Bogart LM, Wagner GJ, Green Jr HD, et al. Medical mistrust among social network members may contribute to antiretroviral treatment nonadherence in African Americans living with HIV. Social Science & Medicine 2016;164:133-40.
15. Odibo IN, Wendel PJ, Magann EF. Telemedicine in obstetrics. Clinical obstetrics and gynecology 2013;56:422-33.
16. Magann EF, McKelvey SS, Hitt WC, Smith MV, Azam GA, Lowery CL. The use of telemedicine in obstetrics: a review of the literature. Obstetrical & gynecological survey 2011;66:170-78.
17. Logsdon MC, Pinto-Foltz MD, Stein B, Usui W, Josephson A. Adapting and testing telephone based depression care management intervention for adolescent mothers. Archives of women's mental health 2010;13:307.
18. Kim EW, Teague-Ross TJ, Greenfield WW, Williams DK, Kuo D, Hall RW. Telemedicine collaboration improves perinatal regionalization and lowers statewide infant mortality. Journal of Perinatology 2013;33:725-30.
19. Hirshberg A, Sammel MD, Srinivas SK. Text message remote monitoring reduced racial disparities in postpartum blood pressure ascertainment. American Journal of Obstetrics & Gynecology 2019;221:283-85.
20. Limaye MA, Lantigua-Martinez M, Trostle ME, et al. Differential uptake of telehealth for prenatal care in a large New York City academic obstetrical practice during the COVID-19 pandemic. American Journal of Perinatology 2021;38:304-06.
21. Kern-Goldberger AR, Srinivas SK. Telemedicine in obstetrics. Clinics in Perinatology 2020;47:743-57.
22. Ferraz dos Santos L, Borges RF, de Azambuja DA. Telehealth and breastfeeding: an integrative review. Telemedicine and e-Health 2020;26:837-46.
23. DeNicola N, Grossman D, Marko K, et al. Telehealth interventions to improve obstetric and gynecologic health outcomes: a systematic review. Obstetrics and gynecology 2020;135:371.
24. Hauspurg A, Lemon LS, Quinn BA, et al. A postpartum remote hypertension monitoring protocol implemented at the hospital level. Obstetrics and gynecology 2019;134:685.
25. Janssen MK, Demers S, Srinivas SK, et al. Implementation of a text-based postpartum blood pressure monitoring program at 3 different academic sites. American journal of obstetrics & gynecology MFM 2021;3:100446.
26. Sawyer MR, Jaffe EF, Naqvi M, Sarma A, Barth Jr WH, Goldfarb IT. Establishing better evidence on remote monitoring for postpartum hypertension: a silver lining of the coronavirus pandemic. American Journal of Perinatology Reports 2020;10:e315-e18.
27. Khosla K, Suresh S, Mueller A, et al. Elimination of racial disparities in postpartum hypertension follow-up after incorporation of telehealth into a quality bundle. American Journal of Obstetrics & Gynecology MFM 2022;4:100580.
28. Gmunder KN, Ruiz JW, Franceschi D, Suarez MM. Demographics associated with US healthcare disparities are exacerbated by the telemedicine surge during the COVID-19 pandemic. Journal of telemedicine and telecare 2021:1357633×211025939.
29. Chunara R, Zhao Y, Chen J, et al. Telemedicine and healthcare disparities: a cohort study in a large healthcare system in New York City during COVID-19. Journal of the American Medical Informatics Association 2021;28:33-41.
30. Bray SR, McLemore MR. Demolishing the myth of the default human that is killing Black mothers. Frontiers in public health 2021;9:630.
| 36496115 | PMC9726646 | NO-CC CODE | 2022-12-08 23:18:17 | no | Am J Obstet Gynecol MFM. 2022 Dec 7;:100831 | utf-8 | Am J Obstet Gynecol MFM | 2,022 | 10.1016/j.ajogmf.2022.100831 | oa_other |
==== Front
One Health
One Health
One Health
2352-7714
The Authors. Published by Elsevier B.V.
S2352-7714(22)00103-3
10.1016/j.onehlt.2022.100471
100471
Article
One health system supporting surveillance during COVID-19 epidemic in Abruzzo region, southern Italy
Di Lorenzo Alessio
Mangone Iolanda ⁎
Colangeli Patrizia
Cioci Daniela
Curini Valentina
Vincifori Giacomo
Mercante Maria Teresa
Di Pasquale Adriano
Radomski Nicolas
Iannetti Simona
Istituto Zooprofilattico Sperimentale dell'Abruzzo e del Molise “G. Caporale”, Campo Boario, 64100 Teramo, Abruzzo, Italy
⁎ Corresponding author.
7 12 2022
6 2023
7 12 2022
16 100471100471
1 9 2022
6 12 2022
6 12 2022
© 2022 The Authors
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The Istituti Zooprofilattici Sperimentali (IZSs) are public health institutes dealing with the aetiology and pathogenesis of infectious diseases of domestic and wild animals. During Coronavirus Disease 2019 epidemic, the Italian Ministry of Health appointed the IZSs to carry out diagnostic tests for the detection of SARS-CoV-2 in human samples. In particular, the IZS of Abruzzo and Molise (IZS-Teramo) was involved in the diagnosis of SARS-CoV-2 through testing nasopharyngeal swabs by Real Time RT-PCR. Activities and infrastructures were reorganised to the new priorities, in a “One Health” framework, based on interdisciplinary, laboratory promptness, accreditation of the test for the detection of the RNA of SARS-CoV-2 in human samples, and management of confidentiality of sensitive data. The laboratory information system – SILAB – was implemented with a One Health module for managing data of human origin, with tools for the automatic registration of information improving the quality of the data. Moreover, the “National Reference Centre for Whole Genome Sequencing of microbial pathogens - database and bioinformatics analysis” – GENPAT – formally established at the IZS-Teramo, developed bioinformatics workflows and IT dashboard with ad hoc surveillance tools to support the metagenomics-based SARS-CoV-2 surveillance, providing molecular sequencing analysis to quickly intercept the variants circulating in the area. This manuscript describes the One Health system developed by adapting and integrating both SILAB and GENPAT tools for supporting surveillance during COVID-19 epidemic in the Abruzzo region, southern Italy. The developed dashboard permits the health authorities to observe the SARS-CoV-2 spread in the region, and by combining spatio-temporal information with metagenomics provides early evidence for the identification of emerging space-time clusters of variants at the municipality level. The implementation of the One Health module was designed to be easily modelled and adapted for the management of other diseases and future hypothetical events of pandemic nature.
Keywords
Bioinformatics analysis
COVID-19 surveillance
Laboratory information system
GIS
Metagenomics
One health
==== Body
pmc1 Introduction
The novel coronavirus (CoV) called SARS-CoV-2 is responsible for the coronavirus disease 19 (COVID-19) causing the current pandemic [1]. According to the International Committee on Taxonomy of Viruses, SARS-CoV-2 belongs to the species severe acute respiratory syndrome-related virus, in a clade within the order Nidovirales, the family Coronaviridae, genus β-coronavirus, subgenus Sarbecovirus [2], together with other human infecting viruses, like the SARS CoV virus (2003) [3] and Middle East respiratory syndrome coronavirus (MERS-CoV) (2013). The scientific community greatly agrees on the possible animal origin of these viruses [1,2]. In the spillover to humans, bat species are considered the natural animal reservoir of these beta-coronaviruses, with the possible role of intermediate hosts [4,5].
COVID-19 was first reported in December 2019 in humans in connection with the Huanan seafood wholesale market where various species of farmed and wild animals are usually sold (Wuhan, Hubei Province, China) [[5], [6], [7], [8]]. Since 31 December 2019 and as of week 2022–22, 531,470,423, cases of COVID-19 (in accordance with the applied case definitions and testing strategies in the affected countries) were reported, including 6,318,391 deaths [9]. In the absence of effective drugs and a vaccine, in a fully susceptible population, from the starting of the epidemic in less than one year, SARS-CoV-2 resulted in >30 million confirmed cases (2 million in Europe) of infection worldwide and over 900,000 deaths (185,000 in Europe) [10,11]. On 18th February the first Italian case of COVID-19 due to secondary transmission outside China was identified in Codogno, Lombardia region, northern Italy [12]. The unexpected pandemic of COVID-19 caused a never-seen-before disaster in terms of hospitalizations and deaths, overloading the National Health Care System (Servizio Sanitario Nazionale, SSN), particularly in the north of the country, where the epidemic spread with more intensity. From 31 December 2019, >290,000 confirmed cases of SARS-CoV-2 infection and >35,000 deaths were reported in nine months [13]. Besides harmful impacts on workload and organization of hospitals [14] and medical clinics [15], as well as on health [16] and employment [[17], [18], [19]] of populations, the SSN had to perform thousands of daily tests [7,8]. Lockdown and other emergency measures, such as restriction on mobility, social distancing and the closure of all non essential services, were applied.
COVID-19 surveillance in Italy was officially established by the MoH starting from 22 January 2020, setting out the first criteria and methods for reporting cases of SARS-CoV-2 infection [20]. In the Abruzzo region, Ordinance no. 104 of 25 November 2020 [21], obliged the analysis laboratories to use the Swabs Tracing Application of the Abruzzo Region (ATTRA), with the aim of better coordinating surveillance activities and tracing contacts.
In this framework, the Italian Ministry of Health (MoH) appointed the Istituti Zooprofilattici Sperimentali (IZSs), to carry out diagnostic tests for the detection of SARS-CoV-2 in human samples with the aim of supporting the SSN by increasing the total capacity of the analysis laboratories [22].
In central-southern Italy, the IZS of Abruzzo and Molise (IZS-Teramo, headquarter), was involved in the diagnosis of SARS-CoV-2 in the territories under its jurisdiction (Abruzzo and Molise regions) through testing nasopharyngeal swabs by Real Time RT-PCR, under the authority of the Italian MoH.
By late March 2020, Villa Caldari, a small village of the municipality of Ortona (Chieti province, Abruzzo region), registered an incidence rate of COVID-19 cases ten times greater than the overall municipality and was declared a red area [23]. The IZS-Teramo in collaboration with the Local Health Authority (LHA) of Chieti supported in diagnostics and epidemiological investigations, obtaining the suspension of the emergency measures within one month [23].
To face these challenges, the IZS-Teramo reorganised the activities and infrastructures to adapt them to the new priorities, within a broader approach, beyond animal health and food safety.
The IZS-Teramo supported COVID-19 surveillance also as “National Reference Centre for Whole Genome Sequencing of microbial pathogens: database and bioinformatics analysis” (GENPAT). The Decree of the MoH 30 May 2017 formally established the GENPAT at the IZS-Teramo headquarter [24], with the main objective of developing a national platform dedicated to whole genome sequencing (WGS) of microbial pathogens.
In response to the on-going global pandemic, requiring fast graphical assessment of SARS-CoV-2 epidemiological clusters from large numbers of samples, the GENPAT was reorganised in units dedicated to wet-lab, dry-lab, bioinformatics and information technology (IT). During COVID-19 burden, GENPAT contributed to SARS-CoV-2 surveillance by developing bioinformatics workflows, metagenomics databases and informatics systems.
Basic and analytical applications of GIS in epidemiology can help in visualising and analysing geographic distribution of diseases through time, thus revealing space-time trends, patterns, and relationships that would be more difficult or obscure to discover in tabular or other visualization formats [25]. During epidemic and pandemic emergencies, real-time mapping of cases is critical in tracking and controlling the spread of infection. When disease can spread quickly, information has to move even faster. This is where map-based dashboards become crucial [26] to efficiently monitor the spread of infection at a variety of suitable scales [25] and promptly inform decision-makers about the spatiotemporal development of disease outbreaks [27].
This manuscript describes the adaptation and integration of the laboratory information and management system (LIMS) of IZS-Teramo, named SILAB, with the development of a “One Health” module, specially implemented for the management of samples of human origin delivered for the diagnosis of COVID-19 and the implementation of ad hoc tools for collecting human data and information, to support surveillance activities during COVID-19 epidemic. Besides, the manuscript present GENPAT implementations in terms of wet-lab, dry-lab and IT dashboard to observe the geographical spread of the infection for each lineage, in the framework of a unified shotgun metagenomics-based dashboard helping SARS-CoV-2 surveillance in Italy.
2 Materials and methods
2.1 Sample flow and infrastructure reorganisation
During the epidemic, most of the samples came from hospitals of the Abruzzo region, including Teramo, Pescara, and L'Aquila. Usually, control and identification of samples take place in a dedicated laboratory, named Sample Acceptance and Control Unit, at the IZS-Teramo, where the samples delivered are checked for analytical activities, and specialised data entry operators carry out the correct recording in SILAB of the data reported on the samples submission forms and identify the samples with a barcode label in order to make it anonymous. At the Sample Acceptance and Control Unit, COVID-19 samples were delivered by special couriers in triple-wrapped cases, with the sample submission forms outside the cases. For safety reasons, only the operation of data entry was carried out upon delivery of the samples, while samples were checked and identified directly in the Biosafety Level 3 laboratory (BSL3), where each test tube containing the swab was individually identified in biosecurity with a barcode label.
The significant increase of the amount of samples analysed required a significant expansion of the IZS IT infrastructures, i.e. dedicated servers, hardware and hard disks, as well as extension of space for all the acknowledgements of certified email.
2.2 SILAB implementation
SILAB was adapted for collecting data on human samples, by developing a “One Health” module, specially implemented for the management of samples of human origin delivered for the diagnosis of COVID-19. This module was focused on the aspects of privacy and confidentiality, as required by current legislation [28], and on the reduction of the time for the operation of data entry, at the moment of sample acceptance, and of results reporting, given the state of emergency. This “One Health” module was activated in a parametric way (i.e. each item of the system can be modelled to be adapted to new needs), starting from a dedicated area of SILAB to collect data on human health, reserved for appropriately profiled operators, authorised to manage this data.
Moreover, the format of the final test report elaborated by SILAB was appropriately revised, adapting it to show human data and health information (i.e. priority level), as well as processing it in English language, in case of request.
Three ways to automate the sample acceptance process were identified:1. Prior acceptance: a function for quickly entering in SILAB the data of the patients, by collecting most of them automatically in a single form, reading them from parameter tables (Fig. 1). After confirmation, the sample itself was automatically assigned to the analysis to be performed.Fig. 1 Prior acceptance: simplified acceptance form. The figure shows the form used for insert a sample in SILAB. Mandatory data are indicated using the * symbol.
Fig. 1
This way was experienced in the first phase of the epidemic, and during the epidemiological investigations carried out in Villa Caldari.2. Automatic data entry through interoperability between Regional portal and SILAB: after the issuance of the Regional Ordinance [29], according to which the laboratories must use the ATTRA, the connection between this application and SILAB was ensured via Web Service by a unique code assigned to each sample by ATTRA itself. The data of the samples delivered to the IZS-Teramo (i.e. name and surname, date of birth, address of residence and fiscal code of the patient], identified with the ATTRA code, were acquired automatically in SILAB by reading the ATTRA code through a barcode reader. In the same way, the test reports were sent to ATTRA using Web Service technology.
3. Massive acceptance from file: given the large number of samples conveyed to the IZS-Teramo from Bergamo, it was necessary to implement a function of massive acceptance. An Excel file structured with a defined template was populated by the public health authorities of Bergamo with all data required for sample acceptance. This file was uploaded by the operator in a specific section of SILAB, and each row was automatically checked and loaded producing a registration number for each sample and its assignment to the laboratory tests to be carried out.
As for test results, regardless of the ways of sample data collection and recording, described above, in order to speed up results acquisition, and to grant quick response times, the laboratory results were acquired and selected from files in a parametric way. In particular, the results of the Real Time RT-PCR were entered by the laboratory staff in an appropriately prepared sheet of an Excel® file (named “schema RT-PCR Sars-CoV-2 - geni ORF1ab, N protein e S protein”], using Excel formulas the final result in terms of positive/ negative/ doubt to the Real Time RT-PCR was calculated, taking into account each specific exam related to the test for the detection of the Sars-CoV-2 (ORF1ab, S protein, N protein]. At the same time, another sheet of this Excel file produced, through appropriate formulas, as rows as there were the samples, each of them containing the sample acceptance number, type of sample, species, date of starting and ending of the analysis, and the result to the Real Time RT-PCR. This latter sheet was uploaded in SILAB, all rows were automatically loaded and checked to verify their consistency and, if no error occurred, the final test report for each tested sample was then elaborated and confirmed.
2.3 Samples processing and sequencing
All the samples were processed and analysed using a specific Real Time RT-PCR test as described in Lorusso et al. 2020. SARS-CoV-2 samples positive with PCR threshold cycle (Ct] <25 were sequenced for genomic surveillance, essential for monitoring the emergence and global spread of viral variants. RNA purified was processed by means of several methods. The first included a metagenomics approach by the combination of the sequence-independent single-primer amplification (SISPA) with Nextera DNA Flex Library Prep (Illumina Inc., San Diego, CA, USA) [8]. The second protocol provided for a targeted approach by the enrichment of some SISPA libraries using myBaits Expert Virus—SARS-CoV-2 kit (Arbor Biosciences, Ann Arbor, MI, USA). Other samples were processed by the targeted-amplicon approach “ARTIC protocol” [30], according to which the cDNA was synthesised with random hexamers and amplified using two Artic v3 primer pools specific for whole SARS-CoV-2 genome [31]. Library preparation was carried out using DNA Flex Library Prep (Illumina Inc., San Diego, CA USA) Finally, an amplicon-based commercial protocol namely Illumina COVIDSeq Test (Illumina Inc., USA) were used to sequence the most of SARS-CoV-2 samples. This kit combines ARTIC multiplex PCR protocol with Illumina sequencing technology [32]. Deep sequencing of all libraries was performed on the Illumina platforms (300-cycles and standard 150 bp paired-end reads).
2.4 GENPAT implementation
GENPAT implemented a next generation sequencing (NGS)-based workflow dedicated to SARS-CoV-2 from the assembly to the lineage identification and the variants analysis [33]. The first step of the workflow was the reads refinement using Trimmomatic (version 0.36, parameters: illuminaclip:2:30:10, leading:25 trailing:25 sliding windows:20:25, minlen: 36 [34] followed by an assembly step using Snippy (version 4.5.1, default parameters). The reference genome Wuhan-Hu-1/2019 NC_045512) was used for read mapping and variant calling analysis. Consensus sequences were derived using iVar (version 1.3, parameters: minimum length of read to retain after trimming m = 1, minimum quality threshold for sliding window to pass q = 20) [35]. The lineage assignment was implemented using the algorithm pangoLEARN from the workflow PANGOLIN 2.0 [36]. The results coming from the pipeline were imported in the GENPAT database through an automatic procedure.
2.5 Connecting SILAB and GENPAT: a Dashboard for real-time surveillance
An Oracle nightly scheduled procedure filled in a table in the GENPAT database with geo-referenced data provided with latitude and longitude; original information was read from a SILAB database view containing always-current data, including patient information, sampling reason, administrative unit, municipality and domicile coordinates and sampling date. In data transfer to the table, all samples from the same municipality were aggregated in the same row and received a conventional couple of coordinates (SDO_GEOMETRY type field) falling into the municipality centroid for privacy reasons.
The obtained spatial table was used to feed a Representational State Transfer (ReST) geo web service deployed with the ArcGIS Enterprise platform, developed by Esri, which made available the database information (in GeoJSON format) for a dashboard web application (Fig. 2 ).Fig. 2 Dashboard architecture and data flow.
Fig. 2
The data originated from the described procedure were structured to integrate information coming from the sampling activities and the results of the analysis carried out on each sample with location information.
The dashboard front-end (Fig. 3 ) used for data filtering and investigation was developed using JavaScript open source libraries and consumes the ReST geo web service shared by the back-end infrastructure through asynchronous JavaScript and XML (AJAX) calls. The map and functionalities of the geographical information system (GIS) were built using Open Layers, while interactive charts and table views were realised with Chart.js and Tabulator respectively. The general look and feel of the whole application was defined through the bootstrap cascading style sheets (CSS) framework. Finally, Parceljs bundler was used to build a production version of the application with minified and bundled code.Fig. 3 Dashboard user interface. The figure shows the main interface used for data filtering and investigation.
Fig. 3
3 Results
SILAB was adapted to offer different data entry methods functional to different situations: massive acquisition from file, prior acceptance, and automatic acceptance through application cooperation with ATTRA via Web Service. The operations in SILAB, i.e. data entry and selection of the tests to be carried out on the samples delivered, were simplified and reduced to a minimum, thus allowing training quickly new staff necessary to cover, in some periods, even 3 laboratory shifts a day, and to reduce the time for sample registration, reaching over 4000 daily admissions.
Overall, the entire diagnostic process was automatized. Consequently, the possibility of manual error was reduced at minimum, and a higher quality and homogeneity of the data were obtained.
During 2020 and 2021, under COVID-19 emergency a total of 322,950 and 277,554 nasopharyngeal swabs were tested respectively. The largest number of nasopharyngeal swabs was delivered in the last months of both years, reflecting the actual trend of the epidemic. Of the total samples tested in 2020, the 2.2% (seven thousand seventy-seven samples) came from Bergamo.
Over 88.0% of the test results were reported within 24 h from the delivery and registration of the samples in SILAB.
During 2021, the IZS-Teramo also provided molecular sequencing analysis, to quickly intercept the variants circulating in the area. All the sampling data archived in the database were geographically aggregated at municipality level, and made available for the dashboard (Fig. 4 ). The built graphical user interface was composed of different functional views focusing on several specific aspects of the information (Fig. 4).Fig. 4 Content of each tab in the lower right tabbed area of the dashboard user interface. The top left panel shows the distribution of samples per lineages, the top left panel shows the spread of the lineages per months, the bottom left panel shows the number of sample per day, the bottom right panel shows the metadata of the samples.
Fig. 4
The map view in the left half of the user interface showed circular clusters for each municipality under surveillance (Fig. 4). The radius of each cluster is directly proportional to the number of collected samples in the corresponding municipality, through the considered time window, and for all the SARS-CoV-2 lineages (Fig. 4). The map view provides also useful tools to interact with the geographic component of the information (e.g. map navigation, zoom and data selection). From top to bottom of the right half of the dashboard, summarised information is accessible inside cards reporting the number of samples, lineages and municipalities, as well as the last update of the underlying database (Fig. 4). A series of filters located under the summary cards, allow selection of single or multiple lineages, provinces and ranges of sampling dates in order to display specific combined data on the dashboard (Fig. 4). For instance, the pie chart on the right of the filter panel, showed the percentage of the samples collected in each province of the considered geographic area. The bar charts in the lower tabbed area of the graphical user interface showed the number of samples per lineage, the number of lineages per month and the number of samples per day.
The last tab contains a table view of data accessible for downloading as a comma-separated values (CSV) file (Fig. 4).
Using the selection tool available on the upper left corner of the map view (Fig. 3), an area of interest focusing on virus distribution can be drawn (Fig. 3). All the summary badges and charts (Fig. 3) reacts immediately to the drawn area by filtering the displayed information.
4 Discussion
From the start of COVID-19 surveillance, data on laboratory-confirmed SARS-CoV-2 infections are provided on a daily basis to the National Health Institute (ISS) by all Regions and Autonomous Provinces. The ISS processes and analyses the data making them available to ensure monitoring of the epidemic across the country [37].
The outputs of the One Health system described in this paper provided timely information to public health authorities and to the general population on the evolution of the epidemic at regional level, thus contributing to the continuous re-assessment of risk related to transmission and impact of the epidemic, and to the surveillance of COVID-19 in Italy [38].
The dashboard made it possible to observe the geographical spread of the infection for each lineage, integrating spatial and metagenomics information stored in the database and combining them to create an easy to read web application (Fig. 3). Indeed, the spatial component of the information we collected is displayed on the map as circular clusters aggregating the number of samples at the municipality level (Fig. 3). This approach is very effective because of its easy and fast data acquisition during the sampling activity, allowing quick reactions in cases of new rapidly spreading lineages.
The One Health module implemented in SILAB was used by other IZSs (i.e. IZS of Sicila and IZS of Puglia and Basilicata), further harmonising the data collection and reporting of COVID-19 cases to the regions of their territorial competence (Sicilia, Puglia and Basilicata - south Italy) and, thus, to the ISS, contributing to the whole surveillance system in the view of a One Health approach.Moreover, a similar One Health module was funded by FAO for African countries, and installed in the 12 countries which already used SILAB for diagnostic activities in animal health and food safety [39], helping harmonise data collection in these countries.
It is worth emphasising that the flow developed for data collection and for the storage and use is independent of the disease of interest. The One Health module is based on SILAB web application which is already strongly parametric and is easy to add and to configure new fields related with specific requirements of the new disease. As described in the methods, the One Health module offers a dedicated “test report model” which also includes these additional fields. Thus, the module can be applied to other human diseases in restricted time. As for the test report, operative procedures managing patient data uploading and the test results (i.e. prior acceptance, massive acceptance from files) can be easily reused and adapted to new needs.
One of the limits of the dashboard is the aggregation on the municipal centroid of the spatial component of the data, necessary for reasons of privacy protection. This did not allow an extremely precise representation of the distribution of the infection on the target territory in space and over time. It should be emphasized, however, that given the emergency context, our approach remains very useful because of its easy and fast data acquisition during the sampling activity allowing quick reactions in cases of new rapidly spreading lineages. The use of non-aggregated spatial data would in any case represent an improvement capable of adding further capacity for analysis and extrapolation of useful information, such as the identification of space-time clusters. Tracking the changes in the distribution of infections would help epidemiologists and authorities to predict where the next hotspot will appear, and thus attempt to prevent it by ordering lockdowns before the rate of infection increases [40].
5 Conclusions
One Health Surveillance focuses on activities across multiple sectors including human, animal and environmental health to promote health for all [10,41]. This paper highlights how a greater spectrum of experts, comprising front-line healthcare workers (veterinarians or clinicians), epidemiologists, information technology specialists, and laboratory personnel, are needed for health surveillance and preparation for disease control and treatment, in particular in the view of future pandemic. The key to fight potential pandemic threats, such as zoonotic disease, is having well-integrated surveillance systems, capable of adapting to new challenges, like has been our system for COVID-19 surveillance during the main months of the epidemic. It is essential that information be timely exchanged between systems collecting data and information related to different sectors, in order to alert authorities to unusual disease outbreaks so that they can take appropriate action. At the global level, countries around the world intensified their efforts in the establishment of advanced systems for genomic surveillance such as the Belgian genomic surveillance consortium [42], the The Coronavirus Disease 2019 (COVID-19) Genomics UK Consortium (COG-UK) [43], the FAO EMPRES-i [44] and the GLEWS [45], and serve as early Warning One Health systems able to deal with pandemic threats.
SILAB implementation with the One Health module devoted for managing data of human origin, with automatic registration of several information, avoiding errors and improving the quality of the data, contributed substantially to the management of the epidemic providing the public health authorities with timely reports of the laboratory results. The developed dashboard, thanks to the connection to SILAB, has proven to be a very useful tool for daily surveillance of emerging space-time clusters of the SARS-CoV-2 variants at the municipality level.
In the light of the experience reported in this paper, this approach has proven to be the most effective in responding to the pandemic, and perhaps the only feasible one to effectively detect, respond and prevent future zoonoses or other public health risks.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
CRediT authorship contribution statement
Alessio Di Lorenzo: Conceptualization, Writing – original draft, Data curation, Formal analysis, Visualization, Methodology, Software. Iolanda Mangone: Conceptualization, Writing – original draft, Formal analysis, Methodology, Software. Patrizia Colangeli: Conceptualization, Investigation, Data curation, Supervision. Daniela Cioci: Conceptualization, Writing – original draft, Data curation, Formal analysis, Methodology, Software. Valentina Curini: Formal analysis, Writing – review & editing. Giacomo Vincifori: Writing – review & editing. Maria Teresa Mercante: Writing – review & editing. Adriano Di Pasquale: Conceptualization, Investigation, Supervision. Nicolas Radomski: Conceptualization, Writing- Original draft preparation, Formal analysis. Simona Iannetti: Conceptualization, Writing – original draft, Data curation, Methodology.
Declaration of Competing Interest
The authors whose names are listed immediately below certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers' bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.
Data availability
No data was used for the research described in the article.
==== Refs
References
1 Coronaviridae Study Group of the International Committee on Taxonomy of Viruses The species severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2 Nat. Microbiol. 5 4 2020 Apr 536 544 10.1038/s41564-020-0695-z 32123347
2 Walker P.J. Siddell S.G. Lefkowitz E.J. Mushegian A.R. Dempsey D.M. Dutilh B.E. Harrach B. Harrison R.L. Hendrickson R.C. Junglen S. Knowles N.J. Kropinski A.M. Krupovic M. Kuhn J.H. Nibert M. Rubino L. Sabanadzovic S. Simmonds P. Varsani A. Zerbini F.M. Davison A.J. Changes to virus taxonomy and the international code of virus classification and nomenclature ratified by the International Committee on Taxonomy of Viruses (2019) Arch. Virol. 164 9 2019 Sep 2417 2429 10.1007/s00705-019-04306-w 31187277
3 Yang Y. Peng F. Wang R. Yange M. Guan K. Jiang T. Xu G. Sun J. Chang C. The deadly coronaviruses: The 2003 SARS pandemic and the 2020 novel coronavirus epidemic in China J. Autoimmun. 109 2020 May 10.1016/j.jaut.2020.102434 102434. Epub 2020 Mar 3. Erratum in: J Autoimmun. 2020 Jul;111:102487
4 Zhou P. Yang X.L. Wang X.G. Hu B. Zhang L. Zhang W. Si H.R. Zhu Y. Li B. Huang C.L. Chen H.D. Chen J. Luo Y. Guo H. Jiang R.D. Liu M.Q. Chen Y. Shen X.R. Wang X. Zheng X.S. Zhao K. Chen Q.J. Deng F. Liu L.L. Yan B. Zhan F.X. Wang Y.Y. Xiao G.F. Shi Z.L. A pneumonia outbreak associated with a new coronavirus of probable bat origin Nature 2020 10.1038/s41586-020-2012-7
5 Wu D. Tu C. Xin C. Xuan H. Meng Q. Liu Y. Yu Y. Guan Y. Jiang Y. Yin X. Crameri G. Wang M. Li C. Liu S. Liao M. Feng L. Xiang H. Sun J. Chen J. Sun Y. Gu S. Liu N. Fu D. Eaton B.T. Wang L.F. Kong X. Civets are equally susceptible to experimental infection by two different severe acute respiratory syndrome coronavirus isolates J. Virol. 79 4 2005 Feb 2620 2625 10.1128/JVI.79.4.2620-2625.2005 15681462
6 Decaro N. Lorusso A. Novel human coronavirus (SARS-CoV-2): a lesson from animal coronaviruses Vet. Microbiol. 244 2020 May 108693 10.1016/j.vetmic.2020.108693
7 Lorusso A. Calistri P. Petrini A. Savini G. Decaro N. Novel coronavirus (SARS-CoV-2) epidemic: a veterinary perspective Vet. Ital. 56 1 2020 Apr 24 5 10 10.12834/VetIt.2173.11599.1 32048818
8 Lorusso A. Calistri P. Mercante M.T. Monaco F. Portanti O. Marcacci M. Cammà C. Rinaldi A. Mangone I. Di Pasquale A. Iommarini M. Mattucci M. Fazii P. Tarquini P. Mariani R. Grimaldi A. Morelli D. Migliorati G. Savini G. Borrello S. D’Alterio N. A “one-health” approach for diagnosis and molecular characterization of SARS-CoV-2 in Italy One Health 10 2020 Apr 19 100135 10.1016/j.onehlt.2020.100135
9 ECDC COVID-19 Situation Update 502 Worldwide, as of Week 22, Updated 9 June 2022 https://www.ecdc.europa.eu/en/geographical504distribution-2019-ncov-cases Accessed on 13/06/2022
10 World Health Organization WHO Coronavirus Disease (COVID-19) Dashboard https://covid19.who.int/ Accessed on 13/06/2022
11 ECDC COVID-19 Pandemic https://www.ecdc.europa.eu/en/covid-19-pandemic accessed on 13/06/2022
12 Castaldi S. Maffeo M. Rivieccio B.A. Zignani M. Manzi G. Nicolussi F. Salini S. Micheletti A. Gaito S. Biganzoli E. Monitoring emergency calls and social networks for COVID-19 surveillance. To learn for the future: the outbreak experience of the Lombardia region in Italy Acta Biomed 91 9–S 2020 Jul 20 29 33 10.23750/abm.v91i9-S.10038
13 Istituto Superiore di Sanità Dati della Sorveglianza integrata COVID-19 in Italia https://www.epicentro.iss.it/coronavirus/sars-cov-2-dashboard Accessed on 13/06/2022
14 Klein M.G. Cheng C.J. Lii E. Mao K. Mesbahi H. Zhu T. Muckstadt J.A. Hupert N. COVID-19 models for hospital surge capacity planning: a systematic review Disaster Med. Public Health Prep. 16 1 2022 Feb 390 397 10.1017/dmp.2020.332 32907668
15 Ather A. Patel B. Ruparel N.B. Diogenes A. Hargreaves K.M. Coronavirus disease 19 (COVID-19): implications for clinical dental care J. Endod. 46 5 2020 May 584 595 32273156
16 Huang C. Huang L. Wang Y. Li X. Ren L. Gu X. Kang L. Guo L. LiuM Zhou X. Luo J. Huang Z. Tu S. Zhao Y. Chen L. Xu D. Li Y. Li C. Peng L. Li Y. Xie W. Cui D. Shang L. Fan G. Xu J. Wang G. Wang Y. Zhong J. Wang C. Wang J. Zhang D. Cao B. 6-month consequences of COVID-19 in patients discharged from hospital: a cohort study Lancet. 397 10270 2021 Jan 16 220 232 10.1016/S0140-6736(20)32656-8 33428867
17 Fana M. Torrejón Pérez S. Fernández-Macías E. Employment impact of Covid-19 crisis: from short term effects to long terms prospects J. Ind. Bus. Econ. 47 2020 391 410
18 Bartik A.W. Bertrand M. Cullen Z. Glaeser E.L. Luca M. Stanton C. The impact of COVID-19 on small business outcomes and expectations Proc. Natl. Acad. Sci. U. S. A. 117 30 2020 Jul 28 17656–66
19 Nicola M. Alsafi Z. Sohrabi C. Kerwan A. Al-Jabir A. Iosifidis C. The socio-economic implications of the coronavirus pandemic (COVID-19): A review Int. J. Surg. 78 2020 Jun 185 193 10.1016/j.ijsu.2020.04.018 32305533
20 Ministero della Salute Norme, circolari e ordinanze https://www.salute.gov.it/portale/nuovocoronavirus/archivioNormativaNuovoCoronavirus.jsp accessed on 28/07/2022
21 disposizioni sui test molecolati antigienici di igiene e sanità pubblica O.P.G.R. n. 104 del 25.11.2020 https://www.regione.abruzzo.it/content/opgr-n-104-del-25112020#:~:text=Disposizioni%20sui%20test%20molecolari%2C%20antigenici,di%20igiene%20e%20sanit%C3%A0%20pubblica 2022
22 Nota del Ministero della Salute DGSAF Prot 0005086-P-02/03/2020. Attività di controllo ufficiale per gli obiettivi di sanità pubblica veterinaria e di sicurezza alimentare per rischio Coronavirus COVID 19 https://www.seremi.it/sites/default/files/Documento_Principale_0005086-02_03_2020-DGSAF-MDS-P.pdf 2022
23 Cito F. Amato L. Di Giuseppe A. Danzetta M.L. Iannetti S. Petrini A. Lorusso A. Bonfini B. Leone A. Salini R. Mancinelli A. Torzi G. Savini G. Migliorati G. Schael T. D’Alterio N. Calistri P. A COVID-19 hotspot area: activities and epidemiological findings Microorganisms. 8 11 2020 Oct 31 1711 10.3390/microorganisms8111711 33142840
24 Ministero della Salute Decreto 30 maggio 2017. Istituzione di un Centro di referenza nazionale nel settore veterinario. G.U. n 196 del 26 agosto 2017
25 Meade M.S. Earickson R. Medical Geography 2nd ed. 2005 Guilford New York
26 Kamel Boulos M.N. Geraghty E.M. Geographical tracking and mapping of coronavirus disease COVID-19/severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemic and associated events around the world: how 21st century GIS technologies are supporting the global fight against outbreaks and epidemics Int. J. Health Geogr. 19 1 2020 Mar 11 8 10.1186/s12942-020-00202-8 32160889
27 Hohl A. Delmelle E.M. Desjardins M.R. Lan Y. Daily surveillance of COVID-19 using the prospective space-time scan statistic in the United States Spat. Spatiotemporal Epidemiol. 34 2020 Aug 10.1016/j.sste.2020.100354 100354
28 European Parliament and Council Regulation (EU) 2016/679 of the European Parliament and Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation). OJ L 119 4.5.2016 1 88
29 Ordinanza del presidente della Giunta Regionale (O.P.G.R.) n. 104 del 25 novembre 2020. Misure urgenti per la prevenzione e gestione dell'emergenza epidemiologica da COVID-2019 – Ordinanza ai sensi dell'art. 32, comma 3, della legge 23 dicembre 1978, n. 833 in materia di igiene e sanità. Disposizioni sui test molecolari, antigenici e sierologici per tracciatura campioni biologici presso laboratori pubblici e privati e su gruppi di popolazione a rischio più elevato - Ordinanza ai sensi dell'art. 32, comma 3, della legge 23 dicembre 1978, n. 833 in materia di igiene e sanità pubblica https://www.regione.abruzzo.it/content/opgr-n-104-del-25112020 2022 accessed on 13/06/2022
30 Amato L. Jurisic L. Puglia I. Di Lollo V. Curini V. Torzi G. Di Girolamo A. Mangone I. Mancinelli A. Decaro N. Calistri P. Di Giallonardo F. Lorusso A. D’Alterio N. Multiple detection and spread of novel strains of the SARS-CoV-2 B.1.177 (B.1.177.75) lineage that test negative by a commercially available nucleocapsid gene real-timeRT-PCR Emerg. Microbes. Infect. 10 1 2021 Dec 1148 1155 10.1080/22221751.2021.1933609 34019466
31 Itokawa K. Sekizuka T. Hashino M. Tanaka R. Kuroda M. Disentangling primer interactions improves SARS-CoV-2 genome sequencing by multiplex tiling PCR PLoS One 15 9 2020 Sep 18 10.1371/journal.pone.0239403 e0239403
32 Bhoyar R.C. Jain A. Sehgal P. Divakar M.K. Sharma D. Imran M. Jolly B. Ranjan G. Rophina M. Sharma S. Siwach S. Pandhare K. Sahoo S. Sahoo M. Nayak A. Mohanty J.N. Das J. Bhandari S. Mathur S.K. Kumar A. Sahlot R. Rojarani P. Lakshmi J.V. Surekha A. Sekhar P.C. Mahajan S. Masih S. Singh P. Kumar V. Jose B. Mahajan V. Gupta V. Gupta R. Arumugam P. Singh A. Nandy A. Jha R.M. Kumari A. Gandotra S. Rao V. Faruq M. Kumar S. Reshma G.B. Varma G.N. Roy S.S. Sengupta A. Chattopadhyay S. Singhal K. Pradhan S. Jha D. Naushin S. Wadhwa S. Tyagi N. Poojary M. Scaria V. Sivasubbu S. High throughput detection and genetic epidemiology of SARS-CoV-2 using COVIDSeq next-generation sequencing PLoS One 16 2 2021 Feb 17 10.1371/journal.pone.0247115 e0247115
33 Di Pasquale A. Radomski N. Mangone I. SARS-CoV-2 surveillance in Italy through phylogenomic inferences based on hamming distances derived from pan-SNPs, -MNPs and -InDels BMC Genomics 22 1 2021 Oct 30 782 10.1186/s12864-021-08112-0 34717546
34 Bolger A.M. Lohse M. Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data Bioinforma Oxf. Engl. 30 15 2014 Aug 1 2114–20
35 Grubaugh N.D. Gangavarapu K. Quick J. Matteson N.L. De Jesus J.G. Main B.J. Tan A.L. Paul L.M. Brackney D.E. Grewal S. Gurfield N. Van Rompay K.K.A. Isern S. Michael S.F. Coffey L.L. Loman N.J. Andersen K.G. An amplicon-based sequencing framework for accuratelymeasuring intrahost virus diversity using PrimalSeq and iVar Genome Biol. 20 1 2019 Jan 8 8 10.1186/s13059-018-1618-7 30621750
36 GitHub Inc https://github.com/cov-lineages/pangolin accessed on 13/06/2022
37 Riccardo F. Andrianou X. Bella A. Del Manso M. Urdiales A.M. Fabiani M. Bellino S. Boros S. D'Ancona F. Rota M.C. Filia A. Punzo O. Siddu A. Vescio M.F. Di Benedetto C. Tallon M. Ciervo A. Pezzotti P. Stefanelli P. (ISS). COVID-19 Integrated Surveillance System https://www.epicentro.iss.it/en/coronavirus/sars-cov-2-integrated-surveillance-data accessed on 28/07/2022
38 Amato L. Candeloro L. Di Girolamo A. Savini L. Puglia I. Marcacci M. Caporale M. Mangone I. Cammà C. Conte A. Torzi G. Mancinelli A. Di Giallonardo F. Lorusso A. Migliorati G. Schael T. D’Alterio N. Calistri P. Epidemiological and genomic findings of the first documented Italian outbreak of SARS-CoV-2 alpha variant of concern Epidemics 39 2022
39 Colangeli P. Del Negro E. Molini U. Malizia S. Scacchia M. “SILAB for Africa”: an innovative information system supporting the veterinary African laboratories Telemed. J. E Health 25 12 2019 Dec 1216 1224 10.1089/tmj.2018.0208 30767711
40 Al-Kindi K.M. Alkharusi A. Alshukaili D. Spatiotemporal assessment of COVID-19 spread over Oman using GIS techniques Earth Syst. Environ. 4 2020 797 811 10.1007/s41748-020-00194-2 34723076
41 Napoli C. Iannetti S. Rizzo C. Vector borne infections in Italy: results of the integrated surveillance system for West Nile disease in 2013 Biomed. Res. Int. 2015 2015 10.1155/2015/643439. PMID: 25874224; PMCID: PMC4385594 643439
42 Cuypers L. Dellicour S. Hong S.L. Potter B.I. Verhasselt B. Vereecke N. Lambrechts L. Durkin K. Bours V. Klamer S. Bayon-Vicente G. Vael C. Ariën K.K. De Mendonca R. Soetens O. Michel C. Bearzatto B. Naesens R. Gras J. Vankeerberghen A. Matheeussen V. Martens G. Obbels D. Lemmens A. Van den Poel B. Van Even E. De Rauw K. Waumans L. Reynders M. Degosserie J. On Behalf of COVID-19 Genomics Belgium Consortium Maes P. André E. Baele G. Two years of genomic surveillance in Belgium during the SARS-CoV-2 pandemic to attain country-wide coverage and monitor the introduction and spread of emerging variants Viruses 14 10 2022 10.3390/v14102301 2301
43 The COVID-19 Genomics UK (COG-UK) consortium https://www.cogconsortium.uk/ 2022
44 Claes F. Kuznetsov D. Liechti R. Von Dobschuetz S. Dinh Truong B. Gleizes A. …Dauphin G. The EMPRES-i genetic module: a novel tool linking epidemiological outbreak information and genetic characteristics of influenza viruses Database 2014 2014
45 The Joint FAO–OIE–WHO Global Early Warning System for health threats and emerging risks at the human–animal–ecosystems interface http://www.glews.net/ 2022
| 36507072 | PMC9726647 | NO-CC CODE | 2022-12-16 23:18:10 | no | One Health. 2023 Jun 7; 16:100471 | utf-8 | One Health | 2,022 | 10.1016/j.onehlt.2022.100471 | oa_other |
==== Front
Heliyon
Heliyon
Heliyon
2405-8440
The Author(s). Published by Elsevier Ltd.
S2405-8440(22)03436-3
10.1016/j.heliyon.2022.e12148
e12148
Research Article
Gender-based depression factors of older adults living alone during the COVID-19 pandemic: A cross-sectional and secondary data approach
Jung SuJung ∗
Semyung University, Jecheon, Chungcheongbuk-do, South Korea
∗ Corresponding author.
7 12 2022
7 12 2022
e1214811 8 2022
29 10 2022
29 11 2022
© 2022 The Author(s)
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The coronavirus disease 2019 (COVID-19) pandemic negatively affected the mental health of older adults living alone. This study aimed to examine the differences in factors that influence depression among older adults based on gender. This study was a cross-sectional study employing the secondary data of 3581 older adults living alone at the early stage of COVID-19, collected from the 2020 Korea Community Health Survey, and used multiple linear regression analyses to identify factors associated with depression. We found that women had a higher level of depressive status than men. Low subjective health status was most significantly related to depression in both older men and older women. For women, body mass index and more changes in daily life due to COVID-19 were predictors of depression. Conversely, for men, a lower level of monthly income and smoking were significant predictors of depression. Depressive status caused by COVID-19 was likely to be frailer for older women who were living alone. There were differences in the factors related to depression due to COVID-19 by gender.
Depression; COVID-19; Living alone; Gender.
Keywords
Depression
COVID-19
Living alone
Gender
==== Body
pmc1 Introduction
In March 2020, the World Health Organization (WHO) declared coronavirus disease 2019 (COVID-19) a pandemic (WHO, 2021). The pandemic led to increased physical and mental health issues (Morrow-Howell et al., 2020). In particular, mental health was aggravated by social distancing. Depression is one of the most common mental disorders. Based on the epidemiological findings, older adults from Koreahave higher prevalence rates of major depressive disorder than those from Western countries and other Eastern countries (Park and Kim, 2011). Even before the pandemic, researchers had found that depression had higher rates among older adults living alone due to a lack of family relationships and social support than among those who were not living alone (Bruce and McNamara, 1992; Kim et al., 2018). Many studies on depression have been conducted among older adults living alone, especially women with long life expectancy and relatively high depressive tendencies (Lee and Kim, 2016; Won and Lee, 2016; Kim, 2015; Hong et al., 2021).
Living alone in late life has been a long-standing concern as life expectancy increases worldwide (Esteve et al., 2020; Reher and Requena, 2018). Among the related factors with the prevalence of older adults living alone, national economic status is also one of the associated predictos (Reher and Requena, 2018). Moreover, mostly women older adults living alone are considered as an matter in later life due to live long than men (Esteve et al., 2020). In Korea, the proportion of older adults living alone among the total older adult population gradually increased from 16% in 2000 to 19.5% in 2022 (Statistics Korea, 2022a, 2022b). With the number of older adults increasing, 35.1% of households now comprise older adults living alone (Statistics Korea, 2021).
Among older single-person households, the proportion of older men living alone is expected to rise from 28.3% in 2021 to 35.9% in 2047, while the proportion of older women living alone is expected to decrease from 71.7% in 2021 to 64.1% in 2047 (Statistics Korea, 2021). The degree of depression among older men living alone is significantly higher than that among older women living alone (Girgus et al., 2017; Ko et al., 2019). The COVID-19 pandemic has been prolonged, and 65.8% of older Koreans living alone responded that they were depressed (Namkung, 2021). COVID-19 induced restrictions affected the interactions and inclusion of older adults living alone in society. Although the ratio of the older adults living alone in Korea continues to change and has been exposed to a long-term pandemic situation, the latest evidence for gender-based depression in the older adults living alone is still a lacking.
Older Koreans living alone had different risk factors for depression compared to those living with others (Kim et al., 2018). Previous studies report conflicting results with factors related to depression among older adults living alone. Specifically, these include psychosocial factors such as social network satisfaction and subjective health status in older men living alone (Choi and Lee, 2022). Similarly, in a study of older adults living alone, Kim (2015) found that subjective health status was significantly correlated with depression. Conversely, Hong et al. (2021) reported that subjective health status was only related to depression in older women living alone, and there were differences in factors depending on gender. Many studies have only focused on what variables predict depression in a particular gender (Girgus et al., 2017). Although gender is a factor related to depression in older adults living alone (Kim et al., 2018), it is difficult to draw a concrete conclusion about what other factors are related to depression in older adults based on gender.
Older adults living alone experience a disconnection from social networks, suffer from economic poverty, and are at high risk of being isolated from the public security system (Kwon, 2019). Such isolation increases depression, which can lead to further isolation from society by homeboundness (Kim and Jung, 2022). The lack of resources and limited support also leaves health care vulnerable (Haslbeck et al., 2012). The main purpose of this study was to examine the differences in factors associated with depression in older Koreans living alone during the COVID-19 era. The specific objectives of study were to assess (a) the differences in depression depending on gender and other participant characteristics and (b) determinants of depression among older adults living alone by gender. This study’s findings contribute to future improvement of gender-based interventions for the mental health of older adults living alone.
2 Methods
2.1 Design
This cross-sectional study used secondary data from the 2020 Korea Community Health Survey to identify the factors related to depression, based on gender, through the COVID-19 experiences of older adults living alone. The Korea Community Health Survey is organized by the Korea Disease Control and Prevention Agency. It is conducted annually to calculate the health statistics necessary for establishing local health and medical plans (KDCA, 2021). This survey typically targets about 900 local residents of the community (aged 19 and over) per public health center, of which there are 225 total. In the 2020 survey, 142 items across 18 categories including temporary questions related to COVID-19 experiences were investigated. Thorough quarantine regulations were followed due to the COVID-19 pandemic (Appendix 1). The Korea Community Health Survey was conducted from August 16 to October 31, 2020, in which a trained surveyor directly visited the sample household and conducted one-on-one interviews using computer-assisted (laptop) personal interviewing. The interviews took an average of 20–30 min per participant.
2.2 Study participants
In this study, data from 842 men to 2739 women aged 65 years or older and who were living alone were extracted from the 229,269 subjects who participated in the 2020 Korea Community Health Survey.
2.3 Measurements
2.3.1 Background characteristics
Demographic, social and economic conditions (age, education, and monthly income), and physical characteristics (hypertension, diabetes, and subjective health status) were evaluated to determine the background characteristics of the participants. Age, education, and monthly income were measured on a continuous scale. Additionally, the chronic diseases that can be investigated through the 2020 Korea Community Health Survey are hypertension and diabetes, which are categorized as either diagnosed by a doctor and not diagnosed. We used a single item to measure subjective health status: “How would you rate your current health status, in general?” WHO recommended five options: very good, good, fair, bad, and very bad (de Bruin et al., 1996) The higher the score, the better the subjective health status.
2.3.2 Depression
To measure depression, the Korean version of the Patient Health Questionnaire-9 (PHQ-9; Choi et al., 2007) tool was used in the 2020 Korea Community Health Survey. The PHQ-9 evaluates nine items on a scale of 0–3, with a total score of 27 points; higher scores indicate higher levels of depression. The Cronbach’s alpha in this study was 0.828.
2.3.3 Health-related behaviors
“Never smoking,” “no heavy drinking,” and “not obese” were measured to assess health-related behaviors (KDCA, 2021). Current nonsmoking was defined as respondents’ self-reporting of smoking 100 cigarettes or less during their lifetime or having smoked at the time of the survey. Heavy drinking was defined as more than 7 drinks for men and 5 drinks for women per drinking session at least twice a week during the previous year. Body mass index (BMI) was used to evaluate normal body weight based on self-reported height and weight. BMI was classified as less than 25.0 kg/m2 and 25kg/m2 and more (KDCA, 2021; Statistics Korea, 2022a, 2022b).
2.3.4 COVID-19 experiences
Self-quarantine or admission due to COVID-19 and daily life changes were measured. Daily life changes were investigated using the question “What is your current state of life, with 100 points being the state of daily life before COVID-19 and 0 points being the complete suspension of daily life?” This was a question to verify how much the average change in recent daily life compared to before the COVID-19 pandemic’s outbreak in January 2020 (KDCA, 2021). A lower score indicated more negative changes in daily life due to COVID-19.
2.4 Ethical considerations
The 2020 Korea Community Health Survey was obtained without personal identifying information after approval from the Korea Institute for Health and Social Affairs. In addition, this study was approved by the Institutional Review Board (IRB No.: SMU-EX-2022-07-003) of Semyung University.
2.5 Data analysis
IBM’s SPSS Statistics (version 23.0) was used for the statistical analyses in this study. To explain the characteristics of the participants, descriptive statistics included means with standard deviations (SDs) for numeric variables and proportions with percentages for categorical variables. To examine the differences in depression according to participant characteristics by gender, an independent t-test and analysis of variance (ANOVA) were used for numeric data and Pearson’s χ 2 test for categorical data, respectively. For ANOVA, the Bonferroni post-hoc test was used. Multiple linear regression was used to identify the factors related to depression by sex. Before running the multiple linear regression, we conducted a correlation analaysis. In multiple linear regression, covariates (age, educational level, mothly income, diagnosed hypertension or diabetes, subjective health status, smoking, drinking, BMI, daily life change due to COVID-19) were adjusted. Data analyses were conducted in accordance with the guidelines provided by the Korea Disease Control and Prevention Agency (KDCA, 2021). By applying listwise deletion, less than 10% of missing cases were excluded from the analysis (Hair et al., 2006; KDCA, 2021). p values <.05 were considered statistically significant.
3 Results
3.1 Participants' characteristics based on gender
The participants' characteristics according to gender are presented in Table 1 . Among the older adults living alone, women were more depressed than men. Furthermore, women were older, had a lower education level, and were poorer than men. Regarding physical characteristics, women had more higher rates of hypertension and lower subjective health status than men. On the other hand, in health-related behaviors, men included more current smokers and heavy drinkers than women. Moreover, regarding BMI, there were more women than men weighing at 25.0 kg/m2 and more. Women’s daily life changes were higher than men’s due to COVID-19.Table 1 Participants' characteristics according to gender.
Table 1Classification Variables Categories Total (n = 3581) n (%) Men (n = 842) Women (n = 2739) χ2/t(p)*
Background characteristics Age 65–74 1749 (48.8) 506 (60.1) 1243 (45.4) 55.799 (.000) *
≥75 1832 (51.2) 336 (39.9) 1496 (54.6)
Mean (SD) 75.13 (6.690) 73.41 (6.510) 75.65 (6.656) -8.618 (.000) *
Education None (a) 516 (14.4) 36 (4.3) 480 (17.5) 397.760 (.000) *
Elementary ≥ (b) 1418 (39.6) 183 (21.8) 1235 (45.1)
Middle (c) 677 (18.9) 199 (23.7) 478 (17.5)
≥High (d) 968 (27.0) 422 (50.2) 546 (19.9)
Monthly income (KRW10,000) Mean (SD) 94.93 (84.771) 106.41 (95.800) 91.47 (80.854) 4.023 (.000) *
Hypertension Yes 2005 (56.0) 401 (47.6) 1604 (58.6) 31.698 (.000) *
No 1575 (44.0) 441 (52.4) 1134 (41.4)
Diabetes Yes 872 (24.4) 217 (25.8) 655 (23.9) 1.503 (.472)
No 2708 (75.6) 625 (74.2) 2083 (76.0)
Subjective health status Excellent (a) 120 (3.4) 58 (6.9) 62 (2.3) 58.534 (.000) *
Very good (b) 937 (26.2) 252 (29.9) 685 (25.0)
Good (c) 1478 (41.3) 330 (39.2) 1148 (41.9)
Fair (d) 799 (22.3) 153 (18.2) 646 (23.6)
Poor (e) 247 (6.9) 49 (5.8) 198 (7.2)
Health-related behaviors Smoking Yes 299 (8.3) 236 (28.0) 63 (2.3) 557.088 (.000) *
No 3282 (91.7) 606 (72.0) 2676 (97.7)
Drinking Yes 87 (2.4) 65 (7.7) 22 (0.8) 129.969 (.000) *
No 3494 (97.6) 777 (92.3) 2717 (99.2)
BMIa <25 2435 (68.0) 605 (71.9) 1830 (66.8) 18.247 (.000) *
≥25 1042 (29.1) 229 (27.2) 813 (29.7)
Mean (SD) 23.54 (3.144) 23.44 (2.878) 23.57 (3.224) -1.133 (.257)
COVID-19 experiences Self-quarantine or admission due to COVID-19 Yes 8 (0.2) 0 8 (0.3) 2.465 (.116)
No 3573 (99.8) 842 (100.0) 2731 (99.7)
Daily life changeb 0 82 (2.3) 27 (3.2) 55 (2.0) 20.410 (.026) *
10 108 (3.0) 21 (2.5) 87 (3.2)
20 182 (5.1) 42 (5.0) 140 (5.1)
30 322 (9.0) 61 (7.2) 261 (9.5)
40 247 (6.9) 52 (6.2) 195 (7.1)
50 973 (27.2) 219 (26.0) 754 (27.5)
60 324 (9.0) 74 (8.8) 250 (9.1)
70 432 (12.1) 97 (11.5) 335 (12.2)
80 385 (10.8) 101 (12.0) 284 (10.4)
90 219 (6.1) 58 (6.9) 161 (5.9)
100 247 (6.9) 76 (9.0) 171 (6.2)
Mean (SD) 56.18 (24.311) 58.02 (25.438) 55.61 (23.930) 2.416 (.016) *
Depression Mean (SD) 3.00 (3.913) 2.72 (3.781) 3.09 (3.949) -2.428 (.015) *
Note.a1.0% (8 cases) missing data in men, 3.5% (96 cases) missing data in women, b1.7% (60 cases) missing data in total, 1.7% (14 cases) missing data in men, 1.7% (46 cases) missing data in women, *p < .05.
3.2 Differences in depression depend on gender by participants' characteristics
Table 2 displays the differences in depression depending on gender by participants' characteristics. Both women and men were more depressed with lower levels of education, poor subjective health status, and current smoking. However, only women with older age, hypertension and diabetes were more likely to be depressed.Table 2 Differences in depression depending on gender and participants' characteristics.
Table 2Classification Variables Categories Depression Mean (SD)
Men (n = 842) t/F(p)* Women (n = 2739) t/F(p)*
Background characteristics Age 65–74 2.71 (3.812) -.100 (.920) 2.76 (3.686) -3.948 (.000) *
≥75 2.73 (3.739) 3.36 (4.136)
Education None (a) 3.69 (4.458) 4.500 (.004) b > d 3.87 (4.618) 9.683 (.000) * a>b > c > d
Elementary ≥ (b) 3.26 (4.158) 3.09 (3.793)
Middle (c) 2.99 (4.100) 2.86 (3.759)
≥High (d) 2.26 (3.325) 2.60 (3.720)
Hypertension Yes 2.86 (3.777) 1.074 (.283) 3.24 (4.047) 2.330 (.020) *
No 2.58 (3.784) 2.88 (3.799)
Diabetes Yes 2.85 (3.520) .596 (.552) 3.28 (4.220) 2.784 (.005) *
No 2.67 (3.869) 2.97 (3.850)
Subjective health status Excellent (a) 0.98 (1.395) 58.783 (.000) a<b < c < d < e 1.34 (3.219) 143.117 (.000) * a<b < c < d < e
Very good (b) 1.40 (2.016) 1.80 (2.642)
Good (c) 2.32 (2.910) 2.32 (2.857)
Fair (d) 4.72 (4.986) 4.65 (4.631)
Poor (e) 7.98 (5.932) 7.48 (5.834)
Health-related behaviors Smoking Yes 3.36 (4.364) 2.798 (.005) 4.22 (4.467) 2.304 (.021) *
No 2.47 (3.500) 3.06 (3.933)
Drinking Yes 2.46 (3.496) -.565 (.572) 3.41 (2.754) .380 (.704)
No 2.74 (3.805) 3.09 (3.957)
BMIφ <25 2.67 (3.611) .435 (.663) 2.86 (3.782) -1.574 (.116)
≥25 2.80 (4.071) 3.12 (3.967)
COVID-19 experiences Self-quarantine or admission due to COVID-19 Yes 0 - 1.88 (2.416) -.872 (.383)
No 2.72 (3.781) 3.09 (3.952)
Note.φ2.9% (104 cases) missing data, *p < .05.
3.3 Factors related to depression in men and women living alone
The models of older adults living alone were statistically significant and explained 20.7% and 15.4% of the variance in depression in men and women, respectively (Table 3 ). The results of the regression analysis showed that low subjective health status was a significant predictor of depression in men and women living alone; moreover, subjective health status was the strongest impact factor among the predictors in both men and women with depression. Conversely, a lower level of monthly income and current smoking were significant predictors of depression in men living alone. For women living alone, less than 25.0 kg/m2 and more changes in daily life due to COVID-19 were significant predictors of depression.Table 3 Factors related to depression in men and women living alone.
Table 3Predictors β SE t p* β SE t p*
Men (n = 792) Women (n = 2638)
(Constant) 1.851 5.576 .000 1.905 5.003 .000
Background characteristics Age -.023 .020 -.705 .481 .032 .012 1.661 .097
Education (None = 0) Elementary .030 .646 .440 .660 -.028 .205 -1.092 .275
Middle .022 .642 .306 .760 -.020 .251 -.806 .420
High -.014 .623 -.174 .862 -.014 .254 -.539 .590
Monthly income -.096 .001 -2.901 .004* -.031 .001 -1.627 .104
Hypertension (Yes = 0) .005 .248 .142 .887 .001 .141 .049 .961
Diabetes (Yes = 0) .029 .285 .879 .380 -.003 .160 -.165 .869
Subjective health status -.413 .128 -12.399 .000* -.369 .080 -19.696 .000*
Health-related behaviors Smoking (Yes = 0) -.103 .278 -3.142 .002* -.034 .475 -1.868 .062
Drinking (Yes = 0) .005 .468 .168 .866 -.012 .786 -.679 .497
BMI (25≥ = 0) .002 .153 .075 .940 .057 .052 3.097 .002*
COVID-19 experiences Daily life change -.040 .005 -1.243 .214 -.050 .003 -2.769 .006*
R2 = .207, F = 18.214, p = .000 R2 = .154, F = 41.131, p = .000
Note. Missing data were excluded listwise; Men) Durbin-Watson: 423, VIF 1.023–6.589, Women) Durbin-Watson: 303, VIF 1.018–2.102, *p < .05.
4 Discussion
This study examined the differences in factors that influence depression among older Korean adults living alone based on gender during COVID-19. We showed that depressive status was likely to be frailer for older women living alone than older men living alone. With related factors to depression, there were differences by gender due to COVID-19. Nevertheless, low subjective health status was the most significantly related factor to depression in both older men and older women.
Social distancing is a preventive measure against COVID-19. However, it isolates older adults from the external environment, thus leading to emotional dysfunction such as depression (Tang et al., 2021). Owing to prolonged COVID-19, 65.8% of older adults living alone responded that they felt depressed (Namkung, 2021). Although the number of older adult men living alone is gradually increasing, evidence of depression-related factors that differ by gender has been insufficient. This study’s findings provide some significant factors of depression identified among older adults living alone at the early stage of COVID-19.
In this study, older women living alone experienced more severe depressive status than older men living alone during COVID-19. Our findings are similar to those of Kim et al. (2018). In contrast, some studies have reported higher levels of depressive symptoms in men than in women (Ko et al., 2019; Hong et al., 2021). However, some researchers have proposed that a gender difference in living alone might be a predictor of the gender difference in depression (Girgus et al., 2017). Thus, to interpret the conflicting findings, further investigation is required, considering the cause of living alone (e.g., widowed or divorced) and the duration of living alone. In addition, it is necessary to confirm COVID-19’s effect on depressive status through longitudinal data.
Similar to previous studies, our findings confirmed that women who are living alone were older, had lower education, lower monthly income, and poorer subjective health than men (Hong et al., 2021; Lee and Lee, 2021). Conversely, in men, lower monthly income was associated with depression. Although men living alone are considered economically vulnerable because of retirement or unemployment, social network satisfaction and self-esteem have significant mediating effects of socioeconomic status on depression (Choi and Lee, 2022). Therefore, for older men living alone after retiring, interpersonal reserves (social network satisfaction, self-esteem, and perceived health status) are important for preventing their depression (Choi and Lee, 2022). Furthermore, many older adults had to stop working or change their employment patterns due to the COVID-19 pandemic (Namkung, 2021). Taken together, it is necessary to review job policies for older men living alone to ensure adequate income and social network maintenance for reducing and preventing depression.
A low level of subjective health status was a common indicator that increased depressive symptoms in both men and women older adults living alone, which is similar to previous reports (Choi and Lee, 2022; Hong et al., 2021; Kim, 2015). The high subjective health status of older adults during the COVID-19 pandemic was related to positive emotions such as gratitude and contentment (Fingerman et al., 2021). Subjective health status is considered an age-related factor necessary for maintaining independent living in late life (Zivin et al., 2013). In addition, social isolation was associated with poorer subjective health status in older adults (Ward et al., 2019). Older adults who live alone were less likely to see others in person during the pandemic (Fingerman et al., 2021) and therefore, were at risk of isolation. Thus, it is necessary to assess subjective health status at an early stage to intervene and prevent depression in older adults living alone beyond the COVID-19 pandemic.
Among the health-related behaviors predicting depression, smoking was associated with depression in older men living alone. Living arrangements moderated the association between depressive symptoms and current smokers (Kim et al., 2018). Moreover, smoking is considered a determinant of psychosocial health, as smokers are more likely to live alone and increase social isolation (Philip et al., 2022). Owing to COVID-19, there are changes in health-related behaviors among older Koreans (decreased physical activity, decreased sleep duration, and decreased number of meals; see Namkung, 2021). The results of maintaining normal weight in older women living alone were correlated with depression in this study, which may be related to the marked decrease in the number of meals among older women without spouses after COVID-19 (Namkung, 2021).
Older women living alone felt more atrophy in their daily life after COVID-19 than before, and this factor was related to depression. This confirms Ryu et al.’s (2022) results finding that social activity frequency was greater in older women than men living alone during the pandemic in Korea. Older adults living alone may be more reactive to social distancing during COVID-19 than older adults living with others since living alone was associated with more positive emotions when able to engage in in-person contact (Fingerman et al., 2021). In particular, older women living alone had a higher level of depression the less they met close people (Hong et al., 2021; Kim, 2015). In the literature reported during the COVID-19 pandemic, decreased social life and interpersonal interactions were associated with increased depression (Lebrasseur et al., 2021). Therefore, in the face of the long-term pandemic, it is necessary to adopt new ways (such as communication technology) to protect older adults' mental health. The findings of this study with a large sample may help improve interventions that prevent depression for older adults living alone according to gender, thereby maintaining and improving psychological function.
This study had some limitations. First, with cross-sectional data, it was not possible to directly compare the degree of change in depression before and after the pandemic. Second, although the PHQ-9 is a useful tool because of the minimal number of items and easy scoring (Kroenke et al., 2001), there is a need for a depression tool that can assess the depression of older adults living alone. Third, our dataset did not include severity of depression, treatment histories, or medication. Finally, we were unable to include any variables that reflected the details of living alone, such as duration of or reasons for living alone. Further research should consider these variables in longitudinal surveys, since the delayed impact of the COVID-19 on depression needs to be verified. This study was conducted on older adults living alone, a vulnerable group due to the lockdown by COVID-19. Researchers can compare the diversity of depression among groups (older adults living alone versus older adults not living alone, or according to generation) to expand our findings. Social support, government care management systems, and economic support may have positive effects on the mental health of older adults living alone during the COVID-19. Therefore, it is important to conduct further research on and prepare specific community-based non-contact care policies (e.g., communication via ICT technologies) for older adults living alone to prevent depression related to COVID-19 in the future.
5 Conclusion
This study aimed to investigate the factors influencing depression among older Korean adults living alone according to gender during COVID-19. We found that low subjective health status was most significantly associated with depression in both older men and older women living alone. However, BMI and more changes in daily life due to COVID-19 were specifically related to depression in older women living alone. For older men living alone, a lower level of monthly income and smoking status were significant factors of depression. Based on these findings, assessing and intervening with the identified influencing factors is essential for maintaining the psychological function of older adults living alone. In particular, tailored interventions that reflect these factors are necessary to address individual needs by gender, thereby supporting the mental health of older adults living alone in a post-COVID-19 era.
Declarations
Author contribution statement
SuJung Jung: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.
Funding statement
This work was supported by Semyung University [Semyung University Research Grant of 2022].
Data availability statement
The authors do not have permission to share data.
Declaration of interest’s statement
The authors declare no conflict of interest.
Additional information
No additional information is available for this paper.
Appendix A Supplementary data
The following is the supplementary data related to this article:Supplemental file_102922
Supplemental file_102922
Acknowledgements
Not applicable.
==== Refs
References
Bruce M.L. McNamara R. Psychiatric status among the homebound elderly: an epidemiologic perspective J. Am. Geriatr. Soc. 40 6 1992 561 566 1534092
Choi H.S. Lee J.E. Factors affecting depression in middle-aged and elderly men living alone: a cross-sectional path analysis model Am. J. Men’s Health 16 1 2022
Choi H. Choi J. Park K. Joo K. Ga H. Ko H. Kim S. Standardization of the Korean version of Patient Health Questionnaire-9 as a screening instrument for major depressive disorder J. Korean Acad. Fam. Med. 28 2 2007 114 119
de Bruin, A., Picavet, H. S. J., & Nossikov, A. (1996). Health Interview Surveys: towards International Harmonization of Methods and Instruments: World Health Organization, Regional Office for Europe.
Esteve A. Reher D.S. Treviño R. Zueras P. Turu A. Living alone over the life course: cross‐national variations on an emerging issue Popul. Dev. Rev. 46 1 2020 169 189
Fingerman K.L. Ng Y.T. Zhang S. Britt K. Colera G. Birditt K.S. Charles S.T. Living alone during COVID-19: social contact and emotional well-being among older adults J. Gerontol. B Psychol. Sci. Soc. Sci. 76 3 2021 e116 e121 33196815
Girgus J.S. Yang K. Ferri C.V. The gender difference in depression: are elderly women at greater risk for depression than elderly men? Geriatrics 2 4 2017 35 31011045
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. (2006). Multivariate Data Analysis (sixth ed.). Upper Saddle River, NJ: Pearson University Press
Haslbeck J.W. McCorkle R. Schaeffer D. Chronic illness self-management while living alone in later life: a systematic integrative review Res. Aging 34 5 2012 507 547
Hong E. Kim Y. Park J. Kim H. A comparative study of factors associated with geriatric depression between two sex groups living alone J. Korean Acad. Psychiatr. Ment. Health Nurs. 30 4 2021 411 423
KDCA. (2021). The Guide for Using 2020 Korea Community Health Survey Raw Data. Available from: https://chs.kdca.go.kr/chs/igm/igmMain.do. (Accessed 15 May 2022).
Kim E. Comparison of the factors related to depression of the female elderly living alone by region Korean J. Hum. Ecol. 24 6 2015 811 827
Kim H. Kwon S. Hong S. Lee S. Health behaviors influencing depressive symptoms in older Koreans living alone: secondary data analysis of the 2014 Korean longitudinal study of aging BMC Geriatr. 18 1 2018 1 11 29291720
Kim Y.R. Jung H.S. Effects of social interaction and depression on homeboundness in community-dwelling older adults living alone Int. J. Environ. Res. Publ. Health 19 6 2022 3608
Ko H. Park Y.H. Cho B. Lim K.C. Chang S.J. Yi Y.M. Noh E.Y. Ryu S.I. Gender differences in health status, quality of life, and community service needs of older adults living alone Arch. Gerontol. Geriatr. 83 2019 239 245 31102926
Kwon H.C. A qualitative study on the social isolation and poverty of the elderly living alone J. Soc. Sci. 26 3 2019 135 160
Kroenke K. Spitzer R.L. Williams J.B. The PHQ-9: validity of a brief depression severity measure J. Gen. Intern. Med. 16 9 2001 606 613 11556941
Lebrasseur A. Fortin-Bédard N. Lettre J. Raymond E. Bussières E.L. Lapierre N. Impact of the COVID-19 pandemic on older adults: rapid review JMIR Aging 4 2 2021 e26474
Lee S.E. Kim B.H. Predictors of depression in community dwelling older women living alone J. Korean Gerontol. Nurs. 18 1 2016 1 11
Lee S.H. Lee K.H. Living alone older adults’ depression symptoms according to social participation and gender J. Korea Contents Assoc. 21 12 2021 607 620
Morrow-Howell N. Galucia N. Swinford E. Recovering from the COVID-19 pandemic: a focus on older adults J. Aging Soc. Pol. 32 4-5 2020 526 535
Namkung E.H. Social and economic experiences and health changes for older persons during the COVID-19 pandemic Health Welf. Policy Forum 2021 10 2021 72 85
Park J.H. Kim K.W. A review of the epidemiology of depression in Korea J. Korean Med. Assoc. 54 4 2011 362 369
Philip K.E. Bu F. Polkey M.I. Brown J. Steptoe A. Hopkinson N.S. Fancourt D. Relationship of smoking with current and future social isolation and loneliness: 12-year follow-up of older adults in England Lancet Reg. Health-Eur. 14 2022 100302
Reher D. Requena M. Living alone in later life: a global perspective Popul. Dev. Rev. 44 3 2018 427 454
Ryu S.I. Park Y.H. Kim J. Huh I. Chang S.J. Jang S.N. Noh E.Y. Impact of COVID-19 on the social relationships and mental health of older adults living alone: a two-year prospective cohort study PLoS One 17 7 2022 e0270260
Statistics Korea, 2021. Social Indicators in Korea 2021. Available from: https://kostat.go.kr/portal/eng/pressReleases/1/index.board?bmode=read&bSeq=&aSeq=418234&pageNo=1&rowNum=10&navCount=10&currPg=&searchInfo=srch&sTarget=title&sTxt=indicators. (Accessed 5 July 2022).
Statistics Korea, 2022a. e-National Indicators. Available from: https://www.index.go.kr/main/do?cate=1. (Accessed 20 May 2022).
Statistics Korea, 2022b. Percentage of older adults living alone. Available from: https://www.index.go.kr/unify/idx-info.do?idxCd=8039. (Accessed 5 July 2022).
Tang F. Liang J. Zhang H. Kelifa M.M. He Q. Wang P. COVID-19 related depression and anxiety among quarantined respondents Psychol. Health 36 2 2021 164 178 32567952
Ward, M., Layte, R., & Kenny, R. A. (2019). Loneliness, Social Isolation, and Their Discordance Among Older Adults. IDS-TILDA.
Won M.R. Lee K.J. A study on the experience of depression in elderly women Living alone J. Korean Acad. Psychiatr. Ment. Health Nurs. 25 3 2016 195 206
World Health Organization. WHO Director-General’s Opening Remarks at the media Briefing on COVID-19 [Internet]. Geneva: World Health Organization Service; 2021[updated 2020 Mar 11; cited 2021 Oct 20]. Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019 (Accessed 5 July 2022).
Zivin K. Wharton T. Rostant O. The economic, public health, and caregiver burden of late-life depression Psychiatr. Clin. 36 4 2013 631 649
| 36510561 | PMC9726648 | NO-CC CODE | 2022-12-12 23:20:05 | no | Heliyon. 2022 Dec 7; 8(12):e12148 | utf-8 | Heliyon | 2,022 | 10.1016/j.heliyon.2022.e12148 | oa_other |
==== Front
Brain Behav Immun
Brain Behav Immun
Brain, Behavior, and Immunity
0889-1591
1090-2139
Elsevier Inc.
S0889-1591(22)00465-2
10.1016/j.bbi.2022.12.007
Article
Adaptor protein MyD88 confers the susceptibility to stress via amplifying immune danger signals
Yao Xia-Ping a
Ye Jian a
Feng Ting a
Jiang Feng-Chao b
Zhou Ping c
Wang Fang adef⁎
Chen Jian-Guo adef⁎
Wu Peng-Fei adef⁎
a Department of Pharmacology, School of Basic Medicine, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
b School of Pharmacy, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China
c Institute of Organ Transplantation, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China
d The Key Laboratory for Drug Target Researches and Pharmacodynamic Evaluation of Hubei Province, Wuhan, China
e The Research Center for Depression, Tongji Medical College, Huazhong University of Science, 430030 Wuhan, China
f Key Laboratory of Neurological Diseases (HUST), Ministry of Education of China, Wuhan, China
⁎ Corresponding authors at: Department of Pharmacology, School of Basic Medicine, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
7 12 2022
2 2023
7 12 2022
108 204220
27 5 2022
27 11 2022
4 12 2022
© 2022 Elsevier Inc. All rights reserved.
2022
Elsevier Inc.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Graphical abstract
Increasing evidence supports the pathogenic role of neuroinflammation in psychiatric diseases, including major depressive disorder (MDD) and neuropsychiatric symptoms of Coronavirus disease 2019 (COVID-19); however, the precise mechanism and therapeutic strategy are poorly understood. Here, we report that myeloid differentiation factor 88 (MyD88), a pivotal adaptor that bridges toll-like receptors to their downstream signaling by recruiting the signaling complex called ‘myddosome’, was up-regulated in the medial prefrontal cortex (mPFC) after exposure to chronic social defeat stress (CSDS) or severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein. The inducible expression of MyD88 in the mPFC primed neuroinflammation and conferred stress susceptibility via amplifying immune danger signals, such as high-mobility group box 1 and SARS-CoV-2 spike protein. Overexpression of MyD88 aggravated, whereas knockout or pharmacological inhibition of MyD88 ameliorated CSDS-induced depressive-like behavior. Notably, TJ-M2010-5, a novel synthesized targeting inhibitor of MyD88 dimerization, alleviated both CSDS- and SARS-CoV-2 spike protein-induced depressive-like behavior. Taken together, our findings indicate that inhibiting MyD88 signaling represents a promising therapeutic strategy for stress-related mental disorders, such as MDD and COVID-19-related neuropsychiatric symptoms.
Abbreviations
AAV, adeno-associated virus
AMPAR, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor
Casp1, cysteinyl aspartate specific proteinase 1
CORT, corticosterone
CSDS, chronic social defeat stress
CP, control peptide
DAMP, damage-associated molecular pattern
EPM, elevated plus maze
FST, forced swim test
GFAP, glial fibrillary acidic protein
GFP, green fluorescent protein
Iba1, ionized calcium binding adaptor molecule 1
IL-1β, interleukin-1β
IκBα, inhibitor of NF-κB
IRAK, IL-1 receptor-associated kinase
MAPK, mitogen-activated protein kinase
MyD88, myeloid differentiation factor 88
MAL, MyD88 adaptor-like protein
MIP, MyD88 inhibitor peptide
mPFC, medial prefrontal cortex
NF-κB, nuclear factor κB
OFT, open field test
PAMP, pathogen-associated molecular pattern
RAGE, the receptor of advanced glycation end-products
rHMGB1, recombinant high-mobility group box 1 protein
SSDS, subthreshold social defeat stress
SIT, social interaction test
SPT, sucrose preference test
Spike RBD, spike receptor-binding domain
TST, tail suspension test
TLR2, Toll-like receptor 2
TRAF6, tumor necrosis factor receptor-associated factor 6
Keywords
MyD88
Proinflammatory
Stress
Depression
Coronavirus disease 2019
==== Body
pmc1 Introduction
A growing body of evidence indicates that neuroinflammation is critically associated with the pathophysiology of stress-related psychiatric diseases, such as major depressive disorder (MDD) (Leng et al., 2018, Norman et al., 2010). However, much less is known about the precise mechanism for neuroinflammation initiation underlying psychiatric diseases. It has been demonstrated that stress-induced release of immune danger signals, such as high-mobility group box 1 (HMGB1), initiate innate immune signaling by engaging toll-like receptors (TLRs) and the receptor for advanced glycation end-products (RAGE) (Frank et al., 2015, Franklin et al., 2018), followed by production of proinflammatory cytokines such as interleukin-1 beta (IL-1β), which increases stress susceptibility (Cheng et al., 2016). Except for endogenous danger-associated molecular patterns (DAMPs), pathogen-associated molecular patterns (PAMPs) such as spike protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) also binds to TLRs and triggers innate immune response (Frank et al., 2021, Khan et al., 2021, Olajide et al., 2021, Zhao et al., 2021). Recent studies have revealed that COVID-19 increases blood–brain barrier (BBB) permeability (Soung et al., 2022), and spike protein of SARS-CoV-2 crosses the murine BBB by adsorptive transcytosis, entries the central nerve system (CNS) and induces neuroinflammation, which may lead to brain alterations and behavioral abnormalities (Frank et al., 2021, Rhea et al., 2021). Considering coronaviruses are neurotropic (Bauer et al., 2022, Kumari et al., 2021, Song et al., 2021), and neuropsychiatric symptoms (Huang et al., 2021a, Huang et al., 2021b) accompanied by the pandemic of coronavirus disease 2019 (COVID-19), TLRs-dependent neuroinflammation may increase mental risk in COVID-19 patients.
TLRs-dependent innate immune signaling requires the self-assembly of signaling proteins into oligomeric complexes (Deliz-Aguirre et al., 2021, Fitzgerald and Kagan, 2020), including inflammasome and myddosome (Gay et al., 2011, Lin et al., 2010), to coordinate signal transduction cascades in time and space. Nearly all TLRs interact with a central cytoplasmic signaling adaptor, myeloid differentiation primary response protein 88 (MyD88), which reorganize myddosome complex to govern immune signaling cross-talk and initiate downstream signal transduction pathways (Jang et al., 2013). Recent studies have revealed that MyD88 expression is associated with the severity of COVID-19 disease, and MyD88 is required for SARS-CoV-2-induced inflammatory response (Khan et al., 2021, Zheng et al., 2021). Notably, it has been recently reported that MyD88-deficient mice exhibit altered stress response (Hosoi et al., 2021), and the expression of MyD88 mRNA is increased in peripheral blood mononuclear cells of MDD patients (Hajebrahimi et al., 2014). However, the exact role of MyD88 in psychiatric disorders is unclear. Herein, we utilized behavioral, molecular, and genetic approaches to clarify that MyD88 as a novel pharmacological target for the treatment of MDD and neuroinflammation-related behavior abnormalities.
2 Materials and methods
2.1 Animals
Adult male C57BL/6J mice (6–7 weeks old, 18–20 g) and male CD-1 (6–8 months old) were purchased from Hunan SJA Laboratory Animal (Changsha, Hunan, China) and Vital River Laboratory Animal Technology (Beijing, China), respectively. MyD88−/− and their wild-type mice were kindly given from Professor Ping Zhou (Institute of Organ Transplantation, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China). Mice were housed in individual ventilation cage (IVC, ≤ 6 mice per cage) and maintained on a 12 h light–dark cycle (lights on at 8:00 a.m.) with free access to food and water all the time under standard laboratory conditions (ambient temperature: 22 ± 1 °C; humidity: 40 % ± 5 %). The study was conducted according to the National Institute of Health Guide for the Care and Use of Laboratory Animals and approved by the Animal Welfare Committee of Huazhong University of Science and Technology [reference number: 2019-S1827]. All behavioral testing and tissue harvesting procedures were performed on male mice and investigators were blind to the treatments.
2.2 Chronic social defeat stress (CSDS) model
CSDS was conducted as described previously (Li et al., 2018). Briefly, a single invading C57BL/6J mice was physically attacked by different CD-1 residents for 10 min once a day for 10 consecutive days. The mice and the CD-1 were then kept overnight in a two-compartment mouse cage separated by a perforated plexiglass divider to provide the pressure of sensory contact. Control mice were placed in equivalent cages with a different littermate every day.
2.3 Subthreshold social defeat stress (SSDS) model
The SSDS paradigm was conducted as previously described (He et al., 2021). For SSDS protocol, mice were exposed to three social defeat sessions (5 min) with a novel CD-1 aggressor, and then the mice were placed on the other side of a perforated divider to provide psychological stress for 15 min. After SSDS session, the mice were single-housed and social interaction test (SIT) was tested at 24 h later.
2.4 BV2 cell culture
The mouse BV2 cell line, a model of microglia, was obtained from the Jennio Biotech Co. Ltd (Cat No. JNO-598; Guangzhou, China) and cultured in Dulbecco’s Modified Eagle Medium (DMEM, Gibco Laboratories, Cat No. C11995500BT; Grand Island, NY, USA) supplemented with 10 % fetal bovine serum (FBS) (Gibco Laboratories, Cat No. 10099141C; Grand Island, NY, USA) and 1 % penicillin/streptomycin (Gibco Laboratories, Cat No. 2441838; Grand Island, NY, USA) in a humidified incubator of 5 % CO2 at 37 °C. The cells were exposed to corticosterone (CORT, 5 nM, 50 nM and 100 nM) and SARS-CoV-2 spike receptor-binding domain (RBD, 0.1 nM, 1 nM and 10 nM) for 24 h before harvesting.
2.5 Primary microglial culture and virus transfection
The cortical tissue was separated from the brains of neonatal C57BL/6 mice under aseptic conditions, then rinsed and cut into pieces in precooled phosphate buffered saline (PBS). The cortical tissue was digested with 0.25 % trypsin (the final concentration is 0.125 %) for 15 min at 37 °C, and filtered through 70 µm cell strainer after termination of digestion. After centrifugation, the precipitated cells were resuspended in the DMEM/F12 intact medium with 10 % FBS and plated at a final density of 1 × 106 cells/flask in 75 cm2 flasks containing 0.1 g/ml of poly-l-lysine solution, then cultured at 37 °C for 7 to 9 d. Mechanical shaking (37 °C, 6 h) was applied for isolation and purification of glia from mixed glia and neurons culture, then collected supernatant to centrifugation and inoculation. Primary microglia were cultured in the DMEM medium containing 10 % FBS, 1 % penicillin/streptomycin at 37 °C in a 5 % CO2 incubator. The purity of these microglial cells was 99 % as determined by ionized calcium binding adaptor molecule 1 (Iba1) immunoreactivity. Primary microglia were resuspended by DMEM/F12 and seeded into 24-well plates pre-coated with poly-l-lysine at a density of 1.0 × 106 cells/well for overnight cultured in 5 % CO2 at 37 °C. Next, adeno-associated virus 9 (AAV9) vector delivery system containing green fluorescent protein (GFP) was used to overexpress MyD88 (AAV-MyD88, NCBI Reference Sequence: NM-010851.3), and the vector (AAV-GFP) was used as a control. After 8 h of transfection, serum-free transfer solution was replaced by complete medium to culture for 6 d.
2.6 Drugs preparation
Detailed information about drugs is described. CORT (HY-B1618), FPS-ZM1 (HY-19370), JSH-23 (HY-13982) and sparstolonin B (SSnB) (HY-116213) were purchased from MedCheExpress (Monmouth Junction, NJ, USA). MyD88 inhibitor peptide set (MIP) (NBP2-29328) was purchased from Novus Biologicals (Littleton, CO, USA). Recombinant Human HMGB1 protein (rHMGB1) (CF-1690-HMB-050) was purchased from RD systems (Minneapolis, MN, USA). Biotinylated SARS-CoV-2 spike RBD (SPD-C82E9, HEK293-expressed) was purchased from Acrobiosystems (Beijing, China). TJ-M2010-5 (PCT/CN2012/070811) was kindly given from Professor Feng-Chao Jiang (Department of Pharmacy, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China).
CORT was dissolved in phosphate-buffered saline (PBS) to 5 mM concentration. For in vitro study, the final concentrations of CORT were 5 nM, 50 nM, 500 nM, respectively. Control inhibitor peptide (CP) and MIP were dissolved in PBS to 5 mM concentration. For in vivo or in vitro study, the final concentrations of CP and MIP were 50 µM. JSH-23 and SB203580 were dissolved in sodium l-ascorbate and dimethyl sulfoxide (DMSO) to 20 µM and 5 µM as final concentrations, respectively. rHMGB1 protein was dissolved in PBS to 40 µM concentration. For in vivo, the working concentrations of rHMGB1 were 0.2 µM and 2 µM. FPS-ZM1 and SSnB were dissolved in normal saline to 200 nM and 100 µM as final concentrations, respectively. SARS-CoV-2 spike RBD was dissolved in PBS to 7 µM concentration. For in vitro, the final concentrations of SARS-CoV-2 spike RBD were 0.1 nM, 1 nM and 10 nM, respectively. For in vivo, the working concentration of SARS-CoV-2 spike RBD was 10 nM. TJ-M2010-5 was dissolved in distilled deionized water to a stock solution of 5 mg/kg and further diluted to deliver doses of 1.25 mg/kg and 2.5 mg/kg in 0.1 ml. For in vitro, the final concentration of TJ-M2010-5 was 30 µM.
2.7 Behavioral experiments
For all behavioral tests, C57BL/6J mice were transported to behavioral room and adapted for at least 1 h before testing. Mice were assessed in SIT (He et al., 2021), sucrose preference test (SPT) (Luo et al., 2020), tail suspension test (TST) (Zhou et al., 2019), forced swim test (FST) (Deng et al., 2021), open field test (OFT), (Wu et al., 2021) and elevated plus maze test (EPM) (Shen et al., 2019) as previous reports.
2.7.1 SIT
Mice were placed in a novel and open area (42 cm × 42 cm × 42 cm) with a small-animal cage (10 cm × 6 cm) on one side to explore social interaction and corner zones for 2.5 min in the absence or presence of an unfamiliar CD-1 invader mouse, respectively. Movements were video-taped and recorded by the ANY-maze tracking system 5.3 (Stoelting Co., Wood Dale, IL, USA). The social interaction (SI) index was calculated as the total time spent by the mice in the interaction zone with target absent or present or as social interaction ratio (time spent in the interaction with social target present / social target absent) was classified mice as susceptible (ratio < 1) and resilient (ratio ≥ 1).
2.7.2 SPT
Mice was singly housed and adapted to two bottles (50 ml) with one containing 1 % sucrose solution and the other tap water for 48 h, and the position of the bottles was exchanged every 12 h to prevent position preference. After water deprivation for 12 h, mice were exposed to 1 % sucrose solution and tap water for 12 h in the dark phase. Sucrose preference was calculated by dividing the weight of sucrose intake consumed by the total weight of fluid intake (sucrose consumption + water consumption) × 100 %.
2.7.3 TST
Mice were suspended with tape to the suspension bar with the nose 20 cm above the ground. A 6-min test period was recorded by the ANY-maze tracking system 5.3 (Stoelting Co., Wood Dale, IL, USA) and the time of immobility was measured when they exhibited no body movement.
2.7.4 FST
Mice were individually placed into the clear plexiglass cylinder (height 35 cm; diameter 15 cm) filled with water (24 ± 1 °C, 10 cm-depth), and recorded with ANY-maze tracking system 5.3 (Stoelting Co., Wood Dale, IL, USA) for a 6-min session. The immobility time was recorded during the last 4 min. Immobility time was defined as the time spent by the mice floating without struggling in the water away from the wall of a cylinder.
2.7.5 OFT
Mice were placed in a large square open box (50 cm × 50 cm × 40 cm), which was divided into central area of 35 cm × 35 cm and the surrounding marginal zone and allowed to explore the area for 10 min. The movement were tracked with ANY-maze tracking system 5.3 (Stoelting Co., Wood Dale, IL, USA), including the total distance traveled in the open field, the time spent in the central area and the number of entries into the central area.
2.7.6 EPM
The apparatus consisted of two opposing open arms (30 cm × 5 cm × 0.5 cm) and two closed arms (30 cm × 5 cm × 15 cm) extending from a central platform (5 cm × 5 cm) and elevated 40 cm above the ground. Mice were placed in the central platform facing an open arm, and tracked the total distance traveled in the open and closed arms, the number of entries into open and closed arms and the time spent in open and closed arms during the 5-min test period using ANY-maze tracking system 5.3 (Stoelting Co., Wood Dale, IL, USA).
2.8 Western blotting
Cell or tissue samples were homogenized in ice-cold radioimmunoprecipitation (RIPA) lysis buffer with protease and phosphatase inhibitors (150 mM NaCl, 50 mM Tris, 1 % Triton X-100, 0.1 % sodium dodecyl sulfate (SDS), 1 % sodium deoxycholate, protease inhibitor mixture, pH 7.4). The samples were centrifuged at 4 °C with 12,000 g for 20 min and quantified by Bicinchoninic Acid Assay (BCA) protein assay kit (Beyotime Biotechnology, Shanghai, China). All protein samples were heated for 10 min at 95 °C in loading buffer. Equal amounts of protein (30 μg) were separated by 10 % sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes (Merck Millipore, Billerica, MA, USA). The membranes were blocked for 1 h at room temperature in 5 % bovine serum albumin (BSA) (Merck Millipore, Billerica, MA, USA) in Tris-buffered saline containing 0.1 % Tween-20 (TBST). Blots were incubated overnight at 4 °C with the different primary antibodies. anti-MyD88 (ab2064, RRID: AB_302807; 1:1000), anti-MAL (ab133332, 1:1000), anti-TLR2 (ab209216, 1:1000), anti-IL-1β (ab9722, RRID: AB_308765; 1:1000), anti-Iba1 (ab178846, RRID: AB_2636859; 1:1000), anti-GluA1 (ab183797, RRID: AB_2728702; 1:1000) and anti-GluA2 (ab52932, RRID: AB_880226; 1:1000) were purchased from Abcam (Cambridge, MA, USA). anti-IκBα (#4812, RRID: AB_10694416; 1:1000), anti-NF-κB p65 (#59674, RRID: AB_2799570; 1:1000), and anti-phospho-NF-κB p65 (#13346, RRID: AB_2798185; 1:1000) were obtained from Cell Signaling Company (San Francisco, CA, USA). Anti-Caspase-1 (06-503-1, 1:200) was purchased from Merck Millipore (Billerica, MA, USA). Anti-Caspase-1 p10 (sc-56036, RRID: AB_781816; 1:500) and anti-β-actin (sc-47778, RRID: AB_626632; 1:2000) were purchased from Santa Cruz Biotechnology (Waltham, MA, USA). Goat anti-rabbit 800 CW (926-32211, RRID: AB_621843; 1:10000) and Goat anti-mouse 800 CW (926-32210, RRID: AB_621842; 1:10000) were obtained from LI-COR (Nebraska, USA). After washing three times at room temperature with TBST, blots incubated with anti-NFκB p65 (HRP conjugate) were directly reacted with enhanced chemiluminescence substrate (Super Signal West Pico; Pierce Chemical Co., Rockford, IL) and captured by Micro Chemi (DNR Bio-imaging systems, Jerusalem, Israel), and the optical densities of the immunoblots were measured by ImageJ (NIH, Washington, USA) software. Other blots were incubated with secondary antibodies labeled by fluorescent dye in dark at room temperature for 1 h. Odyssey DLx imaging system (Lincoln, Nebraska, USA) and image studio software (NIH, Bethesda, MD, USA) were used to quantify the optical densities of bands. The results were presented as the percentage of control after normalization. All original of key western blots and information were presented in the Figs. S6–S9 and Table S2.
2.9 RNA preparation and real-time PCR
Total RNA was extracted from medial prefrontal cortex (mPFC) using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Then 1 µg total RNA was used for cDNA synthesis by using the RevertAid First Strand cDNA Synthesis kit (Fermentas, Thermo Scientific, Canada). Quantitative real-time PCR was performed in a StepOnePlusTM Real-Time PCR System (Applied Biosystems, Foster City, CA) with SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA). The protocol consisted of denaturation at 95 °C for 10 min followed by 39 cycles of 95 °C for 30 s, 95 °C for 5 s and 60 °C for 30 s. The gene expression was analyzed by ΔΔ Ct method and relative gene expression was normalized to housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primers sequences were listed as follows (5′ to 3′): IL-1β, CCTGCAGCTGGAGAGTGTGGAT (forward), TGTGCTCTGCTTGTGAGGTGCT (reverse); GAPDH, ATGGTGAAGGTCGG TGTG (forward), CATTCTCGGCCTTGACTG (reverse).
2.10 Immunofluorescent staining
Mice were anesthetized with sodium pentobarbital (45 mg/kg, intraperitoneally) and perfused intracardially with 0.9 % saline followed by 4 % paraformaldehyde in PBS through the heart. The brains were post-fixed overnight in paraformaldehyde at 4 °C and placed in 10 %–30 % sucrose gradient at 4 °C for 3 d followed cut into 30 µm thick slices using a freezing microtome (CM1900, Leica Microsystems, Wetzlar, Germany). For immunofluorescence staining of cells, the cells were fixed in 4 % paraformaldehyde for 15 min. The cells or brain sections were washed with PBS and blocked with 5 % BSA containing 0.1 % Triton X-100 at room temperature for 2 h followed by antibodies against CD68 (MAC1957, 1:400; Biorad, California, USA), Iba1 (ab178846, 1:400; ab5076, 1:200; Abcam, Cambridge, MA, USA), TLR2 (ab209216, 1:200; Abcam, Cambridge, MA, USA), glial fibrillary acidic protein (GFAP) (#3670, 1:400; Cell Signaling Company, San Francisco, CA, USA) overnight at 4 °C, then the secondary antibodies labeled with Alexa dye (Invitrogen, Auckland, NEW ZEALAND) and 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) (D9542, 1:8000; Sigma-Aldrich, St. Louis, MO, USA) were incubated. The argon laser lines of 405 nm, 488 nm, 559 nm and 635 nm were used for excitation. The fluorophores were captured with appropriate filters, and images were acquired under 20 × objective with a confocal laser scanning microscope (FV1000, Olympus, Tokyo, Japan). ImageJ (NIH, Washington, USA) was used for immunofluorescence analysis. Images were converted to 8-bit black-and-white images, and background was subtracted. A fixed threshold was set to acquire optimal representation for each staining group. The number of Iba1 positive cells was obtained using the “analyze particles” plugins in ImageJ, and divided by the total area of the acquired field to represent microglia density (microglia/mm2). A mean count per mouse was calculated and used for statistical analysis. For the immunofluorescent staining in vivo, the percentage of CD68-immunostained area (% area) was calculated for each field and each section. For the immunofluorescent staining in vitro, average intensities of the Iba1 or CD68 protein immunostaining were normalized with the number of cells. All results were presented as the percentage of control group following normalization. All analyses were conducted by someone blind to the groups of animals.
2.11 Stereotaxic surgery and cannula infusion
Mice were anesthetized with sodium pentobarbital (40 mg/kg, intraperitoneally) and immobilized in a stereotaxic apparatus (RWD Life Science Co., Shenzhen, Guangdong Province, China). For cannula implantation, mice were bilaterally implanted 22-gauge stainless steel guide cannulas into mPFC (AP = 2.0 mm, ML = ±0.4 mm, DV = −2.2 mm relative to the bregma; AP, ML, and DV denote anteroposterior, mediolateral and dorsoventral distances from bregma, respectively) and recovered for at least 5 d after surgery. Agents were microinjected into mPFC with the microsyringe pump and the injector remained for 2 min after infusion. For stereotaxic viral injection, AAV vector containing the MyD88 or GFP were bilaterally microinjection into mPFC (0.5 µl/side) at a slow rate of 50 nl/min followed by an additional 10 min of rest to diffusion. Behavioral test was commenced 3 weeks after viral injection.
2.12 Statistical analysis
Animals were randomly assigned to different groups. The experimenters were blinded to groups during experiments and quantitative analyses. Sample sizes were determined according to those used in similar studies from our group (He et al., 2021, Luo et al., 2020, Zhou et al., 2019) and justified by the power analyses (http://www.powerandsamplesize.com/). For power analyses, α was 0.05, the desired power (1-β) was 0.8, and the anticipated effect size and standard deviation were based on the result of our preliminary experiments. The information for power analyses were presented in Supplementary materials and methods. Data were expressed as the mean ± SEM and analyzed using GraphPad 8.0 software (GraphPad Software, Inc., USA). We performed statistical comparisons between two groups using Student’s t-tests. One or two-way analysis of variance (ANOVA) and Bonferroni post hoc analyses were used in analyses with multiple experimental groups, where appropriate. p < 0.05 was considered statistically significant. Exact p values, F values and other detailed statistical information for each figure are provided in the Table S1.
3 Results
3.1 Inducible expression of MyD88 by stress primes neuroinflammation in the mPFC
Glucocorticoid-dependent processes underlie the development of chronic stress-induced disorders, in particularly, MDD. We found that the expression of MyD88 (F(3, 53) = 4.452, p = 0.0073) in BV2 cells were elevated by ∼20 % after incubation with CORT (50 nM and 500 nM) for 24 h (Fig. 1 A). Then, primary microglia were pretreated with CP (50 µM) or MIP (50 µM) to interfere with the formation of MyD88 homodimer, followed by exposure to CORT (50 nM) for 24 h. Our results found that CORT (50 nM) changed morphology of primary microglia into ameboid-like shape with an increased immunostaining for CD68 (a marker of phagocytic activity) and Iba1 (a marker for microglia), which was abolished by MIP (CD68: F(1, 21) = 11.02, p = 0.00331, Vehicle + CP vs CORT + CP, p = 0.0002; CORT + CP vs CORT + MIP, p = 0.1486; Iba1: F(1, 21) = 9.492, p = 0.0057; Vehicle + CP vs CORT + CP, p = 0.0214; CORT + CP vs CORT + MIP, p = 0.0010) (Fig. 1B). Meanwhile, transfection of primary microglia with AAV-MyD88 increased the immunofluorescence intensity for CD68 (t = 2.487, p = 0.0418) and Iba-1 (t = 3.280, p = 0.0135) by 128.4 % and 109.8 % respectively. And MyD88-transfected microglia displayed a more ‘‘amoeboid’’ morphology, characterized by hypertrophic bodies, with fewer and shorter processes (Fig. 1C).Fig. 1 MyD88 promotes stress-induced microglial priming and neuroinflammation. (A) Incubation with CORT (50 nM, 500 nM) increased MyD88 expression in BV2 cells (n = 12–15 wells/group). (B) Representative images (left) of immunostaining for CD68 (red) and Iba1 (grey) in primary microglia. Quantitative analyses (right) showed that MIP (50 µM) significantly blocked CORT-induced microglial activation. Scale bar, 50 µm (n = 6–7 wells/group). (C) Representative images (left) of immunostaining for CD68 (red) and Iba1 (grey) in primary microglia. Quantitative analyses (right) showed that exposure of primary microglia to AAV-MyD88 resulted in microglial activation. Scale bar, 50 µm (n = 4–5 wells/group). (D) Representative images (left) of immunostaining for CD68 (green) and Iba1 (red) in the mPFC. Quantitative analyses (right) showed that CSDS increased the number of Iba1+ cell and immunofluorescence intensity of CD68 in microglial. Scale bar, 20 µm (n = 4 mice/group). (E) Representative western blots showing a decreased IκBα level (n = 13–14 mice/group), an increased phospho-NF-κB p65 (n = 13–14 mice/group) and IL-1β (n = 7–8 mice/group) expression in the mPFC of susceptible mice. (F) IL-1β mRNA was increased in the mPFC of susceptible mice (n = 8 mice/group). (G) The expression of MyD88 was upregulated in the mPFC (n = 8–13 mice/group), but not Hippo (n = 5–11 mice/group) and NAc (n = 4–11 mice/group) of susceptible mice. (H) The levels of TLR2 and MAL were elevated in the mPFC of susceptible mice (n = 6–12 mice/group). (I) Schematic timeline of CSDS, CP/MIP treatment and behavioral tests. (J) MIP reversed CSDS-induced upregulation of IL-1β expression in the mPFC (n = 5 mice/group). Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, ***p < 0.001, by one-way (A, G, H) or two-way (B, J) analysis of variance followed by Bonferroni’s post hoc test, Student’s t test (C–F). See also Fig. S1.
Then, we evaluated the effect of stress on microglia in vivo using a CSDS model (Fig. S1A). Immunostaining results showed that CSDS resulted in an approximately 46.1 % increase in the number of Iba1 + microglia (t = 4.255, p = 0.0054) and 46.7 % increase in the immunofluorescence intensity of CD68 (t = 2.969, p = 0.0250) in the mPFC, indicating that CSDS promotes microglial activation (Fig. 1D). In addition, CSDS increased neuroinflammation in the mPFC, as indicated by IκBα level reduced by 17.4 % (t = 2.880, p = 0.0080), with an 22.9 %, 36.0 % and 66.8 % increase in the level of NF-κB p65 phosphorylation (t = 2.099, p = 0.0461), pro-IL-1β and mat-IL-1β protein, respectively (pro-IL-1β, t = 2.484, p = 0.0274; mat-IL-1β, t = 5.065, p = 0.0002) (Fig. 1E). IL-1β mRNA expression was raised by ∼ 70 % (t = 5.324, p = 0.0001) in the mPFC of stressed mice (Fig. 1F).
We next examined the impact of CSDS on MyD88 expression in the mPFC. Exposure to CSDS resulted in an increase in MyD88 expression in the mPFC of susceptible mice, but not resilient mice (F(2, 30) = 22.15, p < 0.0001; Control vs Susceptible, p < 0.0001; Control vs Resilient, p = 0.1031). However, little effect of CSDS on MyD88 expression in the hippocampus (Hippo) (F(2, 22) = 2.021, p = 0.1564; Control vs Susceptible, p > 0.9999) and nucleus accumbens (NAc) (F(2, 22) = 0.8071, p = 0.4589; Control vs Susceptible, p > 0.9999) were observed (Fig. 1G). In addition, CSDS also increased the levels of other elements involved in myddosome signaling, including TLR2 (F(2, 26) = 3.509, p = 0.0447; Control vs Susceptible, p = 0.0450) and MyD88 adaptor-like protein (MAL) (F(2, 26) = 3.804, p = 0.0355; Control vs Susceptible, p = 0.0428) (Fig. 1H) in the mPFC of susceptible mice. Immunostaining results revealed that in the mPFC of both control and stressed mice, TLR2 was mainly expressed in the microglia, but not in the astrocyte (Fig. S1B), suggesting that innate immune responses of microglia may confer CSDS-induced neuroinflammation.
Considering that exposure of microglia to MIP significantly attenuated CORT-induced microglial activation, we next sought to determine the effect of MIP on stress-induced inflammation in vivo (Fig. 1I). We found that consecutive intracerebroventricular (i.c.v.) administration of MIP (50 µM, 1 µl/side) in the mPFC for 3 d reversed CSDS-induced increase in the expression of IL-1β (pro-IL-1β: F(1, 16) = 7.522, p = 0.0144; CSDS + CP vs CSDS + MIP, p = 0.0020; mat-IL-1β: F(1, 16) = 6.368, p = 0.0226; CSDS + CP vs CSDS + MIP, p = 0.0038) (Fig. 1J). Together, these results suggest that inducible expression of MyD88 promotes stress-induced microglial priming in the mPFC.
3.2 Targeting MyD88 in the mPFC blocks CSDS-induced depressive-like behaviors
We further used a viral expression approach to confirm whether inducible expression of MyD88 in the mPFC could generate a pro-depressive effect. Immunofluorescence staining confirmed that AAV-GFP and AAV-MyD88 was localized and distributed in the mPFC, and the expression of MyD88 protein was increased by 89.1 % in the MyD88-overexpressed mice compared with AAV-GFP group (t = 5.759, p < 0.0001) (Fig. S2A and Fig. 2 A). Overexpression of MyD88 in the mPFC was sufficient to increase the number of microglia (t = 4.988, p = 0.0076; AAV-GFP = 265 ± 19.74; AAV-MyD88 = 482.6.0 ± 38.91) and the expression of IL-1β protein (t = 5.500, p = 0.0002; AAV-GFP = 1.000 ± 0.04011; AAV-MyD88 = 1.605 ± 0.1099), which implicates an inflammatory activation of microglia (Fig. 2B-C). Furthermore, overexpression of MyD88 induced a significant increase in behavioral despair than that of AAV-GFP group, indicating by a prolonging duration of immobility in TST from 67.56 ± 9.802 s to 115.3 ± 10.08 s (t = 3.215, p = 0.0026) and FST from 52.66 ± 6.318 s to 92.37 ± 10.74 s (t = 2.760, p = 0.0088) (Fig. 2D). Meanwhile, MyD88-overexpressed mice developed anxiety-like behaviors in the EPM test, including a 35.4 % decrease in time (t = 2.111, p = 0.0416) and 42.9 % decrease in distance (t = 2.254, p = 0.0299) in the open arm (Fig. 2E), but exerted little effect on the time spent in the interaction zone with target (F(1, 40) = 0.2269, p = 0.6364; AAV-GFP vs AAV-MyD88, p > 0.9999; Fig. S2B), sucrose preference (t = 0.9311, p = 0.08757; AAV-GFP = 88.37 ± 2.441 %; AAV-MyD88 = 88.12 ± 1.543 %; Fig. S2C), as well as motor function in the OFT (t = 0.8054, p = 0.4255; AAV-GFP = 54.67 ± 2.577 m; AAV-MyD88 = 57.13 ± 1.793 m; Fig. S2D).Fig. 2 MyD88 in the mPFC mediates susceptibility to stress. (A) Representative photomicrographs (left) of injection site in the mPFC. Scale bars, 200 µm. Quantitative western blot assay (right) the expression of AAV-mediate MyD88 in the mPFC (n = 9 mice/group). (B) Quantitative analyses showed that MyD88 overexpression increased the number of Iba1+ cells in the mPFC. Scale bar, 50 µm (Left), 20 µm (Right) (n = 3 mice/group). (C) Representative western blots showing an increase in IL-1β expression in the mPFC of MyD88-overexpressed mice (n = 6–7 mice/group). (D) MyD88 overexpression increased the immobility time in TST and FST in mice (n = 16–25 mice/group). (E) MyD88 overexpression decreased the time (n = 16–23 mice/group) and distance (n = 16–25 mice/group) in open-arm in the EPM. (F) Schematic illustration of virus injection, SSDS and behavioral tests. (G-I) SSDS increased social avoidance (G), decreased sucrose preference (H) and prolonged the immobility time in TST and FST (I) in MyD88-overexpressed mice (n = 18–22 mice/group). (J) Schematic illustration of virus injection, CSDS and behavioral tests. (K-M) MyD88-overexpressed mice aggravated CSDS-induced decrease in the social interaction time (K) and sucrose preference (L), and prolonged the immobility time in TST (M) (n = 8–10 mice/group). (N and O) MyD88−/− mice prevented CSDS-induced decrease in the social interaction time (N) and social interaction ratio (O) (n = 3–6 mice/group). (P–Q) MIP (50 µM) reversed CSDS-induced social avoidance (P, n = 10–15 mice/group) and sucrose preference deficits (Q, n = 10–14 mice/group). Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, ***p < 0.001, by two-way (G, K–Q) analysis of variance followed by Bonferroni’s post hoc test, Student’s t test (A–E, H, I). See also Fig. S2.
We further explored directly the potential role of MyD88 in stress vulnerability (Fig. 2F). As expected, SSDS lead to depressive-like behaviors in MyD888-overexpressed mice, which was characterized by increased social avoidance (F(1, 76) = 4.724, p = 0.0329; AAV-GFP(Target) vs AAV-MyD88(Target), p = 0.0171; Fig. 2G), reduced sucrose preference from 81.74 ± 3.299 % to 57.37 ± 5.396 % (t = 3.649, p = 0.0008; Fig. 2H), and prolonged immobility time in TST from 97.64 ± 14.22 s to 144.5 ± 14.28 s (t = 2.300, p = 0.0271) and FST from 45.15 ± 6.611 s to 93.14 ± 9.073 s (t = 4.105, p = 0.0002; Fig. 2I), without changing the locomotor activity in the OFT (t = 1.675, p = 0.1042; AAV-GFP = 26.08 ± 2.298 m; AAV-MyD88 = 20.32 ± 2.513 m; Fig. S2E), suggesting that MyD88 may aggravate stress vulnerability. We further verified the effect of MyD88 on CSDS-induced depressive-like behaviors (Fig. 2J). Under the CSDS paradigm, MyD88-overexpressed mice showed less social interaction (F(1, 32) = 0.6651, p = 0.4208; CSDS + AAV-GFP vs CSDS + AAV-MyD88, p = 0.0035; Fig. 2K) and sucrose preference (F(1, 32) = 2.427, p = 0.1291; CSDS + AAV-GFP vs CSDS + AAV-MyD88, p = 0.0444; Fig. 2L) than that of AAV-GFP group, with an increase in about 22.6 % in the immobility time in TST (F(1, 32) = 0.7303, p = 0.3991; CSDS + AAV-GFP vs CSDS + AAV-MyD88, p = 0.1274; Fig. 2M). These results indicate that MyD88 in the mPFC determines the susceptibility to stress.
To further verify the role of MyD88 in stress susceptibility, MyD88 knockout (MyD88−/−) mice were treated by the CSDS paradigm. MyD88−/− mice were found to be resistant to stress-induced depressive-like behaviors under the SIT paradigm, such as increased social interaction time with target (F(1, 14) = 3.022, p = 0.1041; CSDS + WT vs CSDS + MyD88-/-, p = 0.0312; Fig. 2N) and social interaction ratio (F(1, 14) = 8.298, p = 0.0121; CSDS + WT vs CSDS + MyD88-/-, p = 0.0092; Fig. 2O). Furthermore, to examine the effect of MIP on depressive-like behaviors, we implanted cannulations into the mPFC of mice, and microinjected with CP (50 µM, 1 µl/side) or MIP (50 µM, 1 µl/side) for 3 d. We found that the local administration of MIP in the mPFC reversed CSDS-induced deficits in social interactions with target (F(1, 45) = 11.97, p = 0.0012; CSDS + CP vs CSDS + MIP, p < 0.0001; Fig. 2P) and sucrose preference (F(1, 43) = 5.237, p = 0.0271; CSDS + CP vs CSDS + MIP, p = 0.0234; Fig. 2Q). Taken together, these findings suggest that targeting MyD88 in the mPFC may serve as a potential strategy for treatment of depression.
3.3 Overexpression of MyD88 in the mPFC facilitates danger signal-associated behavioral abnormalities
Previous studies have shown that DAMPs, such as HMGB1, may contribute to neuroinflammatory priming (Fonken et al., 2016, Frank et al., 2015), and we found that microinjection with rHMGB1 (2 µM, 1 µl/side) into the mPFC followed by SSDS induced depression-related behaviors, including increased social avoidance (F(2, 23) = 3.728, p = 0.0396; Vehicle vs 2 μM, p = 0.0244; Fig. S3A) and prolonged immobility time in TST (F(2, 24) = 4.512, p = 0.0217; Vehicle vs 2 μM, p = 0.0171; Fig. S3B), but exerted little effect on the locomotor activity in the OFT (F(2, 24) = 1.242, p = 0.3068; Vehicle vs 2 μM, p = 0.2563; Fig. S3C). We further found that microinjection with rHMGB1 (2 µM, 1 µl/side) into the mPFC (Fig. 3 A) only increased the immobility time of FST in MyD88-overexpressed mice (F(1, 71) = 2.127, p = 0.1491; AAV-MyD88 + Vehicle vs AAV-MyD88 + rHMGB1, p = 0.0434; Fig. 3B), but not in GFP-injected mice, with no significant effect on locomotor activity measured by the OFT (F(1, 72) = 1.112, p = 0.2952; AAV-MyD88 + Vehicle vs AAV-MyD88 + rHMGB1, p > 0.9999; Fig. 3C). These results indicate that overexpression of MyD88 in the mPFC aggravates HMGB1-induced behavioral despair.Fig. 3 Overexpression of MyD88 in the mPFC facilitates DAMPs-induced behavioral abnormalities. (A) Schematic timeline of viral injection, cannula implantation, drugs treatment and behavioral tests. (B and C) Intra-mPFC infusion of rHMGB1 (2 µM) increased the immobility time in FST (B) with no changes total distance traveled (C) in the OFT in MyD88-overexpressed mice (n = 18–20 mice/group). (D-F) Exposure with FPS-ZM1 (200 nM) in the mPFC reduced the immobility time in TST (D) and FST (E) with no effect on total distance traveled (F) in the OFT in MyD88-overexpressed mice (n = 7–12 mice/group). (G-I) Intra-mPFC injection of SSnB (100 µM) blocked the prolongation of immobility time in TST (G, n = 7–10 mice/group) and FST (H, n = 16–21 mice/group) without changing total distance traveled (I, n = 8–10 mice/group) in the OFT in MyD88-overexpressed mice. Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, by two-way (B-I) analysis of variance followed by Bonferroni’s post hoc test. See also Fig. S3.
We next investigated whether MyD88 overexpression mediates neuroinflammation via engaging TLR2/4 and RAGE. It was found that intra-mPFC injection of FPS-ZM1 (200 nM, 1 µl/side), a RAGE inhibitor (Deane et al., 2012), for 3 d, prevented MyD88 overexpression-induced behavioral despair, indicated by reduced immobility time in TST (F(1, 34) = 5.205, p = 0.0289; AAV-MyD88 + Vehicle vs AAV-MyD88 + FPS-ZM1, p = 0.0399; Fig. 3D) and FST (F (1, 34) = 7.199, p = 0.0112; AAV-MyD88 + Vehicle vs AAV-MyD88 + FPS-ZM1, p = 0.0101; Fig. 3E), without changing motor function in the OFT (F(1, 34) = 0.1956, p = 0.6611; AAV-MyD88 + Vehicle vs AAV-MyD88 + FPS-ZM1, p > 0.9999; Fig. 3F). Next, intra-mPFC injection of SSnB (100 µM, 1 µl/side), a TLR2/4 inhibitor (Liang et al., 2011), for 3 d, also prevented MyD88 overexpression-induced behavioral despair, such as reduced immobility time in TST (F(1, 32) = 3.234, p = 0.0816; AAV-MyD88 + Vehicle vs AAV-MyD88 + SSnB, p = 0.0284; Fig. 3G) and FST (F(1,66) = 8.355, p = 0.0052; AAV-MyD88 + Vehicle vs AAV-MyD88 + SSnB, p = 0.0102; Fig. 3H), but exerted little effect on locomotor activity in the OFT (F(1, 33) = 0.1020, p = 0.7515; AAV-MyD88 + Vehicle vs AAV-MyD88 + SSnB, p > 0.9999; Fig. 3I). These findings further suggest that inhibition of DAMPs signaling may antagonize MyD88 overexpression-induced behavioral abnormalities.
3.4 MyD88 controls stress susceptibility via p38-MAPK/NF-κB signaling pathway
Our pervious study has reported that caspase-1-IL-1β signaling pathway underlies depression pathogenesis via regulating α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs) (Li et al., 2018). In this study, we found that SSDS resulted in an approximately 32.7 % and 23.9 % increase in the level of activated caspase-1 (Casp 1 p10, t = 3.004, p = 0.0110) and IL-1β (pro-IL-1β: t = 2.735, p = 0.0097), respectively (Fig. 4 A), with a 36.6 % and 25.2 % decrease in the expression of AMPAR subunit glutamate receptor 1 (GluA1, t = 2.799, p = 0.0108) and AMPAR subunit glutamate receptor 2 protein (GluA2, t = 2.496, p = 0.0209) (Fig. 4B) in MyD88-overexpressed mice, respectively, prompting a possible mechanism underlying MyD88-promoted stress susceptibility.Fig. 4 MyD88/p38-MAPK/NF-κB signaling pathway mediates stress susceptibility. (A-B) Representative western blots showing the increased levels of activated Casp1 (A, n = 7 mice/group) and IL-1β (A, n = 17–20 mice/group), as well as decreased levels of GluA1 and GluA2 (B, n = 11–12 mice/group) in the mPFC of MyD88-overexpressed mice exposure to SSDS. (C) Experimental paradigms for viral injection, cannula implantation, drugs treatment and behavioral tests. (D) Representative heatmap of SI behavior from MyD88-overexpressed mice after JSH-23 or SB203580 pretreatment followed by stimulation with SSDS. (E-H) SB203580 (5 µM) or JSH-23 (20 µM) prevented SSDS-induced social avoidance (E), sucrose preference deficits (F), and prolongation of immobility time in TST (G) and FST (H) in MyD88-overexpressed mice (n = 10–12 mice/group). (I) Pretreatment of MyD88-overexpressed mice with JSH-23 or SB203580 followed by stimulation with SSDS produced little effect on total distance traveled in the OFT (n = 12 mice/group). (J and K) Representative western blots showed that not SB203580 (5 µM) (J) but JSH-23 (20 µM) (K) prevented SSDS-induced an increase in the expression of IL-1β in MyD88-overexpressed mice (n = 10 mice/group). Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, ***p < 0.001, by two-way (E-K) analysis of variance followed by Bonferroni’s post hoc test, Student’s t test (A and B).
MyD88 bridges TLRs to their downstream signaling pathways, such as the p38-mitogen-activated protein kinase (MAPK) and NF-κB signaling cascades, which have been involved in the regulation of IL-1β production (Lawrence, 2009, Liu et al., 2019b). We next sought to confirm the molecular pathway that cooperated MyD88 with depressive-like behaviors. MyD88-overexpressed mice were pretreated with SB203580 (i.c.v., 5 µM, 1 µl/side), a specific inhibitor of p38-MAPK (Liu et al., 2020), and JSH-23 (i.c.v., 20 µM, 1 µl/side), a NF-κB inhibitor (Koo et al., 2010), for 24 h, respectively, followed by exposure to SSDS (Fig. 4C). We found that pretreatment with SB203580 and JSH-23 prevented SSDS-induced depressive-like behaviors in MyD88-overexpressed mice, as evidenced by increased social interactions (F(2, 63) = 6.900, p = 0.0020; AAV-MyD88 + Vehicle vs AAV-MyD88 + SB203580, p = 0.0018; AAV-MyD88 + Vehicle vs AAV-MyD88 + JSH-23, p = 0.0054; Fig. 4D-E) and sucrose preference (F(2, 63) = 4.411, p = 0.0161; AAV-MyD88 + Vehicle vs AAV-MyD88 + SB203580, p = 0.0196; AAV-MyD88 + Vehicle vs AAV-MyD88 + JSH-23, p = 0.0473; Fig. 4F), as well as shortened immobility time in TST (F(2, 63) = 4.371, p = 0.0167; AAV-MyD88 + Vehicle vs AAV-MyD88 + SB203580, p = 0.0320; AAV-MyD88 + Vehicle vs AAV-MyD88 + JSH-23, p = 0.0135; Fig. 4G) and FST (F(2, 63) = 5.322, p = 0.0073; AAV-MyD88 + Vehicle vs AAV-MyD88 + SB203580, p = 0.0006; AAV-MyD88 + Vehicle vs AAV-MyD88 + JSH-23, p = 0.0262; Fig. 4H). Neither SB203580 nor JSH-23 affected locomotor activity in the OFT (F(2, 66) = 0.4539, p = 0.6371; AAV-MyD88 + Vehicle vs AAV-MyD88 + SB203580, p > 0.9999; AAV-MyD88 + Vehicle vs AAV-MyD88 + JSH-23, p = 0.6809) in MyD88-overexpressed mice after SSDS (Fig. 4I). Meanwhile, exposure of MyD88-overexpressed mice to SSDS also resulted in an increase in the expression of IL-1β, which was abolished by pretreatment with JSH-23, but not SB203580 (pro-IL-1β: F(1, 36) = 3.384, p = 0.0741; AAV-GFP + Vehicle vs AAV-MyD88 + Vehicle, p = 0.0177; AAV-MyD88 + Vehicle vs AAV-MyD88 + SB203580, p = 0.3367, Fig. 4J; pro-IL-1β: F(1, 36) = 5.550, p = 0.0240; AAV-GFP + Vehicle vs AAV-MyD88 + Vehicle, p = 0.0045; AAV-MyD88 + Vehicle vs AAV-MyD88 + JSH-23, p = 0.0375; Fig. 4K). These results suggest that p38-MAPK/NF-κB signaling pathway may underlie the increase in stress susceptibility in MyD88-overexpressed mice.
3.5 Injection of SARS-COV-2 spike RBD into mPFC induces depressive-like behaviors via MyD88-dependent proinflammatory signaling
Previous studies have reported that the binding of coronavirus spike protein to the TLRs induced the production of proinflammatory cytokine (Aboudounya and Heads, 2021, Lewis et al., 2021). Thus, we investigated whether SARS-CoV-2 spike protein induced microglial activation and depressive-like behaviors. We used SARS-CoV-2 spike RBD to mimic the effect of SARS-CoV-2 on microglia. It was shown that exposure of primary microglia to spike RBD (10 nM, but not 0.1 nM and 1 nM) for 24 h increased the immunofluorescence intensity of CD68 and Iba1 by 65.6 % and 52.0 %, respectively, together with more “amoeboid” morphology (CD68: t = 3.416, p = 0.0027; Iba1: t = 2.103, p = 0.0483; Fig. 5 A; and CD68: F(2, 12) = 0.5785, p = 0.5756; Iba1: F(2, 12) = 0.2968, p = 0.7485; Fig. S4A). Meanwhile, the expression of IL-1β and interleukin-6 (IL-6) in spike RBD (10 nM, but not 0.1 nM and 1 nM, 24 h)-exposed BV2 cells were elevated by about 20 % (pro-IL-1β: t = 2.342, p = 0.0286; IL-6: t = 2.844, p = 0.0174; Fig. 5B; and pro-IL-1β: F(2, 33) = 0.02688, p = 0.9735; mat-IL-1β: F(2, 33) = 0.4513, p = 0.6407; Fig. S4B), indicating that SARS-CoV-2 spike RBD triggers neuroinflammation. The final concentration of 10 nM (1 µl/side) of spike RBD was employed in the following animal experiments.Fig. 5 SARS-COV-2 spike RBD induces microglial activation and increases behavioral abnormalities by activating MyD88 signaling pathway. (A) Representative images (left) of immunostaining for CD68 (red) and Iba1 (grey) in primary microglia. Quantitative analyses (right) showing the activated microglia induced by Spike RBD (10 nM). Scale bar, 50 µm (Left), 20 µm (Right) (n = 11 wells/group). (B) Representative western blots showing increased expression of IL-1β (n = 12 wells/group) and IL-6 (n = 6 wells/group) in BV2 cells exposed to Spike RBD (10 nM). (C) Schematic timeline of cannula implantation, Spike RBD treatment and behavioral tests. (D) Spike RBD (10 nM) increased the immobility time in TST and FST in naïve mice (n = 17–19 mice/group). (E-F) Spike RBD (10 nM) did not alter sucrose preference (E, n = 7–8 mice/group) and total distance traveled (F, n = 17–19 mice/group) in mice. (G) Representative western blots showing the upregulated expression of MyD88 (n = 6–8 mice/group), TLR2 (n = 6–8 mice/group), phospho-NF-κB p65 (n = 9–10 mice/group) and Iba1 (n = 10 mice/group) in mice exposed to Spike RBD. (H) Representative western blots showing the increased IL-1β (n = 12 mice/group) production and reduced expression of GluA1 and GluA2 (n = 6–7 mice/group) in mice exposed to Spike RBD. (I) Schematic timeline of cannula implantation, MIP, Spike RBD treatment and behavioral tests. (J-K) MIP (50 µM) alleviated the increase in immobility time induced by Spike RBD in FST (J), without changing total distance traveled (K) in the OFT (n = 7–9 mice/group). Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, ***p < 0.001, by two-way (J-K) analysis of variance followed by Bonferroni’s post hoc test, Student’s t test (A, B, D-H). See also Fig. S4.
To confirm that spike RBD induced a pro-depressive effect, we infused spike RBD (10 nM) stereotactically in naive mice and tested them 24 h later in behavioral test (Fig. 5C). Intriguingly, spike RBD alone, without priming by SSDS was sufficient to induce behavioral despair of mice, indicating by prolonged immobility time in TST from 80.62 ± 8.285 s to 114.2 ± 13.64 s (t = 2.046, p = 0.0485) and FST from 50.84 ± 3.792 s to 78.60 ± 9.938 s (t = 2.498, p = 0.0175; Fig. 5D), but exerted little effect on sucrose preference (t = 0.2274, p = 0.8236; Vehicle = 87.21 ± 4.004 %; Spike RBD = 88.18 ± 1.855 %; Fig. 5E) and locomotor activity in the OFT (t = 0.4502, p = 0.6554; Vehicle = 40.81 ± 3.304 m; Spike RBD = 42.80 ± 2.965 m; Fig. 5F). Subsequent analysis showed that spike RBD exposure resulted in a 30.2 %, 131.4 %, 200.4 % and 26.8 % increase in the expression of MyD88 (t = 2.373, p = 0.0352), TLR2 (t = 3.032, p = 0.0104) and NF-κB p65 phosphorylation (t = 3.150, p = 0.0058), as well as Iba1 (t = 2.313, p = 0.0327), respectively (Fig. 5G). Furthermore, spike RBD also elevated the expression of pro-IL-1β and mat-IL-1β by 63.5 % and 39.7 % (pro-IL-1β: t = 3.150, p = 0.0046; mat-IL-1β: t = 2.280, p = 0.0326), and decreased the expression of GluA1 (t = 3.061, p = 0.0108) and GluA2 (t = 2.892, p = 0.0146) by about 30 % (Fig. 5H). We next assessed whether MyD88-dependent proinflammatory signaling mediated spike RBD-induced behavioral abnormalities (Fig. 5I). Pretreatment with MIP (i.c.v., 50 µM, 1 µl/side), a classic MyD88 antagonist, for 3 d prevented spike RBD-induced behavioral despair of mice, as indicated by reduced immobility time in FST (F(1, 28) = 11.58, p = 0.0020; Spike RBD + CP vs Spike RBD + MIP, p < 0.0001; Fig. 5J), without altering general locomotion measured the OFT (F(1, 29) = 0.003286, p = 0.9547; Spike RBD + CP vs Spike RBD + MIP, p > 0.9999; Fig. 5K). Taken together, these findings suggest that SARS-CoV-2 spike RBD may work as a precipitating factor that induces behavioral abnormalities.
3.6 Injection of SARS-COV-2 spike RBD into mPFC increases stress susceptibility via MyD88-dependent proinflammatory signaling
Thus, we asked whether injection of SARS-COV-2 spike RBD into the mPFC followed by exposure to SSDS would increase susceptibility to stress (Fig. 6 A). Our results found that spike RBD-treated mice are more susceptible to SSDS, as indicated by an approximately 38.5 % increase in social avoidance (F(1,66) = 11.34, p = 0.0013; Vehicle (Target) vs Spike RBD (Target), p = 0.0155; Fig. 6B), with a 43.2 % decrease in sucrose preference (t = 3.060, p = 0.0099; Fig. 6C), and prolonged immobility time in TST from 81.80 s ± 9.553 s to 126.1 ± 10.56 s (t = 3.061, p = 0.0044) and FST from 45.77 ± 7.804 s to 81.62 ± 9.735 s (t = 2.799, p = 0.0085; Fig. 6D), without altering spontaneous activity in the OFT (t = 1.810, p = 0.0794; Vehicle = 33.22 ± 3.240 m; Spike RBD = 26.84 ± 1.745 m; Fig. 6E). Notably, following SSDS exposure, the expression of MyD88 (t = 5.721, p < 0.0001), TLR2 (t = 2.717, p = 0.0137) and NF-κB p65 phosphorylation (t = 4.045, p = 0.0007) in the mPFC of spike RBD-treated mice were increased by 208.4 %, 86.6 % and 148.1 %, respectively (Fig. 6F). In addition, SSDS resulted in an approximately 221.9 % and 106.2 % increase in the expression of pro-IL-1β and mat-IL-1β (pro-1β: t = 3.229, p = 0.0044; mat-1β: t = 2.765, p = 0.0123), respectively, with a decrease in 54.1 % and 40.5 % in the total levels of GluA1 (t = 3.231, p = 0.0044) and GluA2 (t = 2.627, p = 0.0166) in spike RBD-treated mice (Fig. 6G). Next, we investigated whether MyD88 signaling pathway was required for SSDS-induced behavioral abnormalities in spike RBD-treated mice. Mice were pre-injected with MIP (i.c.v., 50 µM, 1 µl/side) for 3 d followed by treatment with spike RBD and SSDS (Fig. 6H). It was found that MIP prevented SSDS-induced behavioral abnormalities in spike RBD-exposed mice, such as deficits in social interactions (F(1, 29) = 11.44, p = 0.0021; Spike RBD + CP vs Spike RBD + MIP, p = 0.0026; Fig. 6I) and sucrose preference (F(1, 29) = 8.109, p = 0.0080; Spike RBD + CP vs Spike RBD + MIP, p = 0.0295; Fig. 6J), as well as behavioral despair (TST: F(1, 29) = 22.60, p < 0.0001; Spike RBD + CP vs Spike RBD + MIP, p < 0.0001; FST: F(1, 29) = 3.788, p = 0.0614; Spike RBD + CP vs Spike RBD + MIP, p = 0.0430; Fig. 6K-L), with no effect on locomotor activity (F(1, 29) = 0.3645, p = 0.5507; Spike RBD + CP vs Spike RBD + MIP, p > 0.9999; Fig. 6M). Together, these results suggest that spike RBD-increased stress susceptibility is mediated by activation of MyD88 signaling pathway in the mPFC.Fig. 6 MyD88 signaling pathway mediates SARS-CoV-2 spike RBD-increased stress susceptibility. (A) Schematic timeline of cannula implantation, Spike RBD treatment, SSDS and behavioral tests. (B–E) SSDS decreased the social interaction time (B, n = 16–19 mice/group) and sucrose preference (C, n = 6–8 mice/group), as well as increased immobility time in TST and FST (D, n = 16–19 mice/group), without altering total distance (E, n = 16–19 mice/group) in Spike RBD-treated mice. (F) Representative western blots showing the elevated levels of MyD88, TLR2 and phospho-NF-κB p65 in the mPFC of Spike RBD-treated mice after SSDS (n = 10–11 mice/group). (G) Representative western blots showing the increased IL-1β expression, and reduced expression of GluA1 and GluA2 in the mPFC of Spike RBD-treated mice after SSDS (n = 10–11 mice/group). (H) Schematic timeline of cannula implantation, MIP and Spike RBD treatment, SSDS and behavioral tests. (I-M) MIP (50 µM) reversed SSDS-induced social avoidance (I), sucrose preference deficits (J), and prolongation of immobility time in TST (K) and FST (L) in Spike RBD-treated mice, without changing total distance traveled (M) in the OFT (n = 8–9 mice/group). Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, ***p < 0.001, by two-way (B, I–M) analysis of variance followed by Bonferroni’s post hoc test, Student’s t test (C–G).
3.7 Targeting MyD88 dimerization by TJ-M2010-5 alleviates CSDS-/SARS-COV-2 spike RBD-induced depressive-like behaviors and proinflammatory signaling
In our previous studies (Ding et al., 2019, Liu et al., 2019a, Xie et al., 2016, Zou et al., 2020), we designed, synthesized and characterized a novel small MyD88 inhibitor, TJ-M2010-5, which specifically interacts with the MyD88 TIR domain and interferes with MyD88 dimerization. In vitro, we sought to validate the effect of TJ-M2010-5 on MyD88-related inflammatory pathways. It was shown that pretreatment with TJ-M2010-5 (30 µM) inhibited CORT (50 nM)-induced neuroinflammatory activation in BV2 cells, indicating by a reduced level of NF-κB p65 phosphorylation (F(1, 20) = 4.904, p = 0.0386; CORT + Vehicle vs CORT + TJ-M2010-5, p = 0.0301) and IL-1β protein (F(1, 20) = 8.442, p = 0.0087; CORT + Vehicle vs CORT + TJ-M2010-5, p = 0.0048) (Fig. 7 A). Next, we sought to determine the effect of TJ-M2010-5 on behavioral despairin mice and screened the effective dosage of TJ-M2010-5 using TST. We found that exposure of mice to TJ-M2010-5 at 1.25, 2.5 and 5 mg/kg (i.p., once daily) for 1 d (F(3, 32) = 0.6535, p = 0.5867) or 2 d (F(3, 32) = 0.2472, p = 0.8627) exerted little effect on the immobility time in TST (Fig. S5A, B). The results showed that intraperitoneal injection of TJ-M2010-5 (2.5 mg/kg, twice daily) for 2 d reduced the immobility time in TST (F(3, 32) = 4.075, p = 0.0147; Vehicle vs 1.25 mg/kg, p = 0.0295; Vehicle vs 2.5 mg/kg, p = 0.0082; Fig. 7B). Additionally, further analysis indicated that injection of TJ-M2010-5 (2.5 mg/kg, i.p., twice daily) for at least two consecutive days reduced the immobility time in TST (F(2, 35) = 15.50, p < 0.0001; 0 d vs 1 d, p = 0.2385; 0 d vs 2 d, p < 0.0001; Fig. 7C, Fig. S5C) and FST (F(2, 36) = 8.847, p = 0.0007; 0 d vs 1 d, p = 0.8170; 0 d vs 2 d, p = 0.0005; Fig. 7D), indicating a two-day administration was required for the antidepressant effects of TJ-M2010-5.Fig. 7 TJ-M2010-5, a small molecular MyD88 inhibitor, prevents CSDS-/SARS-COV-2 Spike RBD-induced depressive-like behaviors and neuroinflammation in mice. (A) TJ-M2010-5 (30 µM) blocked CORT-induced increase in phospho-NF-κB p65 and IL-1β level (n = 5–7 wells/group). (B) TJ-M2010-5 (1.25, 2.5 and 5 mg/kg, intraperitoneally, twice daily) for 2 d reduced the immobility time in TST in naïve mice (n = 9 mice/group). (C and D) Time-dependent effects of TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) on the immobility time in TST (C, n = 9–19 mice/group) and FST (D, n = 9–20 mice/group). (E) Experimental timeline of CSDS, TJ-M2010-5 administration and behavioral tests. (F-G) TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) for 2 d rescued CSDS-induced social avoidance (F, n = 11–12 mice/group) and sucrose preference deficits (G, n = 10–12 mice/group). (H and I) TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) for 2 d blocked CSDS-induced the increase in the phosphorylation expression of NF-κB p65 (H, n = 11–13 mice/group) and IL-1β production (I, n = 12–14 mice/group) in the mPFC. (J–L) TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) blocked the effect of Spike RBD on immobility time in TST (J, n = 19 mice/group) and FST (K, n = 18–19 mice/group) without affecting total distance traveled (L, n = 19 mice/group) in the OFT. Data are presented as mean ± SEM, with each point representing data from an individual. *p < 0.05, **p < 0.01, ***p < 0.001, by one-way (B-D) or two-way (A, F-L) analysis of variance followed by Bonferroni’s post hoc test. See also Fig. S5.
Then, the mice exposed to CSDS were used to further evaluate the beneficial effects of TJ-M2010-5 (Fig. 7E). We found that intraperitoneal injection of TJ-M2010-5 (2.5 mg/kg, twice daily) for 1 d failed to reverse the social avoidance in susceptible mice (F (1, 17) = 0.5569, p = 0.4657; Fig. S5D), while exposure of susceptible mice to TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) for 2 d rescued CSDS-induced deficits in social interactions (F(1, 42) = 1.736, p = 0.1947; CSDS + Vehicle vs CSDS + TJ-M2010-5, p = 0.0401; Fig. 7F) and sucrose preference (F(1, 41) = 2.319, p = 0.1355; CSDS + Vehicle vs CSDS + TJ-M2010-5, p = 0.0330; Fig. 7G). TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) for 2 d also blocked CSDS-induced NF-κB overactivation (F(1, 42) = 4.235, p = 0.0458; CSDS + Vehicle vs CSDS + TJ-M2010-5, p = 0.0131; Fig. 7H) and IL-1β overproduction (pro-IL-1β: F(1, 47) = 8.487, p = 0.0055; CSDS + Vehicle vs CSDS + TJ-M2010-5, p = 0.0043; mat-IL-1β: F(1, 47) = 9.331, p = 0.0037; CSDS + Vehicle vs CSDS + TJ-M2010-5, p = 0.0419; Fig. 7I).
We next assessed whether TJ-M2010-5 prevented depression-related behaviors in mice exposed to spike RBD (Fig. 7J). Pretreatment of mice with TJ-M2010-5 (2.5 mg/kg, intraperitoneally, twice daily) for 2 d completely prevented the increase in immobility time in TST (F(1, 72) = 11.30, p = 0.0012; Spike RBD + Vehicle vs Spike RBD + TJ-M2010-5, p < 0.0001; Fig. 7K) and FST (F(1, 71) = 10.55, p = 0.0018; Spike RBD + Vehicle vs Spike RBD + TJ-M2010-5, p < 0.0001; Fig. 7L) caused by spike RBD, without altering locomotor activity in the OFT (F(1, 72) = 0.8654, p = 0.3553; Fig. 7M and Fig. S5E). Thus, TJ-M2010-5 may produce antidepressant effects via disrupting MyD88-dependent p38-MAPK/NF-κB/IL-1β signaling in the mPFC, and ameliorate stress- or SARS-CoV-2 spike protein-induced depressive-type behaviors (Fig. 8 ), thereby offering a potential therapy against MDD and neuroinflammation-related behavior abnormalities.Fig. 8 TJ-M2010-5 alleviates CSDS-/SARS-COV-2 Spike protein-related behavior abnormalities via targeting MyD88-dependent neuroinflammation. Stress-triggered DAMPs or SARS-COV-2 Spike protein as PAMPs bind to TLR2/4 or RAGE, leading to assembly of Myddosome, which heightens neuroinflammation by activating p38 MAPK/NF-kB signaling pathways. TJ-M2010-5 specifically interferes with MyD88 dimerization and thus interrupts the assembly of Myddosome, which may alleviate the development of depression by inhibiting neuroinflammation in vivo.
4 Discussion
Evidence from rodents and humans supports that depressive-like behaviors are associated with neuroinflammatory priming (Fenn et al., 2014, Fonken et al., 2018). Emerging evidence also indicates a role of neuroinflammation underlying COVID-19 neuropathology (Costanza et al., 2022, Matschke et al., 2020, Mingoti et al., 2022, Ribeiro et al., 2021). Our results revealed that MyD88, a pivotal adaptor that forms innate immune signaling complex called the ‘myddosome’, controls stress susceptibility via amplifying proinflammatory signaling. Inducible MyD88 expression by stress or SARS-CoV-2 spike protein induced behavioral abnormalities, which may be blocked by a small molecular MyD88 inhibitor, and highlighted MyD88 as a novel potential therapeutic target for depression, especially inflammation-related depression.
Our findings strongly support the emerging view that innate immune signaling contributes to the development of psychiatric disorders. Clinical findings have shown that the levels of the serum proinflammatory cytokines in MDD patients were significantly higher in comparison with controls (Dantzer et al., 2008, Stewart et al., 2009). However, as a noninfectious disease, less is known about how inflammation is activated in MDD. Previous reports have indicated that chronic stress-induced inflammatory responses occur at least partly via DAMPs signaling, and the receptors underlying stress-related DAMP signaling have recently been identified, including RAGE and TLR4 (Gong et al., 2020, Heijmans et al., 2012, Huebener et al., 2015). Except for DAMPs signaling, several lines of evidence have indicated that stress increases the permeability of the BBB by reducing the expression of tight junction proteins, and proinflammatory cytokines from peripheral circulation may entry central nervous system and activate TLRs in the brain of stressed individuals (Dion-Albert et al., 2022, Menard et al., 2017, Welcome and Mastorakis, 2020). Emerging roles of adaptor proteins in the function of TLRs have been clarified (Brown et al., 2011, Chen and Jiang, 2013). MyD88 is a key adaptor of TLRs and RAGE, which initiates key signal transduction pathways to elicit critical inflammatory immune responses by inducing the assembly of signaling complexes termed myddosomes (Deliz-Aguirre et al., 2021). In our study, rHMGB1 was used to mimic the effect of DAMPs, and we found that rHMGB1 only induced behavioral despair in MyD88-overexpressed mice. Meanwhile, both FPS-ZM1, a RAGE inhibitor, and SSnB, a TLR2/4 inhibitor, abolished MyD88 overexpression-induced behavioral despair, highlighting the role of innate immune signaling in stress susceptibility.
Increasing evidence has confirmed the role of inflammation in depression, whereas only a handful of studies have explored mechanisms to mediate neuroinflammation-induced depression-related behaviors. A very late study revealed that neuroinflammation promotes the development of depression-like symptoms during adolescence via excessive microglial engulfment of neuronal spines (Cao et al., 2021). Our previous study demonstrates that activation of IL-1β signaling pathway may generate depressive-like behaviors by downregulating the surface level of AMPARs (Li et al., 2018), a key glutamate receptor involved in depression and antidepressant therapy. Other studies have proved that lipopolysaccharide administration significantly increased IL-1β production and decreased GluA1 expression, which induced depressive-like behaviors (Li et al., 2019, Zhang et al., 2017). Our results suggested that chronic stress and SARS-CoV-2 infection elevated MyD88 expression, which drove IL-1β production via p38-MAPK/NF-κB pathway. It has been demonstrated that the increase in IL-1β led to the downregulation of AMPAR in the mPFC, which may directly generate depressive-like behaviors. In this study, we clarified the upstream signaling of IL-1β-AMPAR and revealed that inducible expression of MyD88 by stress facilitated the pattern recognition to trigger AMPAR deficits, which may represent a basis for inflammation-induced depression.
Our results also indicated that spike-MyD88 signaling may induce behavior abnormalities via amplifying neuroinflammation. The number of reported COVID-19 infection cases had surpassed 601 million, and the number of deaths caused by COVID-19 was more than 6.47 million (WHO operational update). COVID-19 is a multifaceted disease with multi-organ complications, including neurological symptoms. COVID-19-related mood disorders can occur during acute infection and persist or even emerge within 6 months after symptom onset (Huang et al., 2021a). It has been reported that SARS-CoV-2 infects brain vascular endothelial cells and crosses the blood–brain barrier, consequently activates the microglia, leading to a hyperinflammatory response, which may further induce neurological symptoms (Goncalves de Andrade et al., 2021, Zhang et al., 2021). Previous studies indicated that the coronavirus spike protein binds to TLRs and activates proinflammatory signaling (Olajide et al., 2021, Shirato and Kizaki, 2021, Zhao et al., 2021). Our results indicated that the SARS-CoV-2 spike protein generated behavioral abnormalities in vivo, and we proposed that the induction of neuroinflammation by spike protein was mediated through MyD88-dependent proinflammatory signaling, possibly as a result of TLRs activation. However, it should be noted that spike protein-induced behavioral abnormality is a highly limited model system to evaluate how SARS-CoV-2 induces behavior abnormalities. The mechanism involved in the COVID-19-related neurological manifestations are complicated. Based on our results, we hypothesized that MyD88 contributed to neuroinflammation-related depression via two mechanisms: Firstly, inducible MyD88 expression by SARS-CoV-2 spike protein in the mPFC may directly activate neuroinflammation and generate behavioral abnormalities, indicating by prolonged immobility time in TST and FST caused by spike RBD alone, without priming by SSDS (Fig. 5H). Secondly, as a stressful event, COVID-19 has an alarming impact on mental health (Amsalem et al., 2021). Our findings indicate that inducible MyD88 expression increases stress susceptibility. COVID-19 patients might be more vulnerable to mental health risk; thus, MyD88-increased stress susceptibility may be involved. Therefore, MyD88 may serve as a promising therapeutic target for neuroinflammation-related mental disorders.
A greater understanding of the role of neuroinflammation in stress responses and mood disorders may facilitate the development of antidepressants by targeting innate immune signaling. To date, however, no anti-inflammatory drugs have been used in clinical practice to directly improve MDD symptoms. Several peptide MyD88 inhibitors are commercially available, including ST2825 (Wang et al., 2019), and Pepinh-MYD (Sulaiman et al., 2017); however, due to their large molecular weights, pharmacokinetic disadvantages limit their application in neurological diseases. TJ-M2010-5 was one member of a series of 2-aminothiazole-derived MyD88 inhibitors (TJ-M2010 series; WIPO patent application number: PCT/CN2012/070811) in our laboratory. TJ-M2010-5 can bind to the TIR domain of MyD88 and inhibit MyD88 dimerization (Xie et al., 2016), thus disrupting myddosome formation. TJ-M2010-5 has been demonstrated to be effective in the treatment of acute liver injury (Ding et al., 2019), transplant rejection (Li et al., 2017), and colitis-associated colorectal cancer (Xie et al., 2016). Here, we found that TJ-M2010-5 at a moderate dose, but not the highest dose, exerted a better antidepressant effect, indicating that the highest dose of TJ-M2010-5 may produce side effects that counteracted the behavior benefits. Additionally, TJ-M2010-5 ameliorated both stress- and coronavirus spike protein-induced depressive-like behaviors, illustrating the potential value of TJ-M2010-5 in the treatment of psychiatric disorders. Considering the delayed onset of action, limited therapeutic efficacy and modest response rates of currently available antidepressants, and few drugs with demonstrated clinical efficacy for COVID-19-related mental disorders are available, targeting MyD88 dimerization by TJ-M2010-5 may provide a new avenue for stress-related disorders.
It should be noted that our study still has limitations that need to be addressed. First, although an increase in MyD88 expression was observed in CORT-treated BV2 cells and CSDS-treated mice, our current results cannot clarify the mechanism underlying the inducible expression of MyD88. More-detailed mechanisms required further investigation. Second, although overexpression of MyD88 may underlie the behavior effects of spike protein, some inconsistency also existed. For instance, MyD88 overexpression in the mPFC decreased sucrose preference but microinjection of spike protein exerted little effect. Third, although some evidence indicates that a possible role of immune activation in mood disorders due to COVID-19 (Al-Jassas et al., 2022), the level of neuroinflammation in the COVID-19 patients are difficult to evaluate. It should be noted that the spike protein-induced behavioral abnormality was induced by directly injecting the spike protein into the brain, not by natural infection. Although it was very useful to assess how MyD88 signaling contributes to neuroinflammation-induced behaviors, the limitations were obvious for its use in investigations about how SARS-CoV-2 affects behaviors. Additionally, some pattern recognition receptors such as TLR2 could be as well expressed in reactive astrocytes. Although our immunohistological data indicated that in the mPFC, the expression of TLR2 was mainly expressed in microglia, but not astrocyte, both in the normal and stressed mice, the effect of TLR2 and MyD88-related neuroinflammation signaling in astrocyte could not be fully excluded. Thus, the role of reactive astrocytes in inflammation-related depression should be investigated in the next study.
In summary, our results reveal a distinct role for MyD88 in regulating stress and spike protein-induced depressive-like behaviors, indicating that MyD88 may be an attractive therapeutic target for stress- and COVID-19-related mental disorders. We also show that MDD patients or COVID19 patients with mental disorders may benefit from MyD88-targeted therapeutics.
CRediT authorship contribution statement
Xia-Ping Yao: . Jian Ye: . Ting Feng: . Feng-Chao Jiang: . Ping Zhou: . Fang Wang: Funding acquisition. Jian-Guo Chen: Funding acquisition. Peng-Fei Wu: Methodology.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary data
The following are the Supplementary data to this article:Supplementary data 1
Supplementary data 2
Data availability
Data will be made available on request.
Acknowledgments
This work was supported by the Foundation for National Key R&D Program of China (Grant No. 2021ZD0202900 to J-G.C.), 10.13039/501100001809 National Natural Science Foundation of China (Grant No. 82130110 to J-G.C. and Grant No. U21A20363 to F.W.), Innovative Research Groups of National Natural Science Foundation of China (Grant No. 81721005 to J-G.C. and F.W.), 10.13039/501100001809 National Natural Science Foundation of China (No. 82073834 to P.F.W., No. 81971279 to F.W., No. 81973310 to J.G.C.), and PCSIRT (No. IRT13016 to J.G.C).
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbi.2022.12.007.
==== Refs
References
Aboudounya M.M. Heads R.J. COVID-19 and Toll-like receptor 4 (TLR4): SARS-CoV-2 may bind and activate TLR4 to increase ACE2 expression, facilitating entry and causing hyperinflammation Mediators Inflamm. 2021 2021 8874339 10.1155/2021/8874339 33505220
Al-Jassas H.K. Al-Hakeim H.K. Maes M. Intersections between pneumonia, lowered oxygen saturation percentage and immune activation mediate depression, anxiety, and chronic fatigue syndrome-like symptoms due to COVID-19: A nomothetic network approach J. Affect. Disord. 297 2022 233 245 10.1016/j.jad.2021.10.039 34699853
Amsalem D. Dixon L.B. Neria Y. The coronavirus disease 2019 (COVID-19) outbreak and mental health: Current risks and recommended actions JAMA Psychiatry 78 2021 9 10 10.1001/jamapsychiatry.2020.1730 32579160
Bauer L. Laksono B.M. de Vrij F.M.S. Kushner S.A. Harschnitz O. van Riel D. The neuroinvasiveness, neurotropism, and neurovirulence of SARS-CoV-2 Trends Neurosci. 45 2022 358 368 10.1016/j.tins.2022.02.006 35279295
Brown, J., Wang, H., Hajishengallis, G.N., Martin, M., 2011. TLR-signaling networks: An integration of adaptor molecules, kinases, and cross-talk. J. Dent. Res. 90, 417–427. https://doi.org/10.1177/0022034510381264.
Cao P. Chen C. Liu A. Shan Q. Zhu X. Jia C. Peng X. Zhang M. Farzinpour Z. Zhou W. Wang H. Zhou J.N. Song X. Wang L. Tao W. Zheng C. Zhang Y. Ding Y.Q. Jin Y. Xu L. Zhang Z. Early-life inflammation promotes depressive symptoms in adolescence via microglial engulfment of dendritic spines Neuron 109 2021 2573 2589.e9 10.1016/j.neuron.2021.06.012 34233151
Chen H. Jiang Z. The essential adaptors of innate immune signaling Protein Cell 4 2013 27 39 10.1007/s13238-012-2063-0 22996173
Cheng Y. Pardo M. Armini R.S. Martinez A. Mouhsine H. Zagury J.F. Jope R.S. Beurel E. Stress-induced neuroinflammation is mediated by GSK3-dependent TLR4 signaling that promotes susceptibility to depression-like behavior Brain Behav. Immun. 53 2016 207 222 10.1016/j.bbi.2015.12.012 26772151
Costanza A. Amerio A. Aguglia A. Serafini G. Amore M. Hasler R. Ambrosetti J. Bondolfi G. Sampogna G. Berardelli I. Fiorillo A. Pompili M. Nguyen K.D. Hyper/neuroinflammation in COVID-19 and suicide etiopathogenesis: Hypothesis for a nefarious collision? Neurosci. Biobehav. Rev. 136 2022 104606 10.1016/j.neubiorev.2022.104606
Dantzer R. O'Connor J.C. Freund G.G. Johnson R.W. Kelley K.W. From inflammation to sickness and depression: When the immune system subjugates the brain Nat. Rev. Neurosci. 9 2008 46 56 10.1038/nrn2297 18073775
Deane R. Singh I. Sagare A.P. Bell R.D. Ross N.T. LaRue B. Love R. Perry S. Paquette N. Deane R.J. Thiyagarajan M. Zarcone T. Fritz G. Friedman A.E. Miller B.L. Zlokovic B.V. A multimodal RAGE-specific inhibitor reduces amyloid beta-mediated brain disorder in a mouse model of Alzheimer disease J. Clin. Invest. 122 2012 1377 1392 10.1172/JCI58642 22406537
Deliz-Aguirre R. Cao F. Gerpott F.H.U. Auevechanichkul N. Chupanova M. Mun Y. Ziska E. Taylor M.J. MyD88 oligomer size functions as a physical threshold to trigger IL1R Myddosome signaling J. Cell Biol. 220 2021 e202012071 33956941
Deng S.L. Hu Z.L. Mao L. Gao B. Yang Q. Wang F. Chen J.G. The effects of Kctd12, an auxiliary subunit of GABAB receptor in dentate gyrus on behavioral response to chronic social defeat stress in mice Pharmacol. Res. 163 2021 105355 10.1016/j.phrs.2020.105355
Ding Z. Du D. Yang Y. Yang M. Miao Y. Zou Z. Zhang X. Li Z. Zhang X. Zhang L. Wang X. Zhao Y. Jiang J. Jiang F. Zhou P. Short-term use of MyD88 inhibitor TJ-M2010-5 prevents d-galactosamine/lipopolysaccharide-induced acute liver injury in mice Int. Immunopharmacol. 67 2019 356 365 10.1016/j.intimp.2018.11.051 30583234
Dion-Albert L. Cadoret A. Doney E. Kaufmann F.N. Dudek K.A. Daigle B. Parise L.F. Cathomas F. Samba N. Hudson N. Lebel M. Signature C. Campbell M. Turecki G. Mechawar N. Menard C. Vascular and blood-brain barrier-related changes underlie stress responses and resilience in female mice and depression in human tissue Nat. Commun. 13 2022 164 10.1038/s41467-021-27604-x 35013188
Fenn A.M. Gensel J.C. Huang Y. Popovich P.G. Lifshitz J. Godbout J.P. Immune activation promotes depression 1 month after diffuse brain injury: a role for primed microglia Biol. Psychiatry 76 2014 575 584 10.1016/j.biopsych.2013.10.014 24289885
Fitzgerald K.A. Kagan J.C. Toll-like receptors and the control of immunity Cell 180 2020 1044 1066 10.1016/j.cell.2020.02.041 32164908
Fonken L.K. Frank M.G. Kitt M.M. D'Angelo H.M. Norden D.M. Weber M.D. Barrientos R.M. Godbout J.P. Watkins L.R. Maier S.F. The alarmin HMGB1 mediates age-induced neuroinflammatory priming J. Neurosci. 36 2016 7946 7956 10.1523/JNEUROSCI.1161-16.2016 27466339
Fonken L.K. Frank M.G. Gaudet A.D. D'Angelo H.M. Daut R.A. Hampson E.C. Ayala M.T. Watkins L.R. Maier S.F. Neuroinflammatory priming to stress is differentially regulated in male and female rats Brain Behav. Immun. 70 2018 257 267 10.1016/j.bbi.2018.03.005 29524458
Frank M.G. Weber M.D. Watkins L.R. Maier S.F. Stress sounds the alarmin: The role of the danger-associated molecular pattern HMGB1 in stress-induced neuroinflammatory priming Brain Behav. Immun. 48 2015 1 7 10.1016/j.bbi.2015.03.010 25816800
Frank M.G. Nguyen K.H. Ball J.B. Hopkins S. Kelley T. Baratta M.V. Fleshner M. Maier S.F. SARS-CoV-2 spike S1 subunit induces neuroinflammatory, microglial and behavioral sickness responses: Evidence of PAMP-like properties Brain Behav. Immun. 100 2021 267 277 10.1016/j.bbi.2021.12.007 34915155
Franklin T.C. Wohleb E.S. Zhang Y. Fogaca M. Hare B. Duman R.S. Persistent increase in microglial RAGE contributes to chronic stress-induced priming of depressive-like behavior Biol. Psychiatry 83 2018 50 60 10.1016/j.biopsych.2017.06.034 28882317
Gay N.J. Gangloff M. O'Neill L.A. What the Myddosome structure tells us about the initiation of innate immunity Trends Immunol. 32 2011 104 109 10.1016/j.it.2010.12.005 21269878
Goncalves de Andrade E. Simoncicova E. Carrier M. Vecchiarelli H.A. Robert M.E. Tremblay M.E. Microglia fighting for neurological and mental health: on the central nervous system frontline of COVID-19 pandemic Front. Cell. Neurosci. 15 2021 647378 10.3389/fncel.2021.647378
Gong T. Liu L. Jiang W. Zhou R. DAMP-sensing receptors in sterile inflammation and inflammatory diseases Nat. Rev. Immunol. 20 2020 95 112 10.1038/s41577-019-0215-7 31558839
Hajebrahimi B. Bagheri M. Hassanshahi G. Nazari M. Bidaki R. Khodadadi H. Arababadi M.K. Kennedy D. The adapter proteins of TLRs, TRIF and MYD88, are upregulated in depressed individuals Int. J. Psychiatry Clin. Pract. 18 2014 41 44 10.3109/13651501.2013.859708 24168294
He J.G. Zhou H.Y. Xue S.G. Lu J.J. Xu J.F. Zhou B. Hu Z.L. Wu P.F. Long L.H. Ni L. Jin Y. Wang F. Chen J.G. Transcription factor TWIST1 integrates dendritic remodeling and chronic stress to promote depressive-like behaviors Biol. Psychiatry 89 2021 615 626 10.1016/j.biopsych.2020.09.003 33190845
Heijmans J. Buller N.V. Muncan V. van den Brink G.R. Rage mediated DAMP signaling in intestinal tumorigenesis Oncoimmunology 1 2012 1165 1166 10.4161/onci.20929 23170266
Hosoi T. Yamawaki Y. Kimura H. Honda S. Ozawa K. Possible involvement of MyD88 in regulating stress response in mice Front. Neurosci. 15 2021 621446 10.3389/fnins.2021.621446
Huang C. Huang L. Wang Y. Li X. Ren L. Gu X. Kang L. Guo L. Liu M. Zhou X. Luo J. Huang Z. Tu S. Zhao Y. Chen L. Xu D. Li Y. Li C. Peng L. Li Y. Xie W. Cui D. Shang L. Fan G. Xu J. Wang G. Wang Y. Zhong J. Wang C. Wang J. Zhang D. Cao B. 6-month consequences of COVID-19 in patients discharged from hospital: a cohort study Lancet 397 2021 220 232 10.1016/s0140-6736(20)32656-8 33428867
Huang L. Yao Q. Gu X. Wang Q. Ren L. Wang Y. Hu P. Guo L. Liu M. Xu J. Zhang X. Qu Y. Fan Y. Li X. Li C. Yu T. Xia J. Wei M. Chen L. Li Y. Xiao F. Liu D. Wang J. Wang X. Cao B. 1-year outcomes in hospital survivors with COVID-19: a longitudinal cohort study Lancet 398 2021 747 758 10.1016/s0140-6736(21)01755-4 34454673
Huebener P. Pradere J.P. Hernandez C. Gwak G.Y. Caviglia J.M. Mu X. Loike J.D. Schwabe R.F. The HMGB1/RAGE axis triggers neutrophil-mediated injury amplification following necrosis J. Clin. Invest. 125 2015 539 550 10.1172/JCI76887 25562324
Jang S.E. Hyam S.R. Jeong J.J. Han M.J. Kim D.H. Penta-O-galloyl-beta-D-glucose ameliorates inflammation by inhibiting MyD88/NF-kappaB and MyD88/MAPK signalling pathways Br. J. Pharmacol. 170 2013 1078 1091 10.1111/bph.12333 23941302
Khan S. Shafiei M.S. Longoria C. Schoggins J. Savani R.C. Zaki H. SARS-CoV-2 spike protein induces inflammation via TLR2-dependent activation of the NF-kappaB pathway eLife 10 2021 e68563 34866574
Koo J.W. Russo S.J. Ferguson D. Nestler E.J. Duman R.S. Nuclear factor-kappaB is a critical mediator of stress-impaired neurogenesis and depressive behavior Proc. Natl. Acad. Sci. U.S.A. 107 2010 2669 2674 10.1073/pnas.0910658107 20133768
Kumari P. Rothan H.A. Natekar J.P. Stone S. Pathak H. Strate P.G. Arora K. Brinton M.A. Kumar M. Neuroinvasion and encephalitis following intranasal inoculation of SARS-CoV-2 in K18-hACE2 mice Viruses 13 2021 132 10.3390/v13010132 33477869
Lawrence T. The nuclear factor NF-kappaB pathway in inflammation Cold Spring Harb. Perspect. Biol. 1 2009 a001651 10.1101/cshperspect.a001651
Leng L. Zhuang K. Liu Z. Huang C. Gao Y. Chen G. Lin H. Hu Y. Wu D. Shi M. Xie W. Sun H. Shao Z. Li H. Zhang K. Mo W. Huang T.Y. Xue M. Yuan Z. Zhang X. Bu G. Xu H. Xu Q. Zhang J. Menin deficiency leads to depressive-like behaviors in mice by modulating astrocyte-mediated neuroinflammation Neuron 100 551–563 2018 e7
Lewis S.A. Sureshchandra S. Zulu M.Z. Doratt B. Jankeel A. Ibraim I.C. Pinski A.N. Rhoades N.S. Curtis M. Jiang X. Tifrea D. Zaldivar F. Shen W. Edwards R.A. Chow D. Cooper D. Amin A. Messaoudi I. Differential dynamics of peripheral immune responses to acute SARS-CoV-2 infection in older adults Nat. Aging 1 2021 1038 1052 10.1038/s43587-021-00127-2
Li J.M. Liu L.L. Su W.J. Wang B. Zhang T. Zhang Y. Jiang C.L. Ketamine may exert antidepressant effects via suppressing NLRP3 inflammasome to upregulate AMPA receptors Neuropharmacology 146 2019 149 153 10.1016/j.neuropharm.2018.11.022 30496753
Li C. Zhang L.M. Zhang X. Huang X. Liu Y. Li M.Q. Xing S. Yang T. Xie L. Jiang F.C. Jiang H.Y. He W.T. Zhou P. Short-term pharmacological inhibition of MyD88 homodimerization by a novel inhibitor promotes robust allograft tolerance in mouse cardiac and skin transplantation Transplantation 101 2017 284 293 10.1097/TP.0000000000001471 27607533
Li M.X. Zheng H.L. Luo Y. He J.G. Wang W. Han J. Zhang L. Wang X. Ni L. Zhou H.Y. Hu Z.L. Wu P.F. Jin Y. Long L.H. Zhang H. Hu G. Chen J.G. Wang F. Gene deficiency and pharmacological inhibition of caspase-1 confers resilience to chronic social defeat stress via regulating the stability of surface AMPARs Mol. Psychiatry 23 2018 556 568 10.1038/mp.2017.76 28416811
Liang Q. Wu Q. Jiang J. Duan J. Wang C. Smith M.D. Lu H. Wang Q. Nagarkatti P. Fan D. Characterization of sparstolonin B, a Chinese herb-derived compound, as a selective Toll-like receptor antagonist with potent anti-inflammatory properties J. Biol. Chem. 286 2011 26470 26479 10.1074/jbc.M111.227934 21665946
Lin S.C. Lo Y.C. Wu H. Helical assembly in the MyD88-IRAK4-IRAK2 complex in TLR/IL-1R signalling Nature 465 2010 885 890 10.1038/nature09121 20485341
Liu P. Gao Q. Guan L. Sheng W. Hu Y. Gao T. Jiang J. Xu Y. Qiao H. Xue X. Liu S. Li T. Atorvastatin attenuates isoflurane-induced activation of ROS-p38MAPK/ATF2 pathway, neuronal degeneration, and cognitive impairment of the aged mice Front. Aging Neurosci. 12 2020 620946 10.3389/fnagi.2020.620946
Liu Q. Zhang Y. Liu S. Liu Y. Yang X. Liu G. Shimizu T. Ikenaka K. Fan K. Ma J. Cathepsin C promotes microglia M1 polarization and aggravates neuroinflammation via activation of Ca(2+)-dependent PKC/p38MAPK/NF-kappaB pathway J. Neuroinflammation 16 2019 10 10.1186/s12974-019-1398-3 30651105
Liu J. Zhang X. Wang H. Zhang M. Peng Y. Li M. Xie L. Jiang F. Gong Y. Zhao Q. Zhou P. Implication of myeloid differentiation factor 88 inhibitor TJ-M2010-5 for therapeutic intervention of hepatocellular carcinoma Hepatol. Res. 49 2019 1182 1194 10.1111/hepr.13359 31074165
Luo H. Wu P.F. Cao Y. Jin M. Shen T.T. Wang J. Huang J.G. Han Q.Q. He J.G. Deng S.L. Ni L. Hu Z.L. Long L.H. Wang F. Chen J.G. Angiotensin-converting enzyme inhibitor rapidly ameliorates depressive-type behaviors via bradykinin-dependent activation of mammalian target of rapamycin complex 1 Biol. Psychiatry 88 2020 415 425 10.1016/j.biopsych.2020.02.005 32220499
Matschke J. Lütgehetmann M. Hagel C. Sperhake J.P. Schröder A.S. Edler C. Mushumba H. Fitzek A. Allweiss L. Dandri M. Dottermusch M. Heinemann A. Pfefferle S. Schwabenland M. Sumner Magruder D. Bonn S. Prinz M. Gerloff C. Püschel K. Krasemann S. Aepfelbacher M. Glatzel M. Neuropathology of patients with COVID-19 in Germany: a post-mortem case series Lancet. Neurol. 19 2020 919 929 10.1016/s1474-4422(20)30308-2 33031735
Menard C. Pfau M.L. Hodes G.E. Kana V. Wang V.X. Bouchard S. Takahashi A. Flanigan M.E. Aleyasin H. LeClair K.B. Janssen W.G. Labonte B. Parise E.M. Lorsch Z.S. Golden S.A. Heshmati M. Tamminga C. Turecki G. Campbell M. Fayad Z.A. Tang C.Y. Merad M. Russo S.J. Social stress induces neurovascular pathology promoting depression Nat. Neurosci. 20 2017 1752 1760 10.1038/s41593-017-0010-3 29184215
Mingoti M.E.D. Bertollo A.G. Simoes J.L.B. Francisco G.R. Bagatini M.D. Ignacio Z.M. COVID-19, oxidative stress, and neuroinflammation in the depression route J. Mol. Neurosci. 23 2022 1 16 10.1007/s12031-022-02004-y
Norman G.J. Karelina K. Zhang N. Walton J.C. Morris J.S. Devries A.C. Stress and IL-1beta contribute to the development of depressive-like behavior following peripheral nerve injury Mol. Psychiatry 15 2010 404 414 10.1038/mp.2009.91 19773812
Olajide O.A. Iwuanyanwu V.U. Adegbola O.D. Al-Hindawi A.A. SARS-CoV-2 spike glycoprotein S1 induces neuroinflammation in BV-2 microglia Mol. Neurobiol. 59 2021 445 458 10.1007/s12035-021-02593-6 34709564
Rhea E.M. Logsdon A.F. Hansen K.M. Williams L.M. Reed M.J. Baumann K.K. Holden S.J. Raber J. Banks W.A. Erickson M.A. The S1 protein of SARS-CoV-2 crosses the blood-brain barrier in mice Nat. Neurosci. 24 2021 368 378 10.1038/s41593-020-00771-8 33328624
Ribeiro D.E. Oliveira-Giacomelli A. Glaser T. Arnaud-Sampaio V.F. Andrejew R. Dieckmann L. Baranova J. Lameu C. Ratajczak M.Z. Ulrich H. Hyperactivation of P2X7 receptors as a culprit of COVID-19 neuropathology Mol. Psychiatry 26 2021 1044 1059 10.1038/s41380-020-00965-3 33328588
Shen Z.C. Wu P.F. Wang F. Xia Z.X. Deng Q. Nie T.L. Zhang S.Q. Zheng H.L. Liu W.H. Lu J.J. Gao S.Q. Yao X.P. Long L.H. Hu Z.L. Chen J.G. Gephyrin palmitoylation in basolateral amygdala mediates the anxiolytic action of benzodiazepine Biol. Psychiatry 85 2019 202 213 10.1016/j.biopsych.2018.09.024 30454851
Shirato K. Kizaki T. SARS-CoV-2 spike protein S1 subunit induces pro-inflammatory responses via toll-like receptor 4 signaling in murine and human macrophages Heliyon 7 2021 e06187 33644468
Song E. Zhang C. Israelow B. Lu-Culligan A. Prado A.V. Skriabine S. Lu P. Weizman O.E. Liu F. Dai Y. Szigeti-Buck K. Yasumoto Y. Wang G. Castaldi C. Heltke J. Ng E. Wheeler J. Alfajaro M.M. Levavasseur E. Fontes B. Ravindra N.G. Van Dijk D. Mane S. Gunel M. Ring A. Kazmi S.A.J. Zhang K. Wilen C.B. Horvath T.L. Plu I. Haik S. Thomas J.L. Louvi A. Farhadian S.F. Huttner A. Seilhean D. Renier N. Bilguvar K. Iwasaki A. Neuroinvasion of SARS-CoV-2 in human and mouse brain J. Exp. Med. 218 2021 e20202135 33433624
Soung A.L. Vanderheiden A. Nordvig A.S. Sissoko C.A. Canoll P. Mariani M.B. Jiang X. Bricker T. Rosoklija G.B. Arango V. Underwood M. Mann J.J. Dwork A.J. Goldman J.E. Boon A.C.M. Boldrini M. Klein R.S. COVID-19 induces CNS cytokine expression and loss of hippocampal neurogenesis Brain. awac270 2022 10.1093/brain/awac270
Stewart, J.C., Rand, K.L., Muldoon, M.F., Kamarck, T.W., 2009. A prospective evaluation of the directionality of the depression-inflammation relationship. Brain Behav. Immun. 23, 936-944. https://doi.org/10.1016/j.bbi.2009.04.011.
Sulaiman G. Cooke A. Ffrench B. Gasch C. Abdullai O.A. O'Connor K. Elbaruni S. Blackshields G. Spillane C. Keegan H. McEneaney V. Knittel R. Rogers A. Jeffery I.B. Doyle B. Bates M. d'Adhemar C. Lee M.Y. Campbell E.L. Moynagh P.N. Higgins D.G. O'Toole S. O'Neill L. O'Leary J.J. Gallagher M.F. MyD88 is an essential component of retinoic acid-induced differentiation in human pluripotent embryonal carcinoma cells Cell Death Differ. 24 2017 1975 1986 10.1038/cdd.2017.124 28885616
Wang X. Tan Y. Huang Z. Huang N. Gao M. Zhou F. Hu J. Feng W. Disrupting myddosome assembly in diffuse large Bcell lymphoma cells using the MYD88 dimerization inhibitor ST2825 Oncol. Rep. 42 2019 1755 1766 10.3892/or.2019.7282 31432184
Welcome M.O. Mastorakis N.E. Stress-induced blood brain barrier disruption: Molecular mechanisms and signaling pathways Pharmacol. Res. 157 2020 104769 10.1016/j.phrs.2020.104769
Wu P.F. Han Q.Q. Chen F.F. Shen T.T. Li Y.H. Cao Y. Chen J.G. Wang F. Erasing m(6)A-dependent transcription signature of stress-sensitive genes triggers antidepressant actions Neurobiol. Stress 15 2021 100390 10.1016/j.ynstr.2021.100390
Xie L. Jiang F.C. Zhang L.M. He W.T. Liu J.H. Li M.Q. Zhang X. Xing S. Guo H. Zhou P. Targeting of MyD88 homodimerization by novel synthetic inhibitor TJ-M2010-5 in preventing colitis-associated colorectal Cancer J. Natl. Cancer Inst. 108 2016 djv364 10.1093/jnci/djv364 26712311
Zhang J.C. Yao W. Dong C. Yang C. Ren Q. Ma M. Han M. Wu J. Ushida Y. Suganuma H. Hashimoto K. Prophylactic effects of sulforaphane on depression-like behavior and dendritic changes in mice after inflammation J. Nutr. Biochem. 39 2017 134 144 10.1016/j.jnutbio.2016.10.004 27833054
Zhang L. Zhou L. Bao L. Liu J. Zhu H. Lv Q. Liu R. Chen W. Tong W. Wei Q. Xu Y. Deng W. Gao H. Xue J. Song Z. Yu P. Han Y. Zhang Y. Sun X. Yu X. Qin C. SARS-CoV-2 crosses the blood-brain barrier accompanied with basement membrane disruption without tight junctions alteration Signal Transduct. Target. Ther. 6 2021 337 10.1038/s41392-021-00719-9 34489403
Zhao Y. Kuang M. Li J. Zhu L. Jia Z. Guo X. Hu Y. Kong J. Yin H. Wang X. You F. SARS-CoV-2 spike protein interacts with and activates TLR4 Cell Res. 31 2021 818 820 10.1038/s41422-021-00495-9 33742149
Zheng M. Karki R. Williams E.P. Yang D. Fitzpatrick E. Vogel P. Jonsson C.B. Kanneganti T.D. TLR2 senses the SARS-CoV-2 envelope protein to produce inflammatory cytokines Nat. Immunol. 22 2021 829 838 10.1038/s41590-021-00937-x 33963333
Zhou H.Y. He J.G. Hu Z.L. Xue S.G. Xu J.F. Cui Q.Q. Gao S.Q. Zhou B. Wu P.F. Long L.H. Wang F. Chen J.G. A-kinase anchoring protein 150 and protein kinase a complex in the basolateral amygdala contributes to depressive-like behaviors induced by chronic restraint stress Biol. Psychiatry 86 2019 131 142 10.1016/j.biopsych.2019.03.967 31076080
Zou Z. Du D. Miao Y. Yang Y. Xie Y. Li Z. Zhou L. Zhang L. Zhou P. Jiang F. TJ-M2010-5, a novel MyD88 inhibitor, corrects R848-induced lupus-like immune disorders of B cells in vitro Int. Immunopharmacol. 85 2020 106648 10.1016/j.intimp.2020.106648
| 36496170 | PMC9726649 | NO-CC CODE | 2022-12-14 23:52:22 | no | Brain Behav Immun. 2023 Feb 7; 108:204-220 | utf-8 | Brain Behav Immun | 2,022 | 10.1016/j.bbi.2022.12.007 | oa_other |
==== Front
iScience
iScience
iScience
2589-0042
The Author(s).
S2589-0042(22)02015-6
10.1016/j.isci.2022.105742
105742
Article
Stimulation of IFN-β responses by aberrant SARS-CoV-2 small viral RNAs acting as RIG-I agonists
Arai Yasuha 16
Yamanaka Itaru 26
Okamoto Toru 36
Isobe Ayana 1
Nakai Naomi 2
Kamimura Naoko 2
Suzuki Tatsuya 3
Daidoji Tomo 1
Ono Takao 4
Nakaya Takaaki 1
Matsumoto Kazuhiko 4
Okuzaki Daisuke 25
Watanabe Yohei 17∗
1 Department of Infectious Diseases, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan
2 Genome Information Research Center, Research Institute for Microbial Diseases, Osaka University, Osaka 565-0871, Japan
3 Institute for Advanced Co-Creation Studies, Research Institute for Microbial Diseases, Osaka University, Osaka 565-0871, Japan
4 SANKEN, Osaka University, Osaka 567-0047, Japan
5 Single Cell Genomics, Human immunology, WPI Immunology Frontier Research Center, Osaka University, Osaka 565-0871, Japan
∗ Corresponding author (Y. Watanabe)
6 These authors contributed equally
7 Lead Contact
7 12 2022
7 12 2022
10574227 5 2022
3 7 2022
2 12 2022
© 2022 The Author(s)
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Patients with severe COVID-19 exhibit a cytokine storm characterized by greatly elevated levels of cytokines. Despite this, the interferon (IFN) response is delayed, contributing to disease progression. Here, we report that SARS-CoV-2 excessively generates small viral RNAs (svRNAs) encoding exact 5′ ends of positive-sense genes in human cells in vitro and ex vivo, whereas endemic human coronaviruses (OC43 and 229E) produce significantly fewer similar svRNAs. SARS-CoV-2 5′ end svRNAs are RIG-I agonists and induce the IFN-β response in later stages of infection. The first 60-nt ends bearing duplex structures and 5′-triphosphates are responsible for immune-stimulation. We propose that RIG-I activation by accumulated SARS-CoV-2 5′ end svRNAs may contribute to later drive over-exuberant IFN production. Additionally, the differences in the amounts of svRNAs produced and the corresponding IFN response among CoV strains suggest that lower svRNA production during replication may correlate with the weaker immune response seen in less pathogenic CoVs.
Graphical abstract
Keywords
SARS-CoV-2
COVID-19
5′ end small RNA
IFN induction
cytokine storm
==== Body
pmc
| 36507221 | PMC9726650 | NO-CC CODE | 2022-12-08 23:18:17 | no | iScience. 2022 Dec 7;:105742 | utf-8 | iScience | 2,022 | 10.1016/j.isci.2022.105742 | oa_other |
==== Front
Clin Nutr ESPEN
Clin Nutr ESPEN
Clinical Nutrition Espen
2405-4577
Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
S2405-4577(22)01412-7
10.1016/j.clnesp.2022.12.003
Letter to the Editor
Reply to Letter to the Editor to “Clinical Significance of Micronutrient Supplements in Patients with Coronavirus Disease 2019: A Comprehensive Systematic Review and Meta-Analysis”
Beran Azizullah MD a∗
Mhanna Mohammed MD a
Assaly Ragheb MD b
a Department of Internal Medicine, University of Toledo, Toledo, Ohio, USA
b Division of Pulmonary and Critical Care Medicine, University of Toledo, Toledo, Ohio, USA
∗ Corresponding author.
7 12 2022
7 12 2022
2 6 2022
2 12 2022
© 2022 Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Keywords
Micronutrient supplements
vitamin D
mortality
COVID-19
==== Body
pmc
| 0 | PMC9726651 | NO-CC CODE | 2022-12-08 23:18:17 | no | Clin Nutr ESPEN. 2022 Dec 7; doi: 10.1016/j.clnesp.2022.12.003 | utf-8 | Clin Nutr ESPEN | 2,022 | 10.1016/j.clnesp.2022.12.003 | oa_other |
==== Front
Heliyon
Heliyon
Heliyon
2405-8440
The Author(s). Published by Elsevier Ltd.
S2405-8440(22)03413-2
10.1016/j.heliyon.2022.e12125
e12125
Research Article
Dynamics of anti-spike IgG antibody after a third BNT162b2 COVID-19 vaccination in Japanese health care workers
Ikezaki Hiroaki abc∗
Nomura Hideyuki c
Shimono Nobuyuki a
a Department of General Internal Medicine, Kyushu University Hospital, 3-1-1, Maidashi, Higashi-ku, Fukuoka, 8128582, Japan
b Department of Comprehensive General Internal Medicine, Kyushu University Faculty of Medical Sciences, 3-1-1, Maidashi, Higashi-ku, Fukuoka, 8128582, Japan
c Department of Internal Medicine, Haradoi Hospital, 6-40-8, Aoba, Higashi-ku, Fukuoka, 8138588, Japan
∗ Corresponding author.
7 12 2022
7 12 2022
e1212514 5 2022
9 9 2022
28 11 2022
© 2022 The Author(s)
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objectives
Many countries are administering a third dose of some coronavirus disease 2019 (COVID-19) vaccines, but the evaluation of vaccine-induced immunity after boosting in East Asia is insufficient. This study aimed to evaluate anti-spike immunoglobulin G [IgG(S)] titers after the third BNT162b2 vaccination.
Methods
The dynamics of IgG(S) titers were assessed two months following the third BNT162b2 vaccination in 52 participants. All participants received the primary series of vaccination with BNT162b2 and received the third dose eight months after the second vaccination. Associations among the IgG(S) titer, baseline characteristics, and adverse reactions were also evaluated.
Results
The geometric mean titer of IgG(S) one month after the third vaccination was 17,400 AU/ml, which increased by approximately 30 times that immediately before the third vaccination. The rate of IgG(S) titer decline was significantly slower after the third vaccination (35.7%) than after the second vaccination (59.1%). The IgG(S) titer was significantly negatively associated with age (r = −0.31). Participants who had a headache at the time of vaccination showed significantly higher IgG(S) titers than those without a headache.
Conclusions
The IgG(S) titer induced by primary immunization with BNT162b2 waned over time. The third dose of BNT162b2 substantially increased the IgG(S) titer, with a slower rate of decline.
COVID-19; BNT162b2; Anti-spike IgG antibody.
Keywords
COVID-19
BNT162b2
Anti-spike IgG antibody
==== Body
pmc1 Introduction
Although vaccines against coronavirus disease 2019 (COVID-19), including messenger RNA (mRNA) vaccines, were highly effective in preventing COVID-19 at the initiation of vaccination 1, 2, it has become increasingly difficult to prevent the spread of COVID-19 with the first series of vaccines as mutant variants have emerged. On November 26, 2021, the World Health Organization named the new severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) B.1.1.529 variant omicron and classified it as a variant of concern based on a rapid increase in the number of confirmed cases of SARS-CoV-2 infection with this variant in South Africa 3, 4. The omicron variant has increased transmissibility and immune evasion even after natural infection and vaccination because it has a large number of mutations, including multiple mutations in the receptor-binding domain of the spike protein [5]. In addition, early laboratory data indicate that the neutralizing antibody response to the omicron strain is substantially reduced compared to that to the original strain or the delta variant in vaccinated individuals 6, 7, 8, 9. COVID-19 vaccines are highly effective against both symptomatic and severe diseases caused by the original strain and the alpha variant 1, 2. Although waning of anti-spike immunoglobulin G (IgG) titers and protection has been observed with time after vaccination, especially for the delta variant, booster (third) doses provide rapid and robust increases in anti-spike IgG titers and protection against both mild and severe diseases 10, 11, 12.
Based on these findings, many countries have commenced administering a booster (third) vaccine dose to curb the pandemic 10, 11, 12, 13. In addition, in December 2021, Israel became the first country in the world to administer a fourth dose [14]. Booster doses of both BNT162b2 and mRNA-1273 have been approved in Japan since December 2021 for those over 12 years old [15]. However, there are few reports regarding the postbooster anti-spike IgG antibody response and adverse reactions in East Asian populations; therefore, we aimed to evaluate the post-booster anti-spike IgG titers and adverse reactions in the Japanese population.
2 Materials and methods
2.1 Study participants and design
Study participants were recruited from among health care workers at Haradoi Hospital, a mixed-care hospital in Fukuoka 16, 17. Of the 485 health care workers in this hospital, 52 (10.7%) participated in this long-term prospective study. Most of the study participants were nurses, and approximately 85% were women. All three vaccines administered to the participants were the BNT162b2 mRNA COVID-19 vaccine (Comirnaty®: Pfizer/BioNTech). All participants were offered the first, second, and third doses of the vaccine in March, April, and December 2021. All participants provided written informed consent prior to enrollment. This study was carried out in accordance with the principles of the Declaration of Helsinki, as revised in 2008, and approved by the Haradoi Hospital institutional ethics review committee prior to data collection (approval No. 2020-08).
The main objective of this study was to evaluate the dynamics of anti-spike IgG titers, and the anti-spike IgG titers were measured nine times: before the first vaccination; three weeks after the first vaccination (just before the second vaccination); one, two, four, and six months after the second vaccination; before the third vaccination (approximately eight months after the second vaccination); and one and two months after the third vaccination. The secondary objective of this study was to assess the safety of the BNT162b2 mRNA COVID-19 vaccine by blood tests and interviews too solicit information about adverse reactions during vaccination. Participants provided information on their height, weight, smoking habits (current, past, or never), drinking habits (daily, often, or never), allergies, medical history, and medication use; whether they had experienced adverse reactions to the second and third vaccinations (fever, fatigue, headache, and swelling of axillary lymph nodes); and whether they had needed antipyretics.
2.2 Laboratory measurements
Levels of anti-spike IgG were quantified using a SARS-CoV-2 IgG II Quant assay (Abbott Diagnostics, Chicago, IL, USA) [18]. Participants underwent blood testing to quantitatively assess anti-spike IgG nine times (before the first vaccination; just before the second vaccination; one, two, four, and six months after the second vaccination; just before the third vaccination; and one and two months after the third vaccination). The results of anti-spike IgG quantification are expressed as arbitrary units per milliliter (AU/ml) (positive threshold: 50 AU/ml). We also performed qualitative tests for IgG/immunoglobulin M (IgM) antibodies against the SARS-CoV-2 nucleocapsid protein (positive thresholds: 1.40 index [S/C] for anti-nucleocapsid IgG and 1.00 index [S/C] for anti-nucleocapsid IgM) for all participants to exclude the effects of SARS-CoV-2 infection. Participants also had blood tests for total bilirubin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl transpeptidase (γ-GTP), and serum creatinine levels using standard enzymatic methods. The estimated glomerular filtration rate was calculated using the following equation: 194 × serum creatinine−1.094 × age−0.287 (× 0.739 [if female]).
2.3 Statistical analysis
Data are expressed as median values with 25th and 75th percentile values for continuous variables. The geometric mean titers (GMTs) of anti-spike IgG were calculated. Categorical variables are reported as frequencies and percentages. The Mann–Whitney U test was used to compare two groups, and the Kruskal–Wallis test was used to compare three groups. The Tukey–Kramer method was used for each two-group comparison among the three groups. Anti-spike IgG levels, with adjustment for age and sex, were determined by the least means square method. Paired t test was performed to evaluate differences in anti-spike IgG titers across timepoints, using log-transformed IgG titers. McNemar's test was used to evaluate differences between adverse reactions at the second and third vaccinations. The Wilcoxon signed-rank test was used to evaluate differences in laboratory data before and after the vaccinations. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC). A P value less than 0.05 was considered to indicate statistical significance.
3 Results
3.1 Baseline characteristics of the participants
Table 1 shows the baseline characteristics of the 52 participants of this study. The median age was 40.5 years, 84.6% were women, and the median body mass index (BMI) was 20.9 kg/m2. There were no current smokers, 13.5% were daily alcohol drinkers, and 17.3% had an allergy. Among the study participants, 25 (48.1%) had comorbidities. One patient had rheumatoid arthritis but did not take corticosteroids or immunosuppressants. Another had a history of colorectal cancer without any evidence of recurrence at the enrollment of this study. Before the first vaccination, the titers of anti-spike IgG and the anti-nucleocapsid IgG and IgM of all 49 participants were below the positive threshold, indicating that no one had contracted COVID-19 before participating in this study.Table 1 Baseline characteristics of 52 participants†.
Table 1Demographics
Age – years 40.5 [28.5, 49.5]
Sex – no. (%)
Female 44 (84.6)
Male 8 (15.4)
Body mass index‡ – kg/m2 20.9 [18.7, 22.9]
Smoking habit – no. (current/past/never) 0/6/46
Alcohol drinking habit – no. (daily/often/never) 7/30/15
Allergies – no. (%) 9 (17.3)
Comorbidities
Number of comorbidities – no.. 0 [0, 1]
Hypertension – no. (%) 6 (11.5)
Diabetes – no. (%) 3 (5.8)
Dyslipidemia – no. (%) 5 (9.6)
Hyperuricemia – no. (%) 1 (1.9)
Coronary heart disease – no. (%) 0 (0.0)
Arrhythmia – no. (%) 3 (5.8)
Stroke – no. (%) 0 (0.0)
Lung disease – no. (%) 4 (7.7)
Thyroid disease – no. (%) 1 (1.9)
Atopic dermatitis – no. (%) 8 (15.4)
Autoimmune disease – no. (%) 1 (1.9)
Cancer – no. (%) 1 (1.9)
† Continuous variables are presented as median [1st quartile, 3rd quartile], and categorical variables are presented as number (%).
‡ Body mass index was calculated using the following equation: body weight (kg)/height (m)/height (m).
3.2 Dynamics of anti-spike IgG titers
The dynamics of the anti-spike IgG titers are shown in Figure 1 . Black circles and gray lines indicate anti-spike IgG titers and their dynamics in each participant, while orange squares and lines represent the GMTs of anti-spike IgG and their dynamics in all study participants. One participant was excluded from the result for two months after the third vaccination because the anti-nucleocapsid IgM was positive at two months after the third vaccination. After the first vaccination, all participants had an anti-spike IgG level ≥50 AU/ml, and the GMT three weeks after the first vaccination (just before the second vaccination) was 1024.3 AU/ml. The GMT of anti-spike IgG increased 10-fold to 10361 AU/ml one month after the second vaccination. The anti-spike IgG titer peaked at one month after the second vaccination and continued to decrease until eight months after the vaccination (just before the third vaccination) as 4241.6, 1545.3, 838.3, and 578.6 AU/ml, at two, four, six, and eight months after the second vaccination, respectively. The GMT of anti-spike IgG at eight months after the second vaccination represented an average decrease rate of 93.5% from one month after the second vaccination. One month after the third vaccination, the GMT of anti-spike IgG titer was 17400 AU/ml, which increased 30-fold compared to that just before the third vaccination and 1.67-fold compared to that one month after the second vaccination. Similar to those after the second vaccination, the anti-spike IgG levels declined at two months compared to those at one month after the third vaccination, with a GMT of anti-spike IgG of 11185 AU/ml. However, the average rate of decline was 35.7%, a significantly slower pace of decline than the 59.1% rate of decline after the second vaccination (P < 0.01). The anti-spike IgG titers showed statistically significant differences at each timepoint (all P < 0.05).Figure 1 Dynamics of anti-spike IgG titers. The dynamics of the anti-spike IgG titer of each participant (black circles and gray lines) and geometric mean titer of anti-spike IgG (orange squares and lines) are shown. Compared to that at one month after the second vaccination, the geometric mean titer of anti-spike IgG before the third vaccination decreased by approximately one-tenth. One month after the third vaccination, the geometric mean titer of anti-spike IgG increased 30-fold before the third vaccination and 1.7-fold one month after the second vaccination.
Figure 1
3.3 Systemic adverse reactions and laboratory finding changes
Table 2 shows the differences in systemic adverse reactions after the second and third vaccinations. After both the second and third vaccinations, approximately half of the participants experienced fever; more participants had a fever of 38 °C or higher after the third vaccination than after the second vaccination, but there was no significant difference. The proportion experiencing fatigue also did not differ between the second and third vaccinations, with approximately 60% experiencing fatigue at either time. On the other hand, headache and axillary lymphadenopathy were more common after the third vaccination. After the third vaccination, headache occurred twice, and the rate of axillary lymphadenopathy was seven times higher than that after the second vaccination.Table 2 Comparison of adverse reactions at the 2nd and 3rd vaccinations.
Table 2 2nd vaccination 3rd vaccination P value†
Fever – no. (%) 25 (48.1) 23 (45.1) 0.80
Highest fever degree – no. (%)
37.0–37.4 °C 6 (11.5) 6 (11.5) 0.50
37.5–37.9 °C 11 (21.2) 6 (11.5)
≥38.0 °C 8 (15.4) 11 (21.2)
Fatigue – no. (%) 35 (67.3) 31 (59.6) 0.39
Headache – no. (%) 11 (21.2) 21 (40.4) 0.02
Swelling of axillary lymph nodes – no. (%) 3 (5.8) 21 (40.4) <0.01
† The McNemar test was performed for comparisons of discordant pairs.
Table 3 shows the changes in laboratory findings before and after each vaccination. In this study, we measured total bilirubin, AST, ALT, γ-GTP, and serum creatinine levels along with anti-spike IgG levels. Estimated glomerular filtration rates were calculated using the Cockcroft–Gault formula. Blood test results before the first vaccination were within normal limits for most of the study participants. Blood test values hardly changed before and after each vaccination. Values of ALT and γ-GTP levels decreased by 1 IU/ml after the first vaccination, but we do not consider these changes to be of clinical significance.Table 3 Comparison of the laboratory data before and after vaccination.†
Table 3 Before vaccination After vaccination‡ Change of value P value§
1st vaccination
Total bilirubin – mg/dl 0.6 [0.5, 0.9] 0.6 [0.4, 0.8] 0.0 [−0.2, 0.0] 0.15
Aspartate aminotransferase – IU/l 20 [16, 23] 19 [17, 21] 0.0 [−1.5, 1.0] 0.57
Alanine aminotransferase – IU/l 15 [12, 20] 14 [11, 19] −1 [−3, 1] 0.046
γ-Glutamyl transpeptidase – IU/l 16 [14, 25] 16 [13, 28] −1 [−3, 0] 0.01
Serum creatinine – mg/dl 0.63 [0.57, 0.70] 0.62 [0.56, 0.69] 0.00 [−0.03, 0.05] 0.46
eGFR‖ – ml/min/m2 89.2 [78.6, 99.9] 91.3 [76.0, 97.4] 0.0 [−6.3, 5.1] 0.57
2nd vaccination
Total bilirubin – mg/dl 0.6 [0.4, 0.8] 0.6 [0.5, 0.8] 0.0 [−0.1, 0.1] 0.25
Aspartate aminotransferase – IU/l 19 [17, 21] 18 [16, 21] 0 [−2, 1] 0.43
Alanine aminotransferase – IU/l 14 [11, 19] 14 [11, 19] 0 [−2, 2] 0.94
γ-Glutamyl transpeptidase – IU/l 16 [13, 28] 17 [13, 28] 0 [−1, 1] 0.93
Serum creatinine – mg/dl 0.62 [0.56, 0.69] 0.61 [0.55, 0.68] 0.00 [−0.05, 0.02] 0.41
eGFR‖ – ml/min/m2 91.3 [76.0, 97.4] 90.4 [74.7, 101.1] 0.0 [−3.5, 6.1] 0.43
3rd vaccination
Total bilirubin – mg/dl 0.6 [0.5, 0.8] 0.6 [0.5, 0.8] 0.0 [−0.1, 0.1] 0.46
Aspartate aminotransferase – IU/l 18 [16, 21] 19 [17, 22] 0 [−2, 2] 0.30
Alanine aminotransferase – IU/l 14 [11, 19] 16 [11, 21] 0 [−1, 2] 0.47
γ-Glutamyl transpeptidase – IU/l 15 [13, 20] 16 [14, 23] 0 [−1, 2] 0.10
Serum creatinine – mg/dl 0.62 [0.57, 0.69] 0.62 [0.55, 0.70] 0.00 [−0.04, 0.03] 0.50
eGFR‖ – ml/min/m2 90.4 [80.8, 98.1] 89.2 [74.8, 101.3] 0.0 [−3.8, 6.9] 0.44
eGFR, estimated glomerular filtration rate.
† Data are shown as the median [1st quartile, 3rd quartile].
‡ Blood samples were collected three weeks after the 1st vaccination (just before the 2nd vaccination) and one month after the 2nd and 3rd vaccinations.
§ P values were calculated using the Wilcoxon signed-rank sum test.
‖ The parameter was calculated using the following equation: 194 * serum creatinine−1.094 * age−0.287 (* 0.739 [if female]).
3.4 Factors influencing the anti-spike IgG titer after the third vaccination
We examined factors affecting the anti-spike IgG titer, such as age, BMI, sex, habits, comorbidities, and systemic adverse reactions to the vaccination. As we previously reported, age was the most substantial factor affecting the anti-spike IgG titer induced by vaccination, and there was a significantly negative association between age and the anti-spike IgG titer one month after the third vaccination (r = −0.31, P = 0.02). However, this association disappeared two months after the third vaccination (r = −0.22, P = 0.13). BMI, sex, smoking habits, and any comorbidities did not correlate with anti-spike IgG titers either one or two months after the third vaccination. Although we previously reported that daily alcohol drinkers had significantly lower anti-spike IgG titers, there was no significant association between alcohol drinking habits and anti-spike IgG titers after the third vaccination in this study. Regarding systemic adverse reactions, fever and fatigue did not correlate with anti-spike IgG titers; however, those who experienced headache or axillary lymphadenopathy tended to have higher anti-spike IgG titers after the third vaccination. Using antipyretics also did not affect the anti-spike IgG titer. After adjustment for age, only those who experienced a headache at the third vaccination had significantly higher anti-spike IgG titers one month after the third vaccination (29760 vs. 22182 AU/ml, respectively, P = 0.04).
We also examined the associations between anti-spike IgG titers before and after the third vaccination. The participants were divided into two groups based on anti-spike IgG titer one month after the second vaccination. Participants with anti-spike IgG titers below 10,000 AU/ml were grouped into the lower group, and those with anti-spike IgG titers greater than 10,000 AU/ml were grouped into the higher group. As expected, the higher group always had significantly higher anti-spike IgG titers than the lower group after the second vaccination. Although the GMT of anti-spike IgG in the higher group was approximately three times that in the lower group one month after the second vaccination (17,458 vs. 6148.8 AU/ml), the difference just before the third vaccination (eight months after the second vaccination) was approximately 1.7 times (732.1 vs. 438.1 AU/ml), indicating that the difference between the two groups had narrowed. One month after the third vaccination, the anti-spike IgG titer of both groups increased by a factor of approximately 30; the GMTs of the higher group and the lower group were 22,067 and 11851 AU/ml, respectively. Compared with that one month after the second vaccination, the GMT of the higher group increased by a factor of 1.15, but that of the lower group increased by a factor of 1.87, indicating that the third vaccination might be more effective for the lower group than for the higher group.
4 Discussion
The following three findings emerged from this prospective observational study. First, anti-spike IgG titers increased considerably after the third vaccination, and the GMTs of anti-spike IgG after the third vaccination were higher than those after the second vaccination. In addition, the rate of decline in anti-spike IgG titers after the third vaccination was slower than that after the second vaccination, suggesting that the vaccine-induced anti-spike IgG titer may be maintained for a more extended period. Second, low anti-spike IgG titers after the second vaccination correlated with a greater rate of increase after the third vaccination. As a result, the GMT of anti-spike IgG of study participants increased, with smaller differences between participants. Our results suggest that the third dose might be necessary to obtain sufficient immunogenicity with mRNA vaccines. Finally, no severe adverse reactions were observed for the series of vaccinations, including the third dose, indicating that the BNT162b2 vaccine was well tolerated.
In December 2021, Israel became the first country in the world to apply a fourth vaccination dose. A study of health care workers in Israel who received a fourth dose of BNT162b2 or mRNA-1273 four months after the third dose in a series of three BNT162b2 doses showed that IgG titers increased 3 to 4 times before the fourth dose [14]. However, the IgG titers after the fourth vaccination were just slightly higher than those after the third vaccination. The results of this study suggest that there may be an upper limit to the immunity induced by the mRNA vaccines and that the third vaccination may have achieved maximal immunogenicity. In our study, anti-spike IgG titers increased with the third vaccination by a factor of 3.0 compared to those before the third vaccination and by a factor of 1.7 compared to those after the second vaccination. In addition, participants with low pre-boost anti-spike IgG titers had a higher rate of increase after the third vaccination. A study of 61 healthy subjects also reported that low antibody titers before the third vaccination were associated with a higher rate of increase in antibody titers after the third vaccination [19]. Along with previous studies 19, 20, 21, our findings also suggest that the third vaccination restored vaccine effectiveness against SARS-CoV-2-induced disease, but mRNA vaccine immunogenicity may have an upper limit.
Vaccination is thought to be the most effective method of preventing infectious diseases. Since mRNA vaccines were used for the first time in humans, how immunogenic they are and how many doses are required to provide adequate immunity are unclear. And data from East Asia is insufficient to elucidate these questions. A study of 129 healthcare workers in Greece, including previously infected participants, showed that the third dose of BNT162b2 six months after the primary series of BNT162b2 vaccination increased anti-spike IgG titer to 20231 AU/ml from 437 AU/ml [22], similar to our result. Another study of 90 healthcare workers in Indonesia reported that a booster dose mRNA-1273 vaccine increased anti-spike IgG titer 700-fold compared to that six months after primary vaccination with CoronaVac [21]. The result suggests that mRNA vaccines might elicit an antibody response regardless of previous types of vaccine. The efficacy of a booster dose BNT162b2 vaccine among the elderly was also reported [19]. The third dose of BNT162b2 increased anti-spike IgG titer to 25468 AU/ml, 57.8-fold compared to pre-boost IgG titer among 97 elderly Israelis. No significant association was found between IgG titers and comorbidities or age. However, the median age of the participants was 70 years. Further studies on more elderly populations are also needed. In our previous report [16], the GMT of anti-spike IgG declined to approximately 40% from one month to two months after the second vaccination. In the current study, we found that the rate of decline of the GMT became significantly slower after the third vaccination. Our results suggest that mRNA vaccines, similar to inactivated vaccines such as the hepatitis B vaccine, require booster vaccinations to induce adequate immunity.
Fever, headache, fatigue, and pain at the injection site were the most commonly reported adverse reactions to COVID-19 vaccines, and overall, most adverse reactions were mild and short-lived 23, 24. Although very rare, there have been reports of serious adverse reactions. The following four serious adverse reactions to certain types of COVID-19 vaccination have been found: anaphylaxis, thrombosis with thrombocytopenia syndrome, myocarditis and pericarditis, and Guillain–Barré syndrome 25, 26, 27. There are also some reports of death after COVID-19 vaccination. According to the Vaccine Adverse Event Reporting System in the United States, preliminary death rates after the COVID-19 vaccination were 0.0024% [28]. Along with these previous reports, BNT162b2 appeared to be well tolerated in this study. Even at the third dose, blood tests showed that hepatic enzyme and serum creatinine levels did not change. Regarding systemic reactions, the rates of fever and fatigue did not differ between the second and third vaccinations. Although headache and axillary lymphadenopathy were more frequently observed with the third vaccination than with the second vaccination, no severe adverse reactions were observed.
Limitations of this study should be noted. Because the anti-spike IgG titer measured in this study was against the original strain of SARS-CoV-2, it is difficult to estimate the vaccine efficacy against the omicron variant. The observed anti-spike IgG titer might be estimated to be lower against the omicron variant. Moreover, we did not assess cell-mediated immunity. However, a higher anti-spike IgG titer is still considered protective against SARS-CoV-2 infection, including that by the omicron variant. Therefore, our results suggest that the third vaccination might restore vaccine effectiveness against SARAS-CoV-2 infection. In addition, the number of participants was small, and the participants were young and healthy. Additional research with more participants and a more comprehensive age range is necessary to validate our findings. Finally, we could not assess the effect of the third dose of the vaccine on COVID-19 prevention because there was only one participant possibly infected with SARS-CoV-2, and there was no control group. Further analyses with nationwide surveys are necessary to confirm the efficacy of the additional COVID-19 vaccination.
In conclusion, the third dose of BNT162b2 vaccination successfully increased the anti-spike IgG titer, and the efficacy of vaccination might be maintained longer because the rate of decline of anti-spike IgG titers is slower than that after the second vaccination.
Declarations
Author contribution statement
Hiroaki Ikezaki: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data;
Wrote the paper.
Hideyuki Nomura; Nobuyuki Shimono: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.
Funding statement
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availability statement
The data that has been used is confidential.
Declaration of interest's statement
The authors declare no conflict of interest.
Additional information
No additional information is available for this paper.
Acknowledgements
The authors thank Ms. Ryoko Nakashima for managing the dataset and Drs. Kahori Miyoshi, Yuichi Hara, Jun Hayashi, and Hiroshi Hara for scientific advice.
==== Refs
References
1 Pritchard E. Matthews P.C. Stoesser N. Eyre D.W. Gethings O. Vihta K.D. Impact of vaccination on new SARS-CoV-2 infections in the United Kingdom Nat. Med. 27 2021 1370 1378 34108716
2 Paris C. Perrin S. Hamonic S. Bourget B. Roué C. Brassard O. Effectiveness of mRNA-BNT162b2, mRNA-1273, and ChAdOx1 nCoV-19 vaccines against COVID-19 in healthcare workers: an observational study using surveillance data Clin. Microbiol. Infect. 27 2021 1699.e5–e8
3 World Health Organization. Update on Omicron. https://www.who.int/news/item/28-11-2021-update-on-omicron. [accessed April 6, 2022].
4 Viana R. Moyo S. Amoako D.G. Tegally H. Scheepers C. Althaus C.L. Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in southern Africa Nature 603 2022 679 686 35042229
5 European Centre for Disease Prevention and Control. Implications of the further emergence and spread of the SARSCoV-2 B.1.1.529 variant of concern (Omicron) for the EU/EEA – first update. December 2, 2021. https://www.ecdc.europa.eu/sites/default/files/documents/threat-assessment-covid-19-emergence-sars-cov-2-variant-omicron-december-2021.pdf. [accessed April 6, 2022]
6 Cele S. Jackson L. Khoury D.S. Khan K. Moyo-Gwete T. Tegally H. Omicron extensively but incompletely escapes Pfizer BNT162b2 neutralization Nature 602 2022 654 656 35016196
7 Schmidt F. Muecksch F. Weisblum Y. Silva J.D. Bednarski E. Cho A. Plasma neutralization of the SARS-CoV-2 omicron variant N. Engl. J. Med. 386 2022 599 601 35030645
8 Liu L. Iketani S. Guo Y. Chan J.F.W. Wang M. Liu L. Striking antibody evasion manifested by the Omicron variant of SARS-CoV-2 Nature 602 2022 676 681 35016198
9 Karim S.S.A. Karim Q.A. Omicron SARS-CoV-2 variant: a new chapter in the COVID-19 pandemic Lancet 398 2021 2126 2128 34871545
10 Barda N. Dagan N. Cohen C. Hernán M.A. Lipsitch M. Kohane I.S. Effectiveness of a third dose of the BNT162b2 mRNA COVID-19 vaccine for preventing severe outcomes in Israel: an observational study Lancet 398 2021 2093 2100 34756184
11 Bar-On Y.M. Goldberg Y. Mandel M. Bodenheimer O. Freedman L. Kalkstein N. Protection of BNT162b2 vaccine booster against covid-19 in Israel N. Engl. J. Med. 385 2021 1393 1400 34525275
12 Andrews N. Stowe N. Kirsebom F. Toffa S. Rickeard T. Gallagher E. Covid-19 vaccine effectiveness against the omicron (B.1.1.529) variant N. Engl. J. Med. 2022 online ahead of print
13 Patalon T. Gazit S. Pitzer V.E. Prunas O. Warren J.L. Weinberger D.M. Odds of testing positive for SARS-CoV-2 following receipt of 3 vs 2 doses of the BNT162b2 mRNA vaccine JAMA Intern. Med. 182 2022 179 184 34846533
14 Regev-Yochay G. Gonen T. Gilboa M. Mandelboim M. Indenbaum V. Amit S. Efficacy of a fourth dose of Covid-19 mRNA vaccine against omicron N. Engl. J. Med. 2022 online ahead of print
15 Ministry of Health, Labour and Welfare. COVID-19 vaccine booster shots (3rd dose). https://www.mhlw.go.jp/stf/covid-19/booster.html. [accessed April 6, 2022].
16 Ikezaki H. Nomura H. Shimono N. Dynamics of anti-Spike IgG antibody level after the second BNT162b2 COVID-19 vaccination in health care workers J. Infect. Chemother. 28 2022 802 805 35288023
17 Ikezaki H. Hara Y. Hayashi J. Hara H. Nomura H. Shimono Y. Low IgG antibody production in the elderly Japanese population after full BNT162b2 vaccination J. Hosp. Gen. Med. 4 2022 25 28
18 Abbott Diagnostics. Architect SARS-COV-2 IgG II Quant Instructions for Use, H18566R01. Abbott Diagnostics, IL, USA.
19 Goel R.R. Painter M.M. Lundgreen K.A. Apostolidis S.A. Baxter A.E. Giles J.R. Efficient recall of Omicron-reactive B cell memory after a third dose of SARS-CoV-2 mRNA vaccine Cell 185 2022 1875 1887 35523182
20 Eliakim-Raz N. Leibovici-Weisman Y. Stemmer A. Ness A. Awwad M. Ghantous N. Antibody titers before and after a third dose of the SARS-CoV-2 BNT162b2 vaccine in adults aged ≥60 years JAMA 326 2021 2203 2204 34739043
21 Cucunawangsih C. Wijaya R.S. Lugito N.P.H. Suriapranata I. Antibody response after a third dose mRNA-1273 vaccine among vaccinated healthcare workers with two doses of inactivated SARS-CoV-2 vaccine Int. J. Infect. Dis. 118 2022 116 118 35192955
22 Kontopoulou K. Nakas C.T. Papazisis G. Significant increase in antibody titers after the 3rd booster dose of the Pfizer–BioNTech mRNA COVID-19 vaccine in healthcare workers in Greece Vaccines 10 2022 876 35746484
23 Paran Y. Saiag E. Spitzer A. Angel Y. Yakubovsky M. Padova H. Short-term safety of booster immunization with BNT162b2 mRNA COVID-19 vaccine in healthcare workers Open Forum Infect. Dis. 9 2021 ofab656 35165656
24 David S.S.B. Gez S.B. Rahamim-Cohen D. Shamir-Stein N. Lerner U. Zohar A.E. Immediate side effects of Comirnaty COVID-19 vaccine: a nationwide survey of vaccinated people in Israel, December 2020 to March 2021 Euro Surveill. 27 2022 pii=2100540
25 Chong K.M. Yang C.Y. Lin C.C. Lien W.C. Severe immune thrombocytopenia following COVID-19 vaccination (Moderna) and immune checkpoint inhibitor: a case report Am. J. Emerg. Med. 2022 online ahead of print
26 Ogata S. Ishi Y. Asano K. Kobayashi E. Kubota S. Takahashi K. Sensory ataxic Guillain-Barré Syndrome with dysgeusia after mRNA COVID-19 vaccination Intern. Med. 2022 online ahead of print
27 Liu R. Pan J. Zhang C. Sun X. Cardiovascular complications of COVID-19 vaccines Front. Cardiovasc. Med. 9 2022 840929
28 Vaccine Adverse Event Reporting System. https://vaers.hhs.gov. [accessed April 6, 2022].
| 36506407 | PMC9726652 | NO-CC CODE | 2022-12-10 23:15:26 | no | Heliyon. 2022 Dec 7; 8(12):e12125 | utf-8 | Heliyon | 2,022 | 10.1016/j.heliyon.2022.e12125 | oa_other |
==== Front
iScience
iScience
iScience
2589-0042
The Author(s).
S2589-0042(22)02026-0
10.1016/j.isci.2022.105753
105753
Article
Ad26.COV2.S priming provided a solid immunological base for mRNA-based COVID-19 booster vaccination
Geers Daryl 1†
Sablerolles Roos S.G. 2†
van Baarle Debbie 34
Kootstra Neeltje A. 5
Rietdijk Wim J.R. 2
Schmitz Katharina S. 1
Gommers Lennert 1
Bogers Susanne 1
Nieuwkoop Nella J. 1
van Dijk Laura L.A. 1
van Haren Eva 2
Lafeber Melvin 6
Dalm Virgil A.S.H. 7
Goorhuis Abraham 89
Postma Douwe F. 10
Visser Leo G. 11
Huckriede Anke L.W. 3
Sette Alessandro 1213
Grifoni Alba 12
de Swart Rik L. 1
Koopmans Marion P.G. 1
van der Kuy P. Hugo M. 2‡
GeurtsvanKessel Corine H. 1‡
de Vries Rory D. 1‡∗
on behalf of the SWITCH research group
1 Department of Viroscience, Erasmus Medical Center, Rotterdam, the Netherlands
2 Department of Hospital Pharmacy, Erasmus Medical Center, Rotterdam, the Netherlands
3 Department of Medical Microbiology and Infection Prevention, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands
4 Center for Infectious Disease Control, National Institute for Public Health and the Environment, Bilthoven, the Netherlands
5 Department of Experimental Immunology, Amsterdam University Medical Centers, Amsterdam Infection and Immunity Institute, University of Amsterdam, Amsterdam, the Netherlands
6 Department of Internal Medicine, Erasmus Medical Center, Rotterdam, the Netherlands
7 Department of Internal Medicine, Division of Allergy & Clinical Immunology and Department of Immunology, Erasmus Medical Center, Rotterdam, the Netherlands
8 Center of Tropical Medicine and Travel Medicine, Department of Infectious Diseases, Amsterdam University Medical Centers, Amsterdam, the Netherlands
9 Infection & Immunity, Amsterdam Public Health, University of Amsterdam, Amsterdam, the Netherlands
10 Department of Internal Medicine and Infectious Diseases, University Medical Center Groningen, Groningen, the Netherlands
11 Department of Infectious Diseases, Leiden University Medical Center, Leiden, the Netherlands
12 Center for Infectious Disease and Vaccine Research, La Jolla Institute for Immunology (LJI), La Jolla, CA, USA
13 Department of Medicine, Division of Infectious Diseases and Global Public Health, University of California, San Diego (UCSD), La Jolla, CA, USA
∗ Corresponding author & lead contact: Rory D. de Vries
† Equal first author contribution
‡ equal senior author contribution
7 12 2022
7 12 2022
10575325 7 2022
10 11 2022
5 12 2022
© 2022 The Author(s)
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The emergence of novel SARS-CoV-2 variants led to the recommendation of booster vaccinations after Ad26.COV2.S priming. It was previously shown that heterologous booster vaccination induces high antibody levels, but how heterologous boosters affect other functional aspects of the immune response remained unknown. Here, we performed immunological profiling of Ad26.COV2.S-primed individuals before and after homologous or heterologous (mRNA-1273 or BNT162b2) booster. Booster vaccinations increased functional antibodies targeting ancestral SARS-CoV-2 and emerging variants. Especially heterologous booster vaccinations induced high levels of functional antibodies. In contrast, T cell responses were similar in magnitude following homologous or heterologous booster vaccination, and retained cross-reactivity towards variants. Booster vaccination led to a minimal expansion of SARS-CoV-2-specific T cell clones and no increase in breadth of the T cell repertoire. In conclusion, we show that Ad26.COV2.S priming vaccination provided a solid immunological base for heterologous boosting, increasing humoral and cellular responses targeting emerging variants of concern.
Graphical abstract
==== Body
pmc
| 36507223 | PMC9726653 | NO-CC CODE | 2022-12-16 23:21:34 | no | iScience. 2023 Jan 20; 26(1):105753 | utf-8 | iScience | 2,022 | 10.1016/j.isci.2022.105753 | oa_other |
==== Front
Soc Sci Med
Soc Sci Med
Social Science & Medicine (1982)
0277-9536
1873-5347
Published by Elsevier Ltd.
S0277-9536(22)00900-5
10.1016/j.socscimed.2022.115594
115594
Article
Economic risk framing increases intention to vaccinate among Republican COVID-19 vaccine refusers
Zhong Wei ∗
Broniatowski David
Department of Engineering Management and Systems Engineering, The George Washington University, Washington, DC, United States
∗ Corresponding author. Department of Engineering Management and Systems Engineering, The George Washington University, 800 22nd St NW, Washington, DC, 20052, United States.
7 12 2022
7 12 2022
11559418 4 2022
22 11 2022
1 12 2022
© 2022 Published by Elsevier Ltd.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objective
To determine if framing communications about COVID-19 vaccines in economic terms can increase Republicans’ likelihood to get vaccinated.
Methods
We examined Twitter posts between January 2020 and September 2021 by Democratic and Republican politicians to determine how they framed the COVID-19 pandemic. Based on these posts, we carried out a survey study between September and November 2021 to examine whether motivations for COVID-19 vaccine uptake matched message frames that were widely used by these politicians. Finally, we conducted a randomized controlled experiment to examine how these frames (economic vs. health) affected intentions to vaccinate by vaccine refusers in both parties.
Results
Republican politicians were more likely to frame the pandemic in economic terms, whereas Democrats predominantly used health frames. Accordingly, vaccinated Republicans’ choices were more likely to be motivated by economic consideration (β = 0.25, p = 0.02) and personal financial rationales (β = 0.24, p = 0.03). Among vaccine refusers, Republicans exposed to messages using economic rationales to encourage vaccination reported higher vaccination intentions compared to those exposed to messages using public health rationales (F1,119 = 4.16, p = 0.04).
Conclusion
Messages highlighting economic and personal financial risks could increase intentions to vaccinate for vaccine-hesitant Republicans.
Public health implications.
Agencies should invest in developing messages that are congruent with frames that are already widely used by co-partisans. Social media may be helpful in eliciting these frames.
Keywords
COVID-19
Vaccination
Republican
Hesitancy
Economy
Financial risk
Framing
==== Body
pmc1 Introduction
The COVID-19 pandemic remains a global threat to lives, livelihoods, and lifestyles. According to the Centers for Disease Control and Prevention (CDC), the total number of deaths in the United States due to COVID-19 exceeded one million as of November 2022 (CDC, 2022b). The global pandemic has also imposed high economic and social costs on individuals, institutions, businesses, and communities. The economic burden associated with unmitigated COVID-19 is estimated to be a cumulative $1.4 trillion by 2030 for the United States, assuming that 60% percent of the population will be infected between 2020 and 2023 (Chen, 2021).
COVID-19 vaccine uptake is critical for mitigating and slowing not only the impact of the pandemic but also the risks of COVID-19 variants in the United States. To date, although more than half (68.5%) of the US population has fully vaccinated against COVID-19, less than half (49.1%) has received a booster dose, and only 8.4% have gotten an updated booster, according to the CDC (CDC, 2022a). While these vaccines have led to steep declines in COVID-19 cases and deaths, vaccine hesitancy and refusal still pose a severe threat, undermining efforts to control the pandemic. Moreover, COVID-19 may become an endemic disease, perhaps with seasonal epidemic peaks. Ongoing manifestations of severe disease combined with high levels of infection could, in turn, foster the future evolution of the virus (Telenti, 2021). Renewed efforts to increase vaccine uptake are therefore critical to limiting transmission and achieving long-term herd immunity.
Political orientation continues to be strongly associated with people's appraisal of the seriousness of COVID-19, and COVID-19 vaccine uptake (Bruine de Bruin et al., 2020; Fridman, 2021; Khubchandani, 2020; Khubchandani, 2021; Ruiz & Bell, 2021). For instance, Republicans remain more skeptical of COVID-19 vaccines than Democrats, and make up an increasingly disproportionate share of those who remain unvaccinated and or only partially vaccinated (KFF, 2022a).
Scholars have explored why Republicans are more resistant to COVID-19 vaccines and how to increase Republicans’ intentions to vaccinate. Explanations for vaccine hesitancy include a higher prevalence of misconceptions about COVID-19 among Republicans, which might drive vaccine skepticism (Kreps, 2021), increased acceptance of conspiracy theories (Ruiz & Bell, 2021), and greater exposure to anti-vaccine content from prominent political figures (Hornsey et al., 2020). Pink et al. (2021), found that cues from partisan elites can effectively increase vaccine intentions by invoking the partisan nature of vaccine opposition; however, Sylvester (2022) contend that vaccine hesitancy would only be reduced among moderately partisan audiences. These approaches suggest that vaccine hesitancy among Republicans would remain high unless Republican elites decide to explicitly endorse vaccination; however, these endorsements have been limited in practice (KFF, 2022b).
Moving beyond endorsements from political figures, we seek to determine whether COVID-19 vaccine hesitancy associated with political partisanship can be overcome using tailored messaging strategies. Since partisanship may also exert effects on future public health efforts, we expect our findings to generalize to future health communication efforts. We therefore seek to understand the relationship between partisanship and vaccine hesitancy.
Numerous studies demonstrate that politicized and divergent party narratives about the pandemic on social media might help explain the observed partisan gap (Feng & Shao, 2022; Panda, 2020). Although prior authors examining traditional media sources such as newspapers and television have noted that coverage surrounding discussions of COVID-19 from March to May 2020 was highly politicized (Hart, 2020), studies focusing on Twitter in particular show that the Democratic party put more emphasis on public health, whereas the Republican party put more focus on national unity, China's alleged culpability for the pandemic, and the impacts of the pandemic on business (Jing & Ahn, 2021). Thus, as the pandemic unfolded, politicians used Twitter to help the public interpret events with responses increasingly divided across political ideological lines.
The highly politicized and polarized rhetoric of COVID-19 influenced views and attitudes toward COVID-19, due to differences in “issue framing.” Issue framing, increasingly used in political communication (Borah, 2011) and health communication (Guenther, 2021), describes a process by which people develop a particular conceptualization of, or reorient their thinking about, an issue (Chong & Druckman, 2007). Issue framing affects the attitudes and beliefs of audiences, leading to behavior changes (Ajzen, 1985, 1991).
Recent research has documented the effects of issue framing on attitudes and intentions toward COVID-19 vaccines (Borah, 2022; Borah, 2021; Huang and Liu, 2022; Reinhardt & Rossmann, 2021; Yousaf, 2022); however, this literature primarily examines interventions that emphasize health-related information. Specifically, these studies suggest that framing of COVID-19 vaccination in terms of health-related risks and/or benefits may increase vaccination rates (Ashworth, 2021; Borah, 2021; Hallsworth, 2021; Hornsey et al., 2020; Jordan, 2021; Motta, 2021; Palm, 2021). For instance, messages about vaccine safety (Palm, 2021; Van der Linden et al., 2015), risks or benefits to self (Ashworth, 2021) and others (e.g., family, friends, or community members) (Duquette, 2020), and vaccines allowing life to return to normal (Hallsworth, 2021) have been found to increase vaccination rates and intentions. In practice, messages using these frames appear to have had limited efficacy among Republican audiences, as indicated by the partisan vaccination gap. Although some of these prior studies tested the effects of messages on small samples of self-identified Republicans, comparatively little work has examined what issue framing strategies might encourage COVID-19 vaccine uptake among Republicans, in particular.
We posit that economic-related messages might be effective for this purpose. Our rationale is as follows: compared to Democratic politicians who primarily framed vaccination in terms of reducing public health threats, Republican politicians were more likely to discuss the economic and financial costs of the pandemic (e.g., its impacts on small business, and the need for financial assistance programs). We expect that these framing choices capture the attitudes and behaviors of their partisan fellows. Combined with individual inclinations for motivated reasoning (Kunda, 1990), this influence amplifies belief differences pertaining to COVID-19 vaccination. Consequently, we expect that framing messages in health vs. economic terms would have different effects on vaccination intentions among partisans. Specifically, we posit that Republicans and Republican-leaners were more likely to view vaccination as a solution to an economic and personal financial crisis when compared to Democrats. Hence, we test the efficacy of a pro-vaccination message framed in economic terms which, we posit, might leverage Republicans’ perceptions of economic and personal financial risks about COVID-19 to encourage vaccination. We anticipate that these messages might promote vaccine uptake beyond the effects of more conventional, health-framed messaging.
2 Methods
We conducted three studies to test our hypothesis. First, we conducted a retrospective observational study examining Twitter posts by Democratic and Republican politicians to determine how they framed the pandemic in public discourse. Second, we conducted a correlational study using a survey administered on Amazon's Mechanical Turk service (MTurk) – an online crowdsourcing platform – to examine the effects of economic and public health frames on COVID-19 vaccine motivations. We examined how these motivations varied between partisans. Finally, we conducted a randomized controlled experiment to examine the causal effect of these message frames (economic vs. health) on vaccine refusers among subjects who were affiliated with different political parties.
2.1 Study 1: how did partisan politicians frame the pandemic on twitter?
Using the Social Feed Manager software tool (Wrubel & Kerchner, 2020), we retrieved all available tweets containing at least one vaccine or COVID-19 keyword (see Supplementary Material) from 517 United States Senators and Representatives in the 116th US Congress. We next examined differences in how Democratic and Republican legislators framed the pandemic by comparing the frequencies of the top 15 bigrams (two-word phrases, e.g., “pandemic response”) used by members of each party. We also fit a Latent Dirichlet Allocation (LDA) topic model (Blei, 2003) to the same dataset and compared topic proportions across legislators from each party.
2.2 Study 2: what motivated partisans to get vaccinated?
We next conducted a correlational study in which we surveyed adults in the United States and asked them whether they had received at least one dose of the COVID-19 vaccine. Subjects were recruited using MTurk between September 20, 2021, and November 5, 2021. We examined the subset of subjects who reported having gotten at least one dose of a COVID-19 vaccine and asked them to answer questions indicating their motivations for vaccination. Motivations included personal health risk – “protecting myself/not having to worry about getting sick from the virus; ” public health risk – “preventing more illness and death in America; ” personal financial risk – “going back to work/reducing personal financial loss; ” and economic risk – “getting the economy moving again.” Subjects rated their motivations on an 8-point Likert scale from 0 (not at all) to 7 (very much). The respondents were also asked their party affiliation and other demographic questions (see Supplementary Material). We analyzed these data by fitting ordinary least squares (OLS) linear regression models to each motivation factor to test the hypothesis that Republicans were more motivated by economic and personal financial considerations, whereas Democrats were more motivated by public and personal health considerations. Specifically, we used political party as a categorical independent variable and controlled for other demographic variables, such as age, education, gender, race, and Hispanic ethnicity. We also compared the average motivations of Republicans to those of Democrats using permutation tests. Where relevant, we calculated effect sizes and provided eta squared results (Cohen, 1988; Tabachnick and Fidell, 2007; Thompson, 2006).
2.3 Study 3: can economic framing increase intent to vaccinate among the most resistant Republicans?
We next carried out a randomized controlled experiment to examine whether framing the decision to vaccinate in economic terms would increase Republicans’ intentions to vaccinate. To do so, we invited subjects from Study 2 who had indicated that they had not been vaccinated for COVID-19 to take a follow-up survey, in which they were randomized into one of four conditions using MTurk between November 4, 2021, and November 9, 2021. Specifically, we randomly assigned these subjects into two message framing conditions: economic and health. We also manipulated whether the message included a bottom-line summary (i.e., bottom line vs. no bottom line), with these two factors fully randomized and counterbalanced, constituting a 2 x 2 full-factorial design. In this survey, respondents were shown a message corresponding to their experimental condition. After reading the message, participants were asked again about their willingness to vaccinate against COVID-19 when the vaccine was available at no cost. Next, participants were asked a series of questions designed to index their perceptions regarding personal health and financial risks and threats due to the pandemic. Specifically, we asked them to what extent they agreed with statements indicating that the coronavirus outbreak was a major risk to their personal health and personal financial situations, with all questions answered on a 7-point Likert scale ranging from 1 = strongly disagree to 7 = strongly disagree. All surveys included attention checks to filter out inattentive workers. Table 1 summarizes the elements of our experimental design.Table 1 Description of information treatments and the number of participants in each treatment.
Table 1Treatment Description n
Health frame without bottom line Please read the following statement.
The Covid-19 pandemic has led to a dramatic loss of human life and represents an unprecedented challenge to public health in the United States. From January 3, 2020 to June 21, 2021, there have been about 33.5 million confirmed cases of Covid-19 with 601,000 deaths across the country. Currently, COVID-19 is the number one cause of death in the United States, followed by heart disease and cancer. COVID-19 also negatively affected many people's mental health and created new barriers for people already suffering from mental illness and substance use disorders. In 2020, about 4 in 10 adults in the US have reported symptoms of anxiety or depression.
COVID-19 vaccines are highly effective at reducing disease incidence, protecting against severe illness requiring hospitalization and death due to COVID-19. Experts estimated that if at least 75% of the US population got vaccinated with a vaccine efficacy of 70%, then the epidemic peak can be reduced by more than 99% without other interventions. 78
Health frame bottom line The bottom line: By getting vaccinated, you can help to eradicate this pandemic, preventing illness and saving lives. 72
Economic frame without bottom line Please read the following statement.
The economic impact of the COVID-19 pandemic in the US has been disruptive, affecting travel, financial markets, employment, shipping, small businesses and other industries. In 2020, the economy contracted at its deepest pace since World War 2, and the Gross Domestic Product decreased 3.5%, the biggest drop since 1946. Unemployment neared Great Depression levels. Government stimulus totaled more than $2 trillion, sending deficits to record levels. The pandemic resulted in permanent closure of roughly 200,000 US establishments in 2020. There is a consensus among economists that vaccinations will profoundly shape the course of the economic recovery.
According to a brief by the University of Pennsylvania, doubling the number of vaccine doses administered daily to 3 million would create more than 2 million jobs and boost real GDP by about 1% over the summer of 2021. Business viability requires a healthy workforce. 76
Economic frame bottom line The bottom line: By getting vaccinated, you can help to eradicate this pandemic, ending damage to the economy. 73
To understand the efficacy of our treatment, we compared the distributions of vaccination intention before and after treatments by examining how the distribution of vaccination intention responses shifted. Specifically, we used the Cramér-von Mises (CVM) (Cramér, 1928) two-sample test to examine whether the post-treatment vaccination intention distribution differed significantly from that of the pre-treatment. We also examined whether our messages had different effects on different partisan subgroups (that is, unvaccinated Democrats given the health frame, unvaccinated Democrats given the economic frame, unvaccinated Republicans given the health frame, and unvaccinated Republicans given the economic frame). In order to examine the efficacy of our messages on the most hesitant partisans in our sample, we identified those individuals identifying as Republicans and Democrats whose pre-treatment COVID-19 vaccination intentions were “extremely unlikely,” – the lowest possible rating. We refer to these subjects as the “most resistant” partisans throughout the rest of the paper. We conducted ANOVAs to test whether differences in framing changed intentions to vaccinate. Finally, we tested whether personal financial risk attitude (perceived general risks/threats to the financial conditions of individuals) mediated the relationship between economic issue framing and the likelihood of vaccination against COVID-19, using causal mediation analysis with bootstrap (Preacher & Hayes, 2004; Tingley, 2014). Specifically, we first fitted the mediator model where the mediator, perceived personal financial risks, was a function of the frame treatment. Next, we modeled the post-treatment vaccination intention on the mediator and the treatment. Taking these two models as inputs, we tested the significance of our mediator, perceived personal financial risks, in the relationship between having received an economic frame message, and the COVID-19 vaccination intention.
3 Results
3.1 Study 1
3.1.1 Sample characteristics
We collected 181,407 tweets that were posted between January 1, 2020, and September 30, 2021.123,436 (68%) tweets were from Democratic legislators, and 57,971 (32%) tweets were from Republican legislators.
3.1.2 Democratic and Republican politicians emphasized health care and the economy, respectively
Members of Congress exhibited political polarization in their communications about the pandemic and vaccination, with Democratic members emphasizing public health, health behavior, and direct aid to workers (e.g., the phrase “health care” was used 4226 times more by Democratic Congress members than Republican Congress members, Fig. 1 ). The words most frequently used by Democrats concerned public health and health behavior (e.g., “health care,” “wear the mask,” “covid 19 vaccine,” “keep safe,” “save a life”). In contrast, the words most frequently used by Republicans concerned the Trump administration's successful push for the development of a COVID-19 vaccine and the economic impact of COVID-19, including financial assistance programs and reopening businesses (e.g., “warp speed,” “operation warp,” “back work”, “economic impact”). In terms of topics, Republican politicians mentioned relief bills, economic impact, small business assistance, and operation warp speed more often. In contrast, Democratic elites focused on health crises, health care, COVID-19 vaccination, and community support.Fig. 1 Comparison of bigrams and topics posted on Twitter between Democratic and Republican members of the 116th Congress: January 2020–September 2021.
Fig. 1
Focusing on vaccine-related tweets, we found that Republican politicians were less likely to retweet CDC's vaccination recommendations (44% of retweets). In contrast, 56% of Democrats' retweets were CDC's official tweets, including those about the ongoing importance of COVID-19 vaccines, vaccine distribution and access, and expressing concerns about vaccine hesitancy. (A notable exception occurred in March 2020 when former President Donald Trump declared the novel coronavirus a national emergency. During this month, Republicans elites retweeted CDC posts 86 times more frequently than Democratic politicians.) Proportionally, Republican politicians were about 7.5% more actively engaged in discussion about vaccine development, vaccine breakthroughs, vaccine completion, and vaccine effectiveness, whereas Democratic politicians were 4.5% more focused on vaccine safety, eligibility, vaccination sites and appointments, and getting vaccinated.
Taken together, we found that Democratic elites’ narratives were more concerned with pandemic health risks and vaccine promotion, whereas Republicans were less engaged in discussion about vaccination, instead focusing on COVID-19 testing and the development of COVID-19 vaccines by a Republican administration. More broadly, Republican politicians discussed economic impacts, financial assistance programs, and reopening businesses.
3.2 Study 2
3.2.1 Sample characteristics
A total of 3751 individuals participated in the survey, among whom 2666 (71%) had received at least one dose of a COVID-19 vaccine. More than half of fully or partially vaccinated participants were self-identified Democrats (1614, 61.9%), 21.4% (558) were self-identified Republicans, and 16.7% (434) were self-identified political independents. Detailed sample characteristics can be found in the Supplementary Material.
3.2.2 Republicans and Democrats were motivated by the economy and public health, respectively
Linear regression results showed that vaccinated Republicans were significantly more likely to be motivated by economic (β = 0.25, p = 0.02) and personal financial (β = 0.24, p = 0.03) rationales, and less likely to agree with public health or personal sickness rationales relative to Democrats (see Fig. 2 ). In contrast, compared to Republicans, Democrats were more likely to be motivated by collective public health rationales, such as preventing illness (β = 0.68, p < 0.001) and death, and personal health risks, such as protecting themselves (β = 0.42, p < 0.001) (Fig. 2; see Supplementary Material).Fig. 2 Comparison of vaccination motivations between vaccinated partisan individuals: September–November 2021 (The full regression results can be found in the Supplement Material.).
Fig. 2
A permutation test between Democrats and Republicans shows that Republicans' average self-rated motivations for economic and personal finance considerations were significantly higher than those of Democrats. The differences between Republicans' and Democrats' average economic and personal financial motivation responses were 0.31 points (η2 = 0.012, p = 0.002) and 0.25 points (η2 = 0.013, p = 0.02), respectively. In contrast, Republicans' average motivations for collective health and worry about getting sick were significantly lower than Democrats'. The differences between Republicans' and Democrats’ average responses concerning public health and personal health were −0.68 points (η2 = 0.022, p = 0.002) and −0.42 points (η2 = 0.011, p = 0.002), respectively. For all these comparisons, we conducted post-hoc power analyses assuming a one-tailed test at a 0.01 significance level. Results showed that all the tests achieved a statistical power of above 99%. Detailed results of power analyses can be found in the Supplementary Materials.
Beyond partisan differences, we also observed that other demographic subgroups were more likely to be motivated to vaccinate by economic and financial risk perceptions. Specifically, younger adults (β = −0.17, p < 0.001), more educated individuals (β = 0.2, p < 0.001), African Americans (β = 0.49, p = 0.002), and individuals of Hispanic ethnicity (β = 0.79, p < 0.001) were all more likely to be motivated to vaccinate by economic and financial loss considerations (see Fig. 2). Importantly, most of these groups are under-vaccinated for COVID-19 (Kriss, 2022; Monte, 2021). Moreover, by regressing the motivation responses on vaccination status (partially vaccinated - those who got one dose of two-doses vaccine, and fully vaccinated), we found that regardless of partisanship, economic (β = 0.44, p < 0.001) and personal financial (β = 0.62, p < 0.001) rationales for vaccination were more prevalent among partially vaccinated individuals, whereas public (β = −0.64, p < 0.001) and personal (β = −0.81, p < 0.001) health rationales were more common among fully vaccinated respondents. Finally, we found that there was a strong correlation between personal financial considerations and economic considerations that motivated people to get vaccinated regardless of partisanship (Republicans: β = 0.60, p < 0.001; Democrats: β = 0.69, p < 0.001; Independents: β = 0.55, p < 0.001), suggesting that people who were concerned about macroeconomic conditions were more likely to consider personal financial risks when getting vaccinated.
3.3 Study 3
3.3.1 Sample characteristics
A total of 2017 individuals completed our initial screening survey of whom 400 (20%) self-identified as unvaccinated and were invited to the follow-up survey. 357 (89%) of these participants completed the follow-up, of whom 299 (84%) correctly answered all attention check questions. Roughly half (139, 46.5%) of these participants self-identified as Republican, 61 (20.4%) were Democrats, and the others (99, 33.1%) self-identified as political independents. Further sample characteristics can be found in the Supplementary Material.
3.3.2 Economic framing significantly increased intent to vaccinate among the most resistant Republicans
A balance test verified that groups did not differ significantly in terms of demographic factors, including age, gender, race, education, Hispanic ethnicity, and political ideology. Details of the results can be found in the supplementary materials. We found that both economic (p < 0.001) and health (p < 0.001) frames significantly increased respondents’ reported intentions to vaccinate after having seen the messages. However, we did not observe an effect of including a bottom-line summary sentence (potentially indicating that subjects understood the messages without this summary); therefore, we collapsed across this bottom-line condition on all subsequent analyses. Specifically, we conducted a one-way ANOVA comparing post-treatment vaccination intention between frames, among the most vaccine-resistant partisans (defined as a group of unvaccinated participants whose pre-treatment COVID-19 vaccination intention choice was “extremely unlikely.“).
The most resistant Republicans were more likely to vaccinate after having been exposed to economic (δ= 2.6; p < 0.001) and health (δ= 1.9; p = 0.002) messages (Fig. 3 ). ANOVA results show a significant difference between frames, indicating that the most resistant Republicans were significantly more likely to intend to get a COVID-19 vaccine when shown the economic frame compared to the health frame (F1,119=4.16; η2=0.034, p = 0.04).Fig. 3 Comparison of vaccine intention before and after treatment across partisan groups: November 2021. After the economic frame, the 90% percentile of Republicans' vaccination intention also increased from 1.4 to 4, compared to a smaller shift under the health frame (from 2 to 3.9). The 90% percentile of Democrats' willingness to vaccinate increased from 4 to 5 after the health frame treatment, whereas it barely changed (from 4 to 4.1) after the economic treatment.
Fig. 3
For Democrats, the most resistant of respondents significantly increased their intent to vaccinate when shown the health frame (δ= 1.0; p = 0.007), but we did not observe a statistically significant change for the economic frame (δ= 0.1; p = 0.08). ANOVA results showed that the health frame significantly increased the most resistant Democrats’ intent to get vaccinated relative to the economic frame (F1,24=4.49; η2=0.163, p = 0.045; see the Supplemental Materials).
Upon conducting post-hoc power analyses, assuming a one-tailed test at a 0.01 significance level, we found that all tests achieved a statistical power of above 85%. Detailed results of power analyses can be found in the Supplementary Materials.
3.3.3 Economic framing drives willingness to vaccinate by increasing personal financial risk perception
We next examined the mechanism underlying why economic framing increased vaccination intentions. We found that personal financial risk perceptions mediated the relationship between economic framing and vaccination intentions. Specifically, we found that a one-point increase in perceived personal financial risk significantly increased post-treatment vaccine intent by 14% (p = 0.002). Although one might think that economic and personal financial risk questions indexed different risk perceptions a factor analysis shows that participants’ perceived economic risks and personal financial risks load strongly on the same factor (see the Supplementary Materials).
Furthermore, we found that exposure to the economic frame increased participants' perceived personal financial risk by 0.41 points (p = 0.048), compared to a health frame. Finally, results of a causal mediation analysis with bootstrapping showed that perceived personal financial risk significantly mediated between the economic frame and intention to vaccinate (β = 0.06, p = 0.02; Fig. 4 ). To summarize, the economic frame could impact people's intention to vaccinate by increasing their perceptions of how COVID-19 exposes them to financial risk.Fig. 4 Causal mediation diagram between economic frame, perceived financial risks, and vaccination intention: November 2021.
Fig. 4
4 Discussion
This is the first study, to our knowledge, to provide evidence suggesting that messages using an economic frame may increase vaccine uptake among vaccine-hesitant Republicans. These results are consistent with our retrospective analysis of Democratic and Republican Twitter posts during the pandemic, which shows that politicians’ rhetoric about the pandemic and COVID-19 vaccination utilized very distinct frames. Specifically, we found that Republicans were more likely to discuss the impacts of the pandemic in economic terms, whereas Democrats were more likely to use health frames in their public tweets. Consequently, these diverging narratives highlight a very different “subset of relevant considerations” that could influence their audiences regarding whether or not to get vaccinated.
Subjects’ motivations for getting vaccinated were consistent with this political rhetoric. Among individuals who chose to vaccinate, economic rationales (e.g., getting the economy moving again, going back to work/reducing personal financial loss) for vaccination were more prevalent among Republicans, whereas public health rationales (e.g., preventing more illness and death in America, and worrying about getting sick from the virus) were more common among Democrats.
Finally, we found that messages designed to be concordant with how politicians frame the discourse around COVID-19 can increase partisans' vaccination intentions. Specifically, framing the decision to vaccinate in economic terms increased expressed intentions to vaccinate among the most resistant Republicans. These individuals appear to be concerned about the pandemic's impact on the economy and their personal financial situation. This effect was asymmetric – whereas messages using an economic frame were more effective when convincing the most vaccine-hesitant Republicans to vaccinate, those using a public health frame appear were more effective for convincing the most vaccine-hesitant Democrats.
Our results indicate that economic framing may be effective for several under-vaccinated groups, not just Republicans. These other groups – including young people, racial and ethnic minorities, and highly-educated individuals – may be more motivated by economic concerns because of structural differences. For example, individuals of lower socioeconomic status (which overlaps with age, race, and ethnicity) may suffer greater consequences for missing work, making them more sensitive to economic concerns, and thus less willing to take personal financial risks when compared to health risks. Similarly, individuals with more education may feel that they can take measures to protect their own health effectively, but may feel less control over the economy and its impact on their personal financial situation. Thus, messages targeting Republicans may also increase intentions to vaccinate for several other at-risk groups. Results from this study not only contribute to the global push for COVID-19 vaccination but may also inform future campaigns promoting vaccination and other preventive health measures to the public.
Our study is not without limitations. Specifically, we recruited subjects using MTurk, which may not be fully representative of the US population, and may be especially susceptible to economic rationales. However, this susceptibility cannot explain observed differences between Republicans and Democrats, both of whom were recruited using MTurk. Second, the effects of economic considerations on motivations to vaccinate were relatively small in the observational study (effect size η2 ranges from 0.011 to 0.022), which may suggest limited practical application; however, these small effect sizes could also be due to the presence of several confounding factors that cannot be ruled out using a correlational design. In contrast, our randomized controlled experiment showed a larger effect (effect size η2 = 0.34 for the most hesitant Republicans, N = 120). Given the difficulty in promoting vaccination among the most hesitant groups, the dynamics of exponential growth underlying disease spread mean that strategies with even a small effect could mean the difference between achieving herd immunity and an uncontrolled pandemic. Thus, we should not equate small effect sizes with a lack of importance. Future work may explore the efficacy of economic framing on a more representative and larger sample, including more tailored interventions.
5 Public Health Implications
Health communicators often frame messages about COVID-19 in terms of health outcomes. However, vaccination rates differ sharply along partisan lines. Framing COVID-19 only in terms of health outcomes does not engage with one of the primary ways in which the pandemic has been framed by Republican political elites. During the 2020 presidential-election campaign, Republicans believed that the economy was the most important issue facing the nation – a position that is consistent with a reluctance to curtail their movements during the pandemic (Camobreco and He, 2022). Our results suggest that messages promoting vaccination in economic terms can encourage vaccine uptake, especially among the most resistant Republicans.
In addition to standard techniques, such as surveys and focus groups, our study also emphasizes the importance of social listening using tools such as Twitter to understand the broader discourse in which COVID-19 communication is embedded, and then testing insights from those tools using experimental techniques to derive causal conclusions. Solid data, grounded in a causal mechanism, can inform direct input into health communication campaigns that are targeted and tailored to the specific concerns of vaccine-resistant communities. Policymakers should therefore invest in developing in-house expertise for using social media platforms such as Twitter to a) listen to the concerns of vaccine-hesitant populations, b) translate those concerns into messages that are grounded in empirically-validated theoretical constructs, and c) rapidly test these messages using randomized controlled experiments. These interventions need not wait until new vaccines are developed: economic framing of COVID-19 was widespread since the pandemic began. Thus, the approach suggested here may be used to develop messages that are responsive to the concerns elicited in anticipation of novel therapeutic interventions.
Credit author statement
Wei Zhong: Conceptualization, Methodology, Validation, Formal analysis, Data curation, Writing- original draft, Writing -review & editing, Visualization, David Broniatowski: Conceptualization, Investigation, Writing- Reviewing and Editing, Funding acquisition, Data Curation.
Appendix ASupplementary data
The following is the Supplementary data to this article:Multimedia component 1
Multimedia component 1
Data availability
Research data can be found in the attached files.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.socscimed.2022.115594.
==== Refs
References
Ajzen From Intentions to Actions: A Theory of Planned Behavior. Action Control 1985 Springer 11 39
Ajzen The theory of planned behavior Organ. Behav. Hum. Decis. Process. 50 1991 179 211
Ashworth Emphasize Personal Health Benefits to Boost COVID-19 Vaccination Rates vol. 118 2021 Proc Natl Acad Sci U S A
Blei Latent dirichlet allocation J. Mach. Learn. Res. 3 2003
Borah Seeking more information and conversations: influence of competitive frames and motivated processing Commun. Res. 38 2011 303 325
Borah COVID-19 vaccination attitudes and intention: message framing and the moderating role of perceived vaccine benefits J. Health Commun. 26 2021 523 533 34424140
Borah Message framing and COVID-19 vaccination intention: moderating roles of partisan media use and pre-attitudes about vaccination Curr. Psychol. 2022
Camobreco He The party-line pandemic: a closer look at the partisan response to COVID-19 PS Political Sci. Polit. 55 2022 13 21
CDC COVID-19 Vaccinations in the United States 2022 https://covid.cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total accessed
CDC United States COVID-19 Cases, Deaths, and Laboratory Testing (NAATs) 2022 https://covid.cdc.gov/covid-data-tracker/#trends_dailydeaths accessed
Chen The economic burden of COVID-19 in the United States: estimates and projections under an infection-based herd immunity approach The Journal of the Economics of Ageing 20 2021 100328
Chong & Druckman A theory of framing and opinion formation in competitive elite environments J. Commun. 57 2007 99 118
Cohen Statistical Power Analysis for the Behavioral Sciences Lawrence Earlbaum Associates. 20th– 1988 Lawrence Earlbaum Associates
Cramér On the composition of elementary errors: first paper: mathematical deductions Scand. Actuar. J. 1928 13 74 1928
de Bruin Bruine Political polarization in US residents' COVID-19 risk perceptions, policy preferences, and protective behaviors J. Risk Uncertain. 2020 1 18
Duquette Heard” immunity: messages emphasizing the safety of others increase intended uptake of a COVID-19 vaccine in some groups 1 Covid Economics 39 2020
Feng & Shao Understanding the influence of political orientation, social network, and economic recovery on COVID-19 vaccine uptake among Americans Vaccine 40 2022 2191 2201 35227522
Fridman COVID-19 and vaccine hesitancy: a longitudinal study PLoS One 16 2021 e0250123
Guenther Framing as a concept for health communication: a systematic review Health Commun. 36 2021 891 899 31996044
Hallsworth Four Messages that Can Increase Uptake of the COVID-19 Vaccines: Using Large-Scale Testing to Identify Effective Vaccine Messaging 2021 The Behavioural Insights Team 2021
Hart Politicization and polarization in COVID-19 news coverage Sci. Commun. 42 2020 679 697
Hornsey Donald Trump vaccination The effect of political identity, conspiracist ideation and presidential tweets on vaccine hesitancy J. Exp. Soc. Psychol. 88 2020 103947
Huang Liu Promoting COVID-19 vaccination: the interplay of message framing, psychological uncertainty, and public agency as a message source Sci. Commun. 44 2022 3 29
Jing & Ahn Characterizing partisan political narrative frameworks about COVID-19 on Twitter EPJ Data Sci 10 2021 53 34745825
Jordan Don't get it or don't spread it: comparing self-interested versus prosocial motivations for COVID-19 prevention behaviors Sci. Rep. 11 2021
KFF KFF COVID-19 Vaccine Monitor: September 2022 2022 https://www.kff.org/coronavirus-covid-19/poll-finding/kff-covid-19-vaccine-monitor-september-2022/ accessed
KFF KFF COVID-19 Vaccine Monitor: Who Remains Unvaccinated? Unvaccinated Adults Are Younger, Less Educated, and More Republican 2022 https://www.kff.org/coronavirus-covid-19/dashboard/kff-covid-19-vaccine-monitor-dashboard/ accessed
Khubchandani Masks, gloves, and the COVID-19 pandemic: rapid assessment of public behaviors in the United States Epidemiologia 1 2020 16 22 36417208
Khubchandani COVID-19 vaccination hesitancy in the United States: a rapid national assessment J. Community Health 46 2021 270 277 33389421
Kreps The Relationship between US Adults' Misconceptions about COVID-19 Vaccines and Vaccination Preferences vol. 9 2021 Vaccines Basel)
Kriss COVID-19 vaccination coverage, by race and ethnicity — national immunization survey adult COVID module, United States, december 2020–november 2021 MMWR. Morbidity and Mortality Weekly Report 71 2022 757 763 35679179
Kunda The case for motivated reasoning Psychol. Bull. 108 1990 480 498 2270237
Monte Household Pulse Survey Shows Many Don't Trust COVID Vaccine, Worry about Side Effects 2021 https://www.census.gov/library/stories/2021/12/who-are-the-adults-not-vaccinated-against-covid.html accessed
Motta Encouraging COVID-19 vaccine uptake through effective health communication Frontiers in Political Science 3 2021 1
Palm The effect of frames on COVID-19 vaccine resistance Frontiers in Political Science 3 2021
Panda COVID, BLM, and the Polarization of US Politicians on Twitter 2020 (arXiv: Social and Information Networks
Pink Elite party cues increase vaccination intentions among Republicans Proc. Natl. Acad. Sci. USA 118 2021 e2106559118
Preacher & Hayes SPSS and SAS procedures for estimating indirect effects in simple mediation models Behav. Res. Methods Instrum. Comput. 36 2004 717 731 15641418
Reinhardt & Rossmann Age-related framing effects: why vaccination against COVID-19 should be promoted differently in younger and older adults J. Exp. Psychol. Appl. 2021
Ruiz & Bell Predictors of intention to vaccinate against COVID-19: results of a nationwide survey Vaccine 39 2021 1080 1086 33461833
Sylvester Vaccinating across the aisle: using co-partisan source cues to encourage COVID-19 vaccine uptake in the ideological right J. Behav. Med. 2022
Tabachnick Fidell Using Multivariate Statistics 2007 Pearson Boston, MA
Telenti After the pandemic: perspectives on the future trajectory of COVID-19 Nature 596 2021 495 504 34237771
Thompson Foundations of Behavioral Statistics: an Insight-Based Approach 2006 Guilford Publications New York, NY, US
Tingley Mediation: R package for causal mediation analysis J. Stat. Software 59 2014 1 38
Van der Linden Highlighting consensus among medical scientists increases public support for vaccines: evidence from a randomized experiment BMC Publ. Health 15 2015 1207
Wrubel & Kerchner 116th U.S. Congress tweet ids Harvard Dataverse, V1 2020 10.7910/DVN/MBOJNS
Yousaf Immunity debt or vaccination crisis? A multi-method evidence on vaccine acceptance and media framing for emerging COVID-19 variants Vaccine 40 2022 1855 1863 35153094
| 36508989 | PMC9726654 | NO-CC CODE | 2022-12-13 23:16:27 | no | Soc Sci Med. 2023 Jan 7; 317:115594 | utf-8 | Soc Sci Med | 2,022 | 10.1016/j.socscimed.2022.115594 | oa_other |
==== Front
J Affect Disord Rep
J Affect Disord Rep
Journal of Affective Disorders Reports
2666-9153
The Authors. Published by Elsevier B.V.
S2666-9153(22)00152-4
10.1016/j.jadr.2022.100460
100460
Research Paper
First access to mental health services during COVID-19 pandemic: A multicenter study
Petri Eleonora a⁎
Nardoni Cristina b
Fui Erika a
Gulino Elisa c
Abdelghani Lachheb b
Barone Raffaele c
Miragoli Paolo Angelo Fulvio a
Cardamone Giuseppe b
Ciberti Agnese b
a Department of Mental Health, San Carlo Hospital, ASST Santi Paolo e Carlo, Milan, Italy
b Department of Mental Health of Prato, Azienda USL Toscana Centro, Prato, Italy
c Department of Mental Health of Caltagirone-Palagonia, ASP 3 Catania, Caltagirone-Palagonia, Italy
⁎ Corresponding author at: Department of Mental Health, San Carlo Hospital, ASST Santi Paolo e Carlo, via Pio II 3, 20153, Milan, Italy
7 12 2022
1 2023
7 12 2022
11 100460100460
1 8 2022
8 11 2022
6 12 2022
© 2022 The Authors. Published by Elsevier B.V.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background: The impact of the COVID-19 pandemic on mental health is complex and affects a broad segment of the population. Several studies indicate that depressive, anxious and post-traumatic symptoms are common in people exposed to SARS-Cov2.
Methods: 458 subjects were recruited during their first consultation in outpatient psychiatric services between June 2020 and October 2021. Post-traumatic, depressive and anxious symptoms were assessed using the Impact of Event Scale-Revised (IES-R), the Beck Depression Inventory Scale-second edition (BDI-II), and the Self Rating Anxiety Scale (SAS). A specific set of questions was developed, with the aim of evaluating socio-demographic variables and work, environmental and personal characteristics related to the pandemic.
Results: Prevalence rates of clinically significant depressive, anxious and post-traumatic symptoms were 57.6%, 63.5% and 54.8%, respectively. Female gender, worsening of relationship status and financial consequences due to the pandemic were the conditions most strongly associated with the presence of psychopathology.
Limitation: The cross-sectional design of the study doesn't allow an evaluation over time of the sample. No assumption of causality can be made due to the lack of pre-pandemic assessments for the investigated variables.
Conclusions: The impact of the pandemic involves depressive, anxious and post-traumatic dimensions. The investigated psychopathology correlates with several variables expressing the personal and environmental changes that occurred in the population due to the COVID-19 emergency. The study is multicentric and the recruitment of participants was held in a clinical setting, providing a realistic picture of the consequences of the pandemic in clinical practice within mental health services.
Keywords
Depression
Anxiety
post-traumatic
COVID-19
Pandemic
Mental health services
==== Body
pmc1 Introduction
A national lockdown represented the Italian Government's response to the rapid spread of COVID-19 cases in Italy since February 2020. Following the first peak in cases, further “waves” of the pandemic hit Italy and, as a consequence, non-therapeutic public health interventions such as quarantine and restriction in social and community movements have been cyclically reintroduced. During the COVID-19 outbreak many people became unemployed and experienced financial strain, which led to higher rates of mental instability in the population (Sultana et al., 2021). Given the persistence of the health emergency, a growing literature examined the impact of the pandemic on mental health, both short- and long-term. Data confirmed that negative effects on mental health may be outlasting the pandemic itself (Manchia et al, 2022). Anxiety and depressive symptoms were the most common mental health issues reported in the studies and meta-analyses (Bueno-Notivol et al., 2021; Robinson et al., 2022; Morganstein et al., 2020; Salari et al., 2020). During lockdown, patients with general anxiety disorders and OCD were more likely to access consultations in the emergency department (Capuzzi E et al., 2020; Ramadan M et al., 2022). The comparison of COVID-19 pandemic with natural disasters, such as earthquakes or tsunamis (Morganstein et al., 2020), led to the hypothesis of considering the spread of COVID-19 as a novel form of traumatic experience (Fiorillo and Gorwood, 2020). Thus, some studies evaluated post-traumatic psychopathology, finding substantial PTSD prevalence rates among the general population compared to the average global prevalence pre-COVID-19 (Yunitri et al., 2022) and significant associations with several pandemic-related conditions (Castellini et al., 2021).
Within mental health centers, the majority of psychiatric consultations were required by patients already treated and cared for by the outpatient services (Di Lorenzo et al., 2021), but also among the general population many studies outlined mental health issues arising de novo during the pandemic (Cullen et al., 2020). Studies in the general population often used online surveys, which lead to several selection biases of the sample; while research conducted in outpatient psychiatric settings focused on participants with a previous diagnosis of mental illness (Fleischmann et al., 2021). People who accessed mental health services for a first specialistic consultation represent a specific group of patients. In literature, there is a lack of data regarding the characteristics of this particular population during COVID-19 pandemic.
To our knowledge, this is the first study conducted in psychiatric public services that investigated the characteristics of psychological needs of patients who accessed the mental health services for a first specialistic evaluation. Our study took place after the first wave of COVID-19 outbreak and has a multicenter design.
The aim of the study was to evaluate the prevalence of depressive, anxious and post-traumatic symptoms in a population requiring a first psychiatric or psychological consultation. As a second goal, we investigated the correlation between psychopathological symptoms and a set of social, economic and clinic variables related to COVID-19 pandemic.
2 Methods
2.1 Procedures and study participants
The study is multicenter and cross-sectional, involving three public mental health services located in the north, center and south of Italy. The study was conducted at the Department of Mental Health of San Carlo Hospital in Milan (MI, Lombardy, Italy), the Department of Mental Health of Prato (PO, Tuscany, Italy), and of Caltagirone-Palagonia (CT, Sicily, Italy). The present study was carried out in a real-life clinical setting. Patients who accessed the mental health outpatient services for a first psychiatric or psychological evaluation were recruited for the study and given self-report questionnaires. Enrollment was held between June 2020 and October 2021. Exclusion criteria included poor knowledge of the Italian language or other limits to verbal communication, being cognitively impaired or underage. The recruitment was contextual to the clinical evaluation made by the psychiatrist or psychologist during the consultation. Suitable subjects were asked to provide written informed consent after receiving a complete description of the study, having the opportunity to ask questions. All the recruited subjects were included in the study as they met the inclusion criteria on the basis of a clinical evaluation. No data was collected on subjects who were identified as non-suitable by clinicians.
The study was conducted in accordance with the Declaration of Helsinki and approved by the local Ethics Committees (Milano Area 1, Protocol number 2021/ST/105; Regione Toscana, Azienda Unità Sanitaria Locale Toscana Centro-Protocol number 2251).
2.2 Measures
Data on socio-demographic variables were collected from each participant, including age, gender, marital status, occupation, education level. A specific set of questions investigated information regarding variables related to COVID-19 emergency, including work-related, environmental, personal and clinical features (Table 1 ). Recruited subjects were assessed using self-report questionnaires: post-traumatic stress, depressive as well as anxious symptoms were measured using the Impact of Event Scale-Revised (IES-R), the Beck Depression Inventory Scale-second edition (BDI-II), and the Self Rating Anxiety Scale (SAS), respectively.Table 1 Socio-demographic, work-related, environmental, personal, and clinical COVID-related characteristics in the overall sample (N=458)
Table 1- Variable Total Sample N(%)
Gender Male 205 (44.8%)
Female 253 (55.2%)
Nationality Italian 421 (92.1%)
Not italian 36 (7.9%)
Educational level Elementary school 34 (7.4%)
Secondary school 170 (37.1%)
High school 197 (43%)
College 57 (12.4%)
Occupation Freelance 33 (7.2%)
Employed or retired 253 (55.2%)
Unemployed 172 (37.6%)
Marital status Single or widower 186 (40.6%)
Engaged non-cohabiting 44 (9.6%)
Married or cohabiting 228 (49.8%)
Spent quarantine with Alone 71 (15.5%)
With family members 379 (82.7%)
With roommates 5 (1.1%)
Previous psychiatric visits Yes 192 (41.9%)
No 266 (58.1%)
Change in work characteristics due to COVID-19 Yes 209 (45.6%)
No 249 (54.4%)
Loss of income due to COVID-19 Yes 129 (28.2%)
No 329 (71.8%)
Change of workplace due to COVID-19 Yes 147 (32.1%)
No 311 (67.9%)
Reduction of working hours due to COVID-19 Yes 114 (24.9%)
No 344 (75.1%)
Losing employment due to COVID-19 Yes 48 (10.4%)
No 410 (89.5%)
Worsening of financial situation due to COVID-19 Yes 170 (37.1%)
No 288 (62.8%)
Worsening of relationship status due to COVID-19 Yes 106 (23.1%)
No 352 (76.9%)
Having a house with open spaces Yes 329 (71.8%)
No 129 (28.2%)
Family quarrels due to COVID-19 Yes 113 (24.7%)
No 345 (75.3%)
Undergone testing for COVID-19 Yes 154 (33.6%)
No 304 (66.4%)
Got sick with COVID-19 Yes 28 (6.1%)
No 430 (93.9%)
Hospitalized for COVID-19 Yes 12 (2.6%)
No 446 (97.4%)
Acquaintances got sick with COVID-19 Yes 175 (38.2%)
No 283 (61.8%)
Acquaintances died with COVID-19 Yes 88 (19.2%)
No 370 (80.8%)
Loved ones got sick with COVID-19 Yes 104 (22.7%)
No 354 (77.3%)
The Impact of Event Scale-Revised (IES-R) (Weiss et al., 1997; Craparo et al., 2013) is a 22-item scale measuring three core symptomatological characteristics of PTSD: intrusion, avoidance and hyperarousal. The questionnaire has good internal consistency (Cronbach's α for each subscale: intrusion = 0.87 to 0.94, avoidance = 0.84 to 0.97, hyperarousal = .79 to .91), and high test-retest reliability (r = 0.93). Each item is rated on a five-point likert-like scale (0-4). Total score range from 0 to 88. A score over 33 represents a cutoff for the presence of clinically significant post-traumatic stress symptoms. According to the aim of the study, the items referred to subjective traumatic experiences of lockdown and COVID-19 emergency.
The Beck Depression Inventory Scale-second edition (BDI-II) (Beck et al., 1996) is a 21-item scale, used to measure the cognitive, motivational, affective, and somatic symptoms of depression. Each item is rated on a four-point likert-like scale, ranging from 0 to 3. A total score over 14 indicates at least a mild condition, while higher scores indicate more severe symptoms. Both the original (internal consistency: α = 0.92 in a community sample) and the Italian version (internal consistency: α = 0.87 in community patients) demonstrated excellent psychometric properties (Beck et al., 1996; Sica and Ghisi, 2007).
The Self-Rating Anxiety Scale (SAS) (Zung, 1971; Conti, 2000) is a 20 item scale. The raw scores range from 20 to 80. SAS items measure both affective and somatic symptoms. The raw score of 40 is considered the cut-off for clinically significant anxiety (Zung, 1980; Dunstan and Scott, 2020). The SAS has been shown to have good internal consistency with a Cronbach's alpha of 0.82 (Tanaka-Matsumi et al., 1986).
2.3 Statistical analysis
All statistical analyses were performed using the STATA software (version 13.0). Descriptive analyses were carried out in order to evaluate the distribution of socio-demographic, work-related, environmental, personal and clinical characteristics. Chi-square test was used for comparison between groups for categorical variables and Student's t-test for continuous variables. Multivariate logistic regression analyses were implemented in order to evaluate the correlation between psychopathology (DV) expressed as binary variables (presence or non-presence of clinically significant depressive, anxious and post-traumatic symptoms) and the socio-demographic, work-related, environmental, personal and clinical variables (IV) resulted as statistically significant in the descriptive analysis (chi-square, t-test). Odds ratio (OR) with 95% confidence intervals were used for the observed associations.
3 Results
The study included a total sample of 458 subjects, of which 233 (50.8%) enrolled in Prato, 121 (26.4%) enrolled in Milan, and 104 (22.7%) enrolled in Caltagirone. All subjects were asked if they ever had access to psychiatric consultations prior to that visit: approximately 58% of the subjects (266) indicated that they never had psychiatric visits before. We correlated this finding with the presence of depressive, anxious and post-traumatic symptoms and found no significant associations (p=0.41; p=0.70; p=0.40, respectively). Distribution of socio-demographic variables and work-related, environmental, personal, and clinical characteristics related to COVID-19 pandemic in the sample is shown in Table 1.
3.1 Depressive symptoms
The prevalence of significant depressive symptoms (DEP), expressed by the BDI-II score ≥ 15, in the overall sample was 57.6%. The mean age of the depression group (DEP) was 45.2 (±15.7), the mean age of the non-depression group (nDEP) was 46.6 (±16.5). Being female (62.9% vs 55.2%, p=0.01, X²=6.27) and being single or in a non-cohabiting relationship (60.8% and 70.5% vs 40.6% and 9.6%, p=0.05, X²=6.04) were more frequently associated with depressive symptoms. Participants reporting change in work characteristics (64.1% vs 45.6%, p=0.01, X²=6.60), loss of income (66% vs 28.2%, p=0.01, X²=5.99), financial strain (70.6% vs 37.1%, p=0.00, X²=18.55), family quarrels (69.9% vs 24.7%, p=0.00, X²=9.25) and a worsened relationship status (80.2% vs 23.1%, p=0.00, X²=28.71) due to COVID-19 were more likely to experience significant depressive symptoms. After logistic regression analysis (see Table 2 ) the variables most strongly associated with the presence of significant depressive symptoms were having a worsened financial situation (p=0.005; OR 2.20 [1.27-3.80]), having a worsened relationship status (p=0.000; OR 3.22 [1.85-5.57] and being female (p=0.049; OR 1.48 [1.00-2.20].Table 2 Logistic regression analyses of socio-demographic factors and COVID-related variables on the presence of depressive symptoms (BDI-II score ≥ 15)
Table 2Variables in equations OR p 95%CI
Female gender 1.48 0.049 1.00-2.19
Change in work characteristics due to COVID-19 1.13 0.588 0.72-1.76
Loss of income due to COVID-19 0.81 0.501 0.45-1.47
Worsening of financial situation due to COVID-19 2.2 0.005 1.27-3.80
Worsening of relationship status due to COVID-19 3.2 0.000 1.86-5.58
Family quarrels due to COVID-19 1.28 0.328 0.78-2.10
3.2 Anxiety symptoms
The prevalence of significant anxious symptoms (ANX), expressed by the SAS score ≥ 40, in the overall sample was 63.5%. The mean age was similar in the two groups resulting in 45.8 (±16.1) in the anxiety group (ANX) and 45.7 (±16) in the non-anxiety group (nANX). Female gender was more frequently associated with significant anxiety symptoms (73.1% vs 55.2%, p=0.00, X²=22.41). The anxiety group (ANX) more frequently reported changes in work characteristics (70.3% vs 57.8%, p=0.006, X²=7.66), financial strain (70% vs 59.7%, p=0.027, X²=4.87), a worsened relationship status (81.1% vs 58.2%, p=0.00, X²=18.43) and family quarrels (74.3% vs 60%, p=0.006, X²=7.55) due to COVID-19. Subjects with anxiety were more likely to have undergone tests for COVID-19 (70.1% vs 60.2%, p=0.037, X²=4.35). After logistic regression analysis (see Table 3 ) the variables most strongly associated with the presence of significant anxious symptoms were female gender (p=0.000; OR 2.46 [1.64-3.68] and having a worsened relationship status (p=0.002; OR 2.42 [1.38-4.26].Table 3 Logistic regression analyses of socio-demographic factors and COVID-related variables on the presence of anxious symptoms (SAS score ≥ 40)
Table 3Variables in equations OR p 95%CI
Female gender 2.46 0.000 1.64-3.68
Change in work characteristics due to COVID-19 1.41 0.127 0.90-2.21
Undergone testing for COVID-19 1.37 0.166 0.88-2.12
Worsening of financial situation due to COVID-19 1.16 0.533 0.73-1.84
Worsening of relationship status due to COVID-19 2.4 0.002 1.38-4.26
Family quarrels due to COVID-19 1.35 0.250 0.80-2.26
3.3 Post-traumatic stress symptoms
The prevalence of significant post-traumatic stress disorder symptoms, expressed by the IES-R score ≥ 32, in the overall sample was 54.8%. The mean age was similar in the two groups resulting in 45.8 (±15.9) in the PTSD group (PTSD) and 45.8 (±16.1) in the non-PTSD group (nPTSD). Female gender was more frequently associated with significant PTSD symptoms (58.5% vs 50.2%, p=0.038, X²=1.76). Participants reporting change in work characteristics (63.1% vs 47.8%, p=0.000, X²=3.29), loss of income (65.1% vs 50.7%, p=0.0027, X²=2.78), job loss (68.7% vs 53.17%, p=0.021, X²=2.05), financial strain (62.9% vs 50%, p=0.0036, X²=2.69) and relationship status (76.4% vs 48.3%, p=0.000, X²=5.10), and family quarrels (68.1% vs 50.4%, p=0.0005, X²=3.28) due to COVID-19 were more likely to experience significant post-traumatic symptoms. Participants who underwent COVID testing (64.3% vs 50%, p=0.002, X²=2.9), got sick (71.4% vs 53.7%, p=0.03, X²=1.82), and were hospitalized (83.3% vs 54%, p=0.021, X²=2.01) for COVID-19 or whose loved ones got sick with COVID-19 reported more frequently significant post-traumatic symptoms. After logistic regression analysis (see Table 4 ) the variable most strongly associated with the presence of PTSD was having a worsened relationship status (p=0.000; OR 2.66 [1.57-4.5].Table 4 Logistic regression analyses of socio-demographic factors and COVID-related variables on the presence of post-traumatic symptoms (IES-R score ≥ 32)
Table 4Variables in equations OR p 95%CI
Female gender 1.27 0.229 0.85-1.88
Change in work characteristics due to COVID-19 1.19 0.441 0.75-1.88
Loss of income due to COVID-19 1.25 0.472 0.68-2.29
Losing employment due to COVID-19 1.17 0.685 0.54-2.55
Worsening of financial situation due to COVID-19 1.14 0.610 0.67-1.94
Worsening of relationship status due to COVID-19 2.66 0.000 1.58-4.50
Family quarrels due to COVID-19 1.44 0.138 0.88-2.35
Undergone testing for COVID-19 1.43 0.124 0.90-2.26
Got sick with COVID-19 0.82 0.735 0.27-2.54
Hospitalized for COVID-19 3.41 0.211 0.49-23.34
Loved ones got sick with COVID-19 1.44 0.094 0.93-2.23
4 Discussion
We set out to assess the prevalence of psychopathological outcomes in response to the COVID-19 pandemic, reporting prevalence rates of significant depressive, anxious and post-traumatic symptoms of 57.6%, 63.5% and 54.8%, respectively.
In the existing literature, studies carried out in outpatient psychiatric settings always include patients already known by the mental health services. These studies highlighted the relevant impact of pandemic in groups of patients with specific diagnosis, such as personality disorders, OCD and schizophrenia (Caldiroli A et al., 2022).
In our study, all subjects accessed the territorial mental health services for the first time and about 58% of the sample never had a previous psychiatric consultation prior to that visit. Our recruitment selected a subgroup of subjects in the general population who required specialistic help because of a self need. To our knowledge, available studies do not recruit populations with characteristics similar to our sample. As conceivable, in our research, the prevalence of psychopathology resulted significantly higher than those reported in studies in community samples. In particular, the prevalence of depression in our sample (57.6%) was higher than those reported in the meta-analyses by Luo et al. (2020) (28%), Arora et al. (2022) (22%) and by Salari et al. (2020) (34%) and in the large study of Georgieva et al. (2021) (30.3%).
Around 63.5% of the patients in our study reported clinically significant anxiety symptoms. This rate was higher than those reported in meta-analyses in the general population (32% and 28%) (Luo et al., 2020; Arora et al., 2022).
Regarding traumatic symptoms, we reported a prevalence of 54.8%, that was higher than those estimated in meta-analyses in the general public (33%) (Arora et al., 2022).
Unemployment was found to be one of the main determinants of perceived stress (Codagnone et al., 2020), leading people to require mental health support (Allume et al., 2021; Menculini et al., 2021). This finding is consistent with our results showing a significant correlation between worsened financial condition and the manifestation of depressive, anxious, and traumatic symptoms.
According to previous studies (Mazza et al., 2021; Moccia et al., 2020; Wang et al., 2020), our results indicated that the female gender was a determinant for higher levels of depression, anxiety and PTSD symptoms.
Worsening of relationship status due to the pandemic was shown to be the variable most strongly correlated with the presence of significant depressive, anxious as well as traumatic symptoms. These results are consistent with literature showing the correlation between loneliness and social isolation and the vulnerability to the psychological distress of COVID-19 (Lei et al., 2020; Holmes et al., 2020).
Our data also showed that factors directly associated with COVID-19 infection such as undergoing testing, getting sick and being hospitalized for COVID-19 were only associated with PTSD symptoms.
One of the strengths of our research was the recruitment of participants in a real-life clinical setting which led to a greater heterogeneity of the sample in terms of age, ethnicity, education and economic level than in most clinical studies, providing a realistic picture of the consequences of the pandemic in clinical practice.
The results of the present study are also affected by some limitations. First, it is a cross-sectional study so it is unable to track the psychopathological impact over time of the pandemic in the participants. Second, no assumption of causality can be made due to the lack of pre-pandemic assessments for the investigated variables.
Furthermore, to collect social, environmental, relational and clinical characteristics we used a questionnaire created by the authors specifically for the purpose of this study but non validated in literature, and we only used self-report questionnaires for assessing the presence of psychopathology.
5 Conclusions
The COVID-19 emergency impacted on mental health services, given the need to guarantee care to patients with pre-existing psychiatric disorders, but also to provide prompt response to the general population seeking specialistic support for the first time. In this context, a better understanding of the population's psychological needs is crucial. As confirmed by the results of the present study, the impact of the pandemic involves depressive, anxious as well as post traumatic dimensions. In our clinical sample, psychopathology correlates with several variables expressing personal and environmental changes due to the COVID-19 emergency. Being directly exposed to COVID-19 is associated with the presence of post-traumatic symptoms, while economic and social variables correlate with depressive, anxious as well as post-traumatic dimensions. Particularly, worsening of relationship status is strongly associated with all the investigated psychopathology. This should be taken into consideration in order to include specific interventions aimed at improving social and relational support within mental health services.
Contributors
EP, EF and AC designed the study. EP, EF, EG, AC and CN recruited the participants. CN, EP, EF and EG created the dataset. AL analyzed the data. EP and AC interpreted the data and wrote the manuscript. PAFM, RB, and GC revised the manuscript. All authors read the manuscript and approved the submission to Journal of Affective Disorders.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
The Author(s) declare(s) that there is no conflict of interest.
Acknowledgements
The authors would like to thank Dr. Sara Candotti, Dr. Gloria Faraci, Dr. Rebecca Ranieri, and Dr. Fabio Salvaggio for their help in participant recruitment.
==== Refs
References
Alleaume C. Verger P. Peretti-Watel P. Group COCONEL Psychological support in general population during the COVID-19 lockdown in France: needs and access PLoS One 16 5 2021 e0251707 10.1371/journal.pone.0251707
Arora T. Grey I. Östlundh L. Hubert Lam K.B. M. Omar O. Arnone D The prevalence of psychological consequences of COVID-19: a systematic review and meta-analysis of observational studies J. Health Psychol. 27 4 2022 805 824 10.1177/1359105320966639 33118376
Beck A. Steer R. Brown G. BDI-II: Beck Depression Inventory Manual 2nd ed. 1996 Psychological Corporation TX
Bueno-Notivol J. Gracia-García P. Olaya B. Lasheras I. López-Antón R. Santabárbara J. Prevalence of depression during the COVID-19 outbreak: a meta-analysis of community-based studies Int. J. Clin. Health Psychol. 21 1 2021 100196 10.1016/j.ijchp.2020.07.007
Caldiroli A. Capuzzi E. Tringali A. Tagliabue I. Turco M. Fortunato A. Sibilla M. Montana C. Maggioni L. Pellicioli C. Marcatili M. Nava R. Crespi G. Colmegna F. Buoli M. Clerici M. The psychopathological impact of the SARS-CoV-2 epidemic on subjects suffering from different mental disorders: An observational retrospective study Psychiatry Res. 307 2022 114334 10.1016/j.psychres.2021.114334
Capuzzi E. Di Brita C. Caldiroli A. Colmegna F. Nava R. Buoli M. Clerici M. Psychiatric emergency care during Coronavirus 2019 (COVID 19) pandemic lockdown: results from a Department of Mental Health and Addiction of northern Italy Psychiatry Res. 293 2020 113463 10.1016/j.psychres.2020.113463
Castellini G. Rossi E. Cassioli E. Sanfilippo G. Innocenti M. Gironi V. Silvestri C. Voller F. Ricca V. A longitudinal observation of general psychopathology before the COVID-19 outbreak and during lockdown in Italy J. Psychosom. Res. 141 2021 110328 10.1016/j.jpsychores.2020.110328
Codagnone C. Bogliacino F. Gómez C. Charris R. Montealegre F. Liva G. Lupiáñez-Villanueva F. Folkvord F. Veltri G.A. Assessing concerns for the economic consequence of the COVID-19 response and mental health problems associated with economic vulnerability and negative economic shock in Italy, Spain, and the United Kingdom PLoS One 15 10 2020 e0240876 10.1371/journal.pone.0240876
Conti L. Repertorio delle scale di valutazione in psichiatria [Italian collection of the assessment scales in psychiatry] 2000 SEE Firenze
Craparo G. Faraci P. Rotondo G. Gori A. The Impact of Event Scale - Revised: psychometric properties of the Italian version in a sample of flood victims Neuropsychiatr Dis Treat 9 2013 1427 1432 10.2147/NDT.S51793 24092980
Cullen W. Gulat i G. Kelly B.D Mental health in the COVID-19 pandemic QJM 113 5 2020 311 312 10.1093/qjmed/hcaa110 May 1PMID: 32227218; PMCID: PMC7184387 32227218
Di Lorenzo R. Fiore G. Bruno A. Pinelli M. Bertani D. Falcone P. Marrama D. Starace F. Ferri P. Urgent psychiatric consultations at mental health center during COVID-19 pandemic: retrospective observational study Psychiatr. Q. 92 4 2021 1341 1359 10.1007/s11126-021-09907-w 33772425
Dunstan D.A. Scott N. Norms for Zung's self-rating Anxiety scale BMC Psychiatry 20 1 2020 90 10.1186/s12888-019-2427-6 32111187
Fiorillo A. Gorwood P. The consequences of the COVID-19 pandemic on mental health and implications for clinical practice Eur. Psychiatry 63 1 2020 e32 10.1192/j.eurpsy.2020.35 32234102
Fleischmann E. Dalkner N. Fellendorf F.T. Reininghaus E.Z. Psychological impact of the COVID-19 pandemic on individuals with serious mental disorders: a systematic review of the literature World J. Psychiatry 11 12 2021 1387 1406 10.5498/wjp.v11.i12.1387 35070784
Georgieva I. Lepping P. Bozev V. Lickiewicz J. Pekara J. Wikman S. Loseviča M. Raveesh B.N. Mihai A. Lantta T. Prevalence, new incidence, course, and risk factors of PTSD, depression, anxiety, and panic disorder during the Covid-19 pandemic in 11 countries Healthcare (Basel, Switzerland) 9 6 2021 664 10.3390/healthcare9060664 34204925
Holmes E.A. O'Connor R.C. Perry V.H. Tracey I. Wessely S. Arseneault L. Ballard C. Christensen H. Cohen Silver R. Everall I. Ford T. John A. Kabir T. King K. Madan I. Michie S. Przybylski A.K. Shafran R. Sweeney A. Worthman C.M. Bullmore E. Multidisciplinary research priorities for the COVID-19 pandemic: a call for action for mental health science The Lancet. Psychiatry 7 6 2020 547 560 10.1016/S2215-0366(20)30168-1 32304649
Lei L. Huang X. Zhang S. Yang J. Yang L. Xu M. Comparison of prevalence and associated factors of anxiety and depression among people affected by versus people unaffected by quarantine during the COVID-19 epidemic in Southwestern China Med. Sci. Monit. 26 2020 e924609 10.12659/MSM.924609
Luo M. Guo L. Yu M. Jiang W. Wang H. The psychological and mental impact of coronavirus disease 2019 (COVID-19) on medical staff and general public - a systematic review and meta-analysis Psychiatry Res. 291 2020 113190 10.1016/j.psychres.2020.113190
Manchia M. Gathier A.W. Yapici-Eser H. Schmidt M.V. de Quervain D. van Amelsvoort T. Bisson J.I. Cryan J.F. Howes O.D. Pinto L. van der Wee N.J. Domschke K. Branchi I. Vinkers C.H. The impact of the prolonged COVID-19 pandemic on stress resilience and mental health: a critical review across waves Eur. Neuropsychopharmacol. 55 2022 22 83 10.1016/j.euroneuro.2021.10.864 34818601
Mazza, M.G., Palladini, M., De Lorenzo, R., Magnaghi, C., Poletti, S., Furlan, R., Ciceri, F., COVID-19 BioB Outpatient Clinic Study group, Rovere-Querini, P., Benedetti, F., 2021. Persistent psychopathology and neurocognitive impairment in COVID-19 survivors: Effect of inflammatory biomarkers at three-month follow-up. Brain Behav. Immun. 94, 138–147. doi:10.1016/j.bbi.2021.02.021.
Menculini G. Tortorella A. Albert U. Carmassi C. Carrà G. Cirulli F. Dell'Osso B. Luciano M. Nanni M.G. Pompili M. Sani G. Volpe U. Fiorillo A. Sampogna G. Access to mental health care during the first wave of the COVID-19 pandemic in Italy: results from the COMET multicentric study Brain Sci 11 11 2021 1413 10.3390/brainsci11111413 34827412
Moccia L. Janiri D. Pepe M. Dattoli L. Molinaro M. De Martin V. Chieffo D. Janiri L. Fiorillo A. Sani G. Di Nicola M. Affective temperament, attachment style, and the psychological impact of the COVID-19 outbreak: an early report on the Italian general population Brain Behav. Immun. 87 2020 75 79 10.1016/j.bbi.2020.04.048 32325098
Morganstein J.C. Ursano R.J. Ecological disasters and mental health: causes, consequences, and interventions Front Psychiatry 11 2020 1 10.3389/fpsyt.2020.00001 32116830
Ramadan M. Fallatah A.M. Batwa Y.F. Saifaddin Z. Mirza M.S. Aldabbagh M. Alhusseini N. Trends in emergency department visits for mental health disorder diagnoses before and during the COVID-19 pandemic: a retrospective cohort study 2018-2021 BMC Psychiatry 22 1 2022 378 10.1186/s12888-022-03988-y 35659204
Robinson E. Sutin A.R. Daly M. Jones A. A systematic review and meta-analysis of longitudinal cohort studies comparing mental health before versus during the COVID-19 pandemic in 2020 J. Affect. Disord. 296 2022 567 576 10.1016/j.jad.2021.09.098 34600966
Salari N. Hosseinian-Far A. Jalali R. Vaisi-Raygani A. Rasoulpoor S. Mohammadi M. Rasoulpoor S. Khaledi-Paveh B. Prevalence of stress, anxiety, depression among the general population during the COVID-19 pandemic: a systematic review and meta-analysis Global Health 16 1 2020 57 10.1186/s12992-020-00589-w 32631403
Sica C. Ghisi M. The italian versions of the beck anxiety inventory and the beck depression inventory-II: psychometric properties and discriminant power Lange M.A. Leading-Edge Psychological Tests and Testing 2007 Nova Science Publishers Hauppauge, NY 27 50
Sultana M.S. Khan A.H. Hossain S. Islam T. Hasan M.T. Ahmed H.U. Li Z. Khan J. The association between financial hardship and mental health difficulties among adult wage earners during the COVID-19 pandemic in bangladesh: findings from a cross-sectional analysis Front. Psychiatry 12 2021 635884 10.3389/fpsyt.2021.635884
Tanaka-Matsumi J. Kameoka V.A. Reliabilities and concurrent validities of popular self-report measures of depression, anxiety, and social desirability J. Consult. Clin. Psychol. 54 1986 328 10.1037/0022-006X.54.3.328 3722561
Wang C. Pan R. Wan X. Tan Y. Xu L. Ho C.S. Ho R.C. Immediate psychological responses and associated factors during the initial stage of the 2019 coronavirus disease (COVID-19) epidemic among the general population in China Int. J. Environ. Res. Public Health 17 5 2020 1729 10.3390/ijerph17051729 32155789
Weiss D.S. Marmar C.R. The impact of event scale-revised editors Wilson J.P. Keane T.M. Assessing Psychological Trauma and PTSD 1997 Guilford Press New York 399 411
Yunitri N. Chu H. Kang X.L. Jen H.J. Pien L.C. Tsai H.T. Kamil A.R. Chou K.R. Global prevalence and associated risk factors of posttraumatic stress disorder during COVID-19 pandemic: a meta-analysis Int. J. Nurs. Stud. 126 2022 104136 10.1016/j.ijnurstu.2021.104136
Zung W.W.K. A rating instrument for anxiety disorders Psychosomatics 12 1971 371 379 5172928
Zung W.W.K. How Normal is Anxiety? 1980 Durham Upjohn
| 36506487 | PMC9726655 | NO-CC CODE | 2022-12-15 23:17:46 | no | J Affect Disord Rep. 2023 Jan 7; 11:100460 | utf-8 | J Affect Disord Rep | 2,022 | 10.1016/j.jadr.2022.100460 | oa_other |
==== Front
Heliyon
Heliyon
Heliyon
2405-8440
The Author(s). Published by Elsevier Ltd.
S2405-8440(22)03428-4
10.1016/j.heliyon.2022.e12140
e12140
Research Article
The clinical impact of COVID-19 on patients with cancer in British Columbia: An observational study
Mathews Angela S. ab
Paul Ashley ab
Yu Irene S. ab
McGahan Colleen a
Bhang Eric ab
Villa Diego ab
Gelmon Karen ab
Avina-Zubieta Antonio bc
Gerrie Alina S. ab
Lee Ursula ab
Chia Stephen ab
Woods Ryan R. ad
Loree Jonathan M. ab∗
a BC Cancer Agency, 600 W 10th Ave, Vancouver, BC V5Z 4E6, Canada
b University of British Columbia, Vancouver, BC V6T 1Z4, Canada
c Arthritis Research Canada, 5591 No. 3 Road, Richmond, BC, V6X 2C7, Canada
d Simon Fraser University, Burnaby, BC, V5A 1S6, Canada
∗ Corresponding author.
7 12 2022
7 12 2022
e1214028 1 2022
9 6 2022
29 11 2022
© 2022 The Author(s)
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objective
We evaluated survival outcomes for patients with cancer and COVID-19 in this population-based study.
Methods
A total of 631 patients who tested positive for severe acute respiratory syndrome coronavirus 2 and were seen at BC Cancer between 03/03/2020 and 01/21/2021 were included, of whom 506 had a diagnosis of cancer and PCR-confirmed positive test for coronavirus disease 2019. Patient clinical characteristics were retrospectively reviewed and the influence of demographic data, cancer diagnosis, comorbidities, and anticancer treatment(s) on survival following severe acute respiratory syndrome coronavirus 2 infection were analyzed.
Results
Age ≥65 years (Hazard Ratio [HR] 4.77, 95% Confidence Interval [CI] 2.72–8.35, P < 0.0001), those with Eastern Cooperative Oncology Group Performance Status ≥2 (HR 8.36, 95% CI 2.89–24.16, P < 0.0001), hypertension (HR 3.17, 95% CI 1.77–5.66, P < 0.0001), and metastatic/advanced stage (HR 3.70, 95% CI 1.77–7.73, P < 0.0001) were associated with worse coronavirus disease 2019 specific survival outcomes following severe acute respiratory syndrome coronavirus 2 infection. Patients with lung cancer had the highest 30-day COVID-19 specific mortality (25.0%), followed by genitourinary (18.1%), gastrointestinal (16.0%), and other cancer types (<10.0%). Patients with the highest 30-day coronavirus disease 2019 specific mortality according to treatment type were those on chemotherapy (23.0%), rituximab (22.2%), and immunotherapy (16.7%) while patients on hormonal treatments (2.2%) had better survival outcomes (P = 0.041) compared to those on other anticancer treatments.
Conclusion
This study provides further evidence that patients with cancer are at increased risk of mortality from coronavirus disease 2019 and emphasizes the need for vaccination.
Cancer; COVID-19; SARS-CoV-2; Comorbidities; Chemotherapy; Treatment.
Keywords
Cancer
COVID-19
SARS-CoV-2
Comorbidities
Chemotherapy
Treatment
==== Body
pmc1 Introduction
Worldwide spread of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for the coronavirus disease 2019 (COVID-19) pandemic, and has resulted in over 348 million total cases of COVID-19 and over 5.5 million deaths globally as of January 2022 (COVID-19 dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU), 2021). Patients with cancer may be at increased risk of severe complications and mortality from COVID-19 due to immunosuppression caused by anticancer therapies and/or the underlying cancer (Williamson et al., 2020). The most recent update from the COVID-19 and Cancer Consortium (CCC19) reported 30-day all-cause mortality between 13% and 33% (Garassino et al., 2020; Kuderer et al., 2020; Reboot: COVID-Cancer Project, 2021) in patients with cancer and confirmed SARS-CoV-2 infection, compared with 0.5%–2% in the general population (COVID-19 dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU), 2021; Lau et al., 2020).
Although the effects of COVID-19 on patients with cancer have been studied extensively, the larger studies thus far have been based on multi-institutional efforts and were not population-based. There may be bias in outcomes based on the populations included and differences in access to care between countries and health systems, and thus these studies may not be representative of all patients with cancer within a single institution. The purpose of our study is to systematically examine the clinical effects of SARS-CoV-2 on patients with cancer in British Columbia (BC), providing a clear picture on the state of COVID-19 disease management associated with cancer at a broad population level in our publicly funded health care system. Almost all cancer care in the province is centralized within one network of oncology clinics, allowing a robust assessment of diagnoses, treatments, and outcomes in the province's 5.1 million people.
2 Methods
Patients who had been seen at BC Cancer for care after March 1, 2020 with SARS-CoV-2 confirmed by PCR results between 03/03/2020 and 01/21/2021 from across the province of British Columbia were included in this retrospective cohort study (Figure 1 ). Individual patient data were abstracted from BC Cancer's Electronic Health Record (EHR). Within the EHR, patients with a COVID-19 diagnosis were flagged by provincial testing centers, allowing robust capture of all confirmed cases. Patients were assumed to have died from COVID-19 if there was evidence of this in the EHR through chart review (i.e., COVID-19, virus identified) or if the death was within 30 days of their positive test. Patients were classified as having died due to non-COVID-19-related causes if their date of death was over 30 days from the date of their positive test, and if they died due to other health concerns related to progression of their cancer or other comorbidities (e.g., cardiac arrest, unknown causes, medical assistance in dying due to reasons unrelated to COVID-19, etc.). This study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Board of The University of British Columbia (H20-00892–approved 4/1/2020) for research involving human subjects. A waiver of consent was obtained as this was a retrospective research study.Figure 1
Variables concerning demographic data, cancer diagnosis, comorbidities, anti-cancer treatment(s), smoking status, date of COVID-19 diagnosis and follow up survival data were collected. A cancer diagnosis was assigned if patients had active or prior cancer with scheduled follow ups at BC Cancer, and further classified by cancer stage (early: stages I-III, metastatic: stage IV) for applicable cancers. Patients who were seen for hereditary cancer screening or assessment/treatment of benign tumours but who did not have active or prior cancer diagnoses were classified as having no cancer. Patients with liquid cancers (leukemia and lymphoma) were excluded from analyses involving cancer stage as they differ significantly from conventional classification of cancer stage used by solid tumors. Patients without a malignant cancer diagnosis were seen for hereditary screening or management of benign tumours and were used as a control in survival curves; these patients were classified as having no cancer (Figures 2 and 3 ). Cancer types were categorized as follows: breast, gastrointestinal (GI), lung, lymphoma, genitourinary (GU), gynecologic, leukemia, central nervous system (CNS) and other (including thyroid cancer, skin cancer, liposarcoma, head and neck cancer, carcinoma of unknown primary, etc.). Patients with CLL were classified as leukemia. Patients were classified as on active anticancer therapy if the last dose was received within 12 months prior to a COVID-19 diagnosis, and untreated if over 12 months had passed. Therapy types were assigned to the following categories: chemotherapy, hormone therapy, immunotherapy, radiotherapy, no therapy, other treatment, surgery, and CD20 directed (i.e., rituximab). The list of all anticancer treatments classified under each treatment category is detailed in Table 1 .Figure 2
Figure 3
Table 1 Classification of anticancer treatments.
Table 1Chemotherapy Hormone therapy Immunotherapy Radiotherapy Rituximab Other
5-FU and oxaliplatin Anastrozole Ipilimumab Adjuvant radiation therapy R–CHOP/CHOP-R (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) Brentuximab vedotin
5-FU, irinotecan and bevacizumab Apalutamide Nivolumab Adjuvant radiotherapy Rituximab Dasatinib
Abraxane Bicalutamide Pembrolizumab Brachytherapy Dexamethasone
ABVD/LYABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) Bicalutamide Definitive radiotherapy Gefitinib
AC (Adriamycin, cyclophosphamide) BRAJTAM External beam radiotherapy Herceptin
ACT/ACTW (Doxorubicin, Cyclophosphamide, followed by Paclitaxel or Docetaxel) Degarelix I-131 therapy Ibrutinib
Bendamustine Enzalutamide Radioactive iodine ablation Imatinib
BEP (bleomycin, etoposide, platinum) Exemestane SABR Inotuzumab
Bleomycin Exemestane Salvage radiation Lenalidomide
Capecitabine Faslodex and Ibrance SBRT Olaparib
Carboplatin Goserelin Stereotactic radiotherapy Osimertinib
Cisplatin Letrozole Palbociclib
Cisplatin and etoposide Lupron Panitumumab
Cisplatin and pemetrexed Provera Prednisone
CyBorD (cyclophosphamide, bortezomib, dexamethasone) Tamoxifen Regorafenib
Cyclophosphamide Sorafenib
Docetaxel Trastuzumab
Doxorubicin Vismodegib
Doxorubicin, cyclophosphamide, paclitaxel and trastuzumab
Epirubicin
Etoposide
FLAG (fludarabine + high-dose cytarabine + G-CSF)
Fludarabine
Fluorouracil
FOLFIRI (folinic acid/leucovorin, fluorouracil, irinotecan)
FOLFIRINOX (folinic acid/leucovorin, fluorouracil, irinotecan, oxaliplatin)
Gemcitabine
Gemcitabine and cisplatin
Gemcitabine and nab-paclitaxel
GOENDCAT protocol (carboplatin and paclitaxel)
Hydroxyurea
Ifosfamide, gemcitabine, and vinorelbine
Irinotecan
Leukocorin with fluorouracil, docetaxel, oxaliplatin
Long-acting octreotide and lutetium
methotrexate and mercaptopurine
Oxaliplatin
Paclitaxel
PCV (procarbazine, lomustine (CCNU) and vincristine)
Pegylated liposomal doxorubicin and carboplatin
Pemetrexed
Temozolomide
vincristine, prednisone, L-asparagine, daunomycin, cyclophosphamide, cytarabine, 6 thioguanine, 6 MP, methotrexate, dexamethasone
Vinorelbine
Descriptive statistics were used to summarize demographic characteristics and outcomes. Kaplan-Meier curves were generated using Graphpad Prism 8.4.3. After satisfying the proportional hazards assumption, multivariate models were created using a forward likelihood ratio selection with variables incorporated into the model if P < 0.05 and removed from the model when P > 0.1. Patients with missing data for any variable were excluded. Multivariate analysis was performed in SPSS version 17.0. Overall survival was defined as the time between COVID-19 diagnosis and death. Patients alive at the time of last contact were censored. Patients who died due to non-COVID-19-related causes were censored at their time of death when COVID-19-specific survival was assessed and counted as an event when generating overall survival analyses. Three patients were considered clinical diagnoses and did not include a PCR-confirmed test for COVID-19. These patients were included in Table 2 (n = 631) but excluded from all subsequent analyses (n = 628) so as not to introduce bias in time to event analyses. See Figure 1 for a flow diagram detailing eligibility criteria and methods of selection of patients included in the study and numbers of censored patients.Table 2 Patient demographical and clinical characteristics.
Table 2Metric Category Patients who tested positive for COVID-19 (n = 631) Patients with cancer who tested positive for COVID-19 (n = 509)
Count Percentage (%) Count Percentage (%)
Sex Male 298 47.2 248 48.7
Female 333 52.8 261 51.3
Age, years Median age = 62, IQR1: 49–73 Median age = 66, IQR: 53–76
Comorbidities (Data available in 539 patients total and 496 patients with cancer) Asthma 29 5.4 28 5.6
Chronic Obstructive Pulmonary Disease 34 6.3 34 6.9
Congestive Heart Failure 19 3.5 19 3.8
Diabetes 103 19.1 100 20.2
Hypertension 204 37.9 196 39.5
Rheumatologic Disease 19 3.5 18 3.6
Obstructive Sleep Apnea 25 4.6 24 4.8
Chronic Kidney Disease 25 4.6 25 5.0
History of Solid Organ Transplant 1 0.2 1 0.2
Cirrhosis 5 0.9 5 1.0
Cancer Type Breast 96 15.2 96 18.9
GI2 58 9.2 58 11.4
Lung 50 7.9 50 9.8
Lymphoma 60 9.5 60 11.8
GU3 67 10.6 67 13.2
Gynecologic 32 5.1 32 6.3
Leukemia 25 4.0 25 4.9
CNS4 20 3.2 20 3.9
Other 101 16.0 101 19.8
No cancer 122 19.3 NA NA
Last cancer treatment prior to COVID diagnosis (within 12 months) Chemotherapy 64 10.1 64 12.6
Hormone therapy 55 8.7 55 10.8
Immunotherapy 6 1.0 6 1.2
Radiotherapy 46 7.3 46 9.0
No therapy 410 65.0 292 57.4
Other 28 4.4 26 5.1
Surgery 11 1.7 10 2.0
Rituximab 11 1.7 10 2.0
Smoking status (Data available in 474 patients and 458 patients with cancer) 184 38.8 173 37.8
Number hospitalized (Data available in 626 patients and 504 with cancer) 96 15.3 95 18.8
1 Interquartile Range. 2 Gastrointestinal. 3 Genitourinary. 4 Central Nervous System.
Hazard ratios and 95% confidence intervals (CI) were computed for survival curves and a multivariate analysis comparing each term in the model. Odds ratios (OR), 95% confidence intervals and P values were calculated from two-sided chi-squared tests and estimated from contingency tables looking at comorbidity distributions for each cancer type. Thirty-day COVID-19-specific mortality was estimated from survival curves for each variable being compared. ECOG performance status was retrieved from BC Cancer's EHR through chart review prior to a patient's PCR-confirmed positive test for COVID-19, although this data was not always consistently reported. Data on laboratory values at the time of COVID-19 diagnosis, presentation of COVID-19 symptoms, and hospitalization information was collected but excluded from analysis due to inconsistency in timing and availability of this data.
3 Results
Patient demographic and clinical characteristics are outlined in Table 2 for the 631 BC Cancer patients identified with COVID-19. The median age was 62 (interquartile range 49–73). The cohort included a slightly higher number of females (n = 333, or 52.8%). Five-hundred and nine patients with confirmed COVID-19 diagnosis also had cancer (80.7%), and the median age for patients in this group was 66 years (interquartile range 53–76). The rate of hospitalizations was significantly higher in patients with cancer (18.8%) compared to those without cancer (0.8%, P < 0.0001). Figure 2 shows that 122 out of 628 patients (19.4%) were listed as having no cancer and were seen for hereditary counselling. Among patients who were alive at the end of the study, the median follow-up time was 62 days (n = 561).
Patients with active or prior cancer were more likely to die from COVID-19 than patients without cancer (HR 7.59, 95% CI 3.37–17.12, P = 0.018) after censoring patients who died due to non-COVID-19-related reasons (Figure 2A). Thirty-day COVID-19 specific mortality was significantly higher (P = 0.0008) for patients with cancer (11.6%) compared to those without cancer (1.6%) (Table 3 ).Table 3 30-day mortality of patients in British Columbia, Canada seen at BC Cancer following diagnosis of COVID-19 compared by presence or absence of cancer, type of cancer, and type of treatment.
Table 3Patients with confirmed COVID-19 (n = 628) Patients with confirmed COVID-19 by Cancer Type (n = 628) Patients with cancer and confirmed COVID-19 by Treatment Type (n = 506)
Cohort 30-Day mortality (%) Cohort 30-Day mortality (%) Cohort 30-Day mortality (%)
Cancer (n = 506) 11.6% Breast (n = 96) 7.0% Chemotherapy (n = 64) 23.0%
No cancer (n = 122) 1.6% CNS1 (n = 20) 9.1% Hormonal treatment (n = 55) 2.2%
GI2 (n = 58) 16.0% Immunotherapy (n = 6) 16.7%
GU3 (n = 66) 18.1% Radiation (n = 46) 12.3%
Gynecologic (n = 32) 9.6% Surgery (n = 10) 0.0%
Leukemia (n = 25) 4.5% Rituximab (n = 10) 22.2%
Lung (n = 49) 25.0% Other treatment (n = 26) 7.7%
Lymphoma (n = 60) 8.6% No treatment (n = 292) 10.6%
Other (n = 100) 7.1%
No cancer (n = 122) 1.6%
1 Central Nervous System. 2 Gastrointestinal. 3 Genitourinary.
The effects of cancer type and anticancer treatment are shown in Figure 2B and C. Ninety-six patients had active or prior breast cancer, comprising the largest subset of cancer, followed by genitourinary (n = 66) and other cancers (n = 100). Patients with lung cancer (n = 49) had the highest 30-day COVID-19 specific mortality (25.0%), followed by GU (n = 66; 18.1%) and GI (n = 58; 16.0%), with the remaining cancer types (n = 455) under 10.0% (Table 3). A high proportion of patients with cancer were not on any anticancer treatments (n = 292; 57.4%) (Table 2). Of the anticancer treatments considered, patients on chemotherapy had the worst COVID-19 specific survival outcomes (30-day COVID-19 specific mortality of 23.0%). Patients on, rituximab, immunotherapy, and radiation had 30-day COVID-19 specific mortality of 22.2%, 16.7% and 12.3%, respectively (Table 3).
Factors associated with increased 30-day COVID-19 specific mortality are shown in Figure 4 . These include age ≥65 years of age (HR 4.77, 95% CI 2.72–8.35, P < 0.0001), metastatic cancer stage (HR 3.70, 95% CI 1.77–7.73, P < 0.0001), Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≥2 (HR 8.36, 95% CI 2.89–24.16, P < 0.0001), chronic kidney disease (HR 4.29, 95% CI 1.08–16.89, P < 0.0001), and hypertension (HR 3.17, 95% CI 1.77–5.66, P < 0.0001). There was also an increased COVID-19 specific mortality among male patients (HR 2.00, 95% CI 1.14–3.50, P = 0.0018) compared to female patients. Other variables studied including obesity and other cardiopulmonary conditions did not contribute significantly to worse COVID-19 specific survival outcomes following SARS-CoV-2 infection. On multivariate analysis, only ECOG ≥2 (HR 33.90, 95% CI 4.34–265.08, P = 0.001) was significantly associated with COVID-specific survival after controlling for ECOG (0-1 vs ≥ 2), sex, stage (metastatic vs not), cancer type, cancer treatment, smoking history and age (≥65 vs < 65). For overall survival, ECOG ≥2 (HR 21.74, 95% CI 4.79–98.58, P < 0.0001) and presence of metastatic disease (HR 2.93, 95% CI 1.03–8.32, P = 0.044) were both prognostic after controlling for co-variates. All other variables (sex, stage, cancer type, cancer treatment, smoking history and age for COVID-specific survival; and sex, cancer type, cancer treatment, smoking history and age for overall survival) were controlled for but did not remain in the model as they were not significant.Figure 4
Hypertension was found to be the most common comorbidity among all cancers, with 39.5% of cancer patients diagnosed with COVID-19 also having hypertension. Hypertension was the most prevalent among patients with lung cancer, with 60.0% of lung cancer patients also having hypertension compared with all cancers combined (OR 2.29, 95% CI 1.21–4.16, P = 0.0078). Patients with lung cancer were also found to have the highest incidence of chronic kidney disease, at 15.6% compared to 5.1% in all cancers combined (OR 3.46, 95% CI 1.30–8.53, P = 0.0043), although the number of patients with lung cancer and CKD (n = 7/45) is limited. Patients with lung cancer and COVID-19 were also found to have the highest prevalence of COPD, at 40.0% compared to 6.9% in all cancers combined (OR 9.04, 95% CI 4.39–17.70, P < 0.0001). Diabetes was most common among patients with GI cancers and COVID-19, with a prevalence of 38.2%, compared with 20.2% in all patients with cancer combined (OR 2.44, 95% CI 1.37–4.28, P = 0.0023).
4 Discussion
In this population-based observational study, we demonstrated that the diagnosis of cancer is associated with higher rates of hospitalization (18.8% vs 0.8% in those without cancer, P < 0.0001) and worse survival outcomes followed COVID-19 infection, consistent with previous literature ((WHO), 2020; Onder et al., 2020; Robilotti EV, Babady NE, 2021). Prognostic factors identified related to COVID-19 mortality include age ≥65, hypertension, ECOG ≥2, and stage IV cancer. These results are similar to the risk factors identified in the COVID-19 and Cancer Consortium (CCC19) analysis (Kuderer et al., 2020). Smoking status was also associated with worse 30-day mortality in that study but excluded from our analysis due to inconsistent data availability. Given that malignancies are often diagnosed in older patients and can lead to a deterioration in health, there is likely a complex interplay between a patient's baseline comorbid status and the impact of their cancer on their body's ability to deal with a COVID-19 infection.
Patients with lung cancer had a high prevalence of comorbidities compared to other malignancies, which may be a contributing factor to the differences in survival outcomes observed. In the lung cancer population, 60.0% of patients were found to have hypertension, compared with 39.5% in all cancers combined (OR 2.29, 95% CI 1.21–4.16, P = 0.0078). Several studies have suggested that vascular remodelling and associated accumulation of inflammatory immune cells in the context of lung cancer may occur and this could lead to more damage from inflammation due to severe COVID-19, particularly given that immunotherapy is commonly used to treat thoracic malignancies (Battafarano et al., 2002; Guignabert et al., 2013; Pullamsetti et al., 2017a, 2017b, 2014; Tammemagi, C.M.; Neslund-Dudas, C.; Simoff, M.; Kvale, 2003). Unsurprisingly, patients with lung cancer were also found to have the greatest incidence of chronic obstructive pulmonary disease (COPD), at 40.0% compared to 6.9% in all cancers combined (OR 9.04, 95% CI 4.39–17.70, P < 0.0001). Patients with COPD may have decreased pulmonary reserve and a higher susceptibility to respiratory failure due to COVID-19. Overall, these results are consistent with previous studies regarding COVID-19 in patients with lung cancer and may in part explain the poor survival outcomes seen among lung cancer patients following COVID-19 infection (Figure 2C) (Luo et al., 2020; Maringe et al., 2020; Rogado et al., 2020).
Another notable finding in our analysis is that patients receiving chemotherapy had the worst survival outcomes of anticancer therapies, with a predicted 30-day COVID-19 specific mortality of 23.0%. It will be important to determine whether patients will have a robust and lasting response to vaccines to ensure that patients living through several waves of COVID-19 will remain protected given their high risk of death due to the virus. The administration of rituximab, an anti-CD20 antibody, was associated with the second highest 30-day COVID-19 specific mortality out of all the anticancer treatments (22.2%) However, this is limited by the small number of patients, as only 10 out of 628 surveyed patients with cancer received rituximab within one year prior to COVID-19 diagnosis. The immunosuppressive effects of rituximab have been well-studied, especially in the context of long-term use, where it is associated with B cell depletion and decreased antibody production as a result (Avouac et al., 2021; Bingham et al., 2010; Kos et al., 2020; Mehta et al., 2020). It will be important to ascertain whether rituximab interferes with the maintenance of a robust antiviral immune response against SARS-CoV-2 (Avouac et al., 2021; Mehta et al., 2020) in the many serologic studies that are ongoing.
This study should be interpreted in the context of several limitations. First, this was a retrospective study, and some data elements were not fully available and collected in a regimented fashion, such as laboratory investigations. Second, our analysis does not include variables concerning COVID-19 symptoms as these were difficult to capture from outside hospitals when patients presented. We were constrained by the need to rely on discharge summaries for many of these patients which may have had varying degrees of documentation surrounding the initial presentation. As well, it was difficult to determine the level of medical intervention, COVID-19 vaccination status, and if any COVID-directed therapeutics were used while admitted to an outside hospital. COVID vaccination began in the last 2 months of our chosen cohort so most patients would have either not been vaccinated or only had one injection. Additionally, the classification of anticancer therapy as active if the last dose was received within 12 months prior to a COVID-19 diagnosis could impact the generalizability of these findings as the duration of treatment may be an additional factor affecting the severity of toxicity and side effect profiles associated with anticancer therapies. Finally, our study was only conducted in one province during the early stages of the ongoing pandemic. Sample characteristics of the present study may not be fully generalizable to all patients with cancer, as the type and frequency of cancers observed in our population may not be representative of the cancer population, a large proportion of patients were not receiving any treatment, and a significant proportion of patients had additional comorbidities such as HBP. However, taking these considerations into account allows us to provide an accurate snapshot of the early stages of the ongoing COVID-19 pandemic at BC Cancer. These results may not be extrapolated to all health care systems; however, they were drawn from a population-based cohort within a public health care system, which helps limit differences in outcome based on access to care. While disparities in care may still exist within public health care systems, the ability of our study to drawn from a single health care system that provides care to over five million people living in a Canadian province is an important strength. However, there is a possibility our results may have some bias. For example, we did see that the median age of individuals with COVID was 62, while the median age of patients diagnosed with Canada is 66.9 years. This may be due to differences in COVID exposure or differences in the likelihood of a patient receiving cancer treatment changing as one ages (Statistics Canada - Cancer incidence in The Daily - Cancer incidence in Canada, 2018; 2021).
5 Conclusion
Patients with a cancer diagnosis appear at increased risk of 30-day mortality from SARS-Co-V-2 infection and there was significant variability in outcome based on type of cancer and treatment received. This population is often older and has significant comorbidities that put them at increased risk of serious COVID-19 infection. This study adds further support to aggressive vaccination programs and a low threshold for increased monitoring and close follow-up in suspected COVID-19 among patients living with cancer given the potential for adverse outcomes in this group.
Declarations
Author contribution statement
Angela S. Mathews, BSc; Eric Bhang, BSc; Jonathan M. Loree, MD, MS: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.
Ashley Paul, MD; Irene S. Yu, MD; Colleen McGahan, MSc; Diego Villa, MD, MPH; Karen Gelmon, MD; Antonio Avina-Zubieta, MD, MSc, PhD; Alina S. Gerrie, MD; Ursula Lee, MD; Stephen Chia, MD; Ryan R. Woods, MSc, PhD: Contributed reagents, materials, analysis tools or data; Wrote the paper.
Funding statement
Jonathan M Loree & Angela S. Mathews were supported by Michael Smith Health Professional Investigator Awards. This work was supported by Philanthropic funds from BC Cancer Foundation.
Data availability statement
The data that has been used is confidential.
Declaration of interest’s statement
The authors declare no conflict of interest.
Additional information
Supplementary content related to this article has been published online at [URL].
Uncited reference
Reboot, 2021
==== Refs
References
Avouac J. Drumez E. Hachulla E. Seror R. Georgin-Lavialle S. El Mahou S. COVID-19 outcomes in patients with inflammatory rheumatic and musculoskeletal diseases treated with rituximab: a cohort study Lancet Rheumatol 2021 419 426
Battafarano R.J. Piccirillo J.F. Meyers B.F. Hsu H.S. Guthrie T.J. Cooper J.D. Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer J. Thorac. Cardiovasc. Surg. 123 2002 280 287 11828287
Bingham C.O. Looney R.J. Deodhar A. Halsey N. Greenwald M. Codding C. Immunization responses in rheumatoid arthritis patients treated with rituximab: results from a controlled clinical trial Arthritis Rheum. 62 2010 64 74 20039397
COVID-19 dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU). Johns Hopkins Univ Med 2021. https://coronavirus.jhu.edu/map.html (accessed May 5, 2021).
Garassino M.C. Whisenant J.G. Huang L.C. Trama A. Torri V. Agustoni F. COVID-19 in patients with thoracic malignancies (TERAVOLT): first results of an international, registry-based, cohort study Lancet Oncol. 21 2020 914 922 32539942
Guignabert C. Tu L. Le Hiress M. Ricard N. Sattler C. Seferian A. Pathogenesis of pulmonary arterial hypertension: lessons from cancer Eur. Respir. Rev. 22 2013 543 551 24293470
Kos I. Balensiefer B. Roth S. Ahlgrimm M. Sester M. Schmidt T. Prolonged course of COVID-19-associated pneumonia in a B-cell depleted patient After rituximab Front. Oncol. 10 2020 1 5 32076595
Kuderer N.M. Choueiri T.K. Shah D.P. Shyr Y. Rubinstein S.M. Rivera D.R. Clinical impact of COVID-19 on patients with cancer (CCC19): a cohort study Lancet 395 2020 1907 1918 32473681
Lau H. Khosrawipour T. Kocbach P. Ichii H. Bania J. Khosrawipour V. Estimating mortality from COVID-19: a Scientific brief Pulmonology 2020 5 8
Luo J. Rizvi H. Preeshagul I.R. Egger J.V. Hoyos D. Bandlamudi C. COVID-19 in patients with lung cancer Ann. Oncol. 31 2020 1386 1396 32561401
Maringe C. Spicer J. Morris M. Purushotham A. Nolte E. Sullivan R. The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study Lancet Oncol. 21 2020 1023 1034 32702310
Mehta P. Porter J. Chambers R. Isenberg D. Reddy V. B-cell depletion with rituximab in the COVID-19 pandemic: where do we stand? Lancet 2 2020 E589 E590
Onder G. Rezza G. Brusaferro S. Case-Fatality rate and characteristics of patients dying in relation to COVID-19 in Italy JAMA, J. Am. Med. Assoc. 323 2020 1775 1776
Pullamsetti S.S. Schermuly R. Ghofrani A. Weissmann N. Grimminger F. Seeger W. Novel and emerging therapies for pulmonary hypertension Am. J. Respir. Crit. Care Med. 189 2014 394 400 24401129
Pullamsetti S.S. Kojonazarov B. Storn S. Gall H. Salazar Y. Wolf J. Lung cancer-Associated pulmonary hypertension: role of microenvironmental inflammation based on tumor cell-immune cell cross-Talk Sci. Transl. Med. 9 2017 1 17
Pullamsetti SS, Savai R, Seeger W, Goncharova EA. From Cancer Biology to New Pulmonary Arterial Hypertension Therapeutics Targeting Cell Growth and Proliferation Signaling Hubs 2017b.
Reboot: COVID-cancer Project. Reboot Rx, Inc 2021. https://rebootrx.org/covid-cancer (accessed May 5, 2021).
Robilotti E.V. Babady N.E.M.P. Determinants of severity in cancer patients with COVID-19 illness Nat. Med. 26 2021 1218 1223
Rogado J. Pangua C. Serrano-Montero G. Obispo B. Marino A.M. Pérez-Pérez M. Covid-19 and lung cancer: a greater fatality rate? Lung Cancer 146 2020 19 22 32505076
Tammemagi C.M. Neslund-Dudas C. Simoff M. Kvale P. Impact of comorbidity on lung cancer survival | Enhanced Reader Int. J. Cancer 2003 792 802 12516101
The Daily - Cancer incidence in Canada, 2018. Statistics Canada 2021. https://www150.statcan.gc.ca/n1/daily-quotidien/210519/dq210519b-eng.htm (accessed Jun 8, 2022).
(WHO) WHO. Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19) 2020. https://www.who.int/publications/i/item/report-of-the-who-china-joint-mission-on-coronavirus-disease-2019-(covid-19) (accessed June 14, 2021).
Williamson E.J. Walker A.J. Bhaskaran K. Bacon S. Bates C. Morton C.E. Factors associated with COVID-19-related death using OpenSAFELY Nature 584 2020 430 436 32640463
| 36506364 | PMC9726656 | NO-CC CODE | 2022-12-09 23:15:15 | no | Heliyon. 2022 Dec 7; 8(12):e12140 | utf-8 | Heliyon | 2,022 | 10.1016/j.heliyon.2022.e12140 | oa_other |
==== Front
Microbes Infect
Microbes Infect
Microbes and Infection
1286-4579
1769-714X
Published by Elsevier Masson SAS on behalf of Institut Pasteur.
S1286-4579(22)00151-4
10.1016/j.micinf.2022.105081
105081
Original Article
Association of complement pathways with COVID-19 severity and outcomes
Devalaraja-Narashimha Kishor a∗
Ehmann Peter J. a1
Huang Cong ab12
Ruan Qin a
Wipperman Matthew F. a
Kaplan Theodore a
Liu Chien a
Afolayan Simisola a
Glass David a
Mellis Scott a
Yancopoulos George D. a
Hamilton Jennifer D. a
MacDonnell Scott a
Hamon Sara C. a
Boyapati Anita a
Morton Lori a
a Regeneron Pharmaceuticals, Inc., Tarrytown, NY, USA;
b Wesleyan University, Middletown, CT, USA
∗ Corresponding author. Regeneron Pharmaceuticals, Inc., 777 Old Saw Mill River Road, Tarrytown, NY 10591-6707. Tel.: +1 914 847-7000.
1 Co-second authors (equal contribution).
2 At the time of study.
7 12 2022
7 12 2022
10508119 8 2022
16 11 2022
2 12 2022
© 2022 Published by Elsevier Masson SAS on behalf of Institut Pasteur.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objectives
Complement activation has been implicated in COVID-19 pathogenesis. This study aimed to assess the levels of complement activation products and full-length proteins in hospitalized patients with COVID-19 and evaluated if complement pathway markers are associated with outcomes.
Methods
Longitudinal measurements of complement biomarkers from 89 hospitalized adult patients, grouped by baseline disease severity, enrolled in an adaptive, phase 2/3, randomized, double-blind, placebo-controlled trial and treated with intravenous sarilumab (200 mg or 400 mg) or placebo (NCT04315298) were performed. These measurements were then correlated with clinical and laboratory parameters.
Results
All complement pathways were activated in hospitalized patients with COVID-19. Alternative pathway activation was predominant earlier in the disease course. Complement biomarkers correlated with multiple variables of multi-organ dysfunction and inflammatory injury. High plasma sC5b-9, C3a, factor Bb levels, and low mannan-binding lectin levels were associated with increased mortality. Sarilumab treatment showed a modest inhibitory effect on complement activation. Moreover, sera from patients spontaneously deposited C5b-9 complex on the endothelial surface ex vivo, suggesting a microvascular thrombotic potential.
Conclusion
These results advance our understanding of COVID-19 disease pathophysiology and demonstrate the importance of specific complement pathway components as prognostic biomarkers in COVID-19.
Keywords
complement activation
COVID-19
respiratory insufficiency
SARS-CoV-2
==== Body
pmc1 Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of coronavirus disease 2019 (COVID-19), results in heterogeneous symptoms, ranging from mild disease to severe viral pneumonia with acute respiratory distress syndrome, which may require mechanical ventilatory support in some patients and can be fatal [1, 2]. The exact pathogenesis of COVID-19 is not entirely understood but may include direct damage to vascular and epithelial tissues, activation of inflammatory cytokine cascades, and activation of thrombosis [3, 4].
Evidence, including from murine studies on the related Middle East respiratory syndrome and severe acute respiratory syndrome, identified the role of coronaviruses in complement activation, which could be a key contributor to COVID-19 pathogenesis [[5], [6], [7], [8]]. SARS-CoV-2 and the resulting COVID-19 infection also display specific features, including multi-organ failure [6].
Complement activation plays a key role in innate immunity, recognizing and eliminating invading pathogens. There are three independently activated complement pathways, namely the classical pathway (CP), alternative pathway (AP), and lectin pathway (LP), which result in proteolytic processing of different complement components, including C3, C4, and C5 [9]. This stimulates phagocytic cells to clear pathogenic microorganisms and damaged cells, promotes inflammation, and activates the membrane attack complex (MAC) or C5b-9 complex, which can result in cell death and tissue damage. Dysregulation of immune responses, as well as complement and coagulation pathways, leads to inflammation and is implicated in the tissue damage observed in acute respiratory distress syndrome [5, [10], [11], [12]]. Furthermore, studies have shown that patients with COVID-19 demonstrated activation of the complement pathway, which was related to disease severity [13], a generalized thrombotic microvascular injury [10], and decreased levels of C3 plus high sC5b-9 levels, which were associated with poor prognosis and respiratory failure, respectively [14, 15].
Complement pathway components may therefore provide potential prognostic biomarkers to identify the patients most likely to have severe COVID-19 with poor clinical outcomes. Furthermore, blockade of specific complement pathway components may be a potential therapeutic strategy for COVID-19 [16]. Recent therapeutic options in hospitalized patients with COVID-19 have included blocking cytokine interleukin (IL-) 6 to prevent the “cytokine storm” seen in some patients by using sarilumab or tocilizumab [17, 18]. The complement pathway components targeted include C5 inhibitors (e.g., eculizumab and pozelimab), C3 inhibitors (e.g., AMY-101), or LP inhibitors (e.g., narsoplimab) [[19], [20], [21], [22], [23]].
This study comprehensively assessed the levels of complement activation products and full-length proteins in hospitalized patients with varying COVID-19 disease severity. The study also evaluated whether complement pathway markers are associated with poor outcomes in hospitalized COVID-19 patients, including mortality and the requirement for prolonged supplemental oxygen. Moreover, we assessed the thrombotic potential of COVID-19 patient sera ex vivo using pulmonary arterial endothelial cells [24].
2 Methods
2.1 Study design and participants
We included 89 hospitalized patients grouped by baseline disease severity who enrolled in an adaptive, phase 2/3, randomized, double-blind, placebo-controlled trial. Subjects aged ≥18 years, hospitalized with laboratory-confirmed SARS-CoV-2 infection (within 2 weeks of randomization) and COVID-19 pneumonia requiring supplemental oxygen or assisted ventilation, were treated with intravenous (IV) sarilumab (200 mg or 400 mg) or placebo (NCT04315298), as previously described [25]. Local institutional review boards or ethics committees at each center oversaw trial conduct and documentation. All patients provided written informed consent. We also studied 56 healthy control subjects who consented to research, enrolled in phase 1 normal healthy volunteer study (NCT03115996).
Sera were collected from control and hospitalized patients with COVID-19 for cell-based deposition assays. Plasma was evaluated for complement pathway protein analysis at baseline, Day 4, and Day 7, with limited samples available on Days 15 and 29.
2.2 Complement analyses
Multiple complement biomarkers were measured in serum in controls and patients with COVID-19 using commercially available enzyme-linked immunosorbent assay kits (C1Q, C3, C4, C5 complement factor B [CFB], complement factor H [CFH], C4BP [Abcam], C3a, C4a, C5a, sC5-9, AP biomarker factor Bb [Bb; Quidel, San Diego, USA], MBL [Thermo Fisher Scientific, Waltham, USA], and MASP2 [Hycult Biotech, Wayne, PA, USA]), following the manufacturer's instructions (Fig. 1 ).Fig. 1 (a) Schematic showing the measured complement components. Boxplots of 14 complement biomarkers from (b) the common pathway, (c) the CP/LP pathway, and (d) the alternative pathway for COVID-19 patients on study Day 1 (pre-dose), Day 4, and Day 7, and control subjects (n=56) for reference, versus disease severity. Disease severity is noted by dot color and shape. Median values and samples sizes are shown above and below each boxplot, respectively. AP, alternative pathway; Bb, complement factor Bb; C1q, complement component 1q; C3, complement component 3; C4, complement component 4; C4BP, complement component 4 binding protein; C5, complement component 5; CFB, complement factor B; CFH, complement factor H; COVID-19, coronavirus disease 2019; CP, classical pathway; D, day; IV, intravenous; LP, lectin pathway; MASP2, mannan-binding lectin serine protease 2; MBL, mannan-binding lectin; MSOD, multisystem organ dysfunction; PRE, pre-dose; sC5b-9, soluble complement component 5b-9.
Fig. 1
2.3 C5b-9 deposition assay and immunostaining
Serum-complement deposition assay was initially performed using 20 COVID-19 patient samples and 20 healthy donor samples. Eighty-eight serum samples (average age 58 years; 16 severe, 51 critical, 21 multiple system organ dysfunction [MSOD]) from patients with COVID-19 and 20 healthy donors (average age 55 years) were used as additional samples to confirm the observed increase in C5b-9 deposition in COVID-19 when compared to healthy controls.
Human pulmonary microvascular endothelial cells (PromoCell, Heidelberg, Germany) were plated at 20,000 cells per well in 96-well plates in PromoCell C22020 growth media and incubated overnight at 37°C, 5% CO2. The complete growth medium was discarded and replaced with 50% COVID-19 patient serum or healthy donor serum diluted in 0.5% bovine serum albumin-Dulbecco’s modified Eagle’s medium in the presence or absence of 1 mg/mL REGN3918 (pozelimab, a C5 blocking fully human monoclonal antibody) or isotype control. Human C5 depleted serum (Quidel) was used as a negative control. After 4 hours of incubation at 37°C, 5% CO2 supernatants were removed from each well. The cells were fixed with 4% paraformaldehyde for 30 minutes at room temperature, then immunostained with primary rabbit polyclonal anti-human C5b-9 antibody (Abcam: dilution, 1:100) at 4°C overnight. Cells were washed three times with phosphate-buffered saline with Tween-20, then incubated with secondary donkey anti-rabbit conjugated to Alexa Fluor 488 (Thermo Fisher Scientific); nuclei were counterstained with 4′,6-diamidino-2-phenylindole. Images were captured on an Opera Phenix® High-Content Screening System (PerkinElmer, Waltham, USA) using a 20x objective, with 49 sites per well, and fluorescence was quantified using the Harmony® high-content analysis software (PerkinElmer).
2.4 Statistical analyses
Descriptive statistics are reported as median (interquartile range) for continuous variables and frequency (%) for categorical variables. Baseline complement biomarker levels across disease strata (severe, critical, and MSOD) were compared using Kruskal Wallis and post-hoc Dunn tests. Wilcoxon rank-sum tests were used to compare control subjects and patients with COVID-19. Spearman rank correlation (rho; ρ) and Wilcoxon rank-sum tests were used to assess relationships between one complement biomarker and another and demographic, laboratory, and clinical outcomes on Days 1, 4, and 7 (limited data on Day 29) in patients with COVID-19.
Pharmacodynamic effects of sarilumab (200 mg IV [n=34] or 400 mg IV [n=40]) relative to placebo (n=15) on complement biomarkers were assessed using Wilcoxon rank-sum tests of percent change from baseline on Days 4 and 7. Due to limited data, the sarilumab 200 mg (n=34) and 400 mg (n=40) IV treatment arms were pooled.
A Type-I error rate of α = 0.05 was used as the threshold for all statistical tests, with false discovery rate (FDR) correction using the Benjamini-Hochberg method. Nominal p-values are reported and indicated when a test meets the FDR-adjusted threshold (¶). Analyses were conducted using R version 3.6.1.
3 Results
3.1 Study patient demographics, disease severity, and outcomes
Details of the original study population have been previously reported (NCT04315298) [25]. In summary, the adaptive, phase 2/3, randomized, double-blind, placebo-controlled trial enrolled patients hospitalized with COVID-19 who were stratified by disease severity: severe, critical, and MSOD. Patients with severe COVID-19 required low-flow supplemental oxygen. Patients with critical COVID-19 required supplemental oxygen by nonrebreather mask or high-flow nasal device, noninvasive ventilation, or invasive mechanical ventilation and were admitted to the intensive care unit (ICU). Patients with evidence of MSOD required the use of vasopressors, extracorporeal life support, or renal replacement therapy.
The post-hoc complement biomarker analysis included 89 patients hospitalized with COVID-19 (17 severe, 51 critical, 21 MSOD) from the original trial population; 74.2% were male, 40.4% were White, and the median age was 59 years. In this sample, a median time of 2 days elapsed between a positive test result and study enrollment, and a median time of 7 days elapsed between the onset of pneumonia and study enrollment. Overall, 60.7% of patients required invasive mechanical ventilation, with use highest in the MSOD (95.2%) and critical (64.7%) groups. These two groups also had higher concentrations of inflammatory markers (IL-6 and c-reactive protein [CRP]) and viral load relative to the severe group. A comprehensive summary of baseline clinical and laboratory data and patient clinical outcomes is summarized in Supplementary Table 1.
We also studied 56 control subjects enrolled in a phase 1 healthy volunteer study (NCT03115996); 45% were male, 80% were White, and the median age was 34.5 years.
3.2 Complement biomarkers in healthy controls versus patients with COVID-19
Multiple complement activation products were robustly elevated in patients with COVID-19 compared with healthy controls, suggesting activation of all three complement pathways. A few full-length proteins were also differentially modulated in patients with COVID-19 (Fig. 1, Supplementary Fig. 1 and Supplementary Table 2). Nominal p-values are reported, and it is indicated when a test meets the FDR-adjusted threshold (¶).
There was a significant elevation in complement activation products and full-length proteins in patients with COVID-19 compared with controls, in the common pathway (C3, C3a, and C5a; p < 0.0001¶ at Days 1, 4, 7, and 29, sC5b-9 and C5; p < 0.0001¶ at Days 1, 4, and 7), CP (C1q; p = 0.0072 at Day 4, and p = 0.0003¶ at Day 7), LP (mannan-binding lectin [MBL]; p < 0.03 at Days 1, 4, 7, and 29, mannan-binding lectin serine protease 2 [MASP2] and C4a; p < 0.0001¶ at Days 1, 4, and 7, and p < 0.0001¶ at Day 29 for C4a only), AP, (Bb; p < 0.0001¶ at Days 1, 4, and 7), and pathway inhibitors (CFH; p < 0.0001¶ at Days 1, 4, and 7 and p = 0.0131 at Day 29, and C4BP; p < 0.03 at Days 4, 7, and 29) (Supplementary Table 2). The levels of AP-specific full-length protein, CFB, were significantly lower in patients with COVID-19 compared with controls on Days 4, 7, and 29. However, there was no significant difference in the CP/LP-specific full-length protein C4 on Days 1, 4, and 7. The ratios of split product to full-length proteins were still significantly higher in patients with COVID-19, suggesting disproportionately higher activation.
There were no significant differences in complement activation biomarkers among the three COVID-19 severity categories, except for C4a/C4, CFB, and C3 (p < 0.05), which were nominally associated with increasing disease severity (Supplementary Table 3, Fig. 1, and Supplementary Fig. 1). These tests did not reach FDR-adjusted thresholds. Nonsignificant trends also emerged for elevated C4a (p = 0.07) and MASP2 (p = 0.06) with increasing disease severity.
3.3 Correlation of complement biomarkers with each other and with demographic, clinical, and laboratory variables
Correlations between complement activation biomarkers for control subjects and patients with COVID-19 on Days 1, 4, and 7 are shown in Fig. 2 . The AP biomarker Bb was strongly correlated with common pathways split products C3a (ρ = 0.62; p < 0.0001¶), C5a (ρ = 0.44; p < 0.0001), and sC5b-9 (ρ = 0.33; p < 0.01) at Day 1. At Days 4 and 7, both AP biomarker Bb and CP/LP biomarker C4a were correlated with the common pathway split products. Therefore, AP activation could be predominant in the earlier course of the disease. Interestingly, both negative regulators, CFH and C4BP, were correlated with each other at Days 1, 4, and 7.Fig. 2 Spearman correlation matrices of complement biomarkers for (a) control subjects on study Day 1, (b) COVID-19 patients on study Day 1, (c) percent change from baseline on study Day 4, and (d) percent change from baseline on study Day 7. Spearman rho (ρ) correlation coefficients are shown. AP, alternative pathway; Bb, complement factor Bb; C1q, complement component 1q; C3, complement component 3; C4, complement component 4; C4BP, complement component 4 binding protein; C5, complement component 5; CFB, complement factor B; CFH, complement factor H; COVID-19, coronavirus disease 2019; CP, classical pathway; LP, lectin pathway; MASP2, mannan-binding lectin serine protease 2; MBL, mannan-binding lectin; sC5b-9, soluble complement component 5b-9. Values shown are Spearman rho (ρ) correlation coefficients. * indicates nominal statistical significance (p < 0.05) and † indicates statistical significance remains after FDR correction.
Fig. 2
3.3.1 Demographic variables
CFB protein was negatively correlated with age at Day 1 (ρ = -0.32; p < 0.01), and the split products C3a (ρ = 0.35; p < 0.001) and sC5b-9 (ρ = 0.30; p < 0.01) were positively correlated with age at Day 7 (Supplementary Table 4). The C5a and C5a/C5 ratios were consistently higher in Black or African American patients compared with White patients at Days 1, 4, and 7 (p < 0.0001¶).
3.3.2 Clinical variables
Complement activation biomarkers were correlated with multiple variables associated with clinical outcomes, as well as organ injury and inflammation.
The net C5 and CFB activation negatively correlated with days from positive test or symptom duration (Table 1 ). The sC5b-9 levels were significantly higher at Day 4 (p < 0.01) and Day 7 (p < 0.001) in patients who died compared with those who survived and were consistently higher across all time points in patients who did not show clinical improvement compared with those who did. Patients who died had significantly lower MBL levels (p < 0.05) across all time points compared with those who survived, and MBL levels were also inversely correlated with time to oxygen improvement (Fig. 3 , Table 1, and Supplementary Fig. 2). The levels of C3a and Bb were significantly higher at Day 7 (p < 0.001 and p < 0.05, respectively) in patients who died compared with those who survived (Table 1 and Supplementary Fig. 2).Table 1 Association of complement biomarkers with clinical variables in patients with COVID-19 on study Days 1, 4, and 7.
Table 1 Common pathway CP/LP-specific AP-specific CP-specific LP-specific Inhibitors
C5 sC5b-9 C5a C3 C3a C4 C4a CFB Bb C1q MBL MASP2 CFH C4BP
Day 1
Symptoms Days between positive test and enrollment 0.25a –0.11 –0.27a 0.04 0.17 –0.21 0.13 0.04 0.01 0.17 0.08 0.12 0.15 0.30a
Duration of symptoms prior to enrollment 0.10 –0.14 –0.30b –0.12 –0.03 –0.18 –0.15 –0.23a –0.22a 0.20 0.14 –0.06 –0.01 0.16
Recovery All-cause mortalityˆ a
Clinical improvement ˆ a
Lung function Days of hypoxemia –0.01 0.08 0.19 0.06 0.23a –0.22a 0.21 0.02 0.17 –0.14 0.01 0.06 0.08 –0.19
Time to clinical improvement –0.04 0.19 0.09 0.16 0.27a 0.02 0.12 0.01 0.11 –0.05 –0.21 0.17 0.16 –0.10
ICU treatmentˆ a b
Duration of ventilation 0.04 0.18 0.14 0.09 0.32b –0.15 0.30b 0.13 0.24a –0.18 –0.08 0.13 0.07 –0.10
Invasive ventilationˆ b
Time to O2 improvement –0.02 0.19 0.04 0.16 0.07 0.01 0.09 0.15 0.12 –0.02 –0.22a 0.23a 0.14 –0.01
Day 4
Symptoms Days between positive test and enrollment 0.14 –0.23a –0.15 0.30a –0.07 –0.14 0.07 0.12 –0.26a 0.16 0.08 –0.03 –0.03 0.04
Duration of symptoms prior to enrollment 0.01 –0.15 –0.19 0.17 0.16 –0.11 0.01 –0.08 –0.05 0.13 0.15 –0.07 –0.13 0.07
Recovery All-cause mortalityˆ b a a a
Clinical improvementˆ c a
Lung function Days of hypoxemia 0.00 0.18 0.35c 0.01 0.07 –0.16 0.24a 0.21a 0.29b –0.02 0.00 0.05 0.14 –0.14
Time to clinical improvement 0.09 0.33b 0.25a –0.09 0.13 –0.01 0.21 0.02 0.30b 0.19 –0.24a 0.23a 0.10 –0.02
ICU treatmentˆ
Duration of ventilation 0.05 0.25a 0.34b 0.04 0.00 –0.20 0.27a 0.20 0.32b 0.08 –0.12 0.08a 0.14 –0.08
Invasive ventilationˆ a
Time to O2 improvement 0.03 0.24a 0.19 –0.18 0.05 0.07 0.08 0.01 0.25a 0.19 –0.35b 0.24 0.02 0.04
Day 7
Symptoms Days between positive test and enrollment 0.20 –0.02 –0.05 0.46d –0.04 –0.07 0.12 0.12 –0.05 0.05 0.09 0.07 0.15 –0.17
Duration of symptoms prior to enrollment 0.11 0.05 –0.09 0.09 0.23a –0.06 –0.04 0.10 –0.06 –0.16 0.22a –0.03 –0.05 –0.09
Recovery All-cause mortalityˆ c c a *
Clinical improvementˆ c b b *
Lung function Days of hypoxemia 0.16 0.23a 0.25a 0.02 0.06 –0.20 0.23a 0.26a 0.25a 0.04 0.02 0.03 0.13 –0.27a
Time to clinical improvement 0.08 0.49d 0.24* 0.03 0.36c 0.04 0.24a 0.24a 0.41d –0.01 –0.21 0.14 –0.02 –0.27a
ICU treatmentˆ a a a
Duration of ventilation 0.11 0.36c 0.29b 0.11 0.11 –0.15 0.32b 0.31b 0.34b 0.10 –0.10 0.06 0.12 –0.27a
Invasive ventilationˆ a b
Time to O2 improvement –0.10 0.29b 0.06 0.02 0.16 0.07 0.15 0.10 0.20 0.06 –0.32b 0.15 –0.07 –0.12
Ratios
C3a/C3 C5a/C5 sC5b-9/C5 Bb/CFB C4a/C4
Day 1
Symptoms Days between positive test and enrollment 0.09 –0.38c –0.26a 0.00 0.21
Duration of symptoms prior to enrollment –0.01 –0.35b –0.23a –0.15 –0.05
Recovery All-cause mortalityˆ
Clinical improvementˆ
Lung function Days of hypoxemia 0.18 0.16 0.04 0.15 0.30a
Time to clinical improvement 0.21 0.08 0.20 0.09 0.05
ICU treatmentˆ c
Duration of ventilation 0.25a 0.11 0.12 0.19 0.36
Invasive ventilationˆ d
Time to O2 improvement 0.01 0.05 0.19 0.05 0.01
Day 4
Symptoms Days between positive test and enrollment –0.22a –0.20 –0.37c –0.29b 0.17
Duration of symptoms prior to enrollment 0.05 –0.20 –0.15 0.02 0.03
Recovery All-cause mortality ˆ a
Clinical improvement ˆ c a
Lung function Days of hypoxemia 0.07 0.37c 0.24a 0.11 0.23a
Time to clinical improvement 0.19 0.23a 0.34b 0.23a 0.17
ICU treatment ˆ
Duration of ventilation –0.02 0.32b 0.28b 0.15 0.30b
Invasive ventilation ˆ *
Time to O2 improvement 0.14 0.15 0.22a 0.20 0.03
Day 7
Symptoms Days between positive test and enrollment –0.25a –0.09 –0.14 –0.12 0.17
Duration of symptoms prior to enrollment 0.16 –0.12 –0.01 –0.13 0.04
Recovery All-cause mortality ˆ b c
Clinical improvement ˆ b c a
Lung function Days of hypoxemia 0.07 0.23a 0.15 0.11 0.30b
Time to clinical improvement 0.36c 0.24a 0.43d 0.29b 0.22a
ICU treatment ˆ a
Duration of ventilation 0.08 0.28b 0.29b 0.19 0.37c
Invasive ventilation ˆ
Time to O2 improvement 0.16 0.10 0.30b 0.17 0.10
For continuous variables, spearman rho (ρ) and nominal significance level are shown for each correlation. For categorical variables (ˆ), only nominal significance levels are shown for Wilcoxon rank-sum tests. Highlighted cells indicate where ap < 0.05, bp < 0.01, cp < 0.001, dp < 0.0001.
AP, alternative pathway; Bb, complement factor Bb; C1q, complement component 1q; C3, complement component 3; C4, complement component 4; C4BP, complement component 4 binding protein; C5, complement component 5; CFB, complement factor B; CFH, complement factor H; COVID-19, coronavirus disease 2019; CP, classical pathway; CRP, C-reactive protein; ICU, intensive care unit; IL-6, interleukin-6; LDH, lactate dehydrogenase; LP, lectin pathway; MASP2, mannan-binding lectin serine protease 2; MBL, mannan-binding lectin; NLR, neutrophil-lymphocyte ratio; sC5b-9, soluble complement component 5b-9.
Fig. 3 Clinical outcomes for COVID-19 patients on study Day 1 (pre-dose), Day 4, and Day 7, and control subjects (n=56) for reference, for complement factor levels of (a) sC5b-9 and (b) MBL. COVID-19, coronavirus disease 2019; D, day; IV, intravenous; MBL, mannan-binding lectin; PRE, pre-dose; sC5b-9, soluble complement component 5b-9. P-values are shown for biomarker comparisons at Day 1, Day 4, and Day 7 between patients who survived and died using Wilcoxon rank-sum tests.
Fig. 3
Patients who required treatment in the ICU had high C3a (p < 0.05) and C4a (p < 0.01) levels at Day 1. Patients who required invasive ventilation had high C4a (p < 0.01) and C4a/C4 levels (p < 0.0001¶) at Day 1 (Table 1). The time to clinical improvement and duration of ventilation was positively correlated with Bb, C4a, and sC5b-9 at Day 7 (Table 1 and Supplementary Fig. 3).
3.3.3 Laboratory variables
The common pathway activation marker C3a and the CP/LP activation marker C4a were positively correlated with the renal function marker, blood urea nitrogen (ρ = 0.48; p < 0.01 and ρ = 0.37; p < 0.05, respectively; Table 2 ) and serum creatinine (ρ = 0.35; p < 0.001 and ρ = 0.31; p < 0.05, respectively) at Day 1. At later time points (Days 4 and 7), the AP-specific biomarker Bb showed a positive correlation with serum creatinine (ρ = 0.51; p < 0.0001¶ and ρ = 0.40; p < 0.001, respectively).Table 2 Association of complement biomarkers with laboratory variables in patients with COVID-19 on study Days 1, 4, and 7.
Table 2 Common pathway CP/LP-specific AP-specific CP-specific LP-specific Inhibitors
C5 sC5b-9 C5a C3 C3a C4 C4a CFB Bb C1q MBL MASP2 CFH C4BP
Day 1
Kidney function Urea nitrogen 0.06 0.09 0.02 0.33a 0.48b –0.09 0.37a –0.08 0.13 0.17 0.02 0.28 0.12 0.10
Creatinine –0.01 –0.01 0.15 0.20 0.35c 0.04 0.31b –0.04 0.20 0.06 –0.12 0.27a 0.04 0.06
Coag-ulation Platelets 0.29b –0.15 –0.12 0.20 0.08 –0.02 0.05 0.02 –0.16 0.31b 0.14 0.13 0.20 0.23a
Injury & inflammation IL-6 –0.13 0.20 0.21 –0.08 0.38c –0.27a 0.13 0.06 0.41c –0.33b –0.09 –0.06 –0.25a –0.23a
CRP –0.10 0.16 0.18 0.06 0.52d –0.20 0.27a 0.07 0.35b –0.30b –0.03 0.14 –0.18 –0.12
NLR 0.27a 0.27a –0.18 –0.03 0.15 –0.14 0.16 –0.07 –0.20 –0.01 –0.20 0.48d –0.07 0.09
LDH 0.13 0.45c 0.31a 0.00 0.03 –0.17 –0.10 0.20 0.11 –0.10 –0.10 0.05 0.01 –0.19
Virology Viral load –0.04 –0.05 0.03 0.12 –0.16 0.33b 0.26a –0.04 –0.08 –0.09 –0.20 0.26a 0.01 –0.15
Day 4
Kidney function Urea nitrogen 0.35a 0.10 0.18 0.35a 0.19 0.09 0.12 –0.02 0.08 0.38a –0.11 0.11 0.29 –0.05
Creatinine 0.11 0.19 0.22a 0.15 0.24a 0.01 0.27a 0.00 0.51d 0.28b –0.22a 0.12 0.18 0.06
Coag-ulation Platelets 0.12 –0.18 –0.13 –0.05 –0.10 0.07 –0.20 0.06 –0.13 0.16 0.28b 0.02 0.09 0.22a
Injury & inflammation IL-6 –0.16 0.09 0.41c 0.05 0.15 –0.20 –0.10 –0.12 0.28a 0.19 –0.12 0.18 –0.05 –0.15
CRP 0.05 0.12 0.33b 0.20 0.32b –0.07 0.26a 0.19 0.40c –0.09 0.00 0.02 0.00 –0.17
LDH 0.20 0.41c 0.44c –0.22 –0.01 –0.16 0.11 0.02 0.12 0.13 0.04 –0.07 0.01 0.19
Virology Viral load –0.08 –0.05 –0.14 0.22 –0.06 0.12 0.12 –0.09 –0.01 0.04 0.01 –0.03 0.01 –0.28a
Day 7
Kidney function Urea nitrogen 0.09 0.26 0.13 0.05 0.08 0.01 –0.07 –0.32a 0.28 0.09 0.02 –0.22 0.06 0.15
Creatinine 0.03 0.21a 0.19 0.05 0.16 0.09 0.14 –0.10 0.40c 0.14 –0.22a 0.00 –0.04 0.03
Coagu-lation Platelets 0.16 –0.47d –0.22a 0.17 –0.27a 0.06 –0.26a –0.15 –0.25a 0.21 0.18 0.09 0.08 0.17
Injury & inflammation IL-6 0.01 0.18 0.28a –0.02 0.22 0.05 0.16 0.00 0.33b 0.11 –0.17 –0.05 –0.05 –0.20
CRP 0.20 0.34b 0.39c 0.34b 0.57d 0.04 0.38c 0.42c 0.56d –0.09 –0.02 0.32b 0.20 –0.05
LDH –0.01 0.24a 0.23 –0.10 0.25a –0.03 0.20 0.14 0.30a –0.04 –0.08 0.02 –0.11 –0.04
Virology Viral load 0.11 0.04 0.07 0.03 0.12 0.22 0.18 0.17 0.19 0.02 –0.15 0.18 0.36a –0.06
Ratios
C3a/C3 C5a/C5 sC5b-9/C5 Bb/CFB C4a/C4
Day 1
Kidney function Urea nitrogen 0.39a 0.02 –0.01 0.20 0.42b
Creatinine 0.28b 0.15 0.02 0.23a 0.16
Coag-ulation Platelets –0.06 –0.23a –0.32b –0.18 0.10
Injury & inflammation IL-6 0.38c 0.22a 0.23a 0.37c 0.29b
CRP 0.47d 0.19 0.22 0.29b 0.45d
NLR 0.14 –0.26a 0.10 –0.20 0.21
LDH 0.02 0.24 0.32a 0.02 0.12
Virology Viral load –0.18 0.03 0.03 –0.08 –0.04
Day 4
Kidney function Urea nitrogen –0.03 0.08 –0.14 0.08 –0.07
Creatinine 0.11 0.21 0.16 0.48d 0.18
Coag-ulation Platelets –0.04 –0.15 –0.28b –0.15 –0.23a
Injury & inflammation IL-6 0.13 0.45c 0.26a 0.38b 0.04
CRP 0.21 0.33b 0.13 0.26a 0.31b
LDH 0.17 0.41c 0.34b 0.12 0.26a
Virology Viral load –0.24 –0.14 0.01 –0.01 0.00
Day 7
Kidney function Urea nitrogen 0.06 0.13 0.24 0.36a –0.15
Creatinine 0.17 0.18 0.20 0.43d 0.00
Coag-ulation Platelets –0.34b –0.27a –0.55d –0.17 –0.28b
Injury & inflammation IL-6 0.30a 0.30b 0.20 0.32b 0.10
CRP 0.39c 0.35b 0.23 0.31b 0.37b
LDH 0.29a 0.28a 0.25a 0.30a 0.20
Virology Viral load 0.01 0.02 –0.05 0.15 0.13
For continuous variables, spearman rho (ρ) and nominal significance level are shown for each correlation. For categorical variables (ˆ), only nominal significance levels are shown for Wilcoxon rank-sum tests. Highlighted cells indicate where ap < 0.05, bp < 0.01, cp < 0.001, dp < 0.0001.
AP, alternative pathway; Bb, complement factor Bb; C1q, complement component 1q; C3, complement component 3; C4, complement component 4; C4BP, complement component 4 binding protein; C5, complement component 5; CFB, complement factor B; CFH, complement factor H; COVID-19, coronavirus disease 2019; CP, classical pathway; CRP, c-reactive protein; IL-6, interleukin-6; LDH, lactate dehydrogenase; LP, lectin pathway; MASP2, mannan-binding lectin serine protease 2; MBL, mannan-binding lectin; NLR, neutrophil-lymphocyte ratio; sC5b-9, soluble complement component 5b-9.
Platelet number was used to determine the correlation between complement activation components and coagulation. Platelet number was negatively correlated with activation biomarkers from all complement pathways at Day 7, particularly sC5b-9 (ρ = -0.47; p < 0.0001¶; Table 2). Similarly, the inflammatory markers CRP, neutrophil-lymphocyte ratio, and IL-6 were associated with all activation biomarkers specifically at later time-points (Day 7; Table 2 and Supplementary Figs 4, 5, and 6). Common pathway sC5b-9 levels were consistently positively correlated with injury biomarker lactate dehydrogenase (LDH) at all time points. CP/LP-specific C4 levels at Day 1 (ρ = 0.33; p < 0.01) and pathway inhibitor CFH levels at Day 7 (ρ = 0.36; p < 0.05) were positively correlated with viral load.
3.4 Impact of sarilumab treatment and steroid use on complement pathway
The effect of sarilumab treatment on complement activation was evaluated at Days 4 and 7. Sarilumab treatment significantly reduced C3a and C3 levels, but not C3a/C3 ratio relative to placebo (Supplementary Table 5). A significant effect of sarilumab on CFB, C5a, Bb, and MASP2 levels was also observed (Supplementary Table 5 and Supplementary Fig. 7). Although the levels of C3a, C3, C5a, CFB, Bb, and MASP2 decreased with sarilumab treatment, they did not reach the levels observed in healthy controls. Levels of Bb were significantly lower at Day 1 in patients with steroid use after sarilumab treatment was initiated (Supplementary Table 6 and Supplementary Fig. 8). The effect of steroid use in placebo- and sarilumab-treated groups did not have any observed effects on specific complement pathway components (Supplementary Table 6 and Supplementary Fig. 8). Levels of MBL appeared to be lower in those using steroids in the sarilumab treatment group.
3.5 MAC deposition
Levels of MAC deposition on pulmonary endothelial cells ex vivo were determined. Serum from patients with COVID-19 induced higher MAC deposition compared to healthy control serum (Fig. 4 and Supplementary Fig. 9). However, there was no difference in MAC deposition with sera from patients with different disease severity. Treatment with pozelimab, an anti-C5 monoclonal antibody, significantly reduced MAC deposition induced by COVID-19 or healthy serum.Fig. 4 Increased membrane attack complex deposition from COVID-19 serum versus healthy control serum on primary human pulmonary endothelial cells. C5, complement component 5; COVID-19, coronavirus disease 2019; Ctl, control; depl, depleted.
Fig. 4
4 Discussion
Complement pathway activation plays a vital role in the pathogenesis of COVID-19, with dysregulation leading to inflammation and tissue damage [3, 5, 6]. In this study, we comprehensively evaluated the levels of complement activation products and full-length proteins from the three pathways (CP, LP, and AP) in hospitalized patients with varying COVID-19 severity.
Robust complement activation via all pathways was observed in patients with COVID-19, with multiple biomarkers elevated compared to healthy controls. There were no significant differences in complement activation biomarkers among the three COVID-19 severity categories, except for C4a/C4, CFB, and C3. The elevated C4a and C4a/C4 ratios in patients in the critical and MSOD categories suggest higher CP/LP activation in these groups. The high C5a levels were associated with inflammation in COVID-19 patients with acute respiratory distress syndrome, suggesting a C5a-C5aR1 axis in myeloid cell infiltration, and subsequent lung inflammation and endothelialitis [26]. We observed significantly higher C5a levels in hospitalized COVID-19 patients compared with healthy controls. COVID-19 disproportionally affects Black/African American people [27]. Interestingly, in our study, we observed consistently higher C5a levels in Black or African American people compared with White people.
We investigated whether complement pathway markers are associated with poor outcomes in hospitalized COVID-19 patients, including a requirement for prolonged supplemental oxygen and mortality. Complementary to the above findings, patients requiring treatment in the ICU or invasive ventilation also had higher C4a levels. Here, for the first time, we showed that patients with COVID-19 who died had lower MBL levels. MBL has been shown to inhibit SARS-CoV-1 entry into cells and serves as a first line of defense [28]. Whether MBL is required to inhibit SARS-COV2 entry is unknown. Further studies to investigate MBL polymorphisms in these patients are warranted to understand whether there is any association, as low MBL levels have previously been associated with a specific polymorphism [28]. Low MBL levels in our study may also reflect net consumption due to the activation of LP. Recent data showed that circulatory properdin protein levels were lower in patients with severe COVID-19, whereas properdin gene expression was significantly increased [29]. The authors hypothesized that the tissue deposition of properdin could contribute to lower circulating levels of properdin. Hence, it is also possible that the lower levels of MBL in our study could result from tissue deposition. Interestingly, patients who died also had higher sC5b-9 and C3a. However, Bb levels, but not C4a levels, were higher in patients who died. C3 has been shown to play a role in the recovery of patients with COVID-19 [30] with low levels of C3 and C4 linked to increased mortality risk [31, 32]. In this study, we observed increased levels of C3a and sC5b-9, but not C4, and there was no correlation between C4 and all-cause mortality. C3a and C5a can recruit monocytes and macrophages that secrete cytokines like IL-6, contributing to the cytokine storm in some patients with COVID-19 [33].
Complement activation biomarkers (such as C4a and Bb) were correlated with variables/markers of multi-organ dysfunction and injury inflammatory biomarkers. Early in infection, the CP/LP activation marker C4a was positively correlated with kidney function markers, while the AP-specific biomarker Bb was associated at later time points. Additionally, patient serum induced higher MAC deposition, and the sC5b-9 biomarker was consistently associated with LDH level, suggesting MAC-mediated cell injury and a hyper-functional complement system in patients. The higher MAC deposition on the endothelial cells ex vivo with patient sera could also reflect the thrombotic potential [24]. The inflammatory markers IL-6 and CRP were associated with all the complement activation biomarkers, particularly at later time-points. Elevated plasma levels of IL-6 and CRP are associated with clinical worsening and mortality [34, 35, 25].
No major impact of sarilumab treatment on complement activation was observed. Patients who were on sarilumab treatment showed significantly lower C3a and C3 levels, but not C3a/C3 ratio levels, suggesting that the decrease in C3a levels could be secondary to decreased C3 expression and may not be a direct effect on C3 activation [36]. Our data also demonstrated significant, yet modest, decreases in C5a, CFB, Bb and MASP2 levels. However, it remains unclear how sarilumab impacts complement activation and requires further investigation. A previous study has shown that COVID-19 symptoms improved following treatment with sarilumab, corresponding to a rapid decrease in CRP levels [37]. Taken together, the beneficial effect of sarilumab could partly be attributed to mitigating C3a and C5a-induced inflammation [37].
There are a few limitations to our study. Patient numbers were low for each of the COVID-19 disease severity categories. We also did not evaluate mild/moderate patients in our cohort to compare the complement activation with severe categories of patients. Though we could get some insight through correlations with the symptom duration, more studies need to be conducted at the proximal time points to understand the peak complement activation status in the disease course.
In conclusion, our analyses confirm the activation of multiple complement pathways in patients with COVID-19. While no significant differences were observed in complement activation biomarkers across severity (except for C4a/C4, CFB, and C3), these biomarkers were correlated with multiple variables of multi-organ dysfunction and inflammatory injury. The observation that the patients who died had lower MBL levels is exciting and warrants further investigation. As we have shown in this study, serial measurements are required to comprehensively understand the status of complement activation and its association with clinical outcomes and laboratory variables. These results further advance understanding of the disease pathophysiology and may potentially help develop prognostic biomarkers and new therapeutic strategies for severe COVID-19.
Author contributions
KD and LM conceived the concept. KD, LM, AB, SH, PJE, SM, and MW contributed to the study design, analysis plan, and implementation of the research. CH and QR contributed to sample preparation and laboratory testing. AB contributed to primary data acquisition and cleaning for the COVID-19 clinical trial and study samples. PJE, MW, and SCH had access to all data and verified the data and statistical analysis. All authors participated in data analysis, interpretation, manuscript review, and editing.
Funding
Regeneron and Sanofi supported the collection of COVID-19 samples and clinical data. Regeneron and Sanofi were also involved in the study design, collection, analysis and interpretation of data, the writing of this article, and in the decision to submit this article for publication. Certain aspects of this project have been funded in whole or in part with federal funds from the Department of Health and Human Services, Office of the Assistant Secretary for Preparedness and Response, Biomedical Advanced Research and Development Authority, under OT number: HHSO100201700020C.
Data sharing
Qualified researchers can submit a proposal for access to individual patient or aggregate level data from a Regeneron-sponsored clinical trial through Vivli (https://vivli.org/).
Ethical approval statement
Local institutional review boards or ethics committees at each center oversaw trial conduct and documentation. All patients provided written informed consent. We also studied 56 healthy control subjects who consented to research, enrolled in phase 1 normal healthy volunteer study (NCT03115996).
Declaration of Competing Interest
All the authors are employees of Regeneron Pharmaceuticals, Inc. and own stock or stock options.
Appendix A Supplementary data
The following is the Supplementary data to this article:
Acknowledgments
We thank the study participants; their families; and the investigational site members involved in this trial; Georgia Bellingham and Lisa Boersma for operational support for virology testing; the Biomedical Advanced Research and Development Authority; Sanofi; the members of the independent data and safety monitoring committee; Brian Head, PhD, and Caryn Trbovic, PhD, from Regeneron Pharmaceuticals; and Prime, Knutsford, UK for assistance with the development of the manuscript.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.micinf.2022.105081.
==== Refs
References
1 Guan W.J. Ni Z.Y. Hu Y. Liang W.H. Ou C.Q. He J.X. Clinical characteristics of coronavirus disease 2019 in China N Engl J Med 382 2020 1708 1720 32109013
2 Huang C. Wang Y. Li X. Ren L. Zhao J. Hu Y. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China Lancet 395 2020 497 506 31986264
3 Blanco-Melo D. Nilsson-Payant B.E. Liu W.C. Uhl S. Hoagland D. Moller R. Imbalanced host response to SARS-CoV-2 drives development of COVID-19 Cell 181 2020 1036 1045 32416070
4 Loo J. Spittle D.A. Newnham M. COVID-19, immunothrombosis and venous thromboembolism: biological mechanisms Thorax 76 2021 412 420 33408195
5 Gralinski L.E. Sheahan T.P. Morrison T.E. Menachery V.D. Jensen K. Leist S.R. Complement activation contributes to severe acute respiratory syndrome coronavirus pathogenesis mBio 9 2018 017533-e1818
6 Java A. Apicelli A.J. Liszewski M.K. Coler-Reilly A. Atkinson J.P. Kim A.H. The complement system in COVID-19: friend and foe? JCI Insight 5 2020 e140711
7 Jiang Y. Zhao G. Song N. Li P. Chen Y. Guo Y. Blockade of the C5a-C5aR axis alleviates lung damage in hDPP4-transgenic mice infected with MERS-CoV Emerg Microbes Infect 7 2018 77 29691378
8 Perico L. Benigni A. Casiraghi F. Ng L.F.P. Renia L. Remuzzi G. Immunity, endothelial injury and complement-induced coagulopathy in COVID-19 Nat Rev Nephrol 17 2021 46 64 33077917
9 Merle N.S. Church S.E. Fremeaux-Bacchi V. Roumenina L.T. Complement system part I - molecular mechanisms of activation and regulation Front Immunol 6 2015 262 26082779
10 Magro C. Mulvey J.J. Berlin D. Nuovo G. Salvatore S. Harp J. Complement associated microvascular injury and thrombosis in the pathogenesis of severe COVID-19 infection: a report of five cases Transl Res 220 2020 1 13 32299776
11 Tang N. Li D. Wang X. Sun Z. Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia J Thromb Haemost 18 2020 844 847 32073213
12 Zhou F. Yu T. Du R. Fan G. Liu Y. Liu Z. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Lancet 395 2020 1054 1062 32171076
13 de Nooijer A.H. Grondman I. Janssen N.A.F. Netea M.G. Willems L. van de Veerdonk F.L. Complement activation in the disease course of coronavirus disease 2019 and its effects on clinical outcomes J Infect Dis 223 2021 214 224 33038254
14 Fang S. Wang H. Lu L. Jia Y. Xia Z. Decreased complement C3 levels are associated with poor prognosis in patients with COVID-19: a retrospective cohort study Int Immunopharmacol 89 2020 107070
15 Holter J.C. Pischke S.E. de Boer E. Lind A. Jenum S. Holten A.R. Systemic complement activation is associated with respiratory failure in COVID-19 hospitalized patients Proc Natl Acad Sci U S A 117 2020 25018 25025 32943538
16 Risitano A.M. Mastellos D.C. Huber-Lang M. Yancopoulou D. Garlanda C. Ciceri F. Complement as a target in COVID-19? Nat Rev Immunol 20 2020 343 344 32327719
17 Lescure F.-X. Honda H. Fowler R.A. Lazar J.S. Shi G. Wung P. Sarilumab in patients admitted to hospital with severe or critical COVID-19: a randomised, double-blind, placebo-controlled, phase 3 trial Lancet Respir Med 9 2021 522 532 33676590
18 Luo P. Liu Y. Qiu L. Liu X. Liu D. Li J. Tocilizumab treatment in COVID-19: a single center experience J Med Virol 92 2020 814 818 32253759
19 Araten D.J. Belmont H.M. Schaefer-Cutillo J. Iyengar A. Mattoo A. Reddy R. Mild clinical course of COVID-19 in 3 patients receiving therapeutic monoclonal antibodies targeting C5 complement for hematologic disorders Am J Case Rep 21 2020 e927418
20 Mahajan R. Lipton M. Broglie L. Jain N.G. Uy N.S. Eculizumab treatment for renal failure in a pediatric patient with COVID-19 J Nephrol 33 2020 1373 1376 32981025
21 Mastellos D.C. Pires da Silva B.G.P. Fonseca B.A.L. Fonseca N.P. Auxiliadora-Martins M. Mastaglio S. Complement C3 vs C5 inhibition in severe COVID-19: Early clinical findings reveal differential biological efficacy Clinical immunology (Orlando, Fla 220 2020 108598
22 Rambaldi A. Gritti G. Mico M.C. Frigeni M. Borleri G. Salvi A. Endothelial injury and thrombotic microangiopathy in COVID-19: treatment with the lectin-pathway inhibitor narsoplimab Immunobiology 225 2020 152001
23 Latuszek A. Liu Y. Olsen O. Foster R. Cao M. Lovric I. Inhibition of complement pathway activation with pozelimab, a fully human antibody to complement component C5 PloS one 15 2020 e0231892
24 Bettoni S. Galbusera M. Gastoldi S. Donadelli R. Tentori C. Sparta G. Interaction between multimeric von Willebrand factor and complement: a fresh look to the pathophysiology of microvascular thrombosis J Immunol 199 2017 1021 1040 28652401
25 Sivapalasingam S, Lederer DJ, Bhore R, Hajizadeh N, Criner G, Hosain R, et al. Efficacy and safety of sarilumab in hospitalized patients with COVID-19: a randomized clinical trial. Clin Infect Dis ahead of print DOI: 10.1093/cid/ciac153.
26 Carvelli J. Demaria O. Vely F. Batista L. Chouaki Benmansour N. Fares J. Association of COVID-19 inflammation with activation of the C5a-C5aR1 axis Nature 588 2020 146 150 32726800
27 Centers for Disease Control and Prevention. Hospitalization and death by race/ethnicity. https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigations-discovery/hospitalization-death-by-race-ethnicity.html. (Accessed August 16, 2022).
28 Ip W.K. Chan K.H. Law H.K. Tso G.H. Kong E.K. Wong W.H. Mannose-binding lectin in severe acute respiratory syndrome coronavirus infection J Infect Dis 191 2005 1697 1704 15838797
29 Boussier J. Yatim N. Marchal A. Hadjadj J. Charbit B. El Sissy C. Severe COVID-19 is associated with hyperactivation of the alternative complement pathway J Allergy Clin Immunol 149 2022 550 556 e2 34800432
30 Xiao Y. Shi X. She Q. Chen Q. Pan H. Zhang J. Exploration of turn-positive RT-PCR results and factors related to treatment outcome in COVID-19: a retrospective cohort study Virulence 11 2020 1250 1256 32921249
31 Zhao Y. Nie H.X. Hu K. Wu X.J. Zhang Y.T. Wang M.M. Abnormal immunity of non-survivors with COVID-19: predictors for mortality Infect Dis Poverty 9 2020 108 32746940
32 Zinellu A. Mangoni A.A. Serum complement C3 and C4 and COVID-19 severity and mortality: a systematic review and meta-analysis with meta-regression Front Immunol 12 2021 696085
33 Chouaki Benmansour N. Carvelli J. Vivier E. Complement cascade in severe forms of COVID-19: recent advances in therapy Eur J Immunol 51 2021 1652 1659 33738806
34 Lavillegrand J.R. Garnier M. Spaeth A. Mario N. Hariri G. Pilon A. Elevated plasma IL-6 and CRP levels are associated with adverse clinical outcomes and death in critically ill SARS-CoV-2 patients: inflammatory response of SARS-CoV-2 patients Ann Intensive Care 11 2021 9 33439360
35 Boyapati A. Wipperman M.F. Ehmann P.J. Hamon S. Lederer D.J. Waldron A. Baseline severe acute respiratory syndrome viral load is associated with coronavirus disease 2019 severity and clinical outcomes: post hoc analyses of a phase 2/3 trial J Infect Dis 224 2021 1830 1838 34496013
36 Katz Y. Revel M. Strunk R.C. Interleukin 6 stimulates synthesis of complement proteins factor B and C3 in human skin fibroblasts Eur J Immunol 19 1989 983 988 2473911
37 Montesarchio V. Parrela R. Iommelli C. Bianco A. Manzillo E. Fraganza F. Outcomes and biomarker analyses among patients with COVID-19 treated with interleukin 6 (IL-6) receptor antagonist sarilumab at a single institution in Italy Journal for immunotherapy of cancer 8 2020 e001089
| 36494054 | PMC9726657 | NO-CC CODE | 2022-12-08 23:18:17 | no | Microbes Infect. 2022 Dec 7;:105081 | utf-8 | Microbes Infect | 2,022 | 10.1016/j.micinf.2022.105081 | oa_other |
==== Front
J Theor Biol
J Theor Biol
Journal of Theoretical Biology
0022-5193
1095-8541
Elsevier Ltd.
S0022-5193(22)00370-8
10.1016/j.jtbi.2022.111379
111379
Article
Adaptive behaviors and vaccination on curbing COVID-19 transmission: Modeling simulations in eight countries
Li Zhaowan ab
Zhao Jianguo c
Zhou Yuhao a
Tian Lina a
Liu Qihuai ab
Zhu Huaiping d
Zhu Guanghu ab⁎
a School of Mathematics and Computing Science, Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation, Guilin University of Electronic Technology, Guilin, China
b Center for Applied Mathematics of Guangxi (GUET), Guilin, China
c Guangdong Provincial Institute of Public Health, Guangdong Provincial Center for Disease Control and Prevention, Guangzhou, China
d LAMPS and Centre for Diseases Modeling (CDM), Department of Mathematics and Statistics, York University, Toronto, Canada
⁎ Corresponding author at: School of Mathematics and Computing Science, Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation, Guilin University of Electronic Technology, Guilin, China.
7 12 2022
21 2 2023
7 12 2022
559 111379111379
20 6 2022
13 11 2022
2 12 2022
© 2022 Elsevier Ltd. All rights reserved.
2022
Elsevier Ltd
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Current persistent outbreak of COVID-19 is triggering a series of collective responses to avoid infection. To further clarify the impact mechanism of adaptive protection behavior and vaccination, we developed a new transmission model via a delay differential system, which parameterized the roles of adaptive behaviors and vaccination, and allowed to simulate the dynamic infection process among people. By validating the model with surveillance data during March 2020 and October 2021 in America, India, South Africa, Philippines, Brazil, UK, Spain and Germany, we quantified the protection effect of adaptive behaviors by different forms of activity function. The modeling results indicated that (1) the adaptive activity function can be used as a good indicator for fitting the intervention outcome, which exhibited short-term awareness in these countries, and it could reduce the total human infections by 3.68, 26.16, 15.23, 4.23, 7.26, 1.65, 5.51 and 7.07 times, compared with the reporting; (2) for complete prevention, the average proportions of people with immunity should be larger than 90%, 92%, 86%, 71%, 92%, 84%, 82% and 76% with adaptive protection behaviors, or 91%, 97%, 94%, 77%, 92%, 88%, 85% and 90% without protection behaviors; and (3) the required proportion of humans being vaccinated is a sub-linear decreasing function of vaccine efficiency, with small heterogeneity in different countries. This manuscript was submitted as part of a theme issue on “Modelling COVID-19 and Preparedness for Future Pandemics”.
Keywords
COVID-19
Adaptive behavior
Prevention and control
Vaccination
==== Body
pmc1 Introduction
The novel coronavirus (COVID-19) has become a worldwide pandemic and the infection number is keeping increasing (WHO, 2021). As of December 6, 2021, the total number of confirmed cases of COVID-19 in the world is larger than 265 million, with over 5.24 million deaths (WHO, 2021). COVID-19 is disrupting the global economic, political and social systems, which is posing comprehensive threats to population health around the world. To fight against COVID-19, people change their behaviors and are encouraged to take vaccination. Yet the combined effects of virus mutation, limited efficiency of vaccination and high infectivity are bringing great challenges to the prevention and control. Evaluating the effectiveness of intervention strategies and vaccination patterns is a urgent scientific issue.
Under the media coverage and government guidance, the persistent eruption of human infection in COVID-19 triggers a series of protection behaviors, in which people are forced/willing to wear masks, keep social distancing, wash hands frequently and take vaccination (Levin et al., 2021). For example, the US government strengthened border control and restricted incoming travelers since January 2020. After declaring a public health emergency in March, more drastic measures are implemented, including closing school, postponing/canceling big gathering, avoiding international travel, home quarantine. Wearing masks was encouraged in public since April, and mask mandate was enforced and then cancelled during 2021 and 2022. The non-pharmaceutical intervention (NPI) measure were gradually released Since Jun 2020. The UK government imposed stay-at-home order for 6 weeks during March and April, 2020, and many behavioral and social interventions were implemented sine then. From July 2021, UK moved to the final stage of easing restrictions. It was found that all countries have implemented or are implementing intervention strategies in varying degrees to combat COVID-19. These interventions mainly refers to NPIs, including mask wearing, adaptation or closure of school/business, travel restrictions, limits and restrictions on public and private gatherings (Levin et al., 2021). How to design optimized prevention and control measures (considering psychological factors (Brzezinski et al., 2021, Nowak et al., 2020, Petrocchi et al., 2020), behavioral changes (Teslya et al., 2020, Tang et al., 2021) and vaccination strategies (Matrajt et al., 2021, Moore et al., 2021, Chhibber et al., 2022, Han et al., 2021)) have attracted extensive attention recently. Recent studies indicated that (1) surgical masks can prevent the spread of droplets from infectious individuals (Leung et al., 2020); (2) maintaining social distance can reduce the risk of interpersonal communication associated with COVID-19 (Teslya et al., 2020, Anderson et al., 2020, Viner et al., 2020); (3) NPIs can alleviate infection intensity and slow or even contain the variants in COVID-19 (Zhao et al., 2022); and (4) cleaning, hygiene and hand washing can effectively keep away coronavirus (WHO, 2021). However, the estimated effectiveness of vaccination seems to be inconsistent (Moore et al., 2021, Chhibber et al., 2022, Han et al., 2021). Here we went a further step to combine the complex interplay between human protection behavior, vaccination and disease transmission, aiming at providing reasonable intervention strategies under different circumstances. We focused on the following key issues: (1) how to build dynamic equations to describe the interplay, so as to accurately describe the influence of adaptive protection behavior on the mutual checks and balances of the pandemic? (2) how to determine the optimal vaccination coverage, and integrate vaccination and NPI for disease prevention?
To tackle the above issues, we developed an ordinary differential system to simulate the transmission process with two routes (susceptible–exposed–infected–recovered or susceptible–immune), in which vaccination is reflected by shifting parts of susceptible people to those with immunity after a time delay, and adaptive protection behavior is considered by modifying the transmission strength. Here adaptive protection behavior represents the performance people conduct to avoid infections, in which its role is regulated by human awareness, and it is magnified when disease worsens and people become alarmed. Hence it can be taken as adaptive NPI activity. We then validated the model by Markov Chain Monte Carlo (MCMC) method to examine the spread patterns in eight countries with most reported cases (i.e., America, India, South Africa, Philippines, Brazil, UK, Spain and Germany). We finally revealed the influences of adaptive behaviors and vaccination by numerical analysis.
2 Method and materials
2.1 Study area and data
Our study focused on the COVID-19 infection in eight countries, that is the United States (America), India, South Africa, Philippines, Brazil, the United Kingdom (UK), Spain and Germany. The reason of choosing these countries lies in that they have reported the highest numbers of COVID-19 case in the world, and different intervention strategies they adopted can make good comparison. Their total population numbers are 3,267, 1,354, 57, 106, 210, 66, 82 and 46 million, respectively. These countries distribute in North/South America, Asia, Africa, and Europe.
The information of COVID-19 infections and vaccination reported as well as demography in the eight countries during Marth 2020 and October 2021 was used in this study. The cumulative daily numbers of clinical infections in each country were download from the Humanitarian Data Exchange (https://data.humdata.org/dataset/novel-coronavirus-2019-ncov-cases). The daily records of COVID-19 vaccinations and demography in these countries were extracted from the Our World in Data (https://ourworldindata.org/) and Bing surf.
2.2 Modeling framework
In view of the existing knowledge about the clinical progression of COVID-19 infection in humans, as well as the recently implemented control strategies, we developed a new COVID-19 transmission model by ordinary differential equations. The model is based on the following assumptions
• People are divided into the epidemiologically relevant stages for COVID-19 transmission: susceptible (S), latent (E), preclinical infectious (Ip), subclinical infectious (Is), clinical infectious (Ic), recovered (R) and immune due to vaccination (V). Here Ip and Is are inapparent infections, in which Is results in few or no symptoms, and symptoms appear when people evolve from Ip to Ic (Davies et al., 2020). Unlike Ic, they are unlikely to be ascertained by syndromic surveillance (Davies et al., 2020). The sum of these classes equals the total population size, that is, N=S+E+Ip+Is+Ic+R+V. The human population is assumed to be fully susceptible before the introduction of COVID-19 and is kept constant in size throughout the study period.
• Newly infected individuals are generated by the standard mass action formulation. Susceptible people could be infected at rate λ after effective contacts with those who are infectious (Ip,Is and Ic), and then become exposed (in latent state). After an incubation period 1/η, they become either preclinically infectious (with probability κ) or subclinically infectious. Subclinical infections could not be easily found and treated, but they can self-recover after time 1/γ. The preclinical infections appear symptoms after time 1/δ and then enter clinical class. Clinical infections receive treatment and are cured through time 1/ω. After that, they acquire complete immunity upon recovery. It is assumed that the duration of subclinical infection is equal to the sum of duration of preclinical and clinical infection, i.e., 1/γ=1/δ+1/ω.
• During COVID-19 transmission, people usually modify their behavior and take preventative steps to reduce infection risk, such as wearing masks and reducing travel. Individuals’ self-protection awareness usually intensify as the cases increase, and then they will adopt stricter measure against infection. We called such performances as adaptive protection behaviors, which obviously can alleviate infectivity. Their influence weight on infectivity is quantified by variable f, which is a decreasing function of case number (Eksin et al., 2019). If the function f depends on the cumulative (current) cases, it can be called “long-term (short-term) awareness”. The difference of these two kinds of functions are that people take protective measures according to the information of either total infections or current infection (Eksin et al., 2019). Accordingly, we propose the following functions to measure the impacts of people adaptive behavior on disease transmission (2.1) f1=e−kIc,f2=e−k(Ic+ϕR),f3=1−IcNk,f4=1−Ic+ϕRNk.
Here f1 and f3 (f2 and f4) correspond to “short-term awareness” (“long-term awareness”). Function selection in each country is based on their performance in model fitting.
• People can acquire immunity by vaccination, and then they remove from susceptible state to vaccinated state. Since vaccine potency survey showed that people reach protection usually after two weeks of their second injection (Baden et al., 2021), we introduce a time delay to account for the interval between the date that susceptible individual is vaccinated to the date that his/her immune system starts working. Moreover, given that all of the available COVID-19 vaccines cannot completely prevent infection, which may show 94.1% efficacy at preventing Covid-19 illness (Baden et al., 2021), we provide a vaccine protection rate to measure its effectiveness. Hence, at time t, this part of susceptible people ϑθS(t−τ) enters vaccinated class, where ϑ, θ and τ represent vaccine protection rate, vaccination rate and immune delay, respectively.
Accordingly, the essential features of the transmission process are depicted in Fig. 1. The governing equations for simulating the transmission dynamics of COVID-19 are illustrated as follows: (2.2) dS(t)dt=−λfS(t)NIp(t)+αIs(t)+βIc(t)−ϑθS(t−τ),dE(t)dt=λfS(t)NIp(t)+αIs(t)+βIc(t)−ηE(t),dIp(t)dt=κηE(t)−δIp(t),dIs(t)dt=(1−κ)ηE(t)−γIs(t),dIc(t)dt=δIp(t)−ωIc(t),dR(t)dt=ωIc(t)+γIs(t),dV(t)dt=ϑθS(t−τ).
The parameters are presented in Table 1.Fig. 1 Flow diagram of COVID-19 transmission among humans. Related parameters are defined in Table 1.
Table 1 Description of model parameters, with time unit as day or per day.
Parameters Definitions Value Source
1/η Duration of incubation period 7.76 Qin et al. (2020)
λ Dominant infection rate fitting
κ Proportion of the exposed evolving preclinical infections 0.82 Mizumoto et al. (2020)
1/δ Time span from illness onset to be treated 1.5 Kerkhov (2020)
ϑ Vaccine protection rate 0.94 Baden et al. (2021)
θ Vaccination rate varied
1/ω Duration of treatment for symptomatic patients 19 Woodruff (2020)
α Relative infectivity of inapparent infections 1/3 Chen et al. (2020)
β Relative infectivity during treatment 1 Chen et al. (2020)
τ Time from vaccination to vaccine works 14 Baden et al. (2021)
2.3 Model validation
We validated the proposed model by fitting the infection data in eight countries for inferring their transmission patterns. Here we used the MCMC algorithm to estimate the uncertain parameters (i.e., Φ=(λ,k,ϕ,E(0),Ip(0),Ic(0)) in our model. The cumulative cases of reporting and modeling are separately denoted by vectors I=(I1,I2,…,Im) and Λ=(Λ1,Λ2,…,Λm), where m=577 is the study period. Their relationship can be written as Λ=I+ϵ, in which ɛ∼N(0,Σm) is the error matrix with a diagonal matrix of Σm=diag(σ12,σ22,…,σm2). Given the prior information of Φ, the time series of Λ is generated by running the model, and then the likelihood function is computed by: P(Λ|Φ)=(2π)−m/2|Σ|−1/2exp(Λ−I)′(Λ−I)2Σ.
Since the parameters in Φ are conditionally independent, their joint posterior distribution can be written as: (2.3) P(Λ|Φ)∝Πt=1mP(Λ|Φ)P(k)P(ϕ)P(λ)P(E(0))P(Ip(0))P(Ic(0))
We implemented the MCMC algorithm as follows: (1) parameters in Φ were initialized based on their prior information; (2) the predictive cases were produced by running the model, and the posterior distribution P(Ω|Y) was estimated by (2.3); and (3) the values and distribution of Φ were updated through sampling, and the predictive cases were also updated by running the model with a new Φ. In this case, the new Φ is accepted with the probability of min1,P(Ω∗|Y)Q(Ω|Ω∗)P(Ω|Y)Q(Ω∗|Ω),
where the Q(Φ∗|Φ) was the adaptive proposal distribution. After 100,000 iterations, the posterior distributions of Φ were inferred from the final 70% iterations. We examined the performances of the model with different quantification of adaptive behavior, and selected the expression of function fi by the criterion of least fitting error Er=∑t=1m(It−Λt)2/N. The above analysis was realized by employing the deSolve and FME packages in R language software.
To identify the contributions of parameters to the variability of model output, we performed global sensitivity analysis by using Latin Hypercube Sampling (LHS) and partial rank correlation coefficient (PRCC) technique for the input parameters against the simulated infections (Marino et al., 2008). We defined the baseline values of each parameter (see table S1 in Supplementary Information) based on their posterior distribution, and then conducted LHS sampling procedure, which yields 1000 non-overlapping equiprobable intervals for each parameter. Thus, LHS matrix was generated with 1000 rows for the number of simulation (simple size) and 8 columns corresponding to the number of varied parameters. After checking the monotonic relationship between input parameters and output infections, we plug each row of the LHS into the model and conduct simulation, yielding 1000 times series of human infections. Subsequently, we computed the PRCCs between each parameter and the cumulative infection size at each day, and then averaged them over time. Parameters with high absolute PRCC values close to 1 are said to be highly correlated with model output, in which those with negative (positive) PRCC values means that their increase can enlarge (reduce) infections (Marino et al., 2008).
After validating the proposed model, we quantified the transmission patterns of COVID-19 under different scenarios of vaccination and collective behaviors, by assigning the model parameters with different values.
3 Results
Since March 2020, COVID-19 disease began to spread in the studied countries, i.e., America, India, South Africa, Philippines, Brazil, UK, Spain and Germany (the following data are presented in this order), and the infection numbers kept rapidly increasing since then. As of December 2021, the attack rates in these countries were 14%, 3%, 5%, 2%, 10%, 12%, 11% and 5%, respectively.
Based on biological significance, the initial conditions of the model (2.2) were set to be nonnegative, and the right-hand side of the model ensures that its solutions will always stay in the set Ω={(S,E,Ip,Is,Ic,R,V)∈R+7∣0≤S,E,Ip,Is,Ic,R,V≤N}.
The basic reproduction number, R0, is one of the most important theoretical concepts in epidemiology, which can quantify infection potential (Van den Driessche and Watmough, 2002). R0 is interpreted as the average number of secondary cases that are produced by a single primary case in a fully susceptible population (Van den Driessche and Watmough, 2002). We calculated the basic reproduction number R0 by using the theory of next generation matrix, written as R0=ρ(FV−1), where F is the rate of occurring new infections, V is the rate of transferring individuals outside the original group, and ρ represents the spectral radius of matrix (Van den Driessche and Watmough, 2002). Direct calculation yields that F=0λfαλfβλf000000000000,V=η000−κηδ00(κ−1)η0γ00−δ0ω.
It follows from the characteristic equation of FV−1 that the basic reproduction number is given by (3.1) R0=ρ(FV−1)=λfκγω+αλf(1−κ)δω+βλfκδγδγω.
The three components of R0 are separately contributed by the infections in preclinical, subclinical, and clinical states. It is observed that R0 has no connection with vaccination. Based on the fitting parameters (see in the last column of Table 2), the basic reproduction numbers in these countries were estimated at around 1.5–3, with the minimum value (1.46) in the Philippines and the maximum value (3.02) in South Africa.
The fitting results are shown in Table 2, Figs. 2, 4 and S1. It is found that the estimated parameters enable to draw a good fitting capacity of reported cases in these countries, in which the model accounts for larger than 91% of variation in daily data. It is estimated that the infection rate λ is between 0.08 and 0.17, and most people in these countries (except Spain and Germany) mainly exhibited short-term protective behaviors.
The results of sensitivity analysis are shown in Fig. 3. It is observed that the most sensitive parameters are vaccination rate (θ), infection rate (λ) and behavior parameter (k), followed by vaccine protection rate (ϑ). While the initial conditions and long-term behavior parameter (ϕ) have no significant effect on model output. Specifically, the vaccination (infection) rate has overwhelming negative (positive) relation with case number, and such correlation is consistent in all these countries. The behavior parameter in Germany is more sensitive in determining the modeling infection.Fig. 2 The fitting results of the COVID-19 cases in eight countries.
Table 2 Behavioral function, parameter value and error of the best fitting results in eight countries.
Country Function λ k ϕ Error R0
America f3 0.1021 57.579 – 5.505e＋12 2.27
India f1 0.1258 4.2895e−7 – 8.831e＋12 2.27
South Africa f1 0.1677 1.0217e−5 – 2.854e＋10 3.02
Philippines f1 0.0793 1.7474e−6 – 2.029e＋10 1.46
Brazil f3 0.1463 189.014 – 1.602e＋12 2.64
UK f3 0.0929 32.6316 – 4.888e＋11 1.68
Spain f4 0.1078 94.4445 1.3667e−6 7.325e＋10 1.94
Germany f2 0.0965 1.2878e−6 0.033334 1.869e＋10 1.74
Fig. 4 shows the total infection number with the vaccination and adaptive protective behaviors that have already been adopted in these countries. It is observed that if without adaptive behaviors, the total number of human infections (as of October 9, 2021) in America, India, South Africa, Philippines, Brazil, the United Kingdom, Spain and Germany could reach 163.29, 887.72, 44.34, 11.23, 156.64, 13.43, 27.42 and 30.49 million. These were 3.68, 26.16, 15.23, 4.23, 7.26, 1.65, 5.51 and 7.07 times of the reported cases. If without implementing vaccination, human infection number would increase by 65.33%, 14.72%, 2.56%, 13.78%, 4.95%, 137.07%, 42.57% and 42.60%, respectively. It is found that the protective efficacy of adaptive behavior (vaccination) is much more significant in India, South Africa and Brazil (UK and America).Fig. 3 Sensitivity of model parameters to the cumulative cases as indicated by PRCC values, in which only parameters with small P-values (<0.05) are shown.
Fig. 5 shows the effects of different vaccination rates on curbing COVID-19 transmission in eight countries, in case of adaptive behaviors. It is observed that (1) if without vaccination, as of May 2022, the cumulative numbers in these countries would keep increasing and reach 80.84, 60.39, 3.96, 4.74, 34.31, 18.15, 8.67 and 7.18 million, respectively; (2) if the vaccination is carried out from August 2020, with θ=0.001 (i.e., there are 0.1% susceptible individuals to be vaccinated every day), it would not stop the infection in these countries (except Germany), but after 21 months it can reduce total infections by 43.30%, 38.30%, 26.11%, 74.53%, 31.71%, 53.77%, 41.88% and 85.31% (compared to the situation without vaccination), respectively; (3) if the vaccination is implemented from August 2020 and is lasted for 12 months with θ=0.002, it would stop the infection in America, India, Philippines, UK and Spain; and (4) if the vaccination rate θ is larger than 0.003, the infection in all these countries could be prevented after 3−10 months. It is found that the vaccination takes effect more quickly in Germany, Philippines and UK.Fig. 4 The number of cumulative cases as of October 9, 2021, in eight countries in different scenarios.
Fig. 6 shows the impacts of vaccine protection rates on disease evolution with daily vaccination rate θ=0.001. It is observed that the increase of protection rate definitely yields less cases and slower transmission. However, it is impossible to prevent disease transmission in case of small protection rate of vaccine (ϑ<0.2), no matter how long the vaccination is implemented. The epidemic situation in the Philippines and Germany could be controllable when ϑ≥0.2, and in other countries it may need ϑ≥0.4.Fig. 5 Estimated prevalence of COVID-19 infections with different vaccination rate θ. There are two timings for vaccination: from August 2020 or April 2021. The effective protection rate of the vaccine was ϑ=94.1%.
Fig. 7 and Table 3 show how many people should to be vaccinated or acquire immunity in case of different vaccine efficiency (ϑ) and adaptive behaviors, for full control of COVID-19. It is observed that the proportion of population with immunity (denoted by ɛ) is roughly the same with different ϑ in each country. In case of adaptive behaviors, parameter ɛ in these countries should be 90%, 92%, 86%, 71%, 92%, 84%, 82% and 76%, respectively. If without adaptive behaviors, parameter ɛ is a little larger, especially in South Africa and Germany. When vaccine efficiency is ϑ=0.65 and the population has adaptive behavior, the proportions of vaccinated population in these countries should be 119%, 138%, 126%, 106%, 127%, 110%, 111% and 109%, respectively. These data larger than 100% means that some of them have to be vaccinated twice. When vaccine efficiency increases to ϑ=0.85, the above proportions drop to 91%, 105%, 96%, 81%, 98%, 85%, 85% and 83%, respectively. As shown in Fig. 8, it is further observed a sub-linear decrease of vaccination proportion (denoted by ξ) as the vaccine efficiency ϑ increases. Their relationship can be written as quadratic function by regression method, in which ξ is taken the biggest value in India and the smallest value in the Philippines.Fig. 6 Estimated prevalence of COVID-19 infections with different effective protection rate of vaccination. The cumulative numbers of human cases are shown with vaccination from January 2021 and vaccination rate θ=0.001.
Fig. 7 Required proportion of human population that is vaccinated or has immunity in case of different vaccine efficiency and collective behaviors. The simulation time is from March 2020 to October 2021, with vaccination starting from January 2021. The evaluation standard is that the number of new cases in October 2021 will not exceed 100.
Table 3 Requirements for vaccination rate and number of vaccinations (million) in eight countries under different protection rates. Group behavior has self-protection consciousness. X (Y) represents the proportion (number) of population to be vaccinated.
ϑ 60% 65% 70% 75% 80% 85% 90% 95%
America X 1.29 1.19 1.10 1.03 0.96 0.91 0.86 0.81
Y 42.07 38.78 36.06 33.59 31.52 29.69 28.05 27.00
India X 1.49 1.38 1.28 1.20 1.12 1.05 0.99 0.94
Y 202.54 187.05 173.64 161.99 151.85 142.59 134.95 127.93
South Africa X 1.36 1.26 1.17 1.09 1.02 0.96 0.91 0.86
Y 7.83 7.22 6.71 6.26 5.87 5.53 5.22 4.94
Philippines X 1.14 1.06 0.98 0.92 0.86 0.81 0.76 0.72
Y 12.21 11.32 10.50 9.80 9.17 8.63 8.15 7.74
Brazil X 1.38 1.27 1.18 1.10 1.04 0.98 0.92 0.87
Y 29.13 26.89 24.99 23.30 21.86 20.58 19.42 18.40
UK X 1.20 1.10 1.03 0.96 0.90 0.85 0.80 0.76
Y 7.98 7.35 6.85 6.38 5.99 5.63 5.32 5.04
Spain X 1.20 1.11 1.03 0.96 0.90 0.85 0.80 0.76
Y 5.58 5.15 4.78 4.46 4.19 3.94 3.72 3.53
Germany X 1.18 1.09 1.01 0.95 0.88 0.83 0.79 0.75
Y 9.72 8.97 8.33 7.78 7.28 6.86 6.48 6.16
Fig. 8 Relationship between vaccine efficiency and the proportion of human population to be vaccinated for completely preventing COVID-19.
4 Discussion
We have developed a delay differential system to simulate COVID-19 evolution dynamics. We evaluated the roles of adaptive protection behaviors and vaccination situations in curbing COVID-19 infections, by focusing on recent outbreaks in eight countries with most reported cases.
Our model is in accordance with deterministic and compartmental principle, which captures the intrinsic rules of people’s transition across different states. In this model, we measured vaccination by shifting people from susceptible state to immune state. Such tackling technique is similar to existing studies (Moore et al., 2021, Han et al., 2021). Other studies also considered that vaccination can potentially reduce the probability of developing COVID-19 symptoms upon infection and the infectiousness of vaccinated individuals (Matrajt et al., 2021). Here we added a time delay to account for the duration between vaccination and immunity production. Moreover, we quantified the influence of protective behaviors by inserting an activity function f into the infectivity, in which f is a non-linear decreasing expression of infection scale. Since f has time-space heterogeneity and there is no standard reference, inspired by existing studies (Eksin et al., 2019, Xiao et al., 2015), we proposed four formulas of f and chose the specific one according to fitting goodness. The protection behavior can be viewed as adaptive NPI performance. Knowing that adopted intervention would modify the transmission parameters, most existing studies modeled it by a time-varied infectivity (i.e., a piecewise function of time), in which intervention can reduce the infectivity (Tang et al., 2021, Bertuzzo et al., 2020). In such case, more uncertain parameters are needed, which may produce extra fitting difficulty. Here we measured the intervention by activity function f with less uncertain parameters and obtained good fitting, indicating the reliability of our model.
By validating the proposed model to fit the surveillance data in eight countries by MCMC algorithm, we clarified the influencing mechanism of adaptive behavior and vaccination, which offers the following insights for guiding COVID-19 control.
First, the adaptive protection behaviors play a significant role in preventing human infection of COVID-19. It should be noted that all these countries have implemented similar NPI measures from early 2020, and gradually reduced interventions since 2021 to different levels. Our results indicated that if without adaptive behaviors, the total infection in these countries could be 3.68, 26.16, 15.23, 4.23, 7.26, 1.65, 5.51 and 7.07 times as large as reported. Our estimation of NPI outcome is consistent with existing research (Teslya et al., 2020, Tang et al., 2021, Shabat et al., 2021). Hence it is very necessary to maintain media publicity and government guidance during epidemic transmission. Changing adaptive protection behavior may lead to second explosive outbreak. Yet we estimated that the protection yields the feature of short-term response in many countries, which is only dependent on current infections. More comprehensive protection behaviors could results in less cases.
Second, the pattern of vaccination with different efficiency in alleviating COVID-19 infection is further clarified. Promoting vaccination and enhancing vaccine efficiency can quickly and efficiently suppress human infection. To obtain herd immunity, all people should be vaccinated if vaccine efficiency is less than 70%. The average proportions of people with immunity in these countries should be larger than 84% with adaptive protection behaviors or 89% without protection behaviors. Hence no matter how non-pharmaceutical intervention is implemented, collective immunity must reach a high level for disease prevention. Existing studies also claimed that dynamic allocation of vaccines could be a key factor in reducing COVID-19 burden (Matrajt et al., 2021, Chhibber et al., 2022, Han et al., 2021). Moreover, we found that vaccination rate is heavily dependent on vaccine efficiency. That is similar to recent work (Matrajt et al., 2021), in which they indicated that optimal allocation of vaccine vitally depends on the single-dose efficacy (Matrajt et al., 2021). We further clarified the relationship between vaccination rate and vaccine efficiency by quadratic polynomials (see Fig. 8).
Our paper has the following limitations: (1) as an average reflection of collective transmission pattern, the model with fixed parameters is based on deterministic compartmental principle, and the stochastic features originated from parameter diversity and individual difference were not considered. (2) the adaptive behavior function and model parameters could not entirely capture the diversity and heterogeneity of human behaviors and disease transmission in space and time; and (3) our model did not take into account all potential factors (such as virus variant, reinfection, age composition, birth and death), and our analysis relied on fitting results, which may yield certain deviation from reality. It should be noted that reinfection has been detected in some case reports (Ren et al., 2022), and some studies took this into account by using SEIRS model framework (Krueger et al., 2022). Base on current limited information of reinfection (i.e., small reinfection rate, and large duration between the first infection and reinfection), it seems that reinfection has little influence on the transmission process simulated by our model (see Figure S2). Another reason could be the short time of our simulation. Moreover, the vaccination coverage level in our study is usually larger than the R0-dependent threshold (i.e., (1−1/R0)/ϑ). There are two indicators accounting for this, that is, (1) modeling framework: our model considered the time lag between vaccination and immune response, and it involved a nonlinear function to match the effects of human protection behaviors, which played a big role in preventing human infections, but this function became zero when linearizing the model in calculating the next generation matrix; and (2) computation method: the criteria to judge disease elimination in our simulation is that the number of new cases in October 2021 will not exceed 100, which certainly require more vaccination coverage to achieve this criteria in such short time.
In summary, we have constructed an epidemic model for identifying the transmission patterns of COVID-19 in eight countries, with essential impacts of adaptive protection activity and vaccination. We concluded that people behavior and vaccination rate/efficiency are always the key elements that shape the complex occurrence mode and its future trends. People protection behaviors can hold back and prevent infection (particularly in early stage), but vaccination is still the best strategy in the long run.
CRediT authorship contribution statement
Zhaowan Li: Methodology, Validation, Visualization, Data curation, Writing – original draft. Jianguo Zhao: Validation, Data curation, Writing – review & editing. Yuhao Zhou: Software, Investigation, Writing – review & editing. Lina Tian: Data curation, Visualization. Qihuai Liu: Methodology, Resources, Funding acquisition. Huaiping Zhu: Conceptualization, Project administration, Writing – review & editing. Guanghu Zhu: Conceptualization, Methodology, Resources, Funding acquisition, Supervision, Writing – review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary data
The following is the Supplementary material related to this article. MMC S1
Fitting results, parameter distributions used in sensitivity analysis, and simulation with reinfection.
Acknowledgments
This work was jointly supported by the 10.13039/501100001809 National Natural Science Foundation of China (1217116 and 82041021), and the Guangxi Science and Technology Base and Talent Special Project (2021AC06001), China, as well as NSERC of Canada .
Appendix A Supplementary material related to this article can be found online at https://doi.org/10.1016/j.jtbi.2022.111379.
==== Refs
References
Anderson R.M. Heesterbeek H. Klinkenberg D. Hollingsworth T.D. How will country-based mitigation measures influence the course of the COVID-19 epidemic? Lancet 395 10228 2020 931 934 32164834
Baden L.R. El Sahly H.M. Essink B. Kotloff K. Frey S. Novak R. Diemert D. Spector S.A. Rouphael N. Creech C.B. McGettigan J. Khetan S. Segall N. Solis J. Brosz A. Fierro C. Schwartz H. Neuzil K. Corey L. Gilbert P. Janes H. Follmann D. Marovich M. Mascola J. Polakowski L. Ledgerwood J. Graham B.S. Bennett H. Pajon R. Knightly C. Leav B. Deng W. Zhou H. Han S. Ivarsson M. Miller J. Zaks T. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine N. Engl. J. Med. 384 5 2021 403 416 33378609
Bertuzzo E. Mari L. Pasetto D. Miccoli S. Casagrandi R. Gatto M. Rinaldo A. The geography of COVID-19 spread in Italy and implications for the relaxation of confinement measures Nature Commun. 11 1 2020 1 11 31911652
Brzezinski A. Kecht V. Van Dijcke D. Wright A.L. Science skepticism reduced compliance with COVID-19 shelter-in-place policies in the United States Nat. Hum. Behav. 5 11 2021 1519 1527 34646034
Chen Y. Wang A.H. Yi B. Ding K.Q. Wang H.B. Wang J.M. Shi H.B. Wang S.J. Xu G.Z. Epidemiological characteristics of infection in COVID-19 close contacts in Ningbo city Zhonghua Liuxingbingxue Zazhi 41 5 2020 667 671 32447904
Chhibber A. Kharat A. Duong K. Nelson R.E. Samore M.H. Wilson F.A. Chaiyakunapruk N. Strategies to minimize inequity in COVID-19 vaccine access in the US: Implications for future vaccine rollouts Lancet Reg. Health Americas 7 2022 100138
Davies N.G. Kucharski A.J. Eggo R.M. Gimma A. Edmunds W.J. Effects of non-pharmaceutical interventions on COVID-19 cases, deaths, and demand for hospital services in the UK: A modelling study Lancet Public Health 5 7 2020 e375 e385 32502389
Eksin C. Paarporn K. Weitz J.S. Systematic biases in disease forecasting–The role of behavior change Epidemics 27 2019 96 105 30922858
Han S. Cai J. Yang J. Zhang J. Wu Q. Zheng W. Shi H. Ajelli M. Zhou X.H. Yu H. Time-varying optimization of COVID-19 vaccine prioritization in the context of limited vaccination capacity Nature Commun. 12 1 2021 1 10 33397941
Kerkhov V. Report of the WHO-China Joint Mission on Coronavirus Disease 2019(COVID-19)[EB/OL] WHO-China Joint Mission on COVID-19 2020
Krueger T. Gogolewski K. Bodych M. Gambin A. Giordano G. Cuschieri S. Czypionka T. Perc M. Petelos E. Rosińska M. Szczurek E. Risk assessment of COVID-19 epidemic resurgence in relation to SARS-CoV-2 variants and vaccination passes Commun. Med. 2 1 2022 1 14 35603280
Leung N.H.L. Chu D.K.W. Shiu E.Y.C. Chan K.H. McDevitt J.J. Hau B.J.P. Yen H.L. Li Y. Ip D.K.M. Peiris J.S.M. W.H. Seto Leung G.M. Milton D.K. Cowling B.J. Respiratory virus shedding in exhaled breath and efficacy of face masks Nat. Med. 26 5 2020 676 680 32371934
Levin R. Chao D.L. Wenger E.A. Proctor J.L. Insights into population behavior during the COVID-19 pandemic from cell phone mobility data and manifold learning Nat. Comput. Sci. 1 9 2021 588 597
Marino S. Hogue I.B. Ray C.J. Kirschner D.E. A methodology for performing global uncertainty and sensitivity analysis in systems biology J. Theoret. Biol. 254 1 2008 178 196 18572196
Matrajt L. Eaton J. Leung T. Dimitrov D. Schiffer J.T. Swan D.A. Janes H. Optimizing vaccine allocation for COVID-19 vaccines shows the potential role of single-dose vaccination Nature Commun. 12 1 2021 1 18 33397941
Mizumoto K. Kagaya K. Zarebski A. Chowell G. Estimating the asymptomatic proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship, Yokohama, Japan, 2020 Sci. Adv. 25 10 2020 2000180
Moore S. Hill E.M. Tildesley M.J. Dyson L. Keeling M.J. Vaccination and non-pharmaceutical interventions for COVID-19: A mathematical modelling study Lancet Infect. Dis. 21 6 2021 793 802 33743847
Nowak B. Brzóska P. Piotrowski J. Sedikides C. Żemojtel-Piotrowska M. Jonason P.K. Adaptive and maladaptive behavior during the COVID-19 pandemic: The roles of dark triad traits, collective narcissism, and health beliefs Pers. Individ. Differ. 167 2020 110232
Petrocchi S. Levante A. Bianco F. Castelli I. Lecciso F. Maternal distress/coping and children’s adaptive behaviors during the COVID-19 lockdown: Mediation through children’s emotional experience Front. Public Health 8 2020
Qin J. You C. Lin Q. Hu T. Yu S. Zhou X.H. Estimation of incubation period distribution of COVID-19 using disease onset forward time: A novel cross-sectional and forward follow-up study Sci. Adv. 6 33 2020 eabc1202 32851189
Ren X. Zhou J. Guo J. Hao C. Zheng M. Zhang R. Huang Q. Yao X. Li R. Jin Y. Reinfection in patients with COVID-19: A systematic review Global Health Res. Policy 7 1 2022 1 20
Shabat M. Shafir R. Sheppes G. Flexible emotion regulatory selection when coping with COVID-19-related threats during quarantine Sci. Rep. 11 1 2021 1 10 33414495
Tang B. Xia F. Bragazzi N.L. McCarthy Z. Wang X. He S. Sun X. Tang S. Xiao Y. Wu J. Lessons drawn from China and South Korea for managing COVID-19 epidemic: Insights from a comparative modeling study ISA Trans. 2021
Teslya A. Pham T.M. Godijk N.G. Kretzschmar M.E. Rozhnova G. Impact of self-imposed prevention measures and short-term government-imposed social distancing on mitigating and delaying a COVID-19 epidemic: A modelling study PLoS Med. 17 7 2020 e1003166
Van den Driessche P. Watmough J. Reproduction numbers and sub-threshold endemic equilibria for compartmental models of disease transmission Math. Biosci. 180 1–2 2002 29 48 12387915
Viner R.M. Russell S.J. Croker H. Packer J. Ward J. Stansfield C. Mytton O. Bonell C. Booy R. School closure and management practices during coronavirus outbreaks including COVID-19: A rapid systematic review Lancet Child Adolesc. Health 4 5 2020 397 404 32272089
WHO Coronavirus disease (COVID-19) pandemic 2021 https://www.who.int/emergencies/diseases/novel-coronavirus-2019
Woodruff A. COVID-19 follow up testing J. Infection 81 4 2020 647 679
Xiao Y. Tang S. Wu J. Media impact switching surface during an infectious disease outbreak Sci. Rep. 5 1 2015 1 9
Zhao S. Wang K. Chong M.K.C. Musa S.S. He M. Han L. He D. Wang M.H. The non-pharmaceutical interventions may affect the advantage in transmission of mutated variants during epidemics: A conceptual model for COVID-19 J. Theoret. Biol. 542 2022 111105
| 36496185 | PMC9726658 | NO-CC CODE | 2022-12-14 23:54:37 | no | J Theor Biol. 2023 Feb 21; 559:111379 | utf-8 | J Theor Biol | 2,022 | 10.1016/j.jtbi.2022.111379 | oa_other |
==== Front
J Am Med Dir Assoc
J Am Med Dir Assoc
Journal of the American Medical Directors Association
1525-8610
1538-9375
Published by Elsevier Inc. on behalf of AMDA -- The Society for Post-Acute and Long-Term Care Medicine.
S1525-8610(22)00921-5
10.1016/j.jamda.2022.11.024
Original Studies
T-cell mediated response after primary and booster SARS-CoV-2 mRNA vaccination in nursing homes residents
Schiavoni Ilaria 1∗
Palmieri Annapina 2∗
Olivetta Eleonora 3
Leone Pasqualina 1
Di Lonardo Anna 2
Mazzoli Alessandra 4
Cafariello Carmine 4
Malara Alba 5
Palamara Anna Teresa 1
Incalzi Raffaele Antonelli 6
Onder Graziano 2
Stefanelli Paola 1§
Fedele Giorgio 1ˆ§
on behalf of the
the GeroCovid Vax CMI Study GroupAmici Lucia
Berardi Francesca
Bernardi Riccardo
Cardillo Mario
Cobani Anila
Confessore Ida
Fiorucci Claudia
Guerriero Serena
Kountsevitch Liudmila
Leccese Vincenzo
Ruocco Federica
Sabino Pasquale
Sciarretta Antonio
Spaccaferro Deborah
Spinelli Luciana
Ursino Rita
Viotti Romina
1 Department of Infectious Diseases, Istituto Superiore di Sanità, Rome, Italy
2 Department of Cardiovascular, Endocrine-Metabolic Diseases and Aging, Istituto Superiore di Sanità, Rome, Italy
3 National Center for Global Health, Istituto Superiore di Sanità, Rome, Italy
4 Geriatrics Outpatient Clinic and Territorial Residences, Italian Hospital Group, Rome, Italy
5 ANASTE Humanitas Foundation, Rome, Italy
6 Geriatrics Unit, Department of Medicine, Campus Bio-Medico University and Teaching Hospital, Rome
ˆ Corresponding author: Giorgio Fedele, PhD. VPD – Reference Labs unit. Department of Infectious Diseases. Istituto Superiore di Sanità. Viale Regina Elena 299, 00161 Rome, Italy. +39 0649902890;
∗ Equally contributed to this work as Co-first authors -
§ Equally contributed to this work as Co-Last authors
7 12 2022
7 12 2022
29 7 2022
21 11 2022
29 11 2022
© 2022 Published by Elsevier Inc. on behalf of AMDA -- The Society for Post-Acute and Long-Term Care Medicine.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objectives
Nursing home (NH) residents have been significantly affected by the COVID-19 pandemic. Studies addressing the immune responses induced by COVID-19 vaccines in NH residents have documented a good post-vaccination antibody response and the beneficial effect of a third booster vaccine dose. Less is known about vaccine-induced activation of cell mediated immune response in frail elderly subjects in the long term. The aim of the present study is to monitor mRNA SARS-CoV-2 vaccine-induced T-cell responses in a sample of Italian NH residents who received primary vaccine series and a third booster dose and to assess the interaction between T-cell responses and humoral immunity.
Design
Longitudinal cohort study.
Setting and Participants
Thirty-four residents vaccinated with BNT162b2 mRNA SARS-CoV-2 vaccine between February and April 2021 and who received a third BNT162b2 booster dose between October and November 2021 were assessed for vaccine-induced immunity 6 (pre-booster) and 12 (post-booster) months after the first BNT162b2 vaccine dose.
Methods
Pre- and post-booster cell-mediated immunity was assessed at by intracellular cytokine staining of peripheral blood mononuclear cells stimulated in vitro with peptides covering the immunodominant sequence of SARS-CoV-2 Spike protein. The simultaneous production of IFN-γ, TNF-α and IL-2 was measured. Humoral immunity was assessed in parallel by measuring serum concentration of anti-trimeric Spike IgG antibodies.
Results
Before the booster vaccination, 31 out of 34 NH residents had a positive cell-mediated immunity (CMI) response to Spike. Post-booster, 28 out of 34 had a positive response. Residents without a previous history of SARS-CoV-2 infection who had a lower response prior the booster administration, showed a greater increase of T-cell responses after the vaccine booster dose Humoral and cell-mediated immunity were, in part, correlated but only before booster vaccine administration.
Conclusions and Implications
The administration of the booster vaccine dose restored Spike-specific T-cell responses in SARS-CoV-2 naïve residents who responded poorly to the first immunization, while a previous SARS-CoV-2 infection had an impact on the magnitude of vaccine-induced cell-mediated immunity at earlier time-points. Our findings imply the need for a continuous monitoring of the immune status of frail NH residents to adapt future SARS-CoV-2 vaccination strategies.
Key words
SARS-CoV-2
COVID-19 vaccines
nursing homes
cell-mediated immunity
vaccine booster
==== Body
pmcFunding: The GeroCovid Vax Study was funded by a grant from the Italian Medicines Agency (Agenzia Italiana del Farmaco, AIFA – resolution n 14 – feb 4, 2021)
Conflict of interest: None
Brief summary
Nursing home residents showed a durable cell-mediated immunity after the receipt of mRNA COVID-19 vaccine. The benefit of the third booster dose was especially evident in residents without a previous history of infection.
GeroCovid Vax CMI Study Group:
Lucia Amici, Francesca Berardi, Riccardo Bernardi, Mario Cardillo, Anila Cobani, Ida Confessore, Claudia Fiorucci, Serena Guerriero, Liudmila Kountsevitch, Vincenzo Leccese, Federica Ruocco, Pasquale Sabino, Antonio Sciarretta, Deborah Spaccaferro, Luciana Spinelli, Rita Ursino, Romina Viotti:
Italian Hospital Group, Via Tiburtina, 188 - 00012; Guidonia-Montecelio (RM), Italy
Roberta Granata, Manuela Stefanelli:
Villa Sacra Famiglia - Italian Hospital Group, Largo Ottorino Respighi, 6 - 00135 Rome, Italy
| 0 | PMC9726683 | NO-CC CODE | 2022-12-08 23:18:17 | no | J Am Med Dir Assoc. 2022 Dec 7; doi: 10.1016/j.jamda.2022.11.024 | utf-8 | J Am Med Dir Assoc | 2,022 | 10.1016/j.jamda.2022.11.024 | oa_other |
==== Front
Cancer Epidemiol
Cancer Epidemiol
Cancer Epidemiology
1877-7821
1877-783X
Elsevier Ltd.
S1877-7821(22)00218-1
10.1016/j.canep.2022.102313
102313
Article
Japanese cancer screening programs during the COVID-19 pandemic: changes in participation between 2017–2020
Machii Ryoko ⁎
Takahashi Hirokazu
Division of Quality Assurance Programs, Institute for Cancer Control, National Cancer Center, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan
⁎ Corresponding author. Fax: +03-3547-5350.
7 12 2022
7 12 2022
1023136 10 2022
21 11 2022
5 12 2022
© 2022 Elsevier Ltd. All rights reserved.
2022
Elsevier Ltd
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
The impact of the coronavirus disease 2019 (COVID-19) pandemic on cancer screening participation is a global concern. A national database of screening performance is available in Japan for population-based cancer screening, estimated to cover approximately half of all cancer screenings.
Methods
Utilizing the fiscal year (FY) 2017–2020 national database, the number of participants in screenings for gastric cancer (upper gastrointestinal [UGI] series or endoscopy), colorectal cancer (fecal occult blood test), lung cancer (chest X-ray), breast cancer (mammography), and cervical cancer (Pap smear) were identified. The percent change in the number of participants was calculated.
Results
Compared with the pre-pandemic period (FY 2017–2019), in percentage terms FY 2020 recorded the largest decline in gastric cancer UGI series (2.82 million to 1.91 million, percent change was -32.2%), followed by screening for breast cancer (3.10 million to 2.57 million, percent change was -17.2%), lung cancer (7.92 million to 6.59 million, percent change was -16.7%), colorectal cancer (8.42 million to 7.30 million, percent change was -13.4%), cervical cancer (4.26 million to 3.77 million, percent change was -11.6%), and gastric cancer via endoscopy (1.02 million to 0.93 million, percent change was -9.0%).
Conclusion
The number of participants in population-based screenings in Japan decreased by approximately 10%–30% during the pandemic. The impact of these declines on cancer detection or mortality should be carefully monitored.
Keywords
Cancer Screening
Population-based Cancer Screening
COVID-19
SARS-CoV-2
Coronavirus Disease
Cancer Prevention
==== Body
pmc1 INTRODUCTION
Cancer screening in Japan is mainly divided into population-based cancer screenings conducted by local governments and worksite-based cancer screenings conducted by business owners for employees. A large-scale questionnaire survey conducted by the central government every three years estimated that about the same number of people who underwent worksite-based cancer screenings participated in population-based cancer screenings [1]. Currently, population-based cancer screening is the only national cancer control program, and it is implemented based on the guidelines of the Ministry of Health, Labor and Welfare (MHLW). Only the results of the population-based cancer screenings are collected annually by the MHLW from the local government and published as a national database on their website [2].
In a population-based cancer screening, the following programs are recommended by the MHLW [3]: gastric cancer screening (upper gastrointestinal [UGI] or endoscopy, every 2 years for individuals aged ≥50 years; annual UGI is also available for individuals aged 40–49 years), colorectal cancer screening (annual fecal occult blood test [FOBT] for individuals aged ≥40 years), lung cancer screening (annual chest x-ray for individuals aged ≥40 years), breast cancer screening (mammography, every 2 years for individuals aged ≥40 years), and cervical cancer screening (Pap smear, every 2 years for individuals aged ≥20 years). Population-based cancer screening is divided into two types of screening system, one is conducted at large-scale facilities for large groups (mass screening), and the other at local medical facilities (individual screening). Local governments set limits on the number of participants for each type according to screening capacity, and residents are free to choose either type.
A national database of population-based screenings is constructed every fiscal year [FY]. This database includes the number of participants, and this is identifiable by the national total, local government, sex, age, and type of screening (mass screening, individual screening).
Globally, the number of cancer screening participants decreased significantly during the pandemic period, raising concerns about delays in cancer diagnosis and treatment [4], [5], [6]. In FY 2020, the Japanese government declared a state of emergency nationwide only once from April 7 to May 25 [7]. The MHLW had requested local governments to temporarily suspend or postpone population-based screenings during the declaration period, and compliance was dependent on each municipality. The aim of this report was to evaluate the changes in the number of population-based cancer screening participants during the COVID-19 pandemic in Japan, using national data.
2 METHODS
2.1 Data sources
Based on the national database [2], we identified the total number of participants nationwide from FY 2017 to FY 2020. The pandemic period and control period were considered to be FY 2020 (April 2020–March 2021) and FY 2017–2019 (April 2017–March 2020), respectively. The subjects of analysis were five cancer screenings recommended by the MHLW, and gastric cancer screening was performed separately by UGI and endoscopy (making six screening types in total). Furthermore, the number of participants nationwide for each cancer was identified by sex, age group, and screening type.
This study used publicly available open data sources [2], and thus did not require informed consent. As mentioned below in the limitations of this study, because it is currently not possible to calculate the actual cancer screening rate in Japan, we used the number of participants in this study.
2.2 Descriptive statistics
The number of participants in each year was summarized by the national total, sex, age group, and type of screening. The percentage change in the number of participants in each screening during the pandemic period (FY 2020), compared with the control period (FY2017-2019), was calculated as follows:(Number of participants in FY 2020 – Average number of participants in the past 3 years) / Average number of participants in the past 3 years.
Participant numbers may have changed significantly in FY2017-2019, for sensitivity analysis, the percent change in participant numbers from 2019 to 2020 was calculated as follows:(Number of participants in FY 2020 – Number of participants in FY 2019) / Number of participants in FY 2019.
3 RESULTS
Table 1 shows the number of participants in FY 2020 and the average number of participants in the previous 3 years by sex, age group, and screening type. The number of participants in FY 2017–2019 was almost the same for each cancer type (Supplementary file). Compared with that before the pandemic, the total number of participants in FY 2020 decreased from 2.82 million to 1.91 million (gastric cancer, UGI), 1.02 million to 0.93 million (gastric cancer, endoscopy), 8.42 million to 7.30 million (colorectal cancer), 7.92 million to 6.59 million (lung cancer), 3.10 million to 2.57 million (breast cancer), and 4.26 million to 3.77 million (cervical cancer). The percentage change was the largest for gastric cancer (UGI, -32.2%), followed by breast cancer (-17.2%), lung cancer (-16.7%), colorectal cancer (-13.4%), cervical cancer (-11.6%), and gastric cancer (endoscopy, -9.0%) ( Fig. 1).Table 1 Participation in cancer screening programs in Japan during fiscal year (FY) 2017-2020 (percentages in parentheses).
Table 1 FY 2017-2019, avg. FY 2020 Percentage change
Gastric cancer (UGI or Endoscopy)
Total 3837143 2837083 -26.1
Sex Male 1678154 (43.7) 1263891 (44.5) -24.7
Female 2158989 (56.3) 1573192 (55.5) -27.1
Age, y 40-49 411252 (10.7) 280183 (9.9) -31.9
50-59 539951 (14.1) 423897 (14.9) -21.5
60-69 1217547 (31.7) 814228 (28.7) -33.1
70-79 1327201 (34.6) 1046644 (36.9) -21.1
80- 341191 (8.9) 272131 (9.6) -20.2
Type of screening Mass screening 2143680 (55.9) 1412663 (49.8) -34.1
Individual screening 1693463 (44.1) 1424420 (50.2) -15.9
Gastric cancer (UGI)
Total 2819623 1910660 -32.2
Sex Male 1248063 (44.3) 863923 (45.2) -30.8
Female 1571561 (55.7) 1046737 (54.8) -33.4
Age, y 40-49 411252 (14.6) 280183 (14.7) -31.9
50-59 392302 (13.9) 273295 (14.3) -30.3
60-69 895957 (31.8) 549058 (28.7) -38.7
70-79 910524 (32.3) 659832 (34.5) -27.5
80- 209588 (7.4) 148292 (7.8) -29.2
Type of screening Mass screening 2125243 (75.4) 1396841 (73.1) -34.3
Individual screening 694380 (24.6) 513819 (26.9) -26.0
Gastric cancer (Endoscopy)
Total 1017519 926423 -9.0
Sex Male 430091 (42.3) 399968 (43.2) -7.0
Female 587428 (57.7) 526455 (56.8) -10.4
Age, y 50-59 147649 (14.5) 150602 (16.3) 2.0
60-69 321590 (31.6) 265170 (28.6) -17.5
70-79 416677 (41.0) 386812 (41.8) -7.2
80- 131603 (12.9) 123839 (13.4) -5.9
Type of screening Mass screening 18437 (1.8) 15822 (1.7) -14.2
Individual screening 999082 (98.2) 910601 (98.3) -8.9
Colorectal cancer (FOBT)
Total 8424776 7298673 -13.4
Sex Male 3321617 (39.4) 2893898 (39.6) -12.9
Female 5103159 (60.6) 4404775 (60.4) -13.7
Age, y 40-49 856762 (10.2) 693443 (9.5) -19.1
50-59 953545 (11.3) 816647 (11.2) -14.4
60-69 2368211 (28.1) 1802854 (24.7) -23.9
70-79 3086488 (36.6) 2874921 (39.4) -6.9
80- 1159770 (13.8) 1110808 (15.2) -4.2
Type of screening Mass screening 3569051 (42.4) 2787198 (38.2) -21.9
Individual screening 4855726 (57.6) 4511475 (61.8) -7.1
Lung cancer (Chest X-ray)
Total 7918697 6593528 -16.7
Sex Male 3198703 (40.4) 2679184 (40.6) -16.2
Female 4719994 (59.6) 3914344 (59.4) -17.1
Age, y 40-49 710592 (9.0) 556306 (8.4) -21.7
50-59 778708 (9.8) 639820 (9.7) -17.8
60-69 2189666 (27.7) 1571519 (23.8) -28.2
70-79 3010602 (38.0) 2684477 (40.7) -10.8
80- 1229129 (15.5) 1141406 (17.3) -7.1
Type of screening Mass screening 4409525 (55.7) 3163019 (48.0) -28.3
Individual screening 3509172 (44.3) 3430509 (52.0) -2.2
Breast cancer (Mammography)
Total 3100533 2565900 -17.2
Age, y 40-49 879936 (28.4) 727345 (28.3) -17.3
50-59 681034 (22.0) 586869 (22.9) -13.8
60-69 835959 (27.0) 633753 (24.7) -24.2
70-79 607644 (19.6) 530876 (20.7) -12.6
80- 95960 (3.1) 87057 (3.4) -9.3
Type of screening Mass screening 1495147 (48.2) 1105214 (43.1) -26.1
Individual screening 1605385 (51.8) 1460686 (56.9) -9.0
Cervical cancer (Pap smear)
Total 4260272 3767370 -11.6
Age, y 20-29 380885 (8.9) 397995 (10.6) 4.5
30-39 781530 (18.3) 702443 (18.6) -10.1
40-49 939360 (22.0) 831412 (22.1) -11.5
50-59 722492 (17.0) 655134 (17.4) -9.3
60-69 800719 (18.8) 618666 (16.4) -22.7
70-79 551198 (12.9) 484993 (12.9) -12.0
80- 84088 (2.0) 76727 (2.0) -8.8
Type of screening Mass screening 1195295 (28.1) 856012 (22.7) -28.4
Individual screening 3064977 (71.9) 2911358 (77.3) -5.0
Fig. 1 The percentage decrease in the number of screening participants during the pandemic (FY 2020) compared to the FY 2017-2019 average.
Fig. 1
Before and during the pandemic, the number of female participants was approximately 1.2 to 1.5 times higher than that of male participants. The decrease in the number of participants during the pandemic period was larger in females for gastric, colorectal, and lung cancers (percentage change ranged from -30.8% to -7.0% for males and -33.4% to -10.4% for females). By age group, the largest number of participants were in their 70 s (gastric, colorectal, and lung) and 40 s (breast and cervical) both before and during the pandemic. The decrease in the number of participants during the pandemic period was the largest for individuals in their 60 s for all cancers (percentage change ranged from -38.7 to -17.5% for those in their 60 s and -31.9% to 4.5% for other age groups). By screening type, before and during the pandemic, the participation rate in mass screening was the highest for gastric cancer via UGI (73.1–75.4%). The decrease in the number of participants during the pandemic period was larger for mass screening of all cancers (percentage change ranged from -34.3 to -14.2% in mass screening and -26.0 to -2.2% in individual screening). Among all population-based cancer screenings, the largest reduction in the number of participants was noted in mass screening for gastric cancer via UGI (percentage change was -34.3%).
Supplementary Table shows the result of sensitivity analysis. During FY 2017-2019, the total number of participants decreased from 3.04 million to 2.59 million (gastric cancer, UGI), 8.47 million to 8.35 million (colorectal cancer), 7.94 million to 7.87 million (lung cancer), and 4.29 million to 4.23 million (cervical cancer), while the total number of participants increased from 0.90 million to 1.11 million (gastric cancer, endoscopy) and 3.08 million to 3.11 million (breast cancer). The percentage of decrease in the number of screening participants during the pandemic (FY 2020) compared to FY 2019 was the largest for gastric cancer (UGI, -26.1%), followed by breast cancer (-17.5%), gastric cancer (endoscopy, -16.6%), lung cancer (-16.2%), colorectal cancer (-12.6%), and cervical cancer (-10.8%).
4 DISCUSSION
This study revealed that the COVID-19 pandemic was associated with an approximately 10–30% decrease in the number of participants in Japanese population-based screenings. In percentage terms, gastric cancer screening had the greatest decrease in the number of participants. According to a previous study using a hospital-based cancer registry covering 70% of newly diagnosed cancers in Japan, of cancers diagnosed by screening in 2020 (by population-based screening and other screening types), gastric cancer decreased the most since the pre-pandemic period [8]. In our report, the number of participants declined the most in mass screenings for gastric cancer UGI. This is probably because this number was large even before the pandemic, and many local governments were unable to implement the necessary COVID-19 protective precautions in a short period of time to safely carry out mass screenings. The recommendation to temporarily postpone mass gastric cancer screenings by related academic societies may also have influenced the decisions of local governments [9].
Participants in some cancer screenings have been declining before 2020, so part of the decrease in FY2020 would probably also have taken place without COVID-19. Sensitivity analysis comparing FY 2020 and FY 2019 were generally similar to the main results (FY 2020 vs FY2017-2019): reductions in the number of participants for five cancers ranged from approximately 10-26%, with gastric cancer showing the largest percentage reduction.
The gap of the percentage reduction in the two comparisons (FY2020 vs FY2017-2019, FY2020 vs FY2019), was largest for gastric cancer screening. Compared to FY 2019 and FY 2017-2019, the percentage of decrease in 2020 were -26.1% and -32.2% for UGI and -16.6% and -9.0% for endoscopy, respectively. These gaps are due to a decrease in participation in UGI screening and an increase in participation in endoscopic screening over the three years to 2020. Since endoscopic screening started in 2016 in addition to UGI screening, which started in the 1980s, participants may have shifted from UGI screening to endoscopic screening.
There is no clear explanation regarding why the number of female participants and participants in their 60 s declined during the pandemic. Reasons for reduced participation in these strata should be further assessed through questionnaires. In Japan, many workers retire between the ages of 60 and 65 years and lose their eligibility for worksite-based cancer screening. Local governments do not individually identify residents who are disqualified from worksite-based cancer screening each year, so retirees are not called for population-based screening. If retirees wish to participate in population-based cancer screening, they must be referred to the local government and make an appointment. Those who retired during the COVID-19 pandemic may have been more reluctant to access population-based cancer screening than before.
In all cancers, mass screening resulted in greater attrition in participants than individual screening. Compared to individual screening, it is more difficult to maintain distance during the test in mass screening, which may increase the risk of infection among participants. Therefore, it is possible that local governments actively suspended mass screenings, or that residents avoided participating in mass screenings.
The strength of this paper is that it is the first to report changes in the number of cancer screening participants during the COVID-19 pandemic using Japanese national data. There are three limitations to our report. First, it is not clear whether the decline in participation in cancer screening is due to the cancellation of cancer screening by local governments or the refusal of residents to participate in cancer screening. If the number of participants in cancer screening decreased significantly at the time when the MHLW requested local governments to temporarily suspend screening (April-May 2020), cancellation of screening by the local government may have contributed to the decrease in the number of participants. However, this hypothesis could not be confirmed because the database of population-based screening programs does not provide monthly patient numbers (only annual totals are provided). During the first period of the state of emergency (April-May 2020), the Japanese government asked its citizens to strictly limit going outdoors. However, since June, behavioral restrictions had been limited to particularly high-risk behaviors and did not include participation in screening. The impact of these policies on changes in participation was not evaluated.
The second limitation is that the study relied on comparing the total number of screenings between different years and did not account for age standardization—changes in the size or age structure of the population between 2017 and 2020 could influence the results somewhat. Currently it is not possible to calculate the actual cancer screening rate in Japan, therefore we used the number of participants in this study. It is a serious drawback of the current Japanese cancer screening system that the actual measurement of the screening rate is not clear. There are two main reasons for this. One is that the local governments do not precisely identify those eligible for population-based cancer screening (e.g., unemployed persons, employees of companies that do not offer worksite-based cancer screening, retirees, etc.). Another reason is that there is no legal basis for worksite-based cancer screenings, and most of them are currently conducted as opportunistic screenings, and these results (such as the number of those eligible and participants) are not collected. All cancer screening rates (estimated value, including population-based screening and work-site screening) are identified only through questionnaire survey every 3 years.
The third limitation of this study is that changes in the number of participants in worksite- based screenings were not assessed due to the lack of a publicly available database as mentioned above. Differences in pandemic impacts should be assessed based on the data from large screening sites covering both population-based and workplace screenings. In the future, national screening programs should be integrated, and a comprehensive database should be constructed to improve cancer screening.
5 CONCLUSION
Based on a national database, the number of participants in population-based screenings in Japan decreased by approximately 10–30% during the pandemic. The impact of these declines on cancer detection or cancer mortality should be carefully monitored. As an intermediate step, it is important to monitor cancer screening attendance in the post-COVID-19 pandemic period to see if attendance returns to pre-COVID-19 pandemic levels.
Funding
This work was supported by Health and Labour Sciences Research Grants, and Research for Promotion of Cancer Control Programmes.
CRediT authorship contribution statement
Ryoko Machii and Hirokazu Takahashi contributed to the design and implementation of the research, to the analysis of the results and to the writing of the manuscript.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary material
Supplementary material.
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.canep.2022.102313.
==== Refs
References
1 Ministry of Health, Labour and Welfare, Comprehensive survey of living conditions. 〈https://www.mhlw.go.jp/english/database/db-hss/cslc-index.html〉 (accessed 2022 September 28).
2 Ministry of Health, Labour and Welfare. Report on Regional Public Health Services and Health Promotion, Services Portal Site of Official Statistics of Japan Website. 〈https://www.e-stat.go.jp/stat-search/files?page=1&toukei=00450025&kikan=00450&tstat=000001030884〉 (in Japanese) (cited 2022 July 7).
3 Ministry of Health, Labour and Welfare, Guideline for population based cancer screening program. 〈https://www.mhlw.go.jp/bunya/kenkou/gan_kenshin.html〉 (in Japanese) (cited 2022 July 7).
4 Bakouny Z. Paciotti M. Schmidt A.L. Lipsitz S.R. Choueiri T.K. Trinh Q.D. Cancer screening tests and cancer diagnoses during the COVID-19 pandemic JAMA Oncol 7 2021 458 460 10.1001/jamaoncol.2020.7600 33443549
5 Chen R.C. Haynes K. Du S. Barron J. Katz A.J. Association of cancer screening deficit in the United States with the COVID-19 pandemic JAMA Oncol 7 2021 878 884 10.1001/jamaoncol.2021.0884 33914015
6 Gathani T. Clayton G. MacInnes E. Horgan K. The COVID-19 pandemic and impact on breast cancer diagnoses: what happened in England in the first half of 2020 Br. J. Cancer 124 2021 710 712 10.1038/s41416-020-01182-z 33250510
7 Cabinet Secretariat, COVID-19 information and resources. 〈https://corona.go.jp/news/news_20200421_70.html〉 (in Japanese) (cited 2022 July 7).
8 Okuyama A. Watabe M. Makoshi R. Takahashi H. Tsukada Y. Higashi T. Impact of the COVID-19 pandemic on the diagnosis of cancer in Japan: analysis of hospital-based cancer registries Jpn J. Clin. Oncol. 2022 hyac129 10.1093/jjco/hyac129 35909325
9 The Japanese Society of Gastrointestinal Cancer Screening, Response to the new coronavirus infection (COVID-19) for gastrointestinal cancer screening. 〈https://www.jsgcs.or.jp/importants/archives/36〉 (in Japanese) (cited 2022 July 7).
| 36508964 | PMC9726684 | NO-CC CODE | 2022-12-09 23:15:18 | no | Cancer Epidemiol. 2023 Feb 7; 82:102313 | utf-8 | Cancer Epidemiol | 2,022 | 10.1016/j.canep.2022.102313 | oa_other |
==== Front
Asian J Surg
Asian J Surg
Asian Journal of Surgery
1015-9584
0219-3108
Asian Surgical Association and Taiwan Robotic Surgery Association. Publishing services by Elsevier B.V.
S1015-9584(22)01404-X
10.1016/j.asjsur.2022.09.159
Article
"If you're going through hell, keep going": Return to practice helped dental students cope with the (COVID-19) pandemic
Gaballah Kamis
Depart of oral and Craniofacial Health Sciences, College of Dental Medicine. University of Sharjah, Building M28, Office 217 University of Sharjah, United Arab Emirates
7 12 2022
7 12 2022
19 9 2022
26 9 2022
© 2022 Asian Surgical Association and Taiwan Robotic Surgery Association. Publishing services by Elsevier B.V.
2022
Asian Surgical Association and Taiwan Robotic Surgery Association
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmcTo the editor.
The COVID-19 pandemic disrupted dental services and affected urgent and elective services and education. The dental services and training resumed within months, despite the disease remaining active. This return was associated with burnout, lethargy, and concern in dentists and trainees. Most studies focused on student worries and anxiety were undertaken when most dental schools were on lockdown.1 No research has examined how students perceive returning to clinical training during the pandemic. This study assessed students' worries about the pandemic and its impact on their education during the national quarantine. It also focused on how returning to clinical training affected their perspectives and perceptions. During the lockdown, out of 420 students contacted, 364 (86.7%) responded. There were 250 female and 114 male participants in the sample, a ratio of 2.2:1. Most of the students (N = 346, 95.05%) responded to the second questionnaire sent after the return to onsite training. During the lockdown, 76% of both sexes expressed fear and anxiety about catching COVID-19. The students' return to training significantly reduced the number of students who indicated fear during the lockdown (n = 204, 56%) (P = 0.00). Most students (n = 313, 85.9%) were eager to perform dental procedures on patients. Only 260 (71.4%) students showed concern after practice resumed. Most students feared transferring the virus from training clinics to their families. The return to onsite instruction reduced that concern from 93.7% to 87.4%. Most participants (n = 305, 83.8%) were concerned about the pandemic's impact on dental education. All study groups (n = 312, 85.7%) were unnerved by returning to clinical training. Given the pandemic's various risks, only 10% of participants regretted choosing dentistry as a career; When questioned about online instruction, students rated it 3 out of 5 (33% of 120 participants) to 2 out of 5 (8.5% of 31 participants). Over a third of students rated online learning as 2 out of 5 after returning to on-campus training (Fig. 1 ). The first phase of the survey was conducted during the lockdown. Stress and fear generated by quarantine restrictions may have affected answers.2 Participants feared infecting their families. This was similar to health workers' concerns about infecting their families.3 Dental education emphasizes the continuity of clinical and practical training. Short-term quarantine or protracted lockout might impair training and induce anxiety. North American Dental Education Commission and British General Dental Council implemented temporary flexibility measures during the initial stage of the pandemic. These include: Alternate assessment methods, including patient simulation units; curriculum content or requirements are modified or reduced; and program duration or portion duration is shortened. Dental institutions modified educational delivery and student academic advancement to comply with community health regulations. E-learning and online learning have proven effective in the health professions, including dentistry. Blended learning blends classroom and distance learning to enhance autonomy, engagement, and cooperation. Students' lack of preparation for online learning led to mixed comments on their remote learning experience. Many students, particularly seniors, down rated online learning when they returned to training and experienced a more integrated teaching-learning model. Bridges and coworkers recommended mixing learning technology with face-to-face education to boost knowledge and visualization.4 The level of the study affects the student's preferred instruction method. Juniors favor online classes over seniors. Studies suggested that younger students adapt to e-learning more easily.5 The author found that the return to practice reduced stress and anxiety among the same students they expressed during quarantine. The stringent cross-infection policy and inoculation against the disease facilitated this excellent outcome. Students are understandably anxious about obtaining work after graduation, considering the high unemployment rate caused by COVID-19. Seniors echoed this perspective. However, most students were okay with choosing a dental career. This study focuses on one dental school in one country. Students in areas where COVID-19 is more prevalent may give different replies. Finally, the study's findings may inspire dental institutions to examine the emotional impact of the pandemic, monitor students' healthy return to everyday routines, and learn how to employ hybrid education and face future obstacles.Fig. 1 A) Summary of the student's responses to the various questions related to the emotional impact of the pandemic and the impact of the pandemic on the quality of training and education; B) The COVID-19–related responses of the participants after return to training; C) The change in the pattern of student rating of the online teaching and learning experience during the lockdown (red) and after return to training (yellow). The rating key: 1 Strongly disagree; 2 Disagree; 3 Neither agree nor disagree; 4 Agree; 5 Strongly agree.
Fig. 1
Declaration of Competing interest
The author declares no potential financial and non-financial conflicts of interest.
Appendix A Supplementary data
The following are the Supplementary data to this article:Multimedia component 1
Multimedia component 1
Multimedia component 2
Multimedia component 2
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.asjsur.2022.09.159.
==== Refs
References
1 Kharma M.Y. Koussa B. Aldwaik A. Assessment of anxiety and stress among dental students to return to training in dental College in COVID-19 era Eur J Dent 14 S 01 2020 S86 S90 33032336
2 Brooks S.K. Webster R.K. Smith L.E. The psychological impact of quarantine and how to reduce it: rapid review of the evidence Lancet 395 10227 2020 912 920 32112714
3 Hung M. Licari F.W. Hon E.S. In an era of uncertainty: impact of COVID-19 on dental education J Dent Educ 85 2 2021 148 156 32920890
4 Bridges S.M. Botelho M.G. Tsang P.C. PBL.2.0: blended learning for an interactive, problem-based pedagogy Med Educ 44 11 2010 1131 20946496
5 Amir L.R. Tanti I. Maharani D.A. Student perspective of classroom and distance learning during COVID-19 pandemic in the undergraduate dental study program Universitas Indonesia BMC Med Educ 20 1 2020 392 33121488
| 0 | PMC9726685 | NO-CC CODE | 2022-12-08 23:18:18 | no | Asian J Surg. 2022 Dec 7; doi: 10.1016/j.asjsur.2022.09.159 | utf-8 | Asian J Surg | 2,022 | 10.1016/j.asjsur.2022.09.159 | oa_other |
==== Front
Intensive Crit Care Nurs
Intensive Crit Care Nurs
Intensive & Critical Care Nursing
0964-3397
1532-4036
Published by Elsevier Ltd.
S0964-3397(22)00175-6
10.1016/j.iccn.2022.103372
103372
Article
How to guarantee the core competencies in redeployment during the patient surge from COVID-19 or other emerging infectious diseases
Kang Sunjoo a⁎
Young Seo Jin b
Lee Moonju c
Kima Hayun a
Yang Sun-Yi d⁎
a Yonsei University Graduate School of Public Health, Seoul, Republic of Korea
b School of Nursing, Hunter College, The City University of New York, New York, USA
c School of Nursing, The University of Texas Health Science Center at San Antonio, San Antonio, Texas, USA
d College of Nursing, Konyang University, Daejeon, Republic of Korea
⁎ Corresponding authors at: Daejeon Medical Campus, 158 Gwanjeodong-ro, Seo-gu, Daejeon, Republic of Korea, 50-1 Yonsei-ro, Seodaemun-gu, Seoul, Republic of Korea
7 12 2022
7 12 2022
10337216 9 2022
4 12 2022
5 12 2022
© 2022 Published by Elsevier Ltd.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmcDear Editor
When nurses are redeployed, a supplementary training program is recommended for those without prior experience caring for patients with emerging infectious diseases (EID) (San Juan et al., 2022). It guarantees patient safety by boosting nurses’ confidence (Lee and Lee, 2020). The training modules should be developed with real patient scenarios and problem-solving approaches to enhance nurses’ self-directed learning (Dziurka et al., 2022, Guttormson et al., 2022). These efforts to ensure their competencies could be adopted in developing countries via an online platform that compensates for resource scarcity and contributes to the prevention of EID transmission globally (Walker et al., 2020).
Methods: The ethics approvals were obtained for conducting the in-depth interview for developing training modules and the pilot test in 2020 and 2022. The modules were developed via content analysis of interviews with thirty-nine registered nurses from South Korea and the United States, working in intensive care or coronavirus disease 2019 (COVID-19) isolated units. They had three to five years of nursing experience prior to caring for COVID-19-infected patients for a minimum of one the year 2020–2021. A quasi-experimental study was conducted for the pilot test in seven general hospitals in South Korea.
Results: Fifty-five nurses participated in the pilot test among intervention (n=25) and control (n=30) groups (Supplement material 1). The modules covered issues from admission to discharge, and the training was provided to the intervention group. The modules dealt with the contents of infection control and prevention, ventilator management, end-of-life care, and teamwork. Five real case-based scenarios were added for the improvement of problem-solving competence (Supplement material 2, 3). The developed questionnaire to assess the change of confidence has high reliability (Cronbach α .8760): The results of the intervention group showed increased confidence in COVID-19 patient care, infection prevention and control, end-of-life care, and teamwork, and presented statistically significant differences in the pretest and posttest (Table 1 ).Table 1 Comparison of dependent variables between intervention and control groups (n=55)
Variables Groups Pre-intervention Post-intervention Post-Pre Difference Analysis of covariance Pairedt-test
M(SD) M(SD) M(SD) t(p) F(p) t(p)
Nursing care Experimental group 2.76(0.83) 4.44(0.51) 1.68(1.029) 8.15(<0.001) 1.84(<.001) 6.28(<.001)
Control group 3.08(0.91) 3.36(0.86) 0.28(1.32) -.82(0.415)
Infection prevention and control Experimental group 2.76(0.83) 4.64(0.49) 1.88(0.88) 10.66(<0.001) 1.61(<.001) 6.73(<.001)
Control group 3.36(0.81) 3.36(0.86) 0.0867(0.96) 0.49(0.625)
End-of-life care Experimental group 3.28(0.74) 4.60(0.64) 1.32(0.94) 6.98(<0.001) 0.94(<.001) 5.16(<.001)
Control group 3.04(0.88) 3.44(1.08) 0.30(1.23) -1.47(0.153)
Teamwork Experimental group 3.00(0.76) 4.56(0.50) 1.56(0.96) 8.16(<0.001) 4.22(<.001) 4.77(<.001)
Control group 2.96(1.02) 3.32(0.90) 0.066(1.23) 0.29(0.769)
M: mean; SD: standard deviation
Discussion: This study targeted nurses without prior experience caring for EID patients. The training modules gave the participants the adaptability to care for EID patients, and aligned with the findings of the previous studies for enhanced competencies (Vera San Juan et al., 2022). The result indicates its effectiveness in informing global health security from EID, and in improving nurses’ competency and capabilities, as recommended by the previous research findings (Lee and Lee, 2020). Nursing facilities in developed or underdeveloped countries may differ; thus, certain adjustments should be made to the modules when applying the developed training data, main lectures, and case discussions. Additions or exclusions to the online or offline training could be considered the existing education levels of nurses in the health facilities and include selective training.
Implication and Limitation: The developed training program proved effective and improved of core nursing competencies required for caring for EID patients in this study. Therefore, this program can confidently be recommended as supplementary training for nurses prior to their assignment to EID units. Further studies are required to determine the outcomes of this training in different settings.
Financial support: This research was supported by the Global Korean Nursing Foundation’s (GKNF) Mo-Im Kim COVID-19 Pilot Research Grant (research period: 2020.10––2022.03)
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
==== Refs
References
Dziurka M. Machul M. Ozdoba P. Obuchowska A. Kotowski M. Grzegorczyk A. Clinical training during the COVID-19 pandemic: Experiences of nursing students and implications for education Int. J. Environ. Res. Public Health. 19 2022 6352 10.3390/ijerph19106352, PubMed PMID 35627889
Guttormson J.L. Calkins K. McAndrew N. Fitzgerald J. Losurdo H. Loonsfoot D. Critical care nurses’ experiences during the COVID-19 pandemic: A US national survey Am. J. Crit. Care. 31 2022 96 103 34704108
Lee N. Lee H.J. South Korean nurses’ experiences with patient care at a COVID-19-designated hospital: Growth after the frontline battle against an infectious disease pandemic Int. J. Environ. Res. Public Health. 17 2020 9015 33287343
San Juan N.V. Clark S.E. Camilleri M. Jeans J.P. Monkhouse A. Chisnall G. Vindrola-Padros C. Training and redeployment of healthcare workers to intensive care units (ICUs) during the COVID-19 pandemic: A systematic review BMJ Open. 12 2022 e050038
Walker P.G.T. Whittaker C. Watson O.J. Baguelin M. Winskill P. Hamlet A. The impact of COVID-19 and strategies for mitigation and suppression in low- and middle-income countries Science. 369 2020 413 422 10.1126/science.abc0035, PubMed PMID: 32532802, PubMed Central PMCID: PMC7292504 32532802
| 0 | PMC9726686 | NO-CC CODE | 2022-12-08 23:18:18 | no | Intensive Crit Care Nurs. 2022 Dec 7;:103372 | utf-8 | Intensive Crit Care Nurs | 2,022 | 10.1016/j.iccn.2022.103372 | oa_other |
==== Front
Teach Learn Nurs
Teach Learn Nurs
Teaching and Learning in Nursing
1557-2013
1557-3087
Published by Elsevier Inc. on behalf of Organization for Associate Degree Nursing.
S1557-3087(22)00149-4
10.1016/j.teln.2022.12.003
Research
How students learn in small group through online mode during the coronavirus pandemic: Descriptive narratives
KAN Crystal Wai Yee MScNursRN 1#
WONG Florence Mei Fung DNMNBTSNRN 2⁎
1 School of Health Sciences, Caritas Institute of Higher Education, 2 Chui Ling Lane, Tseung Kwan O, New Territories
2 School of Nursing, Tung Wah College, 16/F, Tower 2, Kowloon Commerce Centre, 51 Kwai Cheong Road, Kwai Chung, Kowloon, Hong Kong
⁎ Corresponding author: Telephone: (852) 3468 6838
# Telephone: (852) 2702 4262
7 12 2022
7 12 2022
5 12 2022
© 2022 Published by Elsevier Inc. on behalf of Organization for Associate Degree Nursing.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Small-group work is a teaching method which promotes students’ collaborative attributes. Completing small-group work online can be challenging to students. This study aimed to understand students’ learning through online small-group work. Narrative responses to five open-ended questions were collected from students after they accomplished an online small-group work. These narratives were content analyzed by two independent researchers.
Based on the narratives of 199 students, five major themes related to students’ learning through interactive online group work were identified, including a) Essential communication and interaction, b) A collaborative team, c) Knowledge enrichment, d) Development of intellectual skills, and e) Tutor involvement. These five themes are the essential elements that promote effective learning though online learning. The success of group work greatly depends on these five major factors to promote students’ independent and collaborative learning. To make small-group work online more achievable, nurse educators should pay more attention to these five factors so students’ learning outcomes can be maximized.
Keywords
Online small group work
student learning
narratives
nursing education
==== Body
pmcIntroduction
Small-group work is a beneficial teaching–learning method employed in nursing education to foster individual inventive learning, and collaborative attributes, as well as to enhance the acquisition and retention of knowledge (Mennenga & Smyer, 2020; Wong, 2018). Group work allows students to actively engage in their learning and enhances both personal and professional development (Edmunds & Brown, 2010; Feichtner & Davis, 2016). Through self-directed learning and interactions, students can broaden their thinking and promote transferable skills. These skills include problem-solving skills, time management, interpersonal communication, and critical thinking, which are all conducive to lifelong learning (Langegård et al., 2021; Terzi et al., 2021; Wong, 2020). Thus, small-group work can effectively motivate students to commit to their work and develop interdependence and competence in multiple skills (Shaw et al., 2015; Wong, 2018). The learning outcomes achieved through small-group work is promising, as it promotes mutual learning, understanding, acceptance, and recognition through social networking and trust building (Shaw et al., 2015; Wong, 2020).
Due to the COVID-19 pandemic, online learning has been adopted globally. With advances in communication technology, a variety of online modalities and learning platforms, such as Blackboard, Moodle, and Zoom, are widely employed to maintain the quality of teaching and learning (Antonio et al., 2020). Online learning tends to be more student-directed, and consequently increases students’ individual responsibility and engagement (Langegård et al., 2021; Qamar et al., 2021; Thapa et al., 2021). Nevertheless, students have reported challenges in maintaining collaborative attributes to group achievement while staying in their home environment (Korzycka et al., 2021; Langegård et al., 2021; Suliman et al., 2021). Face-to-face small group work has been reported to have enhanced in students’ competence and achievement (Meo, 2013; Wong, 2018). However, few studies have explored students’ learning through small-group work using an online modality. Therefore, this study aimed to explore how students learned in small groups online using a qualitative approach.
Methodology
Design
In this study, qualitative data was collected from a narrative survey. The survey included five open-ended questions, ‘Share your views about the effectiveness of small group work on your individual studies, ‘What have you learned from your own studies?’, ‘Share your views about the effectiveness of small group work on your group learning’, ‘Share your experience of learning with other resources, such as your tutor, internet resources, journal articles, and textbooks’, and ‘Share your views on how to improve your learning through small group work in the future’.
Theoretical framework
Social interdependence theory developed by Johnson and Johnson (2009), was adapted to this study. This theory covering motivational, social, and cognitive aspects aims to maximize the collaborative potential of a group. It consists of five key elements, including positive interdependence, individual accountability, promotive interaction, interpersonal and social skills, and group processing. Positive interdependence refers to the achievement which an individual perceives when he/she can reach the common goals through collaboration with others. Individual accountability refers to all group members being responsible for and willing to facilitate the accomplishment of work together. Promotive interactions are required for individuals in a group to encourage one another and work together by sharing resources and providing support to further motivate one another. Interpersonal and social skills are methods to enhance trustworthiness through understanding, problem-solving, and decision making so that individuals improve their communication and conflict resolution skills. Group processing refers to an effective working relationship established though the process of discussion, evaluation, and sharing feedback in group.
Sample
Purposive sampling was carried out from 13 January to 31 August 2020. Participants were selected from a cohort of fourth-year students in a 5-year undergraduate nursing program at a professional educational institute. Participants were required to work in small groups through online interactions due to the COVID-19 pandemic.
Tutor
The tutor was a doctoral-prepared registered nurse with 20 years’ experience in nursing. The tutor, a nurse educator for 12 years, has researched group learning among students. she was the sole tutor in this study and acted as a facilitator, monitored learning, and provided guidance and support.
Data collection
In the first week, students formed groups of seven to nine on their own. A guideline comprising students’ roles and responsibilities, instructions of a small-group project, and assessment criteria was introduced. Students could interact using emails, real-time chats, document uploads, and an e-learning discussion forum as the main communication channels. Additionally, other communication platforms, including WhatsApp, Zoom, WeChat, and Skype, were also regarded as appropriate. Each group was assigned a specific problem-based learning scenario and learning objectives. Students were encouraged by the nurse educator to initiate self-directed learning, including setting their own learning objectives. Before each class, students conducted information search, sharing, discussions, collaborated, and accomplished self-paced tasks in order to achieve their common goals.
From the second to sixth weeks, students engaged in an ongoing learning cycle including individual and group learning. Additional case information was provided after each lesson. Students could communicate with their tutor via email, Blackboard, Zoom, and WhatsApp. The tutor provided the students with timely guidance and feedback in each group to facilitate them. A brief report regarding their individual learning and free group discussion was submitted to the tutor before each lesson to monitor the progress of the individual and group. Accordingly, each student's participation, contribution and learning were tracked. In the eighth week, a group presentation as a final product of small-group project was given in class. Finally, students were invited to complete a web-based survey, consisting of five open-ended questions to share their experience in learning through online small-group work.
Data analysis
All 199 of the students’ narratives were analyzed. Two independent researchers, the Co-Investigator (Co-I) and the Research Assistant (RA), reviewed and organized the collected data to become familiar with them. The researchers conducted on-going discussion and re-examined the data to ensure that the research question was answered. The data was coded, and categories and sub-categories were formed. At the end, five main themes related to students’ learning experiences through online small group work were derived.
Rigor and trustworthiness
Each student's narrative was reviewed independently by two researchers to ensure rigor and trustworthiness. Both of them were registered nurses and have been involved in nursing qualitative research for more than two years. The Co-I was a nurse educator with more than eight years of teaching experience. All discrepancies were discussed by the two researchers until they reached a consensus; the Principal Investigator (PI) would otherwise be approached for the final decision.
Ethical considerations
Research ethics approval was obtained from the study institution. All eligible students were invited to voluntarily participate in the study. Consent was obtained before data collection. Students were assured that all data were strictly confidential and anonymous.
Results
Five main themes related to students’ learning through online interactive group project were identified: a) essential communication and interaction, b) a collaborative team, c) knowledge enrichment, d) development of intellectual skills, and e) tutor involvement.
Theme 1: Essential communication and interaction
Online group work poses increasing challenges to a student's learning as it takes time to accommodate the unfamiliar communication methods. A constructive team and effective communications are crucial. First, students preferred forming a group with their friends to facilitate cohesiveness and collaboration. Student 33 shared, ‘We can make a good team if our tutor let us form a group with our friends’. In addition, using familiar channels, including e-learning systems and other online channels were important to facilitate students to receive appropriate learning materials and communicate with their groupmates and the tutor. ‘The e-learning system improves students’ learning and facilitate communication among students as well as students and the tutor’ (Student 61). Students used various modes of communication, such as Blackboard Collaborate Ultra, WhatsApp, Skype, Zoom, and WeChat, to maintain discussion and communication which was vital for problem-solving and decision making. Student 147 mentioned, ‘We usually use WhatsApp and Google drive to discuss and work on the project and we can receive comments from each other’. Student 30 expressed, ‘Active and effective online communication improves the overall quality of the work.’
Theme 2: A collaborative team
An effective team facilitates cooperation and collaboration, and team dynamic, as well as improves work efficiency. Some students agreed that teamwork was essential for nursing, and they learned how to work with others in a group. Student 17 expressed, ‘Teamwork is essential for nurses as we need to cooperate with each other and collaborate with other healthcare providers.’ A collaborative team improves the group dynamic and promotes work efficiency. Students learned to work with others respectfully, contributing to more effective learning outcomes. “I learned to cooperate with different group members to finish the tasks. The work became more efficient when we did it in a group as everyone was responsible for different parts of the project.” (Student 17)
Theme 3: Knowledge enrichment
One of the challenges in online learning is lacking self-motivation due to lack of peer support (Wissing et al., 2022). Conversely, in the small group learning, student's knowledge was enhanced through self-directed learning and deep learning, and brainstorming. Adequate learning resources were crucial to help students learn appropriately. Student 31 expressed, ‘Students could set clearer learning objectives with the e-learning resources provided by the tutor or from relevant websites.’
Students in a small group not only study individually, but also work with their groupmates. During their individual study, students developed self-directed learning ability, allowing them to explore knowledge and learn deeply by themselves. Students indicated that they are more engaged in learning when they work towards a group goal. Some students shared that they overcame difficult situations and learned much more through online learning. Student 112 shared her opinion, ‘It is good to work together because others can help solve the difficulties immediately’. Besides, Student 167 added, ‘Small group work allows us to do more research and explore further on a particular topic compared with doing an individual assignment. Small group work effectively broadens our understanding of a particular topic … Small group work shows its advantages especially in group discussions and brainstorming’.
Theme 4: Development of intellectual skills
Students expressed that they developed various skills, including analytical and problem-solving abilities, and time management, while they collaborate in a small-group team. ‘Our learned knowledge can be shared in a group and can initiate multi-dimensional questions to train our problem-solving ability’ (Student 8). Students 45 and 167 explained that brainstorming created new directions which fostered analytic and problem-solving skills. Students explored new information and integrated their knowledge to enhance their competence. In this sense, students learned problem-solving skills together through identifying problems, searching for reasons and solutions, discussing appropriate actions and strategic plans, applying the strategies, and evaluating their effects. Importantly, students enjoyed group work because it allowed more efficient learning and a more organized schedule. ‘I have learned better time management… to plan self-study, as online learning at home is challenging and mostly needs self-discipline.’ (Student 14).
Some students learned leadership as well. The group leader plays a significant role to facilitate communication and maintain interactions within the group. The leaders learned to coordinate the whole group and guide the members to achieve learning outcomes (Students 199). Other groupmates also agreed that “a leader is important for our discussions and coordination of the team” (Students 24).
Theme 5: Tutor involvement
The tutor played an important role in facilitating small-group work. Students learned from the tutor's guidance and feedback. They reflected that their tutor always made good and informative responses to their questions and direct them the way to improve themselves (Students 31 & 80). The interactions between students and the tutor fostered the students’ learning because students could consolidate the knowledge they had learned and apply it to real situations appropriately (Students 8, 38, 64, 129, & 167). Student 40 expressed, ‘My tutor joined our group and discussed with us. It was really helpful to our studying’.
Discussion
Five main themes concerning the students’ learning and achievement through online small-group work are identified. They are regarded as essential elements of students’ learning, including a) Essential communication and interaction, b) Á collaborative team, c) Knowledge enrichment, d) Development of intellectual skills, and e) Tutor involvement. Today, a variety of innovative technological methods exist. Online learning allows students to experience more autonomy, self-motivation, and interactions with others, which help them further develop collaborative attributes and enhance their professional competence (Qamar et al., 2021; Suliman et al., 2021). These five main themes are the key areas to which tutors should pay more attention. Students can learn through online small-group work only when they accept online learning and they are self-initiative (Anwar et al., 2020; Karaman, 2011; Qamar et al., 2021). This study conveys an important message that virtual group learning is feasible and effective as long as effective communication is maintained.
Effective communication and social interaction are the most significant elements in terms of developing collaboration and intellectual skills (Huter et al., 2020; McCutcheon et al., 2015; Wong 2020). Our study showed that online small-group work provides a high degree of collaborative learning, promotes effective and supportive communication and interactions among students and tutors, improves learning achievement, and increases students’ sense of self-worth (Erford, 2018; Scherling, 2011). In small-group learning, students’ sense of collaboration and knowledge acquisition are enhanced, and they develop intellectual skills for personal and professional development (Wong 2020). Furthermore, effective communication facilitates knowledge exchange and collaborative development (Erford et al., 2018; Korzycka et al., 2021; Langegård et al., 2021; Wong, 2018; Wong et al., 2021).
Based on social interdependence theory, students in this study were able to maximize their collaborative group potential through online small-group work, and thus further develop their personal and professional values and abilities (Langegård et al., 2021). Social interaction plays a significant role in learning (Okita, 2012), including promoting retention of knowledge and deeper learning (Langegård et al., 2021; Thapa et al., 2021). Therefore, to enhance their engagement and maintain interactions, students in an effective team interact with their groupmates using various modes of communications (Wong 2018; 2020). An effective team dynamic motivates students to actively participate in independent learning, which helps develop lifelong learning (Karaman, 2011). Developing group norms before commencing a group project and forming groups with classmates who have similar backgrounds is important (Scherling, 2011; Wong, 2018). A recent scoping review shows that online learning enhances students' responsibilities to learn, and that knowledge retention, self-motivation, and teaching methods are interdependent (Jowseya et al., 2020).
Hence, online learning can lead to enhanced knowledge and skill acquisition, such as self-directed learning, interpersonal communication, problem-solving, and collaboration skills (Scherling, 2011; Suliman et al., 2021; Wong, 2020; Wong et al., 2021). Students gain more learning achievement through the exchange of knowledge and experience, as well as assigning tasks based on individual capabilities (Erford et al., 2018; Korzycka et al., 2021; Langegård et al., 2021; Wong 2018). This form of task distribution improves mutual understanding and promotes team dynamic and collaboration. Constructive discussions and cooperation are necessary, which is consistent with the participants’ responses in this study.
Importantly, tutor involvement is crucial. Tutors play an important role in helping students overcome barriers by giving them adequate support and by enhancing their motivation (Korzycka et al., 2021; Langegård et al., 2021). Students’ performance improves according to the feedback and guidance they receive from their tutors (Antonio et al., 2021; Qamar et al., 2021). To make online learning more favorable to nursing education, tutors should ensure a collaborative team as it will allow students to address problems together and apply their knowledge and skills to real-life nursing situations for optimal patient care (Qamar et al., 2021; Wong, 2020). Clear guidelines and instructions for both tutors and students are essential to setting appropriate direction and achieving teaching and learning outcomes (Langegård et al., 2021; Meo, 2013). While designing learning activity for small-group work, tutors should consider students’ learning needs and the desired learning outcomes as their priority (Scherling, 2011).
Implications for clinical practice
Five major themes are key elements that allow students to learn more effectively through online small-group work. Multidisciplinary collaboration is emphasized in nursing to ensure safe and high-quality patient care (Karaman, 2011; Mennenga & Smyer, 2010). Hence, nursing students are required to be equipped with collaborative attributes, commonly by small-group work, in their learning. During the interactive process, students can learn and adapt to a collaborative environment which will be beneficial in the clinical settings. Through small-group work, students can learn more independently and interact with others to achieve their own learning outcomes (Wong, 2018). This student-directed learning enhances students’ competence, knowledge, and skills for personal and professional development (Edmunds & Brown, 2010). Consequently, small-group work should be considered in the nursing curriculum. Tutors, who play an important role in small-group work, provide adequate support to students particularly in an online mode. Effective communication through various familiar platforms needs to be maintained. An ongoing evaluation of students’ learning in knowledge enhancement and development of skills may be essential with a thorough small-group approach.
Limitations
The content of students’ narratives through a survey was limited and did not provide a comprehensive understanding of their learning experience through online small-group work. A qualitative study with focus group interviews is recommended. Moreover, further studies may be needed to fully understand the differences between students’ learning in face-to-face and online small-group work. This study was conducted in one group of undergraduate nursing students at a single institution, so the results may not be generalizable to students at other levels or other institutions due to the diversity of students’ learning needs in various levels of online programs and degrees of online learning support at institutions. The tutor included in this study was an experienced nurse educator in teaching small groups. Additional support may be required when multiple tutors are involved.
Conclusion
This study has revealed five major themes related to the benefits of students’ learning through online small-group work. Importantly, effective communication is essential to maintaining interaction and leads to more achievement in personal and professional development. Tutor involvement and an effective teaching–learning plan can facilitate students’ learning through online small-group work. The results should increase tutors’ awareness of students’ learning and their roles when online small-group work is adopted.
Conflict of interest
No conflict of interest has been declared by the authors.
Author contributions
Crystal Kan was responsible for making substantial contributions to conception and design, student recruitment, implementation of the intervention, acquisition of data, analysis and interpretation of data, data analysis, data interpretation, drafting manuscript, and verifying final version of the manuscript.
Florence Wong was responsible for making substantial contributions to conception and design, training the Research Assistant for data collection and data entry, student recruitment, data collection, data entry, data interpretation, and verifying final version of the manuscript.
Funding
This study was supported by College Research Grant of Tung Wah College (CRG2018/03).
Conflict of interest
Authors declare no conflict of competing interest
Declaration of Competing Interest
Authors declare no conflict of competing interest
Acknowledgements
None
==== Refs
References
Antonio J.R. César L. José E.M. María R.M. Experiences of nursing students during the abrupt change from face-to-face to e-learning education during the first month of confinement due to COVID-19 in Spain International Journal of Environmental Research and Public Health 17 2020 5519 10.3390/ijerph17155519 32751660
Anwar A. Khan E. Nisar M. Qutub R.D. Azim S.R. Awan T.T. Impact of COVID-19 pandemic on learning of undergraduate medical students: A cross-sectional study from KARACHI Pakistan Armed Forces Medical Journal 70 6 2020 1902 1907 10.51253/pafmj.v70i6.4324
Edmunds S Brown G. Effective small group learning: AMEE guide No. 48 Medical Teacher 32 2010 715 726 10.3109/0142159X.2010.505454 20795801
Erford, B. T. (2018). Group Work: Processes and Applications. 2nd Ed. Taylor & Francis: Routledge.
Feichtner S.B. Davis E.A. Why some groups fail: A survey of students’ experiences with learning groups Journal of Management Education 40 1 2016 12 29 10.1177/1052562915619639
Huter K. Krick T. Domhoff D. Seibert K. Wolf-Ostermann K. Rothgang H. Effectiveness of Digital Technologies to Support, Nursing Care: Results of a Scoping Review Journal of Multidisciplinary Healthcare 13 2020 1905 1926 10.2147/JMDH.S286193 33328736
Johnson D.W. Johnson R.T. An educational psychology success story: Social interdependence theory and cooperative learning Educational Research 38 2009 365 379 10.3102/0013189X09339057
Jowseya, T., Fosterb, G., Cooper-Ioeluc, P., Jacobsd, S. (2020). Blended learning via distance in pre-registration nursing education: A scoping review. Nurse Education in Practice,44, 102775. https://doi.org/10.1016/j.nepr.2020.102775. DOI: 10.1016/j.nepr.2020.102775
Karaman S. Nurses’ perceptions of online continuing education BMC Medical Education 11 2011 10.1186/1472-6920-11-86 86 http://www.biomedcentral.com/1472-6920/11/86
Korzycka M. Bójko M. Radiukiewicz K. Dzielska A. Nałęcz H. Kleszczewska D. Małkowska-Szkutnik A. Fijałkowska A. Demographic analysis of difficulties related to remote education in Poland from the perspective of adolescents during the COVID-19 pandemic Annals of Agricultural and Environmental Medicine 28 1 2021 149 157 10.26444/aaem/133100 33775081
Langegård U. Kiani K. Nielsen S.J. Svensson P. Nursing students’ experiences of a pedagogical transition from campus learning to distance learning using digital tools BMC Nursing 20 2021 23 10.1186/s12912-021-00542-1 33468132
McCutcheon K. Lohan M. Traynor M. Martin D. A Systematic Review Evaluating the Impact of Online or Blended Learning vs. Face-to-face Learning of Clinical Skills in Undergraduate Nurse Education Journal of Advanced Nursing 71 2 2015 255 270 10.1111/jan.12509 25134985
Mennenga H. Smyer T. A model for easily incorporating team-based learning into nursing education International Journal of Nursing Education Scholarship 2010 10.2202/1548-923X.1924 7, 4. DOI
Meo S.A. Basic steps in establishing effective small group teaching sessions in medical schools Pakistan Journal of Medical Sciences 29 4 2013 1071 1076 10.12669/pjms.294.3609 24353692
Okita, S.Y. (2012). Social Interactions and Learning. In: Seel N.M. (eds) Encyclopedia of the Sciences of Learning. Springer, Boston, MA. doi: https://doi.org/10.1007/978-1-4419-1428-6_1770
Qamar, K., Kiran, F., Khan, M.A., Raza, S.N., Iram, M., & Rauf, A. (2021). Challenges of E-Learning faced by medical teachers and students during COVID-19 pandemic. Pakistan Armed Forces Medical Journal, 71 (Suppl-1), S3-9. doi. https://doi.org/10.51253/pafmj.v71iSuppl-1.6191
Scherling, S.E. (2011). Designing and fostering effective online group projects. Adult Learning, 22 (2). Education Collection pg. 13. DOI: https://doi.org/10.1177/104515951102200202
Shaw, J., Mitchell, C., & Del Fabbro, L. (2015). Group work: Facilitating the learning of international and domestic undergraduate nursing students. Education for Health, 28(2), 124-129. doi: 10.4103/1357-6283.170123
Suliman, W.A., Abu-Moghli, F.A., Khalaf, I., Zumot, A.F., & Nabolsi, M. (2021). Experiences of nursing students under the unprecedented abrupt online learning format forced by the national curfew due to COVID-19: A qualitative research study. Nurse Education Today, 100. doi: 10.1016/j.nedt.2021.104829.
Terzi, B., Azizoğlu, F., & Özhan, F. (2021). Factors affecting attitudes of nursing students towards distance education during the COVID-19 pandemic: A web-based cross-sectional survey. Perspectives in Psychiatric Care,1 (9). doi: https://doi.org/10.1111/ppc.12747
Thapa, P., Bhandari, S.L., Pathak, S. (2021). Nursing students’ attitude on the practice of e-learning: A cross-sectional survey amid COVID-19 in Nepal. PLoS ONE, 16(6), e0253651. doi: https://doi.org/10.1371/journal.pone.0253651
Wissing R.O. Hilverda F. Scheepers R.A. Nieboer A.P. Vlolmann M. Peer relationships buffer the negative association of online education with education satisfaction and subsequently with study engagement among undergraduate medical students BMC Medical Education 22 2022 276 10.1186/s12909-022-03337-3 35418086
Wong, F.M.F. (2018). A phenomenological research study: Perspectives of student learning through small group work between undergraduate nursing students and educators. Nurse Education Today, 68, 153-158. DOI: https://doi.org/10.1016/j.nedt.2018.06.013.
Wong M.F.F. Development of Higher-Level Intellectual Skills through Interactive Group Work: Perspectives between Students and Educators Journal of Medical & Clinical Research 5 8 2020 164 169
Wong, F.M.F., Tang, A.C.Y., & Cheng, W.L.S. (2021). Factors associated with self-directed learning among undergraduate nursing students: A systematic review. Nurse Education Today, 104, DOI: https://doi.org/10.1016/j.nedt.2021.104998.
| 36506705 | PMC9726687 | NO-CC CODE | 2022-12-08 23:18:18 | no | Teach Learn Nurs. 2022 Dec 7; doi: 10.1016/j.teln.2022.12.003 | utf-8 | Teach Learn Nurs | 2,022 | 10.1016/j.teln.2022.12.003 | oa_other |
==== Front
Med Clin (Engl Ed)
Med Clin (Engl Ed)
Medicina Clinica (English Ed.)
2387-0206
Elsevier España, S.L.U.
S2387-0206(22)00572-1
10.1016/j.medcle.2022.03.021
Original Article
Survival impact of previous statin therapy in patients hospitalized with COVID-19
Impacto del tratamiento previo con estatinas sobre la supervivencia de los pacientes hospitalizados con COVID-19Barge-Caballero Eduardo a⁎
Marcos-Rodríguez Pedro J. cd
Domenech-García Nieves bc
Bou-Arévalo Germán ce
Cid-Fernández Javier cf
Iglesias-Reinoso Raquel c
López-Vázquez Paula c
Muñiz Javier bg
Vázquez-Rodríguez José M. abc
Crespo-Leiro María G. abc
a Servicio de Cardiología, Complejo Hospitalario Universitario de A Coruña (CHUAC), A Coruña, Spain
b Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain
c Instituto de Investigación Biomédica de A Coruña (INIBIC), A Coruña, Spain
d Servicio de Neumología, CHUAC, A Coruña, Spain
e Servicio de Microbiología, CHUAC, A Coruña, Spain
f Servicio de Inmunología Clínica, CHUAC, A Coruña, Spain
g Departamento de Ciencias de la Salud, Universidad de A Coruña, A Coruña, Spain
⁎ Corresponding author.
7 12 2022
7 12 2022
17 1 2022
22 3 2022
© 2022 Elsevier España, S.L.U. All rights reserved.
2022
Elsevier España, S.L.U.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
Statin therapy might have a beneficial prognostic effect in patients with COVID-19, given its immunomodulative, anti-inflammatory and anti-atherosclerotic properties. Our purpose was to test this hypothesis by using the COVID-19 registry of a Spanish university hospital.
Methods
We conducted a single-center, observational and retrospective study in which hospitalized patients with COVID-19 diagnosed by PCR between March 2020 and October 2020 were included. By means of logistic regression, we designed a propensity score to estimate the likelihood that a patient would receive statin treatment prior to admission. We compared the survival of COVID-19 patients with and without statin treatment by means of Cox regression with inverse probability of treatment weighting (IPTW). The median follow-up was 406 days.
Results
We studied 1122 hospitalized patients with COVID-19, whose median age was 71 years and of which 488 (43.5%) were women. 451 (40.2%) patients received statins before admission. In the IPTW survival analysis, prior statin treatment was associated with a significant reduction in mortality (HR: 0.76; 95% CI: 0.59–0.97). The greatest benefit of previous statin therapy was seen in subgroups of patients with coronary artery disease (HR: 0.32; 95% CI: 0.18–0.56) and extracardiac arterial disease (HR: 0.45; 95% CI: 0.28–0.73).
Conclusions
Our study showed a significant association between previous treatment with statins and lower mortality in hospitalized patients with COVID-19. The observed prognostic benefit was greater in patients with previous coronary or extracardiac atherosclerotic disease.
Introducción
El tratamiento con estatinas podría presentar un efecto pronóstico beneficioso en pacientes con COVID-19, dadas sus propiedades inmunomoduladoras, antiinflamatorias y estabilizadoras de la placa de ateroma. Nuestro propósito fue analizar esta hipótesis tomando como base el registro de COVID-19 de un hospital universitario español.
Métodos
Realizamos un estudio observacional y retrospectivo en el que se incluyeron los pacientes hospitalizados con COVID-19 diagnosticado mediante PCR entre marzo de 2020 y octubre de 2020 en un centro. Mediante regresión logística, diseñamos una puntuación de propensión para estimar la probabilidad de que un paciente recibiese tratamiento con estatinas antes del ingreso. Comparamos la supervivencia de los pacientes con y sin tratamiento con estatinas mediante la regresión de Cox ponderada por la inversa de la probabilidad de recibir el tratamiento (IPT). La mediana de seguimiento fue de 406 días.
Resultados
Estudiamos 1.122 pacientes hospitalizados con COVID-19, cuya mediana de edad era de 71 años y de los cuales 488 (43,5%) eran mujeres. 451 (40,2%) pacientes recibían estatinas antes del ingreso. En el análisis de supervivencia ponderado por la IPT, el tratamiento previo con estatinas se asoció a una reducción significativa de la mortalidad (HR: 0,76; IC 95%: 0,59–0,97). El mayor beneficio del tratamiento previo con estatinas se observó en los subgrupos de pacientes con enfermedad arterial coronaria (HR: 0,32; IC 95%: 0,18–0,56) y enfermedad arterial extracardiaca (HR: 0,45; IC 95%: 0,28–0,73).
Conclusiones
Nuestro estudio mostró una asociación significativa entre el tratamiento previo con estatinas y una menor mortalidad en pacientes hospitalizados con COVID-19. El beneficio pronóstico observado fue mayor en los pacientes con enfermedad aterosclerótica coronaria o extracardiaca previa.
Keywords
Statins
Survival
COVID-19
SARS-CoV-2
Palabras clave
Estatinas
Supervivencia
COVID-19
SARS-CoV-2
==== Body
pmcIntroduction
The disease caused by the new type 2 coronavirus responsible for severe acute respiratory syndrome (SARS-CoV-2), known as COVID-19, is a major challenge for the health system. Despite the advances made in its prevention and treatment, COVID-19 continues to cause significant morbidity and mortality. There is, therefore, a need to look for new therapies to improve the prognosis of patients affected by this disease.
Statin therapy has been proposed as a potential source of significant clinical benefit in patients with COVID-19.1 The rationale behind this working hypothesis is based on its known immunomodulatory, anti-inflammatory and antioxidant properties.2 It has been suggested that statin therapy may limit cytokine release and disease-associated lung damage3; in addition, due to its stabilizing effect on atherosclerotic plaque4 it could also reduce the significant risk of cardiovascular complications in these patients.5
Information from several observational studies and their corresponding meta-analyses suggests that patients with COVID-19 who are pre-treated with statins have a lower risk of mortality and serious disease-related complications.6, 7, 8, 9, 10, 11, 12 Other authors, however, have not been able to confirm this alleged clinical benefit.13 It should be noted that there is significant heterogeneity among the published studies, which limits the validity of pooled analyses to draw reliable conclusions. Moreover, the vast majority of studies have focused only on clinical outcome during the hospital phase, and there is little data on the possible impact of statin therapy on longer-term prognosis.
In view of previous information pointing to a possible benefit of statin therapy in patients with COVID-19 and, at the same time, in view of the fact that there are still gaps in knowledge regarding this working hypothesis, we set out to explore this hypothesis further, using as a basis the information contained in a clinical registry of patients with COVID-19 from a Spanish tertiary level university hospital.
Methods
Study description
The information presented in this manuscript was obtained through an anonymised download from the database of the clinical registry of patients with COVID-19 of the Complejo Hospitalario Universitario de A Coruña (CHUAC), belonging to the Servicio Galego de Saúde (SERGAS). This is an observational and retrospective registry that included all patients diagnosed with COVID-19 using any of the validated microbiological methods (PCR, serology, rapid antigen test) in our health area from March 2020 onwards. The study protocol was approved by the Clinical Research Ethics Committee of the Autonomous Community of Galicia and the of Spanish Agency of Medicines and Medical Devices. Informed consent was obtained from patients, either in writing or verbally, for their inclusion in the study.
The study described in this manuscript only considered patients aged ≥18 years who were hospitalised with a PCR-confirmed diagnosis of COVID-19 for SARS-CoV-2 in a respiratory tract biological specimen between 1 March 2020 and 31 October 2020.
Study objectives
The main objective of the present study was to evaluate a possible positive effect of statin pre-treatment on the survival of patients hospitalized for COVID-19.
Additionally, we aimed to assess the potential impact of maintenance or discontinuation of prior statin therapy on survival of patients hospitalised for COVID-19, as well as to explore the potential prognostic benefit of statin therapy in different clinical subgroups of patients hospitalised for COVID-19.
Statin treatment
The primary independent variable in this study was statin treatment prior to hospitalization with COVID-19. Any statin among those marketed in our country was accepted. High-potency statin therapy was defined as regimens equal to or greater than 40 mg daily of atorvastatin or 20 mg daily of rosuvastatin.14 Patients who were not receiving statins prior to hospitalisation were considered the control group, regardless of whether or not a statin therapy was initiated during admission.
Outcome variables
Patients included in the study underwent follow-up until the date of their death or, alternatively, until October 2021. All-cause mortality was the primary outcome variable.
In addition to mortality, the incidence of adverse clinical outcomes during the hospital phase was also analysed, such as acute respiratory distress, need for mechanical ventilation (invasive and/or non-invasive), need for admission to critical care units, acute coronary syndrome, acute heart failure, acute kidney injury, deep vein thrombosis, and pulmonary thromboembolism.
Statistical analysis
In this manuscript, categorical variables are shown as number of patients and proportions, while quantitative variables are expressed as mean ± standard deviation (SD).
In some laboratory variables there is a significant number of missing values, which are specified in the corresponding tables. No missing value imputation method has been applied in this study, so the data shown are for the subgroup of patients with known values.
Using multivariate logistic regression analysis, we constructed a propensity score that allowed us to estimate the likelihood of a patient receiving statin therapy prior to hospitalisation with a diagnosis of COVID-19 based on their baseline clinical characteristics. The model included 14 clinical variables – categories of age, sex, smoking history, hypertension, diabetes mellitus, atrial fibrillation, heart failure, coronary artery disease, peripheral arterial disease, cerebrovascular disease, chronic kidney disease, asthma, chronic obstructive pulmonary disease, neoplasms –, 5 variables related to pre-admission prescriptions – anti-platelets, anticoagulants, beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin 2 receptor blockers – and one interaction variable – sex * beta-blocker prescription –. The inclusion of variables in the propensity model was based primarily on clinical and literature criteria, seeking to select relevant clinical characteristics that might influence the likelihood of a patient receiving statins prior to hospital admission. The selection of the interaction variable sex * beta-blocker prescription was based on a statistical criterion, as its inclusion in the model was found to improve the balance between the groups.
The decision to incorporate age as a categorised variable and not as a continuous variable was taken after finding that it did not meet the linearity assumption necessary for its inclusion in the logistic regression model, due to the low frequency of prescription of these drugs at both age ends of the population. Variables relating to additional therapeutic measures taken during admission – e.g., i.v. corticosteroid prescription, admission to critical care – were not taken into account for the estimation of the propensity score, as they reflected clinical decisions taken after statin prescription. For a similar reason, laboratory variables were not included in the model, as these were determined at the time of hospitalisation and thus after exposure to the pharmacological group under study.
The possible influence of statin exposure prior to hospitalization on the risk of death of patients was studied using an inverse probability of treatment weighted (IPTW) Cox regression analysis.15 To do this, we attributed a specific weight to each case that was determined on the basis of their propensity score for receiving statins. Thus, for patients who actually received statin treatment, the individual case weight was calculated as “1/propensity score”, while in the control group the individual case weight was calculated as “1/(1–propensity score)”. To avoid a disproportionate increase in the sample size, the individual weights of each case were stabilized, multiplying their value by the marginal probability of receiving the treatment.16
The balance of baseline characteristics between the statin treatment group and the control group was assessed taking into account the standardized mean difference (SDM). According to Austin’s rule17 and taking into account the sample size of the study, variables with a SDM < 0.10 were considered to be well balanced between the two groups under study.
To strengthen the consistency of the main result of the IPTW survival analysis, we performed two sensitivity analyses with multivariate adjustment models that included age as a continuous variable (model 1) and variables related to therapeutic measures taken during admission that were unbalanced between groups – use of tocilizumab, use of remdesivir, use of lopinavir–ritonavir, use of hydroxychloroquine, admission to the critical care unit, use of invasive mechanical ventilation – (model 2).
Survival curves, both unweighted and IPTW, for patients with and without statin prior treatment were constructed using the Kaplan–Meier method and compared with the log-rank test.
Finally, we performed an exploratory analysis of the effect of prior statin treatment in several subgroups of hospitalised patients with clinically relevant COVID-19, based on age (<70 years vs. 12 ≥ 70 years) and sex, as well as the presence of hypertension, diabetes mellitus or previous history of coronary artery disease or extracardiac arterial disease, by introducing the corresponding interaction terms into the statistical model.
Statistical significance for the hypothesis tests was defined as a p-value < 0.05. Statistical analysis was performed with SPSS 25 and Stata 14.
Results
Patients
The present study population included 1122 patients aged ≥18 years who were hospitalised with COVID-19 at our centre between March 2020 and October 2020. The process of selecting patients for the study is outlined in Fig. 1 .Fig. 1 Flowchart of the study.
Fig. 1
Of the 1122 patients studied, 451 (40.2%) were receiving statin therapy prior to admission. Specifically, 243 (21.7%) patients were receiving atorvastatin, 113 (10.1%) simvastatin, 36 (3.2%) pravastatin, 33 (2.9%) rosuvastatin, 18 (1.6%) pitavastatin, 6 (0.5%) fluvastatin and 1 (0.1%) lovastatin. In total, 241 (21.5%) patients were receiving high-potency statin therapy and 205 (18.3%) were receiving low- or medium-potency statin therapy. In 1 (0.1%) patient the type of statin prescribed was not recorded and in 5 (0.4%) patients the dose was not recorded.
Statin pre-treatment was maintained during hospitalisation for COVID-19 in 182 (40.4%) patients, while 269 (59.6%) patients discontinued the treatment. In addition, during admission, statin treatment was initiated in 17 (2.5%) of the patients who were not previously receiving them.
Table 1 presents the baseline characteristics of patients hospitalised with COVID-19 in our study, based on the presence or absence of prior statin therapy. Numerous baseline clinical variables can be seen to have a significant imbalance between the two groups of patients, detected by the defined criterion of a SDM > 0.10. Thus, patients treated with statins were more frequently male, had a higher mean age, and a higher prevalence of cardiovascular risk factors and comorbidities, such as coronary, cerebrovascular, or peripheral atherosclerotic disease, heart failure, chronic kidney disease, atrial fibrillation, heart disease, chronic obstructive pulmonary disease and neoplasms.Table 1 Clinical characteristics of 1122 hospitalized patients with COVID-19, based on whether or not they received prior statin treatment.
Table 1 Unweighted sample IPTW sample
Without statins (n = 671) With statins (n = 451) SDMa Without statins With statins SDMa
Previous medical history
Age (years), mean ± SD 63.6 ± 18.9 74.7 ± 10.4 +0.730 68.6 ± 17.7 70.5 ± 12.7 +0.127
Age groups, n (%) +0.668 +0.017
18–59 years 274 (40.8%) 39 (8.6%) 185 (27.1%) 111 (25.3%)
60–69 years 122 (18.2%) 98 (21.7%) 134 (19.6%) 91 (20.8%)
70–79 years 127 (18.9%) 164 (36.4%) 175 (25.7%) 118 (26.9%)
80 years or older 148 (22.1%) 150 (33.3%) 188 (27.6%) 118 (26.9%)
Female, n (%) 309 (46.1%) 179 (39.7%) −0.129 219 (42.7%) 280 (41.1%) −0.033
Hypertension 243 (36.2%) 308 (68.3%) +0.678 337 (49.5%) 231 (52.6%) +0.063
Diabetes mellitus 81 (12.1%) 155 (34.4%) +0.547 139 (20.4%) 97 (22.1%) +0.043
Smoking history 179 (26.7%) 173 (38.4%) +0.251 222 (32.6%) 145 (33%) +0.009
Coronary artery disease 23 (3.4%) 80 (17.7%) +0.478 62 (9.1%) 40 (9.1%) −0.003
Heart failure 56 (8.3%) 91 (20.2%) +0.343 88 (12.9%) 58 (13.2%) +0.011
Atrial fibrillation 63 (9.4%) 75 (16.6%) +0.216 79 (11.6%) 54 (12.3%) +0.023
Cerebrovascular disease 42 (6.3%) 58 (12.9%) +0.226 59 (8.7%) 41 (9.3%) +0.022
Peripheral arterial disease 26 (3.9%) 43 (9.5%) +0.228 48 (7%) 28 (6.4%) −0.029
Chronic obstructive pulmonary disease 39 (5.8%) 42 (9.3%) +0.133 53 (7.8%) 32 (7.3%) −0.018
Bronchial asthma 54 (8%) 35 (7.8%) −0.011 49 (7.2%) 29 (6.6%) −0.022
Chronic kidney disease 55 (8.2%) 91 (20.2%) +0.348 96 (14.1%) 59 (13.4%) −0.015
Chronic liver disease 23 (3.4%) 21 (4.7%) +0.068 33 (4.8%) 16 (3.7%) −0.054
Neoplasm 116 (17.3%) 102 (22.6%) +0.134 136 (20%) 92 (21%) +0.027
Other drugs
Beta blockers 52 (8%) 120 (27%) +0.516 118 (17.3%) 70 (15.9%) −0.040
Angiotensin converting enzyme inhibitors 78 (12%) 85 (19%) +0.202 89 (13.1%) 68 (15.5%) +0.070
Angiotensin 2 receptor blockers 103 (15%) 153 (34%) +0.436 160 (23.5%) 103 (23.5%) −0.001
Antiplatelet agents 56 (8%) 151 (34%) +0.649 134 (19.7%) 84 (19.1%) −0.015
Anticoagulants 57 (9%) 73 (16%) +0.235 73 (10.7%) 51 (11.6%) +0.028
SD: standard deviation; SDM: standardized deviation of the means; IPTW: inverse probability of treatment weighted.
a Positive SDM value indicates that the mean of the variable is higher in the statin-treated group than in the control group, while a negative SDM value indicates that the mean of the variable is higher in the control group than in the statin-treated group. The absolute values |SDM| > 0.10, highlighted in black, identify variables that show a significant imbalance between the two groups under study.
Table 2 shows the baseline clinical characteristics of patients with and without prior statin exposure after sample IPTW with stabilised individual weights. The vast majority of the relevant clinical variables in this analysis had an absolute value |SDM| < 0.10, i.e., a good balance between the two groups under study. The only baseline clinical variable that showed a significant imbalance between the study groups was age, analysed as a continuous variable [see previous explanation in the Methodology section]; however, a balanced distribution of patients was achieved according to the defined age categories.Table 2 Clinical status of patients with COVID-19 and therapeutic measures undertaken during admission.
Table 2 Unweighted sample IPTW sample
Without statins (n = 671) With statins (n = 451) SDMa Without statins (n = 681) With statins (n = 439) SDMa
Clinical situation on admissionb
Systolic blood pressure (mmHg), mean ± SD 129.4 ± 21.9 133.1 ± 22.8 +0.166 129 ± 21.8 133 ± 21.9 +0.186
PaO2 (mmHg) 71.9 ± 27.1 67.7 ± 19.1 −0.182 72.1 ± 26.2 69.5 ± 21 −0.108
PaCO2 (mmHg) 35.8 ± 7 36 ± 7.2 +0.027 35.9 ± 7.4 35.8 ± 6.6 −0.014
pH 7.45 ± 0.10 7.44 ± 0.06 −0.099 7.45 ± 0.06 7.45 ± 0.05 −0.051
PaO2/FiO2 281 ± 126.1 260 ± 102.5 −0.183 277.3 ± 117.6 272.4 ± 100.6 −0.045
Laboratory on admissionb
Platelets (×109/l) 214.9 ± 107 202.3 ± 85.3 −0.131 207.4 ± 99.7 201.3 ± 83.3 −0.067
Leukocytes (×109/l) 7.1 ± 4.2 7.5 ± 6.3 +0.068 7.2 ± 4.2 7.2 ± 6 −0.002
Lymphocytes (×109/l) 1.14 ± 0.79 1.18 ± 2.88 +0.018 1.1 ± 0.8 1.2 ± 2.7 +0.054
Neutrophils (×109/l) 5.5 ± 3.7 5.9 ± 3.8 +0.083 5.7 ± 3.6 5.8 ± 3.8 +0.030
Haemoglobin (g/dl) 13.3 ± 1.9 13.1 ± 2.0 −0.132 13.2 ± 2.0 13.3 ± 1.9 +0.096
Creatinine (mg/dl) 0.99 ± 0.54 1.26 ± 1.09 +0.309 1.11 ± 0.71 1.11 ± 0.87 −0.008
GOT (IU/l) 44.0 ± 40.2 47.2 ± 40.8 +0.079 44.6 ± 40 47.2 ± 38.2 +0.067
GPT (IU/l) 46.6 ± 50.0 41.9 ± 34.4 −0.111 44.3 ± 47.5 43.8 ± 32.6 −0.014
ESR (mm) 49.0 ± 27.3 54.7 ± 28.3 +0.204 50.7 ± 29.2 51.1 ± 28.3 +0.012
C-reactive protein (mg/l) 7.9 ± 7.2 8.1 ± 6.8 +0.038 8.0 ± 6.9 8.0 ± 6.9 +0.007
Ferritin (ng/ml) 623.2 ± 702.7 675.7 ± 749.0 +0.072 603.4 ± 658.1 694.5 ± 717 +0.132
D-dimers 1.382.4 ± 2.815.7 2.015.8 ± 7.921.3 +0.107 1.758.6 ± 3.069 1.627.2 ± 7.280.8 −0.023
Interleukin 6 34.8 ± 79.1 33.2 ± 56.2 −0.023 38.8 ± 77 29 ± 43.7 −0.156
Therapeutic measures during admission
Lopinavir–ritonavir, n (%) 275 (41%) 193 (42.8%) +0.037 274 (40.2%) 213 (48.5%) +0.167
Hydroxychloroquine 307 (45.8%) 227 (50.3%) +0.092 322 (47.2%) 239 (54.4%) +0.144
Tocilizumab 59 (8.8%) 63 (14%) +0.163 59 (8.7%) 71 (16.2%) +0.228
Remdesivir 65 (9.7%) 31 (6.9%) −0.102 61 (9%) 27 (6.2%) −0.109
Low molecular weight heparin 609 (90.8%) 410 (90.9%) +0.005 625 (91.8%) 404 (92%) +0.013
IV corticosteroids 353 (52.6%) 258 (57.2%) +0.092 367 (53.9%) 238 (54.2%) +0.005
Admission to critical care unit 64 (9.5%) 62 (13.7%) +0.131 66 (9.7%) 61 (13.9%) +0.130
Non-invasive mechanical ventilation 80 (11.9%) 50 (11.1%) −0.026 85 (12.5%) 47 (10.7%) −0.053
Invasive mechanical ventilation 51 (7.6%) 49 (10.9%) +0.113 48 (7%) 49 (11.2%) +0.146
SD: standard deviation; SDM: standardized difference of means; GOT: glutamic oxaloacetic transaminase; GPT: glutamate-pyruvate-transaminase; IPTW: inverse probability of treatment weighted; ESR: erythrocyte sedimentation rate.
a A positive SDM value indicates that the mean of the variable is higher in the statin-treated group than in the control group, while a negative SDM value indicates that the mean of the variable is higher in the control group than in the statin-treated group. The absolute values |SDM| > 0.10, highlighted in bold font, identify variables that show a significant imbalance between the two groups under study.
b Missing values: systolic blood pressure (n = 779), PaO2 (n = 287), PaCO2 (n = 289), pH (n = 292), PaO2/FiO2 (n = 718), platelets (n = 15), leukocytes (n = 15), lymphocytes (n = 15), neutrophils (n = 15), haemoglobin (n = 14), creatinine (n = 27), GOT (n = 50), GPT (n = 96), ESR (n = 744), C-reactive protein (n = 368), ferritin (n = 615), D-dimers (n = 466), interleukin 6 (n = 343).
Additional Material Appendix B shows the distribution of propensity scores in the study patients, both in the unweighted population and in the IPTW population.
Survival
The median follow-up period of the patients was 406 days (interquartile range 338–548 days). During this period, there were 130 (27.8%) deaths in the group of patients prescribed statins prior to admission and 131 (20%) deaths in the control group.
Fig. 1 shows the unweighted Kaplan–Meier survival curve for both groups. Unweighted Cox regression analysis showed a significant increase in the risk of death among patients receiving pre-hospitalisation statin therapy (hazard ratio [HR]: 1.50; 95% CI: 1.18–1.91).
The IPTW Kaplan–Meier survival curves are shown in Fig. 2 , panel A). Pre-admission statin therapy was associated with a statistically significant reduction in all-cause mortality during follow-up (HR: 0.76; 95% CI: 0.59–0.97) in the IPTW Cox regression analysis.Fig. 2 Kaplan–Meier survival curves in hospitalized patients with COVID-19, based on the presence or absence of a statin prescription prior to admission. (A) Unweighted survival analysis. (B) Inverse probability of treatment weighted (IPTW) survival analysis.
Fig. 2
The protective effect of statin pre-treatment was maintained when the results of the IPTW Cox regression were further adjusted by a first multivariate model including age (HR: 0.75; 95% CI: 0.59–0.96) and by a second multivariate model which, in addition to age, also included variables related to therapeutic measures taken during hospitalisation that showed a significant imbalance between the two study groups – tocilizumab use, remdesivir use, lopinavir–ritonavir use, hydroxychloroquine use, admission to the critical care unit, invasive mechanical ventilation use – (HR: 0.76; 95% CI: 0.59–0.97).
The protective effect of treatment with statins was observed mainly in patients in whom the drug was maintained during admission (HR: 0.64; CI 95%: 0.43–0.94); however, the association between prior statin exposure and lower mortality did not reach statistical significance in the subgroup of patients in whom the drug was discontinued at the time of hospitalization (HR: 0.80; CI 95%: 0.61–1.06). Appendix B Supplementary Table S1 shows the baseline clinical characteristics of both subgroups of patients.
The association between prior statin treatment and a lower risk of death was obtained at the expense of the subgroup of patients treated with low or moderate potency statins (HR: 0.63; CI 95%: 0.45–0.88); however, we did not observe a significant association between prior high-potency statin therapy and the risk of death from any cause (HR: 0.88; CI 95%: 0.65–1.18).
Sub-group analysis
Fig. 2, panel B, shows the observed statistical association between prior statin exposure and risk of all-cause death in different subgroups of hospitalised patients with COVID-19, estimated by IPTW Cox regression analysis.
A statistically significant interaction was detected between the existence of atherosclerotic disease, both coronary (p = 0.002) and extracardiac (cerebral or peripheral) (p = 0.012), and the observed association between statin treatment and survival. Pre-treatment with statins was associated with a marked reduction in the risk of death in the subgroups of patients with coronary artery disease (HR: 0.32; 95% CI: 0.18–0.56) and with extracardiac arterial disease (HR: 0.45; 95% CI: 0.28–0.73).
We did not observe a significant interaction between statin exposure and survival of patients hospitalised for COVID-19 based on age, sex or the presence or absence of hypertension or diabetes mellitus (Fig. 3 ).Fig. 3 Hazard ratio for all-cause mortality in hospitalized patients with COVID-19 receiving statin therapy prior to admission vs. patients without statin treatment: inverse probability of treatment weighted-Cox regression analysis.
Fig. 3
Other adverse clinical outcomes during hospitalization
Table 3 shows the cumulative incidence of different adverse clinical outcomes recorded during the in-hospital phase according to the presence or absence of prior statin treatment.Table 3 Adverse clinical events during hospital admission for COVID-19 in patients previously treated with statins and in patients in the control group, both in the unweighted analysis and the IPTW.
Table 3 Unweighted sample IPTW sample
Without statins (n = 671) With statins (n = 451) OR (95% CI) Without statins (n = 681) With statins (n = 439) OR (95% CI)
Acute respiratory distress 152 (22.7%) 134 (29.7%) 1.44 (1.10–1.89) 163 (23.9%) 119 (27.1%) 1.18 (0.90–1.55)
Acute myocardial infarction 2 (0.3%) 2 (0.4%) 1.49 (0.21–10.6) 1 (0.1%) 1 (0.2%) 1.03 (0.81–13.24)
Acute heart failure 13 (1.9%) 19 (4.2%) 2.23 (1.09–4.05) 14 (2.1%) 14 (3.2%) 1.47 (0.69–3.10)
Stroke 4 (0.6%) 7 (1.6%) 2.63 (0.77–9.03) 4 (0.6%) 4 (0.9%) 1.53 (0.37–6.35)
Pulmonary embolism 25 (3.7%) 21 (4.7%) 1.26 (0.70–2.28) 3. 4. 5%) 23 (5.2%) 1.04 (0.60–1.79)
Deep vein thrombosis 6 (0.9%) 7 (1.6%) 1.75 (0.28–5.73) 10 (1.5%) 11 (2.5%) 1.76 (0.74–4.20)
Acute kidney damage 69 (10.3%) 83 (18.4%) 1.97 (1.39–2.78) 94 (13.8%) 62 (14.1%) 1.02 (0.72–1.44)
IPTW: inverse probability of treatment weighted; OR: odds ratio.
In bold font, statistically significant results.
The unweighted univariate analysis showed that statin-treated patients had a higher cumulative incidence of different adverse clinical outcomes, including acute respiratory distress, acute heart failure and acute kidney damage. However, after the IPTW analysis, no statistically significant differences were observed between the groups for the main adverse clinical outcomes analysed.
Discussion
The present observational study, based on a cohort of adult patients hospitalised with COVID-19 in a Spanish university hospital, suggests that pre-treatment with statins is associated with a decreased risk of all-cause death in this population. According to our results, the benefit of treatment with statins would be greater in patients in whom the drug was maintained during hospitalization and in patients with a history of coronary or extracardiac (cerebrovascular or peripheral) atherosclerotic disease.
First, we would like to highlight the notable differences in the baseline clinical characteristics of hospitalised patients with COVID-19 who received prior statin therapy compared to those who did not, a fact that has been consistent in other previously published studies.18, 19, 20 The statin-treated group included older patients, mainly males and with a higher prevalence of comorbidities in the cardiovascular, renal, bronchopulmonary and neoplastic domains. It is not surprising, therefore, that in the survival analysis without weighting the treated group presented a significantly higher mortality than the control group. It was precisely the large number of baseline clinical variables with a significant imbalance between the two study groups that we chose to use the IPTW method for statistical adjustment of the results, since this technique is known to be appropriate to simultaneously control for the possible confounding effect of multiple covariates in studies of intermediate sample size that provide limited statistical power for the use of other adjustment methods, such as propensity score adjustment or simple multivariable regression.15 Using the IPTW method with stabilised weights17 in our study allowed an adequate balance of the baseline clinical variables considered as potential confounders, with the exception of a discrete age imbalance. As discussed above, IPTW survival analysis models, with or without additional adjustment for age or use of other concomitant treatments, showed a statistically significant and clinically relevant protective effect of statin therapy in patients hospitalised with COVID-19, with a reduction in the risk of death of approximately 25% over a follow-up of slightly more than 1 year after diagnosis of infection.
To date, more than 40 observational studies and several meta-analyses have been published in an attempt to discern whether statin therapy can have a positive prognostic effect in patients with COVID-19.6, 18, 19, 20, 21 The published meta-analyses show significant heterogeneity in terms of the type of studies and patients included and their methodology, and it should be noted that the validity of some of them13 has been openly questioned.22 The recent updated meta-analysis by Vahedian-Azimi et al.9 is probably the most comprehensive meta-analysis published to date, as it included data from 47 observational studies with a total aggregate sample of more than 3 million patients — although more than 90% of the cases came from a single large UK population-based study of patients with COVID-19 and diabetes mellitus.23 In this meta-analysis,9 the authors observed no overall association between statin use and a reduced risk of death or admission to critical care units among patients with COVID-19, but a reduced risk of orotracheal intubation; additionally, the authors describe a significant 46% reduction in all-cause mortality among statin users in whom treatment was effectively maintained during hospitalisation for COVID-19. The meta-analysis by Chow et al.,6 based on 13 observational studies involving more than 110,000 patients with COVID-19, reached a similar conclusion, observing a significant reduction in mortality among statin users where treatment was maintained during hospitalisation, as opposed to all other patients. However, Chow et al.6 did not observe a significant benefit of statin treatment among patients with COVID-19 who required admission to an intensive care unit. Diaz-Arocutipa et al.10 analysed 25 pooled observational studies involving more than 147,000 patients with COVID-19, concluding that statin therapy is associated with a significant reduction in the risk of death in the adjusted analysis, which was observed mainly at the expense of chronic statin users. Other meta-analyses of observational studies7, 8, 11 have shown similar conclusions. The meta-analysis by Zeim et al.,12 based on 8 observational studies using propensity score matching techniques and including more than 14,000 patients with COVID-19, stands out for its methodological quality. The authors of this paper12 concluded that statin therapy appears to be associated with a significant prognostic benefit in patients with COVID-19, with a 28% reduction in the risk of death independent of age, sex or previous history of hypertension or diabetes. Finally, two Spanish multicentre studies19, 20 have also suggested a significant prognostic benefit of statin therapy in patients with COVID-19.
The apparent prognostic benefit of statin therapy in hospitalised patients with COVID-19 in our cohort was observed primarily in the group in which statins were maintained during hospitalisation; however, the effect observed in patients in whom treatment was discontinued at the time of hospitalisation did not reach statistical significance. This observation is consistent with previously published information, since, as we have already discussed, maintenance of prior statin therapy during hospitalisation with COVID-19 has been associated with a significant prognostic impact in these patients more consistently than a history of statin prescription prior to admission.6, 8, 9
Our study revealed a greater prognostic benefit of statin therapy in patients with COVID-19 who had a history of established coronary or extracardiac atherosclerotic disease, i.e. in subgroups of patients with an indication for lipid-lowering therapy in secondary prevention. This result suggests that the hypothetical protective effect of statin therapy in patients with COVID-19 could be due to its stabilising effect on atherosclerotic plaques in patients who are theoretically at higher risk of cardiovascular complications associated with the disease. Similarly, other authors have observed a particularly significant clinical benefit of statin therapy in patients with COVID-19 and coronary artery disease.18 While some authors have suggested that the potential benefit of statin therapy in patients with COVID-19 may be mediated by its immunomodulatory properties, which may mitigate the consequences of the systemic inflammatory response and lung damage associated with the disease,3, 24 we did not observe a reduction in the incidence of respiratory distress in these cases. On the contrary, even after IPTW analysis, patients receiving statins in our cohort had a higher need for mechanical ventilation and admission to critical care units than patients not treated with statins.
Finally, we would like to comment on a striking finding of the present study, which is the apparent differential effect of the intensity of lipid-lowering therapy on the prognosis of patients with COVID-19. Surprisingly, the greatest reduction in the risk of death was observed among patients receiving treatment with low or intermediate potency statins, while the reduction in the risk of death observed among patients receiving high potency statins was quantitatively smaller and did not reach statistical significance. Although this exploratory result should be taken with caution and requires confirmation, it may be related to the increased risk of adverse reactions and drug interactions and, perhaps, a greater tendency to discontinuation of treatment associated with high-intensity lipid-lowering regimens, which could be of particular clinical relevance in a particularly vulnerable clinical period such as the hospital phase of COVID-19. Contrary to our study, other authors18 have found no difference in benefit between high-intensity statin therapy and moderate- or low-intensity statin therapy in patients with COVID-19.
The present study has some limitations. Firstly, it is an observational and retrospective study, which is therefore exposed to inherent selection, information and confounding biases. There are a significant number of missing values in the database used for the study, mainly affecting baseline laboratory variables and, to a lesser extent, the doses and type of statin used. In addition, information on prior prescription of statins and their maintenance or discontinuation at the time of hospital admission was obtained from the medical records, but we do not have data on the actual dispensing and adherence to treatment. In particular, the prescription of statins during hospitalisation to 17 patients in the control group might have limited the ability of the study to demonstrate the proposed objective; however, it is our opinion that this fact probably did not result in a significant modification of the results of the study, given the small number of cases involved. The statistical methodology employed, with hypothesis contrasts weighting and survival analysis by IPTW, allowed a good number of baseline clinical variables to be balanced between the two study groups, with the aim of making them more comparable. However, this technique does not control for possible confounding biases due to unmeasured covariates, nor does it ensure a good balance of measured covariates in the subgroups analysed retrospectively. Finally, it should be acknowledged that the external validity of the study is not guaranteed, as it included patients from a single centre.
In summary, when using IPTW techniques, we observed a statistically significant association between prior statin treatment and a lower risk of death in a retrospective cohort of adult hospitalised patients with COVID-19 in a Spanish university hospital. The prognostic benefit of statin therapy was greater in patients with a history of coronary or extracardiac atherosclerotic disease – cerebrovascular or peripheral – and in those in whom these drugs were maintained during admission due to infection. Our results suggest that, in the absence of contraindication or relevant adverse reactions, pre-treatment with statins should be maintained as far as possible during hospitalisation in patients with COVID-19 who have a well-established indication for statin therapy.
Funding
This research project was funded by a grant from the Fundación Mutua Madrileña for biomedical research focused on coronavirus infection (COVID-19) in its extraordinary call PRI-2020-13 of April 2020. The project was co-funded by the Consellería de Sanidade, the Servizo Galego de Saúde and 10.13039/501100001129 ACIS through the Programa de refuerzo de la investigación sanitaria de Galicia Traslaciona COVID-19 with a charge to social patronage within the campaign “Botemos unha manha” of the Xunta de Galicia (File No. CT850A - 8).
Conflict of interests
No conflict of interest is declared.
Appendix A Supplementary data
The following is Supplementary data to this article:
Appendix A Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.medcle.2022.03.021.
==== Refs
References
1 Lima Martínez M. Contreras M.A. Marín W. Marco D.L. Statins in COVID-19: is there any foundation? Clin Invest Arterioscler 32 2020 278 281
2 Zeiser R. Inmune modulatory effects of statins Inmunology 154 2018 69 75
3 Bifulco M. Gazzero P. Statin therapy in patients with COVID-19: much more than a single pathway Eur Heart J Cardiovasc Pharmacother 6 2020 410 411 10.1093/ehjcvp/pvaa055 32529218
4 Konishi T. Funayama N. Yamamoto T. Hotta D. Nomura R. Nakagaki Y. Stabilization of symptomatic carotid atherosclerotic plaques by statins: a clinico-pathological analysis Heart Vessels 33 2018 1311 1324 29789903
5 Cenko E. Badimon L. Bugiardini R. Claeys M. de Luca G. de Wit C. Cardiovascular disease and COVID-19: a consensus paper from the ESC Working Group on Coronary Pathophysiology & Microcirculation, ESC Working Group on Thrombosis and the Association for Acute CardioVascular Care (ACVC), in collaboration with the European Heart Rhythm Association (EHRA) Cardiovasc Res 117 2021 2705 2729 10.1093/cvr/cvab298 34528075
6 Chow R. Im J. Chiu N. Chiu L. Aggarwal R. Lee J. The protective association between statins use and adverse outcomes among COVID-19 patients: a systematic review and meta-analysis PLoS One 16 2021 e0253576 34166458
7 Kow C.S. Hasan S.S. Meta-analysis of effect of statins in patients with COVID-19 Am J Cardiol 134 2020 153 155 32891399
8 Permana H. Huang I. Purwiga A. Kusumawardhani N.Y. Sihite T.A. Martanto E. In-hospital use of statins is associated with a reduced risk of mortality in coronavirus-2019 (COVID-19): systematic review and meta-analysis Pharmacol Rep 73 2020 769 780 10.1007//s43440-021-00233-3
9 Vahedian-Azimi A. Mohamadi S. Banach M. Beni F.H. Guest P.C. al-Rasadi K. Improved COVID-19 outcomes following statin therapy: an updated systematic review and meta-analysis Biomed Res Int 2021 2021 1901772
10 Diaz-Arocutipa C. Melgar-Talavera B. Alvarado-Yarasca A. Saravia-Bartra Cazorla P. Belzusarri I. Statins reduce mortality in patients with COVID-19: an updated meta-analysis of 147 824 patients Int J Infect Dis 110 2021 374 381 34375760
11 Kollias A. Kyriakoulis K.G. Kyriakoulis I.G. Nitsotolis T. Poulakou G. Stergiou G.S. Statin use and mortality in COVID-19 patients: updated systematic review and meta-analysis Atherosclerosis 330 2021 114 121 34243953
12 Zeim A.F. Sulistiyana C.S. Khasanah U. Wibowo A. Lim M.A. Pranata R. Statin and mortality in COVID-19: a systematic review and meta-analysis of pooled adjusted effect estimates from propensity-matched cohorts Postgrad Med J 2021 10.1136/postgradmedj-2021-140409
13 Hariyanto T.I. Kurniawan A. Statin and outcomes of coronavirus disease 2019 (COVID-19): a systematic review, meta-analysis, and meta-regression Nutr Metab Cardiovasc Dis 31 2021 1662 1670 33838992
14 Stone N.J. Robinson J.G. Lichtenstein A.H. Bairey Merz C.N. Blum C.B. Eckel R.H. American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines J Am Coll Cardiol. 63 2014 2889 2934 24239923
15 Elze M.C. Gregson J. Baber U. Williamson E. Sartori S. Mehran R. Comparison of propensity score methods and covariate adjustment: evaluation in 4 cardiovascular studies J Am Coll Cardiol 69 2017 345 347 28104076
16 Xu S. Ross C. Raebel M.A. Shetterly S. Blanchette C. Smith D. Use of stabilized inverse propensity scores as weights to directly estimate relative risk and its confidence intervals Value Health 13 2010 273 277 19912596
17 Austin P.C. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples Statis Med 28 2009 3083 3107
18 Choi D. Chen Q. Goonewardena S.N. Pacheco H. Mejia P. Smith R.L. Efficacy of statin therapy in patients with hospital admission for COVID-19 Cardiovasc Drugs Ther 2021 10.1007/s10557-021-07263-2
19 Masana L. Correig E. Rodriguez-Borjabad R. Anoro E. Arroyo J.A. Jericó C. Effect of statin therapy on SARS-COV-2 infection-related mortality in hospitalized patients Eur Heart J Cardiovasc Pharmacother 8 2020 157 164 10.1093/ehjcvp/pvaa128
20 Torres-Peña J. Pérez-Belmonte L. Fuentes-Jiménez F. López-Carmona M.D. Pérez-Martínez P. López-Miranda J. Prior treatment with statins is associated to improved outcomes of patients with COVID-19: data from the SEMI-COVID Registry Drugs 81 2021 685 695 33782908
21 Bergqvist R. Ahlqvist V.H. Lundberg M. Hergens M.P. Sundrström J. Bell M. HMG-CoA reductase inhibitors and COVID-19 mortality in Stockholm, Sweden: a registry-based cohort study PLoS Med 8 2021 e1003820 10.1371/Journal.pmed.1003820
22 Tandaju J.R. Ij W. Barati-Boldaji R. Raeisi-Dehkordi H. Meta-analysis of statin and outcomes of coronavirus disease 2019 (COVID-19): reconsideration is needed Nutr Metab Cardiovasc Dis 31 2021 2737 2739 34366177
23 Holman N. Knighton P. Kar P. O’Keefe J. Curley M. Weaver A. Risk factors for COVID-19 related mortality in patients with type 1 and type 2 diabetes in England: a population-based cohort study Lancet Diab Endocrinol 8 2020 823 833
24 Kashour T. Halwani R. Arabi Y.M. Sohail M.R. O’Horo J.C. Badley A.D. Statins as adjunctive therapy for COVID-19: the biological and clinical plausibility Immunopharmacol Immunotoxicol 43 2020 37 50
| 36504601 | PMC9726688 | NO-CC CODE | 2022-12-16 23:18:15 | no | Med Clin (Engl Ed). 2022 Dec 7; doi: 10.1016/j.medcle.2022.03.021 | utf-8 | Med Clin (Engl Ed) | 2,022 | 10.1016/j.medcle.2022.03.021 | oa_other |
==== Front
Aten Primaria
Aten Primaria
Atencion Primaria
0212-6567
1578-1275
Published by Elsevier España, S.L.U.
S0212-6567(22)00272-4
10.1016/j.aprim.2022.102552
102552
Article
Impacto del confinamiento por COVID-19 en la prescripción de benzodiacepinas
Impact of COVID-19 lockdown on the prescription of benzodiazepinesDiez Sara García a⁎
Valdés Míriam De Nicolás a
Varela Cristina Diéguez b
Martínez Paula Fernández c
Gil Patricio Suárez c
Rodríguez Yolanda Navarro d
a Centro de Salud Contrueces, Calle Rio Cares, 20, 33210 Gijón, Principado de Asturias, España
b Centro de Salud Natahoyo, Avenida de Juan Carlos I, 60A, 33212 Gijón, Principado de Asturias, España
c Plataforma de Bioestadística y Epidemiología del Instituto de Investigación Sanitaria del Principado de Asturias (ISPA), Avenida de Roma s/n, 33011 Oviedo, Principado de Asturias. España
d Centro de Salud El Coto, Calle Avelino González Mallada, 29, 33204 Gijón, Principado de Asturias, España
⁎ Autor de correspondencia: calle San Bernardo 20, 3ºD, 33201 Gijón, Principado de Asturias, España
7 12 2022
7 12 2022
10255220 7 2022
29 11 2022
© 2022 Published by Elsevier España, S.L.U.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objetivo: Evaluar el efecto del confinamiento por COVID-19 sobre la prescripción de benzodiacepinas según edad, sexo y zona básica de salud.
Diseño: Estudio observacional longitudinal.
Emplazamiento: Atención primaria. Área V de Salud del Principado de Asturias.
Participantes: Mayores de 15 años a los que se prescribieron benzodiacepinas entre 2017 y 2020.
Mediciones principales: Diferencia de las medias de DHD (dosis diaria definida por 1.000 habitantes) mensual de benzodiacepinas entre el periodo definido como preconfinamiento y el confinamiento. Además, se ajusta la diferencia por edad, sexo y zona básica de salud, así como por la interacción entre ellas.
Resultados: La DHD media preconfinamiento fue 131,3 y 139,5 durante el confinamiento; en el análisis crudo, esta diferencia fue estadísticamente significativa [IC 95% 4,1 a 12,1]. Se objetivó un aumento de DHD media en el grupo de 60-74 años [IC 95% 2,28 a 21,42] y en el de 90 años ó más [IC 95%. 21,31 a 40,63], así como en las mujeres [IC 95% 3,51 a 14,59]. Finalmente, se observó una disminución de DHD media en las zonas básicas V11 [IC 95% -29 a -0,66] y V14 [IC 95% -54,28 a -25,04].
Conclusiones: Determinados subgrupos muestran un cambio en la tendencia de dispensación de benzodiacepinas sin poder atribuirse completamente al confinamiento. Creemos que pueda existir una inercia terapéutica en la prescripción de psicofármacos, según las características biopsicosociales del paciente, que es importante detectar para evitar la medicalización de cuadros psicológicos.
Objective: To evaluate the effect of COVID-19 lockdown on the prescription of benzodiazepines by gender, age and district health departments.
Design: Longitudinal observational study.
Location: Primary care. Asturias (Spain) health district V.
Participants: People over 15 years of age with filled benzodiazepine prescriptions in between 2017 and 2020.
Main measurements: Benzodiazepine DHD (defined daily dose per 1000 habitants) mean difference between the period defined as pre-lockdown and lockdown. Additionally, the difference was adjusted for gender, sex and district health deparment and also with the interaction among them.
Results: DHD mean pre-lockdown was 131.3 and 139.5 in the lockdown; this difference was significant in the global analysis [IC 95% 4.1 a 12.1]. There was an increase in the DHD mean in the 60-74 age group [IC 95% 2.28 a 21.42], in the group over 90 years old [IC 95%. 21.31 a 40.63] and in women [IC 95% 3.51 a 14.59]. Finally, a decrease in the DHD mean of V11 [IC 95% -29 a -0.66] and V14 [IC 95% -54.28 a -25.04] district health departments was observed.
Conclusions: Certain subgroups show a change in the pattern of benzodiazepine prescription without being able to relate this to the lockdown. We believe that there could be some inertia in the prescription of psychiatric medication according to the biopsychosocial characteristics of the patients; it is important to detect this in order to avoid the medicalization of psychological disorders.
Palabras clave (MeSH)
benzodiazepines
coronavirus infections
prescriptions
social isolation
==== Body
pmc
| 0 | PMC9726689 | NO-CC CODE | 2022-12-09 23:15:17 | no | Aten Primaria. 2022 Dec 7;:102552 | utf-8 | Aten Primaria | 2,022 | 10.1016/j.aprim.2022.102552 | oa_other |
==== Front
Intensive Crit Care Nurs
Intensive Crit Care Nurs
Intensive & Critical Care Nursing
0964-3397
1532-4036
The Authors. Published by Elsevier Ltd.
S0964-3397(22)00169-0
10.1016/j.iccn.2022.103366
103366
Research Article
Physical, social, mental and spiritual functioning of COVID-19 intensive care unit-survivors and their family members one year after intensive care unit-discharge: A prospective cohort study
Onrust Marisa a⁎1
Visser Anja b2
van Veenendaal Nadine c3
Dieperink Willem ad4
Luttik Marie Louise ad5
Derksen Mechteld-Hanna G. e
van der Voort Peter H.J. af6
van der Meulen Ingeborg C. ad7
a University of Groningen, University Medical Center Groningen, Department of Critical Care, the Netherlands
b University of Groningen, Faculty of Theology and Religious Studies, Department of Comparative Study of Religion, the Netherlands
c University of Groningen, University Medical Center Groningen, Department of Anesthesiology, the Netherlands
d Research Group Nursing Diagnostics, Hanze University of Applied Science Groningen, Groningen, the Netherlands
e University Medical Center Groningen, Department of Spiritual Care, the Netherlands
f TIAS School for Business and Society, Tilburg University, Tilburg, the Netherlands
⁎ Corresponding author.
1 ORCID: 0000-0003-3875-373X.
2 ORCID: 0000-0003-3956-3129.
3 ORCID: 0000-0002-4098-4644.
4 ORCID: 0000-0003-2738-7471.
5 ORCID: 0000-0002-4361-8703.
6 ORCID: 0000-0002-3486-2843.
7 ORCID: 0000-0002-8823-4025.
7 12 2022
7 12 2022
10336622 4 2022
14 10 2022
4 12 2022
© 2022 The Authors
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objective
To describe the long-term functioning of patients who survived a COVID-19-related admission to the intensive care unit and their family members, in the physical, social, mental and spiritual domain.
Design
A single-centre, prospective cohort study with a mixed-methods design.
Setting
The intensive care unit of the University Medical Center Groningen in the Netherlands.
Main outcome measures
To study functioning 12 months after intensive care discharge several measurements were used, including a standardised list of physical problems, the Clinical Frailty Scale, the Medical Outcomes Study Short-Form General Health Survey, the McMaster Family Assessment Device, the Hospital Anxiety and Depression Scale, and the Spiritual Needs Questionnaire, as well as open questions and interviews with survivors and their family members.
Results
A total of 56 survivors (77%) returned the 12-month questionnaire, whose median age was 62 (inter-quartile range [IQR]: 55.0–68.0). Moreover, 67 family members (66%) returned the 12-month questionnaire, whose median age was 58 (IQR: 43–66). At least one physical problem was reported by 93% of the survivors, with 22% reporting changes in their work-status. Both survivors (84%) and their family members (85%) reported at least one spiritual need. The need to feel connected with family was the strongest. The main theme was ‘returning to normal’ in the interviews with survivors and ‘if the patient is well, I am well’ in the interviews with family members.
Conclusions
One year after discharge, both COVID-19 intensive care survivors and their family members positively evaluate their health-status. Survivors experience physical impairments, and their family members’ well-being is strongly impacted by the health of the survivor.
Keywords
Aftercare
COVID-19
Critical care
Family
Post intensive care syndrome
Post intensive care syndrome-family
Quality of life
Spirituality
Survivors
==== Body
pmc Implications for clinical practice • Despite regaining functioning after one year, COVID-19 intensive care unit-survivors and their family members express the need for sufficient and appropriate follow-up care.
• Within the social domain, returning to work is an important issue to intensive care unit-survivors and is perceived as a sign of recovery and return to normality. This issue should be included in follow up care by healthcare professionals.
• The different experiences of intensive care unit-survivors and their family members in relation to COVID-19-related intensive care unit-admission can lead to different coping processes and may require different methods of follow-up care for them.
• Within the spiritual domain, nurses, doctors, and other healthcare professionals should be aware that the term ‘spirituality’ is often conflated with religion and that it may be helpful to use other words to identify spiritual needs.
Introduction
Since the outbreak of COVID-19, over 450 million people have been affected by the SARS-Cov-2 virus worldwide (WHOb, 2020), leading to a large number of patients requiring Intensive Care Unit (ICU) admission (Our World in Data, 2021). An ICU-admission has potentially long-lasting consequences for ICU-survivors and their family members. ICU-survivors may develop physical, psychological, and/or cognitive problems, also called Post-Intensive Care Syndrome (PICS) (Harvey and Davidson, 2016, Needham et al., 2012). Approximately 70 % of ICU-survivors develop one or more PICS symptoms one year after ICU-discharge (Geense et al., 2021). Furthermore, family members are prone to develop symptoms of anxiety, depression and Post Traumatic Stress Disorder (PTSD) (van Beusekom et al., 2016), also known as Post-Intensive Care Syndrome-Family (PICS-F) (Davidson et al., 2012).
For both ICU-survivors and their family members, symptoms can have a negative influence on aspects of their social domain of daily living, such as return to work, roles and responsibilities within the family system and Quality of Life (QoL) (Kang and Jeong, 2018, Petrinec and Martin, 2018, van Beusekom et al., 2016). Spiritual questions, such as meaning and purpose in life or identity related questions, may arise as a result of the confrontation with the possibility of death (Ho et al., 2018). In addition, the ICU experience of patients is different from the experience of their family members, which may subsequently affect their personal and dyadic coping processes (Fumis et al., 2015). Spirituality can be defined as ‘… the aspect of humanity that refers to the way individuals seek and express meaning and purpose and the way they experience their connectedness to the moment, to self, to others, to nature, and to the significant or sacred’ (Puchalski et al., 2009). It implies that spirituality can be experienced both within and outside religious traditions.
Given the large number of COVID-19-related ICU-admissions, the incidence of PICS and PICS-F is expected to increase (Berger and Braude, 2021). In addition, besides the known risk factors of PICS and PICS-F (Jolley et al., 2016, Mikkelsen et al., 2020 ), COVID-19 has also created other factors that may increase psychological and spiritual problems, such as visiting restrictions, the lack of visibility of healthcare providers due to wearing protective clothing and social distancing from relatives and friends (Berger and Braude, 2021).
Despite increasing knowledge about COVID-19, little is still known about the one-year consequences for ICU-survivors, and even less for their family members. A recently published study described the clinical outcomes of patients one year after COVID-19-related ICU-admission (Heesakkers et al., 2022a). Another study described the psychological consequences faced by the family members of COVID-19 ICU-survivors (Heesakkers et al., 2022b). We herein present the COVID-19 Follow-up Intensive Care Study (COFICS) to describe the long-term functioning of COVID-19 ICU-survivors and their family members in the physical, social, mental and spiritual domain. Previously we reported the three- and six-months results (van Veenendaal et al., 2021). In this paper, we report on their 12-month outcomes.
Methods
Study design
The COFICS is a single-centre, prospective cohort study with a mixed-methods design, performed at the ICU of the University Medical Center Groningen (UMCG) in the Netherlands. The ICU is a mixed ICU and consists of 35 beds with over 3000 admissions per year. Of these, 19 % are cardio-thoracic admissions, 45 % are medical admissions and 36 % are other admissions. Before the COVID-19 pandemic the overall median ICU Length of Stay (LOS) was 2.5 days (Stichting NICE, 2019).
This study was conducted in accordance with the convergent design, where qualitative and quantitative data are being collected simultaneously and only after that, merged and compared (Creswell and Plano Clark, 2017). The STROBE checklist for cohort studies was used to guide the quantitative part of this study.
Participants
Recruitment took place between March 19th and September 30th 2020. All COVID-19 patients who survived the ICU and their family members were eligible to participate. COVID-19 was diagnosed according to World Health Organization (WHO) definition and was confirmed by RNA detection of the SARS-CoV-2 using the polymerase chain reaction (PCR)-based technique. Family members could be partners, children, other family members, or friends who were identified by the patient as important.
Data collection
An experienced research nurse invited eligible ICU-survivors by telephone to participate in the study, three months after ICU-discharge. In addition, the ICU-survivors were asked permission to contact family members for participation. One or more family member(s) per ICU-survivor could participate. For the first part of the study, questionnaires were sent by mail three and six months after ICU-discharge (van Veenendaal et al., 2021). For the current part of the study, questionnaires were sent by mail 12 months after ICU-discharge. All participants received validated questionnaires, combined with open questions. Additionally, participants received an invitation to an interview with a healthcare chaplain under training.
Measures
Fig. 1 shows the measurements used to study the physical, social, mental and spiritual well-being of ICU-survivors and their family members.Fig. 1 The bundle of measurements used in the COFICS at 12 months post ICU-discharge. NB. Questions in de physical domain were only offered to ICU-survivors.
Questionnaires
The Dutch version of the 9-point Clinical Frailty Scale (CFS) (Dieperink et al., 2017) was used to express the overall level of frailty. The CFS is found to be highly correlated with the Frailty Index (Pearson coefficient 0.80, p < 0.01) showing good construct validity and reliability (intraclass correlation coefficient 0.97, p < 0.001). (Rockwood et al., 2005, Rockwood et al., 2015). Physical symptoms were collected using a standardised list based on our own results at six months (van Veenendaal et al., 2021), as well as two other cohort studies among non-COVID-19 and COVID-19 ICU-survivors (Geense et al., 2021, Huang et al., 2021) (Table 3). To study physical and social functioning, the Medical Outcome Study (MOS) Short-Form General Health Survey (SF-20) was used (Kempen, 1992), containing six QoL domains: physical functioning, role functioning, social functioning, mental health, general health perception and pain. The SF-20 is validated in the general Dutch population (Kempen, 1992) and each subscale showed sufficient correlation coefficient scores ranging from 0.030 to 0.73 (p < 0.001) and high test–retest reliability ranging from 0.58 to 0.85 (p < 0.001) (Kempen, 1992, van Rompaey et al., 2009). Overall family functioning was measured using the general functioning scale of the McMaster Family Assessment Device (FAD-GF6 + ) (Boterhoven de Haan et al., 2015, Hamilton and Carr, 2016). The assessment shows satisfying validity scores (a correlation of 0.909, 95 % CI: 0.90 – 0.92) (Hamilton and Carr, 2016). Psychological well-being was measured with the Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983). The validity of the HADS is sufficient with correlation coefficients ranged between 0.67 and 77 whereas the Cronbach’s alpha scores varied for HADS-A from 0.68 to 9.3 and for HADS-D from 0.67 to 0.90 showing satisfying reliability (Bjelland et al., 2002) and is validated in the Dutch population (Spinhoven et al., 1997). To study spiritual needs, the Spiritual Needs Questionnaire (SpNQ) was used, containing four subscales of spiritual needs: religious needs, existential needs, inner peace needs and needs for positive confirmation (Büssing et al., 2010, Büssing et al., 2018, Nolan). The SpNQ has shown to be reliable with Cronhach’s alpha’s ranging from 0.71 to 0.87 (Nolan, 2022). Several versions of the questionnaire are available. In the current samples all subscales except ‘existential needs’ showed good reliability. For an overview of the version used in the COFICS, refer to supplement 1.
Interviews
Open interviews were conducted by two health care chaplains under training (student researchers), in the context of their master theses. The student researchers interviewed ICU-survivors and family members separately and independent from each other. Previously the student researchers received interview training as part of their education. The interviews started with the question ‘How have you experienced the past year, since your (loved one’s) ICU-admission?’ Subsequently participants were asked about three topics, i.e. spiritual experiences, needs regarding peace, religion, connectedness and vulnerability, and resources that helped to fulfil these needs. In addition, the role of spirituality during and after the ICU stay was explored.
All interviews were conducted remotely, either via Microsoft Teams or by telephone, and audio-recorded. Each interview lasted 55 min on average, ranging from 35 to 105 min.
Demographics
Demographic characteristics of the ICU-survivors, including age, gender and clinical data, such as length of hospital- and ICU-stay, comorbidities and delirium, were retrieved from the electronic patient files. Other demographic data, such as educational level and marital status, as well as healthcare consumption and family characteristics, were addressed in the questionnaire.
Data analysis
Quantitative data are reported as a median with inter-quartile range (IQR), mean with associated standard deviation, or number with percentage. Descriptive analyses were performed using IBM SPSS Statistics version 23.0 for Windows.
Interviews were transcribed verbatim by two student researchers using F4 and Microsoft Word. Both researchers also performed the basis of the analysis in accordance with the steps of reflexive thematic analysis (Braun and Clarke, 2006) to allow insight into overarching themes describing the patterns of experience within and between interviews. Therefore, the interview fragments were first deductively categorised based on the three main topics (i.e. experiences, needs and resources). Then, the categorised fragments were read and re-read while the initial ideas were noted. In the coding phase, initial codes that identified specific experiences, needs and resources, and roles of spirituality were written in the margin of the documents and reflected upon in memos. Subsequently, the two student researchers examined how experiences, needs and resources interacted, to understand the initial themes. These were reviewed by a senior researcher (AV), who listened to the interview recordings and also formulated main themes based on this. In the defining and naming phase the initial themes were merged into one main theme for the ICU-survivors, one main theme for the family members and several subthemes for both groups. This outcome was verified by an independent researcher (IvdM) who listened to two interview recordings as well.
Ethical approval
The COFICS is a sub-study of a larger cohort study called the Follow-up Intensive Care Studies (FICS). The local medical ethical committee approved the protocol for the FICS (METc 2018/627) and waived the need for formal evaluation in accordance with the Dutch Medical Research Involving Human Subjects Act (Sorgdrager, 1998). All participants provided written consent by returning the completed follow-up questionnaire. Additional consent for the interviews was obtained via written informed consent and a telephone call. All participants were allowed to withdraw from the study at any time without providing a reason. This study was performed in accordance with the principles of the Declaration of Helsinki (WMA, 2013).
Results
Study population: patients
At 12-months follow-up, 56 (77 %) out of 73 eligible COVID-19 ICU-survivors responded (Fig. 2 ) of whom 38 (68 %) were male with a median age of 62 (IQR: 55.0–68.0). The median length of ICU stay was 18.9 days (IQR: 12.3–32.7). A total of 27 (48 %) ICU-survivors experienced delirium during their ICU stay. After hospital discharge, 23 ICU-survivors (41 %) went home, and an equal number were discharged to a nursing home or rehabilitation centre. Ten ICU-survivors (18 %) were relocated to another hospital. Before their ICU-admission, 36 survivors (64 %) were employed. The characteristics of the ICU-survivors are shown in Table 1A .Fig. 2 Participant recruitment in the COFICS.
Table 1A Characteristics of enrolled COVID-19 ICU-survivors.
Variable ICU-survivors (n=56)
Age, years, median (IQR) 62.0 (55.0-68.0)
Sex, n (%)
Male 38 (68)
Female 18 (32)
Marital status, n (%)
Married/living together 49 (88)
Single 7 (12)
Educational level1,2, n (%)
Low 18 (32)
Middle 21 (38)
High 16 (29)
Employment status2, n (%)
Employed 36 (64)
Unemployed 20 (36)
BMI, kg/m2, at ICU-admission, median (IQR) 29.6 (27.0 -32.4)
BMI, at ICU-admission, n (%)
Normal (18.5-25) 3 (5)
Overweight (25-30) 27 (48)
Obese (30-35) 20 (36)
Extremely obese (>35) 6 (11)
APACHE IV2, total score, median (IQR) 55.0 (45.0-66.0)
Comorbidities, n (%)
Hypertension 18 (32)
Diabetes mellitus 13 (23)
Cardiovascular disease 11 (20)
Cerebrovascular disease 1 (2)
COPD / asthma 6 (11)
Chronic kidney disease 3 (5)
Malignancy 3 (5)
ECMO, n (%)
Yes 3 (5)
No 53 (95)
Length of ECMO, days, mean (SD) 21.7 (2.5))
Mechanical ventilation, days, median (IQR) 16.3 (10.6-27.5)
Delirium during ICU-stay, n (%)
Yes 27 (48)
No 29 (52)
Length of ICU stay, days, median (IQR) 18.9 (12.3-32.7)
Length of hospital stay, days, median (IQR) 29.9 (21.9-44.7)
Discharge location2, n (%)
Home 23 (41)
Other hospital 10 (18)
Nursing home 4 (7)
Rehabilitation center 19 (34)
Abbreviations:
IQR = interquartile range; SD = standard deviation; BMI = Body Mass Index; APACHE IV = Acute Physiology And Chronic Health Evaluation; ECMO = ExtraCorporeal Membrane Oxygenation; ICU = Intensive Care Unit.
1According to the ISCED2011 classification 2Incomplete data; educational level n = 1 missing; APACHE IV n = 5 missing.
Furthermore, 40 ICU-survivors were invited to participate in an interview, of whom ten were initially willing to participate. Since we expected this number to be sufficient, no more invitations were sent. However, two of them withdrew, and two interviews were cancelled for logistical reasons. Of the six ICU-survivors interviewed, four were male with a mean age of 65 years (range: 61–69). The characteristics of the interviewed ICU-survivors are shown in Table 2.
Study population: family members
A total of 102 family members of COVID-19 ICU-survivors were invited to participate, of whom 67 (66 %) returned the 12-month questionnaire (Fig. 2). The participants had a median age of 58 (IQR: 43–66), were mostly females (70 %) and were mainly partners of ICU-survivors (66 %). A total of 37 family members (55 %) were employed before the ICU-admission of their relative. The characteristics of participating family members are shown in Table 1B .Table 1B Characteristics of enrolled family members of COVID-19 ICU-survivors.
Variable Family members (n=67)
Relation to patient, n (%)
Partner 44 (66)
Son/daughter 17 (25)
Parent 4 (6)
Sibling 2 (3)
Age1, years, median (IQR) 58 (43–66)
Sex1, n (%)
Male 19 (28)
Female 47 (70)
Marital status, n (%)
Married/living together 58 (87)
Single 9 (13)
Educational level2, n (%)
Low 15 (22)
Middle 34 (51)
High 18 (27)
Employment status, n (%)
Employed 37 (55)
Unemployed 30 (45)
Abbreviations: IQR = interquartile range.
1Incomplete data (sex n = 3 missing; age n = 1 missing) 2According to the ISCED2011 classification.
Furthermore, 50 family members were invited to participate in an interview, of whom nine agreed to participate. Three interviews were cancelled for logistical reasons. Of the six interviewed family members, five were female with a mean age of 54 years (range: 42–72). Three of the respondents were spouses of ICU-survivors, two were children of ICU-survivors and one was a sibling. Three of the respondents had been infected with COVID-19 as well. The characteristics of the interviewed family members are shown in Table 2 .Table 2 Characteristics of interviewed participants.
Variable Family members (n=67)
Relation to patient, n (%)
Partner 44 (66)
Son/daughter 17 (25)
Parent 4 (6)
Sibling 2 (3)
Age1, years, median (IQR) 57.5 (42.5-65.8)
Sex1, n (%)
Male 19 (28)
Female 47 (70)
Marital status, n (%)
Married/living together 58 (87)
Single 9 (13)
Educational level2, n (%)
Low 15 (22)
Middle 34 (51)
High 18 (27)
Employment status, n (%)
Employed 37 (55)
Unemployed 30 (45)
*Dyads were interviewed seperately by different researchers.
Quantitative results
Physical domain
Overall, 12 months after ICU-discharge, 82 % of the ICU-survivors rated their condition as ‘not frail’ (Table 3 ). However, 93 % of them reported at least one physical symptom, mainly fatigue (63 %), weakened condition (66 %) and pain (54 %). The mean pain score on the SF-20 was found to be 50 (range: 0–100), and both physical functioning and experienced health were found to be moderate, with scores of 66.7 and 60.0, respectively (range: 0–100).Table 3 The physical, social, mental, and spiritual outcome in COVID-19 ICU survivors 12-months post-ICU-discharge.
Domain Variable 12 months (n=56)
Physical Frailty, n (%)
Not frail (1-3) 46 (82)
Mildly frail (4-5) 9 (16)
Frail (6-8) 1 (2)
Self-reported symptoms5, n (%)
Any one of the following symptoms 52 (93)
Fatigue 35 (63)
Weakened condition 37 (66)
Pain in one or more parts of the body 30 (54)
Joint pain 15 (27)
Chest pain 8 (14)
Muscle pain (myalgia) 15 (27)
Nerve pain 10 (18)
Wound pain 1 (2)
Abdominal pain 5 (9)
Cognitive problems 27 (48)
Dyspnoea 21 (38)
Polyneuropathy 20 (36)
Impaired hand functioning 20 (36)
Muscle weakness/stiffness 19 (34)
Tingling or numb sensation in arms or legs 19 (34)
Reduced lung function 18 (32)
Voice problems 14 (25)
Difficulty walking 13 (23)
Difficulty sleeping 12 (21)
Dizziness 10 (18)
Hair loss 8 (14)
Sexual problems 6 (11)
Skin problems 5 (9)
Loss of smell 3 (5)
Swallowing difficulties 3 (5)
Loss of hearing 3 (5)
Loss of taste 3 (5)
Physical functioning1 (range 0-100) 66.7 (50.0-83.3)
Experienced health1 (range 0-100) 60.0 (35.0-75.0)
Pain2 (range 0-100) 50.0 (0.0-75.0)
Social Role activities1 (range 0-100) 100 (0.0-100.0)
Social functioning1 (range 0-100) 80.0 (60.0-100.0)
Family functioning1 (range 1-4) 4.0 (3.2-4.0)
Return to work, n (%)3
No change 7 (24)
Reduced work percentage 5 (17)
Occupation change (often in combination with
reduced work percentage) 12 (41)
Re-integration 5 (17)
Too ill to work 3 (10)
Early retirement 2 (6)
Other 2 (6)
Work percentage (100%=full-time), mean (SD) 38.9 (36.5)
Mental Anxiety (range 0-21) 2.0 (.0.0-5.0)
Depression (range 0-21) 2.0 (0.0-3.3)
Spiritual ‘Yes’ to 1 or more spiritual need(s), n (%) 47 (84)
Religious needs, n (%) - strength, mean (SD) 12 (21) - 0.12 (0.28)
Existential needs, n (%) - strength, mean (SD) 25 (45) - 0.38 (0.54)
Inner peace needs, n (%) - strength, mean (SD) 37 (66) - 0.74 (0.79)
Needs for positive confirmation, n (%) - strength, mean (SD) 45 (80) - 0.80 (0.67)
Need to feel connected with family, n (%) - strength, mean (SD) 38 (68) - 1.44 (1.13)
Spirituality played a role in coping with Covid-19
Yes 9 (16)
No 46 (82)
No answer 1 (2)
Other Health care consumption, n (%)
Yes 48 (86)
Physiotherapist 39 (70)
General practitioner 31 (55)
Pulmonologist 22 (39)
Psychologist 7 (13)
Home care 5 (9)
Chaplain 1 (2)
Pastor/Imam 1 (2)
Other4 17 (30)
All numbers given are the median and interquartile range (IQR), unless otherwise stated.
Abbreviation: ICU = Intensive Care Unit.
1A higher score reflects a better functioning 2A higher score represents pain in a greater extent 3ICU survivors employed before COVID-19 infection n = 29 4Other health care consumption were mentioned for example; social work, cardiologist, (vascular) surgeon, neurologist or otorhinolaryngologists.
Social domain
ICU-survivors rated role activities (MOS-12) as maximum, with a score of 100 (range: 0–100), similar to family functioning (FAD6 + ), with a score of 4 (range: 1–4). Social functioning (MOS-12) was rated with a score of 80 (range: 0–100). Seven ICU-survivors (24 %) reported no occupational changes, whereas 22 (61 %) reported being in a process of re-integration to work or change to their occupation, sometimes in combination with a reduced work percentage. Three ICU-survivors (10 %) indicated that they were still unable to work as a result of their COVID-19 infection.
Family members rated family functioning with a score of 3.8 (range: 1–4), and 29 family members (78 %) reported no occupational change compared to their previous situation. Moreover, four family members (11 %) reported a reduced work percentage, and two (6 %) indicated being in a process of re-integration to work.
Spiritual domain
Of the ICU-survivors, 84 % reported at least one spiritual need on the SpNQ (Table 3). The need for positive confirmation and attention was the most often reported (80 %), with a mean score of 0.80 (SD: 0.67). Two-thirds of the ICU-survivors (68 %) reported the need to feel connected with family and needs for inner peace (66 %), with mean scores of 1.44 (SD: 1.33) and 0.74 (SD: 0.79), respectively. In addition, 46 ICU-survivors (82 %) answered with ‘No’ to the question of whether spirituality played a role in coping with COVID-19.
Among the family members, 85 % reported at least one spiritual need on the SpNQ (Table 4 ). Most of the family members reported needs for inner peace (80 %) and needs for positive confirmation and attention (76 %), with mean scores of 1.03 (SD: 0.82) and 0.87 (SD: 0.73), respectively. Furthermore, over half of the family members (60 %) reported the need to feel connected with family, with a mean score of 1.25 (SD: 1.14). Moreover, 53 family members (79 %) answered with ‘No’ to the question of whether spirituality played a role during the disease process of their relative.Table 4 The physical, social, mental, and spiritual outcome in family members of COVID-19 ICU-survivors 12-months post-ICU-discharge.
Domain Variable 12 months (n=67)
Social Family functioning1 (range 1-4) 3.8 (3.0-4.0)
Return to work2, n (%)
No change 29 (85)
Reduced work percentage 4 (12)
Occupation change 0
Re-integration 2 (6)
Not returned to work 0
Job loss 1 (3)
Unknown 1 (3)
Work percentage (100% = full-time), mean (sd) 74.8 (30.7)
Mental Anxiety (range 0-21) 4.0 (1.0-6.0)
Depression (range 0-21) 2.0 (0.0-5.0)
Spiritual ‘Yes’ to 1 or more spiritual need(s), n (%) 57 (85)
Religious needs, n (%) - strength, mean (SD) 12 (18) - 0.18 (0.53)
Existential needs, n (%) - strength, mean (SD) 31 (46) - 0.36 (0.45)
Inner peace needs, n (%) - strength, mean (SD) 54 (80) - 1.03 (0.82)
Needs for positive confirmation, n (%) - strength, mean (SD) 51 (76) - 0.87 (0.73)
Need to feel connected with family, n (%) - strength, mean (SD) 40 (60) - 1.25 (1.14)
Spirituality played a role in coping with Covid-19, n (%)
Yes 14 (21)
No 53 (79)
No answer
Other Health care consumption, n (%)
Yes 28 (42)
General practitioner 15 (22)
Psychologist 6 (9)
Social work 5 (8)
Physiotherapist 2 (3)
Chaplain 1 (2)
Pastor / Imam 0
Other3 5 (5)
All numbers given are the median and interquartile range (IQR), unless otherwise stated.
1A higher score reflects a better functioning 2Employed family members at 12 months n = 34 3 Other health care consumption were mentioned for example; home care, osteopath or acupuncturist.
Healthcare consumption
In the six months prior to completing the questionnaire, the majority of ICU-survivors visited one or more healthcare professionals, mainly physiotherapists (70 %) and general practitioners (55 %).
Of the family members, 42 % had visited one or more healthcare professionals, usually their general practitioner (22 %).
Qualitative findings
Themes and subthemes found in the interviews with ICU-survivors and their family members are shown in Table 5 . Several subthemes characterized the interviews, with one overarching theme in the interviews with ICU-survivors and one overarching theme in the interviews with family members.Table 5 Qualitative findings of ICU-survivors and family members 12-months post-ICU-discharge.
ICU survivors Family members
MAIN THEME
Returning to normal MAIN THEME
If the patient is well, I am well
SUBTHEMES
Crisis / confrontation with mortality Fear of no goodbye
Memories Hope
Reconstruct experience Own live revolved around relative
Physical recovery first Reassurance
Returning to previous life (work) Sharing information
Family and friends Family and friends
Gratitude Peace and solitude
Religious faith and being prayed for Outdoors
Difficulties finding sufficient support Diary
Independent person Hospital staff
Positive attitude
ICU-survivors
The overarching theme covering the interviews with the ICU-survivors was ‘returning to normal’. ICU-survivors reported about the changes they experienced, which needs surfaced, and what (spiritual) resources they used to try to return to normal or to cope with the realization that return to normal was not possible. The interaction between the different experiences, needs and resources is described below, based on the subthemes we found (Table 5), which are enclosed in quotation marks.
One of the subthemes was ‘crisis’, because overall, ICU-survivors described their experience as such. Sudden, unexpected and disrupting their life goals and worldview, although not all experiences were negative. For example, one ICU-survivor described a near-death experience, which allowed her to find closure for the death of her husband several years earlier.
Another subtheme was ‘memories’, because amnesia or distorted memories due to delirium, often caused unsettling and insecure feelings. To help them process these feelings in their current life, a few ICU-survivors expressed the desire to go back to the hospital to try and ‘reconstruct their experience’.
Residual physical and cognitive limitations were perceived as distressing, challenging the sense of the self and hindering ‘returning to normal’. After returning home, ICU-survivors mostly focused on ‘physical recovery’, as they felt the desire to ‘return to their previous life’. They considered returning to work as a sign that they had achieved ‘returning to normal’:‘Yes, I’m trying to go back to work … I started working again as soon as I could. I really enjoy it. Each time I go to work, I make a step towards getting my old life back.’ (ICU-survivor 37).
Nevertheless, they expressed how their goals in life shifted and how things other than work had become more important, such as being with ‘family and friends’. Several ICU-survivors indicated that they had not yet fully processed how the experience affected their present and future lives:‘After about 6 months, I had a relapse, because then came the psychological part. What has happened to me? When my wife told me she had received several telephone calls from the hospital saying that I might not survive, I really began to realise that I had been out of it, literally and figuratively speaking. It took a lot of effort for me to process this, and I’ve had a lot of support…. At this point I’m still struggling with my future perspective, because I can no longer cognitively do what I used to. So, what am I going to do when I retire?’ (ICU survivor 8).
Another subtheme was ‘gratitude’, because it was addressed in multiple ways. Several ICU-survivors felt the strong need to express their gratitude to the hospital staff. Furthermore, returning home and recovering felt like a second chance for which ICU-survivors felt grateful.
Resources tapped to cope with the experience included instrumental and social support of ‘family and friends’ and ‘religious faith’. Considering the last subtheme, knowing that you are being prayed for, appeared to be an important source of comfort. Some ICU-survivors reported ‘difficulties finding sufficient support’ after hospital discharge:‘I remember a social worker who came to my hospital bed, only three days after I woke up. Everything was strange and confused and I couldn’t engage with that. I really wanted to go home, so they discharged me rather quickly. But in those early days of the pandemic it was difficult to find aftercare. Finding psychological care was most difficult. When things don’t go well, you need someone outside your family and colleagues to discuss your concerns.’ (ICU-survivor 6).
Some ICU-survivors indicated that it could be helpful to have an ‘independent person’ to discuss the experience with, in addition to conversations with ‘family and friends’. Only one ICU survivor actively sought spiritual support from healthcare chaplains during and after hospitalisation to better understand and process the experience.
Maintaining a ‘positive attitude’ and focussing on good things in life were other coping resources in the process of ‘returning to normal’.
Family members
The main theme in the interviews with family members was ‘if the patient is well, I am well’. Family members indicated in the interviews how they experienced the patients’ critical illness, which needs surfaced and what (spiritual) resources were helpful. The interaction between these is described below, based on the subthemes we found (Table 5), which are enclosed in quotation marks.
One of the subthemes was ‘fear of no goodbye’, because family members highlighted that the worst part of the ICU-admission of their relative was the fear that he or she may die without being able to say goodbye:‘What if it had gone wrong? We wouldn’t have been able to say goodbye … That would have been terrible.’ (Family member 26).
The subtheme ‘hope’ was connected to this, because during this period, family members struggled to maintain hope for recovery, as they regularly received information from ICU staff that the situation was critical and that there was a real possibility that their relative would not survive.
Another subtheme was ‘own live revolved around relative’. Once the ICU-survivor started to recover, their family members felt less anxious. When the ICU-survivors recovered well, their family members felt closure and returned to their own lives. However, the lives of the family members of ICU-survivors with many residual symptoms still strongly revolved around their relatives’ well-being.
During the hospitalisation period of the ICU-survivors, family members had a strong need for ‘reassurance’ regarding the situation and maintained regular contact with the hospital staff. The received ‘information was shared’ with ‘family and friends’ through email or WhatsApp, as family and friends were the main sources of support:‘After I’d spoken with the ICU, I would send the children an app about what was discussed. I had contact with the ICU four times a day. That was excellent, really supportive.’ (Family member 38).
At the same time many of the family members felt overwhelmed during this time and expressed the need for ‘peace and solitude’. Most of the family members tended to go hiking or cycling ‘outdoors’ every day during this period to clear their mind and enjoy the beauty of nature:‘We go hiking a lot. To stay in shape, but also for what is more important, what I call ‘defocussing’. Hiking gives you a feeling of space around you and that provides a lot of peace.’ (Family member 8).
Most family members kept a ‘diary’, in the form of written text, photos or videos, to share with their relatives once they had recovered enough. However, at the time of the interviews, most of the diaries had been looked at only once.
All family members highlighted that they felt strongly reassured by and grateful to the ‘hospital staff’, who have actively kept them informed, asked about their well-being, treated the patient as a person and were always available for them:‘They saw my father as a person, not as a number … They asked for pictures and information about the family relationship and what father thought was important in life. So even when he was sedated, they could tell him “your daughter called”’. (Family member 23).
Some family members prayed and lit candles for their relatives to find strength and peace. However, most of the coping resources they used, cannot be classified as spiritual.
Discussion
With the COFICS, we aimed to report on the long-term outcomes of COVID-19 ICU survivors and their family members in the physical, social, mental and spiritual domain. The results obtained in this study show that the majority of COVID-19 ICU-survivors experience at least one physical impairment one year after ICU-discharge. In addition, most of the previously employed ICU-survivors have not fully returned to work, whereas most of their family members have. Furthermore, the majority of ICU-survivors as well as their family members reported at least one spiritual need. Moreover, several healthcare professionals were visited by ICU-survivors and their family members.
Although we found that the physical functioning of ICU-survivors increased over time, most of them reported at least one physical impairment 12 months after ICU-discharge, comparable to the results that we found after three and six months (van Veenendaal et al., 2021). Symptoms such as fatigue, weakened condition and pain were the most prominent. A study that reported on the 12-month outcomes of patients after a COVID-19 related hospital admission (Betschart et al., 2021) showed, among other findings, a decrease in health-related QoL after 12 months. They also found slight-to-moderate physical limitations in 29 % of the participants, scored using the novel Post-COVID-19 Functional Status (PCFS) scale (Klok et al., 2020, Machado et al., 2021). In another study (Tessitore et al., 2021), participants reported fatigue (27 %) to be most prominent 12 months after hospital discharge, similar to the participants in our study. ICU-survivors in our study also reported pain to be an important impairment, with the most prominent types of pain being joint pain, muscle pain and nerve pain. Overall, the novel PCFS scale (Klok et al., 2020, Machado et al., 2021) includes pain, depression and anxiety and translates these to a grade of functioning, which may be useful in assessing the impact of pain, depression and anxiety on the daily lives of COVID-19 ICU-survivors. Future research should consider using this instrument to help compare the results.
In our study, both ICU-survivors and their family members described their family functioning as good and reported no or hardly any symptoms of depression or anxiety. These results are different from those obtained with other COVID-19 cohorts (Betschart et al., 2021, Heesakkers et al., 2022a, Heesakkers et al., 2022b, Tessitore et al., 2021). Although our previous findings (van Veenendaal et al., 2021) showed that 63 % of family members were psychologically affected by the physical distancing required during the first COVID-19 wave, it is likely that the exceptional circumstances during the first wave influenced the commitment of healthcare professionals and affected the experience of family members (Liang et al., 2021).
Overall, two-thirds of the ICU-survivors in our study reported a change of occupation, often in combination with a reduced work percentage. In a systematic review evaluating return to work in previously employed ICU-survivors (Kamdar et al., 2020), delayed return to work and unemployment were found to be common and persistent issues. Several studies in this review found that unemployment is associated with depression, anxiety and poor QoL. Improved mental health and QoL were found after returning to work. We found similar results in the interviews conducted with ICU-survivors. They particularly experienced returning to work as a sign of recovery. Such coherence between work and living a meaningful life has been described in several studies (Duffy and Sedlacek, 2010, Judge and Watanabe, 1993, Ward and King, 2017). However, ICU-survivors in our study also indicated that work is no longer their main priority in life and that their relationships with friends and family have become more important. Although research focussing on returning to work in family members of ICU-survivors is rare, this topic has been mentioned in some studies on PICS-F (van Beusekom et al., 2016). In the interviews we conducted, several family members mentioned that their return to work or other activities was conditional upon the health and well-being of the patient. Notably, most of the family members had returned to such activities at the time of the interview. Barriers obstructing returning to work were studied in patients after elective coronary bypass surgery (Blokzijl et al., 2021) and identified as personal related, healthcare related, work related and law and regulation related. If patients undergoing elective surgery experience barriers in this domain, it is likely that patients with acute conditions and their family members meet even more difficulties. Hence, healthcare professionals throughout the care chain should add the notion of returning to work to their follow-up care. Further research is also required to gain more insights into the origins of problems and the necessities to support and facilitate returning to work.
In our study, only a few ICU-survivors and their family members indicated that spirituality has played a role in coping with COVID-19. However, most of them reported at least one spiritual need on the SpNQ. The participants may have associated the term ‘spirituality’ with religious faith and religious activities, such as prayer and lighting candles, but not with experiences of inner peace, connection to others and receiving confirmation from others. This implies that using the word ‘spirituality’ with patients and family members may not be sufficient to meet the needs they might have in this dimension. This finding has also been highlighted in other studies (Garssen et al., 2015, Uwland-Sikkema et al., 2018). Overall, this knowledge can influence daily patient care and how existing spiritual needs are identified (NANDA, 2021). In the Dutch context, the term ‘spirituality’ may not be unambiguous enough, and it may be useful to refer to more specific items, such as feelings, thoughts, behaviours and forms of support, instead of using the term ‘spirituality’ in general. Future research on this topic should also use this knowledge.
Although the ICU-survivors and their family members visited healthcare professionals in the six months prior to completing the questionnaire, they also reported that it was difficult to find sufficient support. In addition, although most of the needs seem to have been met in the year following ICU-discharge, many of the interviewed patients indicated that they had not fully processed the experience. They highlighted that discussing their experience with an independent person would be of additional value. Whereas in another study reading the diary with experiences of the ICU admission was perceived as positive and reread by the ICU-survivor (Jones et al., 2010) our findings show the diary was read only once. Perhaps this is because the diaries were written by their family members instead of the healthcare staff and were therefor seen as less important. Family members described how different the ICU experience and the process of recovery have been for them compared to their relative. For the patient, the coping process started once they began to recover, whereas the coping process of the family members ended at that point. Furthermore, since the ICU-survivors had been sedated for a long time, while their family members could remember the entire period, the coping processes of the ICU-survivors and their family members did not harmonise. Similar findings are described in other studies (Fumis et al., 2015). Currently, both ICU-survivors and their family members seem to have found ways to receive support, though not always satisfactory.
The results obtained herein show the importance of follow-up care focussing on more than just the physical domain. Facilitating an open conversation between families about the experience (Broekema et al., 2021, Naef et al., 2020), providing guidance and support in the process of returning to work or a similar meaningful occupation (Duffy and Sedlacek, 2010, Judge and Watanabe, 1993, Ward and King, 2017) and identifying other (spiritual) needs may be just as important.
Strengths and limitations
One of the strengths of this study is that we used a mixed-methods design to assess spiritual outcomes, which allowed us to obtain more in-depth results on this rarely researched topic for ICU-survivors and their family members. Another strength of this study is that we used an inter-disciplinary approach, which helped provide a more holistic view. Moreover, nurses, doctors, healthcare chaplains and students worked together on this project. In addition, the SpNQ was translated thoroughly and precisely, which facilitates its use in other Dutch studies. Finally, this study involved long-term follow-up with a satisfactory response rate.
One of the limitations of this study is its single-centre design with a relatively small sample size. Another limitation is the fact that the interviews were conducted remotely. Although remote interviews using digital tools have become increasingly common, this method may cause feelings of insecurity for the interviewed participant. Finally, it is important to highlight that our study focuses on ICU-survivors and their family members during the first COVID-19 wave. Hence, it is likely that no suitable standardised care was available.
Conclusion
The results obtained in this study show that COVID-19 ICU-survivors and their family members evaluate their health status positively-one year after being discharged from the ICU. In addition, ICU survivors experience minor physical impairments and express the need for follow-up care in aspects such as the social and spiritual domain. Finally, the well-being of the family members of ICU-survivors is strongly influenced by the health status of the patient, and most of them experience feelings of closure after one year.
Ethical statement
The COFICS is a sub-study of a larger cohort study called the Follow-up Intensive Care Studies (FICS). The local medical ethical committee approved the protocol for the FICS (METc 2018/627) and waived the need for formal evaluation in accordance with the Dutch Medical Research Involving Human Subjects Act. All participants provided written consent by returning the completed follow-up questionnaire. Additional consent for the interviews was obtained via written informed consent and a telephone call. All participants were allowed to withdraw from the study at any time without providing a reason. This study was performed in accordance with the principles of the Declaration of Helsinki.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary data
The following are the Supplementary data to this article:Supplementary data 1
Acknowledgements
We gratefully acknowledge all the ICU-survivors and family members for participation in the COFICS. Additionally, we thank all (research) nurses who were involved with the study: Desirée Meertens-Demandt, Elza van den Berg, Hester Tamminga, Hetty Kranen, Janneke van der Veen, Marthe Flanderijn and Roos Mensink. Furthermore, we are grateful to Annette de Vries and Judith van ‘t Hof for their contribution to the qualitative data collection and -analysis of this study.
Role of the Funding Source
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Preregistration
The trial registration can be found on: ClinicalTrials.gov, NCT04460170.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.iccn.2022.103366.
==== Refs
References
Berger P. Braude D. Post–intensive care syndrome: A crash course for general practice Austr. J. General Pract. 50 9 2021 647 649 10.31128/AJGP-07-20-55491
Betschart M. Rezek S. Unger I. Ott N. Beyer S. Böni A. Gisi D. Shannon H. Spruit M.A. Sieber C. One year follow-up of physical performance and quality of life in patients surviving COVID-19: a prospective cohort study Swiss Med. Wkly. 151 2021 10.4414/SMW.2021.W30072
Bjelland I. Dahl A.A. Haug T.T. Neckelmann D. The validity of the Hospital Anxiety and Depression Scale: An updated literature review J. Psychosom. Res. 52 2 2002 69 77 10.1016/S0022-3999(01)00296-3 11832252
Blokzijl F. Onrust M. Dieperink W. Keus F. van der Horst I.C.C. Paans W. Mariani M.A. Reneman M.F. Barriers that obstruct return to work after coronary bypass surgery: a qualitative study J. Occup. Rehabil. 31 2 2021 316 322 10.1007/s10926-020-09919-6 32803466
Boterhoven de Haan K.L. Hafekost J. Lawrence D. Sawyer M.G. Zubrick S.R. Reliability and validity of a short version of the general functioning subscale of the McMaster family assessment device Fam. Process 54 1 2015 116 123 10.1111/FAMP.12113 25385473
Braun V. Clarke V. Using thematic analysis in psychology Qual. Res. Psychol. 3 2 2006 77 101 10.1191/1478088706QP063OA
Broekema S. Paans W. Oosterhoff A.T. Roodbol P.F. Luttik M.L.A. Patients’ and family members’ perspectives on the benefits and working mechanisms of family nursing conversations in Dutch home healthcare Health Soc. Care Community 29 1 2021 259 269 10.1111/HSC.13089 33034928
Büssing A. Balzat H.J. Heusser P. Spiritual needs of patients with chronic pain diseases and cancer – Validation of the spiritual needs questionnaire Eur. J. Med. Res. 15 6 2010 266 273 10.1186/2047-783x-15-6-266 20696636
Büssing A. Recchia D.R. Koenig H. Baumann K. Frick E. Factor structure of the spiritual needs questionnaire (SpNQ) in persons with chronic diseases, elderly and healthy individuals Religions 9 1 2018 10.3390/rel9010013
Creswell, J. W., & Plano Clark, V. L. (2017). Designing and Conducting Mixed Methods Research (3rd revise). Sage Publications Inc. https://doi.org/10.1111/j.1753-6405.2007.00096.x.
Davidson J.E. Jones C. Bienvenu O.J. Family response to critical illness: Postintensive care syndrome–family Crit. Care Med. 40 2 2012 618 624 22080636
Dieperink P. Dijkstra B. Marten-van Stijn G. Postma-Rowden J. van der Hoeven J. van den Boogaard M. Validated Dutch Translation of the Clinical Frailty Scale for ICU Patients and its use in Practice Practice. SL J Anaesth Crit Care 1 1 2017 111
Duffy R.D. Sedlacek W.E. The salience of a career calling among college students: Exploring group differences and links to religiousness, life meaning, and life satisfaction Career Dev. Q. 59 1 2010 27 41 10.1002/j.2161-0045.2010.tb00128.x
Fumis R.R.L. Ranzani O.T. Martins P.S. Schettino G. Schmahl C. Emotional disorders in pairs of patients and their family members during and after ICU stay PLoS One 10 1 2015 e0115332 25616059
Garssen B. Uwland-Sikkema N.F. Visser A. How Spirituality Helps Cancer Patients with the Adjustment to their Disease J. Relig. Health 54 4 2015 1249 1265 10.1007/S10943-014-9864-9 24748130
Geense W.W. Zegers M. Peters M.A.A. Ewalds E. Simons K.S. Vermeulen H. van der Hoeven J.G. van den Boogaard M. New Physical, Mental, and Cognitive Problems 1 Year after ICU-admission : A Prospective Multicenter Study Am. J. Respir. Crit. Care Med. 203 12 2021 1512 1521 10.1164/RCCM.202009-3381OC 33526001
Hamilton E. Carr A. Systematic Review of Self-Report Family Assessment Measures Fam. Process 55 1 2016 16 30 10.1111/FAMP.12200 26582601
Harvey M.A. Davidson J.E. Postintensive Care Syndrome: Right Care, Right Now. and Later Crit. Care Med. 44 2 2016 381 385 10.1097/CCM.0000000000001531 26771784
Heesakkers H. van der Hoeven J.G. Corsten S. Janssen I. Ewalds E. Burgers-Bonthuis D. Rettig T.C.D. Jacobs C. van Santen S. Slooter A.J.C. van der Woude M.C.E. Zegers M. van den Boogaard M. Mental health symptoms in family members of COVID-19 ICU survivors 3 and 12 months after ICU-admission : a multicentre prospective cohort study Intensive Care Med. 48 3 2022 322 331 10.1007/s00134-021-06615-8 35103824
Heesakkers H. van der Hoeven J.G. Corsten S. Janssen I. Ewalds E. Simons K.S. Westerhof B. Rettig T.C.D. Jacobs C. van Santen S. Slooter A.J.C. van der Woude M.C.E. van den Boogaard M. Zegers M. Clinical Outcomes among Patients with 1-Year Survival Following Intensive Care Unit Treatment for COVID-19 JAMA – J. Am. Med. Assoc. 327 6 2022 559 565 10.1001/jama.2022.0040
Ho J.Q. Nguyen C.D. Lopes R. Ezeji-Okoye S.C. Kuschner W.G. Spiritual care in the intensive care unit: a narrative review J. Intensive Care Med. 33 5 2018 279 287 10.1177/0885066617712677 28604159
Huang C. Soleimani J. Herasevich S. Pinevich Y. Pennington K.M. Dong Y. Pickering B.W. Barwise A.K. Clinical characteristics, treatment, and outcomes of critically ill patients With COVID-19: a scoping review Mayo Clin. Proc. 96 1 2021 183 202 33413817
Jolley S.E. Bunnell A.E. Hough C.L. ICU-acquired weakness Chest 150 5 2016 1129 1140 10.1016/J.CHEST.2016.03.045 27063347
Jones C. Backman C. Capuzzo M. Egerod I. Flaatten H. Granja C. Rylander C. Griffiths R.D. RACHEL group T. Intensive care diaries reduce new onset post traumatic stress disorder following critical illness: a randomised, controlled trial Retrieved from 14 5 2010 R168
Judge T.A. Watanabe S. Another look at the job satisfaction-life satisfaction relationship J. Appl. Psychol. 78 6 1993 939 948 10.1037/0021-9010.78.6.939
Kamdar B.B. Suri R. Suchyta M.R. Digrande K.F. Sherwood K.D. Colantuoni E. Dinglas V.D. Needham D.M. Hopkins R.O. Return to work after critical illness: a systematic review and meta-analysis Thorax 75 1 2020 17 27 10.1136/THORAXJNL-2019-213803 31704795
Kang J. Jeong Y.J. Embracing the new vulnerable self: A grounded theory approach on critical care survivors’ post-intensive care syndrome Intensive Crit. Care Nurs. 49 2018 44 50 10.1016/j.iccn.2018.08.004 30193868
Kempen G.I. The MOS Short-Form General Health Survey: single item vs multiple measures of health-related quality of life: some nuances Psychol. Rep. 70 2 1992 608 610 10.2466/pr0.1992.70.2.608 1598378
Klok F.A. Boon G.J.A.M. Barco S. Endres M. Geelhoed J.J.M. Knauss S. Rezek S.A. Spruit M.A. Vehreschild J. Siegerink B. The Post-COVID-19 Functional Status scale: a tool to measure functional status over time after COVID-19 Eur. Respir. J. 56 1 2020 2001494 32398306
Liang Y. Li J. Pan W. Family satisfaction in the intensive care unit: The influence of disease severity, care relationship, patient anxiety and patient pain Intensive Crit. Care Nurs. 63 2021 102995 33349481
Machado F.V.C. Meys R. Delbressine J.M. Vaes A.W. Goërtz Y.M.J. van Herck M. Houben-Wilke S. Boon G.J.A.M. Barco S. Burtin C. van ’t Hul A. Posthuma R. Franssen F.M.E. Spies Y. Vijlbrief H. Pitta F. Rezek S.A. Janssen D.J.A. Siegerink B. Klok F.A. Spruit M.A. Construct validity of the Post-COVID-19 Functional Status Scale in adult subjects with COVID-19 Health Qual. Life Outcomes 19 1 2021 10.1186/s12955-021-01691-2
M.E. Mikkelsen M. Still B.J. Anderson O.J. Bienvenu M.B. Brodsky N. Brummel B. Butcher A.S. Clay H. Felt L.E. Ferrante K.J. Haines M.O. Harhay A.A. Hope R.O. Hopkins M. Hosey C.“.L. Hough J.C. Jackson A. Johnson B. Khan N.I. Lone P. MacTavish J. McPeake A. Montgomery-Yates D.M. Needham G. Netzer C. Schorr B. Skidmore J.L. Stollings R. Umberger A. Andrews T.J. Iwashyna C.M. Sevin Society of Critical Care Medicine’s International Consensus Conference on Prediction and Identification of Long-Term Impairments After Critical Illness 48 11 2020 1670 1679.
Naef R. Massarotto P. Petry H. Family and health professional experience with a nurse-led family support intervention in ICU: A qualitative evaluation study Intensive Crit. Care Nurs. 61 2020 102916 32807604
Nanda I. NANDA International Nursing Diagnoses Definitions & Classification, 2021–2023 2021 Thieme Medical Publishers, Incorporated
Needham D.M. Davidson J. Cohen H. Hopkins R.O. Weinert C. Wunsch H. Zawistowski C. Bemis-Dougherty A. Berney S.C. Bienvenu O.J. Brady S.L. Brodsky M.B. Denehy L. Elliott D. Flatley C. Harabin A.L. Jones C. Louis D. Meltzer W. Muldoon S.R. Palmer J.B. Perme C. Robinson M. Schmidt D.M. Scruth E. Spill G.R. Storey C.P. Render M. Votto J. Harvey M.A. Improving long-term outcomes after discharge from intensive care unit: Report from a stakeholders’ conference Crit. Care Med. 40 2 2012 502 509 21946660
S. Nolan Büssing, A. (Ed.) (2021). Spiritual Needs in Research and Practice: The Spiritual Needs Questionnaire as a Global Resource for Health and Social Care HSCC.
Our World in Data Number of COVID-19 patients in ICU per million 2021 https://ourworldindata.org/grapher/covid-icu-patients-per-million?msclkid=aa6ca826bca211ecad6dea9cc67af265.
Petrinec A.B. Martin B.R. Post-intensive care syndrome symptoms and health-related quality of life in family decision-makers of critically ill patients Palliat. Support. Care 16 6 2018 719 724 10.1017/S1478951517001043 29277171
Puchalski C. Ferrell B. Virani R. Otis-Green S. Baird P. Bull J. Chochinov H. Handzo G. Nelson-Becker H. Prince-Paul M. Pugliese K. Sulmasy D. Improving the quality of spiritual care as a dimension of palliative care: The report of the consensus conference J. Palliat. Med. 12 10 2009 885 904 10.1089/JPM.2009.0142 19807235
Rockwood K. Song X. MacKnight C. Bergman H. Hogan D.B. McDowell I. Mitnitski A. A global clinical measure of fitness and frailty in elderly people CMAJ 173 5 2005 489 495 10.1503/cmaj.050051 16129869
Rockwood K. Theou O. Mitnitski A. What are frailty instruments for? Age Ageing 44 4 2015 545 547 10.1093/ageing/afv043 25824236
Sorgdrager, W. (1998). wetten.nl - Regeling - Wet medisch-wetenschappelijk onderzoek met mensen - BWBR0009408. Overheid. https://wetten.overheid.nl/BWBR0009408/2022-01-31.
Spinhoven P. Ormel J. Sloekers P.P.A. Kempen G.I.J.M. Speckens A.E.M. van Hemert A.M. A validation study of the hospital anxiety and depression scale (HADS) in different groups of Dutch subjects Psychol. Med. 27 2 1997 363 370 10.1017/S0033291796004382 9089829
N.I.C.E. Stichting Data in beeld 2019 https://www.stichting-nice.nl/datainbeeld/public?year=2019&subject=DURATION_OF_TREATMENT&hospital=144&icno=0.
Tessitore E. Handgraaf S. Poncet A. Achard M. Höfer S. Carballo S. Marti C. Follonier C. Girardin F. Mach F. Carballo D. Symptoms and quality of life at 1-year follow up of patients discharged after an acute COVID-19 episode Swiss Med. Wkly. 151 2021 w30093 10.4414/SMW.2021.W30093
Uwland-Sikkema N.F. Visser A. Westerhof G.J. Garssen B. How is spirituality part of people’s meaning system? Psychol. Relig. Spiritual. 10 2 2018 157 165 10.1037/REL0000172
van Beusekom I. Bakhshi-Raiez F. de Keizer N.F. Dongelmans D.A. van der Schaaf M. Reported burden on informal caregivers of ICU survivors: a literature review Crit. Care 20 1 2016 10.1186/s13054-016-1185-9
van Rompaey B. Schuurmans M.J. Shortridge-Baggett L.M. Truijen S. Elseviers M. Bossaert L. Long term outcome after delirium in the intensive care unit J. Clin. Nurs. 18 23 2009 3349 3357 10.1111/j.1365-2702.2009.02933.x 19735334
van Veenendaal N. van der Meulen I.C. Onrust M. Paans W. Dieperink W. van der Voort P.H.J. Six-month outcomes in covid-19 icu patients and their family members: a prospective cohort study Healthcare (Switzerland) 9 7 2021 10.3390/healthcare9070865
Ward S.J. King L.A. Work and the good life: How work contributes to meaning in life Res. Organ. Behav. 37 2017 59 82 10.1016/J.RIOB.2017.10.001
WHOb. (2020). Weekly Epidemiological Update on COVID-19. In World Health Organization (Issue 3 November, p. 1;4). https://www.who.int/publications/m/item/weekly-epidemiological-update-on-covid-19---29-march-2022.
WMA. (2013). World Medical Association declaration of Helsinki: Ethical principles for medical research involving human subjects. In JAMA - Journal of the American Medical Association (Vol. 310, Issue 20, pp. 2191–2194). JAMA. https://doi.org/10.1001/jama.2013.281053.
Zigmond A.S. Snaith R.P. The Hospital Anxiety and Depression Scale Acta Psychiatr. Scand. 67 6 1983 361 370 10.1111/j.1600-0447.1983.tb09716.x 6880820
| 0 | PMC9726690 | NO-CC CODE | 2022-12-16 23:16:07 | no | Intensive Crit Care Nurs. 2022 Dec 7;:103366 | utf-8 | Intensive Crit Care Nurs | 2,022 | 10.1016/j.iccn.2022.103366 | oa_other |
==== Front
Mar Pollut Bull
Mar Pollut Bull
Marine Pollution Bulletin
0025-326X
1879-3363
Elsevier Ltd.
S0025-326X(22)01158-4
10.1016/j.marpolbul.2022.114476
114476
Article
Personal protective equipment (PPE) pollution driven by COVID-19 pandemic in Marina Beach, the longest urban beach in Asia: Abundance, distribution, and analytical characterization
Kannan Gunasekaran a⁎
Mghili Bilal b
De-la-Torre Gabriel Enrique c
Kolandhasamy Prabhu d
Machendiranathan Mayakrishnan e
Rajeswari Mayavan Veeramuthu a
Saravanakumar Ayyappan f
a Centre for Aquaculture, Sathyabama Institute of Science and Technology, Chennai 600 119, Tamil Nadu, India
b LESCB, URL-CNRST N 18, Abdelmalek Essaadi University, Faculty of Sciences, Tetouan, Morocco
c Universidad San Ignacio de Loyola, Av. La Fontana 501, Lima 12, Peru
d Department of Marine Science, Bharathidasan University, Tiruchirappalli 620024, India
e Laboratory for Coastal Ocean Variation and Disaster Prediction, College of Ocean and Meteorology, Guangdong Ocean University, Zhanjiang 524088, Guangdong, China
f Centre of Advanced Study in Marine Biology, Faculty of Marine Sciences, Annamalai University, Parangipettai 608502, Tamil Nadu, India
⁎ Corresponding author.
7 12 2022
1 2023
7 12 2022
186 114476114476
6 11 2022
28 11 2022
4 12 2022
© 2022 Elsevier Ltd. All rights reserved.
2022
Elsevier Ltd
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
COVID-19 pandemic has enforced the use of personal protective equipment (PPE, masks and gloves). However, the mismanagement of litter are exacerbating the increasing plastic issue worldwide. In the present study, we sampled discarded PPE in 10 sites along Marina Beach, India. We characterized the litter types by chemical analysis techniques. A total of 1154 COVID-19-associated PPE items were found on Marina beach. The highest number of items were face masks (97.9 %) and the mean PPE density in the sites studied was 4 × 10−3 PPE m−2. The results demonstrate that poor solid waste management and lack of awareness are the main causes of pollution at Marina beach. FTIR spectroscopy revealed that face masks and gloves were principally made of polypropylene and latex, respectively. The FTIR spectra also showed signs of chemical degradation. Our results suggest that plastic pollution is increasing, possibly becoming more impactful to marine biota. Beach management measures were discussed.
Keywords
Coronavirus
Plastic
Microplastics
Polypropylene
Management
India
==== Body
pmc1 Introduction
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), became a global pandemic in 2020, causing nearly 6 million deaths worldwide (WHO, 2022). During the COVID-19 pandemic, numerous disposable plastic items and disposable PPE were required as a basic precaution to prevent the spread of COVID-19 infection (Ji et al., 2021). PPE items are categorized into many types, including hazard suits, face shields, surgical masks, bouffant caps, gloves high-grade medical masks. This caused the demand for PPE to rise considerably around the world (Prata et al., 2020). It is estimated that about 65 billion gloves and 129 billion face masks are employed worldwide every month (Prata et al., 2020). In the end, adding a large charge to traditional solid waste management systems. There are several studies worldwide on the inappropriate disposal of PPE litter (Ben Haddad et al., 2021; De-la-Torre et al., 2021; Aragaw et al., 2022; Gunasekaran et al., 2022; Mghili et al., 2022; Dioses-Salinas et al., 2022; Ribeiro et al., 2022; Sajorne et al., 2022). Benson et al. (2021) estimated that approximately 3.4 billion face masks were discarded each day. These types of litter will continue to accumulate in the future, potentially exacerbating existing plastic pollution (De-la-Torre et al., 2021). Once discarded, litter can transport from one location to another via wind, streams, and rivers to reach the marine environment (Kutralam-Muniasamy et al., 2022). It was estimated that 1.56 billion facemasks are presently disposed in the world's oceans (OceansAsia, 2020). Scientists have recorded their presence on the beaches of Peru, Brazil, Argentina, Kenya, Morocco, India, Bangladesh, Iran, Ethiopia, and the Philippines (Okuku et al., 2020a; Ardusso et al., 2021; Ben Haddad et al., 2021; De-la-Torre et al., 2022a, De-la-Torre et al., 2022b, De-la-Torre et al., 2022c; Gunasekaran et al., 2022; Rakib et al., 2021; Thiel et al., 2021; Aragaw et al., 2022; Hassan et al., 2022; Sajorne et al., 2022; Mohamadi et al., 2023). Nevertheless, the present data is insufficient to have a global overview of marine pollution by PPE.
Like plastic, PPE could have a significant impact on the environment and marine wildlife. PPE, particularly surgical masks, have been recognized as potential sources of microplastics in the marine environment (Morgana et al., 2021; Saliu et al., 2021). Most PPE consist of synthetic polymers, with high proportions of polypropylene, polyethylene, and polyacrylonitrile and also other polymeric materials such as polyester, polyurethane, nylon, and polystyrene (Ammendolia et al., 2021; Fadare and Okoffo, 2020; Aragaw, 2020). PPE may degrade when exposed to natural factors such as sea waves and sunlight (including UV radiation) (De-la-Torre and Aragaw, 2021; Saliu et al., 2021). The resulting pieces, smaller than 5 mm, are considered microplastics (De-la-Torre et al., 2022b). These microplastics are bioavailable to a large number of marine organisms and can produce toxic effects, the impacts of which can propagate through the food chain (De-la-Torre et al., 2022b). PPE are also able to liberate toxic chemical additives (Hajiouni et al., 2022) and serve as a vector for contaminants (Torres et al., 2021). The ingestion of face masks by marine organisms has also been documented in Brazil and Japan (Neto et al., 2021; Fukuoka et al., 2022). Recently, Hiemstra et al. (2021) and Ammendolia et al. (2022) provided an overview of the impacts of PPE interaction with various types of aquatic and terrestrial animals through entrapment and entanglement. In addition, PPE can perturb the ecosystem stability of marine habitats through the spread of invasive species (De-la-Torre and Aragaw, 2021).
Waste management is an urgent concern in India, especially with the growing population throughout the country (Banerjee et al., 2019). In India, the current COVID-19 pandemic has caused an increase in the demand for single-use plastics, adding pressure to an already out-of-control problem (Singh et al., 2022). It is estimated that the average amount of biomedical waste generated by COVID-19 in December 2021 is nearly 72.8 tons per day (CPCB, 2021). Recent reports have been published indicating that litter related to COVID-19 is not correctly discarded on Indian beaches (Gunasekaran et al., 2022). Ineffective litter management as well as the litter disposal behavior of the population are among the causes of marine litter pollution on beaches. Also, Asian rivers have discharged a huge amount of PPE into the sea and oceans (Peng et al., 2021). Only a few studies have investigated the occurrence and distribution of PPE along the beaches of Tamil Nadu (Gunasekaran et al., 2022). Data on PPE pollution on Indian beaches is still lacking.
Marina beach, located on the Southeast coast of India, is widely knownfor the recreational and fishing activities carried out. Recreational uses like surfing, swimming, horse riding, and picnicking, are likely drivers of plastic pollution. Preliminary marine litter surveys revealed that Marina beach is heavily polluted with marine litter (Arun kumar et al., 2016). Also, a wide range of studies have investigated the occurrence and impact of microplastics on the sediments of Marina beach and the Chennai coast (Karthik et al., 2018; Sathish et al., 2019; Sunitha et al., 2021; Venkatramanan et al., 2022). However, the current state of PPE pollution is yet to be assessed.
In this context, we have conducted wide range of surveys focusing on PPE pollution in the world's second longest beach (Marina beach), Tamil Nadu, India. In addition, a subsample of various types of PPE especially face mask and gloves was collected and analyzed by Fourier transformed infrared (FTIR), which may provide additional information on their chemical composition and degradation in the environment.
2 Materials and method
2.1 Study area
With a shoreline of >7500 km, India has been of the main centers of intense tourism and recreational, fishing, and shipping activities along the shore. The density of its population, the growing plastic consumption, and the presence of large rivers can constitute threatening factors for the increasing plastic pollution in the marine environment. In India, plastic consumption has increased 20-fold between 1990 (0.9 MT) and 2018 (18.45 MT; Indian Plastic Industry Report, 2019). Plastic litter mismanagement continues to be a major problem in India. The growing generation and inappropriate uses of PPE have introduced vast amounts of this type of litter into the environment and aquatic ecosystem. In India, the current average quantity of COVID-19-related biomedical waste generation during May 2022, is estimated at about 3.54 TPD (CPCB, 2022). The government of Tamil Nadu state contributed 0.338 tons per day in May 2022, while COVID-19 biomedical waste kept increasing in June (0.776 tons per day) and July (0.925 TDP) 2022. Particularly, the capital city of Tamil Nadu, Chennai, is the second highest generation of COVID-19 biomedical waste (0.216.82 TDP) during July 2022, took place. Since the outbreak of COVID-19, India and the Tamil Nadu government have proactively taken several measures for containing the disease which are in line with guidance from WHO, CDC, and other international best practices guidance and learning. Despite all these efforts, PPE litter is found on parks, beaches, roads, and sewers.
Chennai Metropolitan city is the fourth largest city in India and the capital city of Tamil Nadu state, located on the southeast coast of India (Fig. 1 ). According to the UN World population prospects, (2022) the Chennai population is estimated 11.5 million. Marina beach (13.05°N, 80.2824°E) is a natural urban beach in Chennai. It is the second largest urban beach in the world, situated by Fort Saint George in the north, Foreshore Estate in the south, and Bay of Bengal in the East. With a length of 6 km, it is India's longest natural city beach (Fig. 1). The mean width of the beach is 300 m with a maximum width of 437 m. It is one of the busiest beaches in the country, attracting nearly 30,000 visitors a day on weekdays and 50,000 visitors per day on weekends and public holidays. The fishing industry is the main activity for the people in the coastal areas, and the urban coastline people are working in industries and government and non-government organizations (Venkatachalapathy et al., 2011). Furthermore, the Adyar and Cooum rivers carry the plastic litter to the shoreline and make Marina beach unhygienic (Gowri et al., 2008). However, solid waste management across the Tamil Nadu state is very poor, resulting in a significant amount of marine litter and plastic debris contaminating coastal areas (Arun kumar et al., 2016). Especially, marine litter and microplastic pollution have been evidenced in the study area (; Arun kumar et al., 2016; Sathish et al., 2019; Ranjani et al., 2022; Venkatramanan et al., 2022).Fig. 1 Map of the sampling sites in Marina beach, Tamil Nadu, India.
Fig. 1
2.2 PPE monitoring
A total of 10 sampling sites were selected in Marina beach (Fig. 1). The selected sites were well distributed along the beach and were representative of different coastal activities, such as recreational and fishing activities (Table 1 ). PPE surveys were carried out from the summer (May & June) to the beginning of the post-monsoon (July) season of 2022. During the sampling campaign, Marina beach was open to the public, thus gathering an enormous number of beachgoers and workers wearing various types of face masks. The sampling method was followed based on the previous studies conducted on the beaches of Morocco and Peru (Ben Haddad et al., 2021; De-la-Torre et al., 2021). In short, a sampling area that covered the entire extent of the beach (from the low-tide line to the upper beach limit) was determined at each location and many transects (parallel to each other) were separated by 8–10 m intervals to cover entire beach areas.Table 1 The major activity, substrate, surveyed area of each sampling site, geographical coordinates Marina beach, number and density of PPE recorded in Marina beach, Tamil Nadu, India.
Table 1Code Site Coordinates Activity Area covered (m2) Number of PPE Density (m−2)
Starts End
S1 MGR memorial spot 13°03′58.77”N; 80°17′22.42″E 13°03′51.63”N; 80°17′18.83″E Recreational 25,678 194 7.55 × 10−3
S2 Marina beach 1 13°03′42.98”N; 80°17′15.26″E 13°03′33.19”N; 80°17′10.87″E Recreational 41,476 275 6.63 × 10−3
S3 Marina beach 2 13°03′31.69”N; 80°17′07.71″E 13°03′24.18”N; 80°17′09.15″E Recreational 30,588 205 6.70 × 10−3
S4 Masi Magam theerthavari spot 13°03′22.78”N; 80°17′06.16″E 13°03′14.01”N; 80°17′05.77″E Recreational 31,798 146 4.59 × 10−3
S5 Mobile restaurant of Tamilnadu fisheries 13°03′07.90”N; 80°17′02.47″E 13°03′00.44”N; 80°17′00.68″E Recreational 25,623 85 3.31 × 10−3
S6 Marina creek 13°02′56.29”N; 80°16′01.88″E 13°02′50.45”N; 80°16′57.02″E Recreational 21,897 74 3.37 × 10−3
S7 Gandhi beach 1 13°02′45.86”N; 80°16′58.20″E 13°02′39.27”N; 80°16′55.03″E Recreational 21,286 76 3.57 × 10−3
S8 Gandhi beach 2 13°02′31.79”N; 80°16′53.68″E 13°02′23.56”N; 80°16′51.12″E Recreational 23,811 49 2.05 × 10−3
S9 Marina beach fish market 13°02′19.87”N; 80°16′52.51″E 13°02′14.10”N; 80°16′50.67″E Fishing 11,424 33 2.88 × 10−3
S10 Foreshore estate beach 13°02′08.06”N; 80°16′50.41″E 13°02′01.45”N; 80°16′48.16″E Fishing 12,687 17 1.33 × 10−3
PPE sampling strategies consist of walking along each transect, visually scanning the environments, and detectable PPE litters were identified, which were categorized as face masks, gloves, face shields, and bouffant caps. Every PPE litter was photographed. Various PPE litters were carefully collected and stored in zip lock bags to be transported to the laboratory for further analysis. In each site, the sampling area was estimated using Google Earth (https://www.google.com/earth/) (Table 1).
2.3 FTIR analysis
The various subsamples of PPE litter (n = 4, 2- weathered surgical face mask, 1 glove, and 1 N-95 face mask) collected from the sampling site of Marina beach were analyzed by Fourier- transform infrared (FTIR) spectroscopy, following De-la-Torre et al. (2022b). In order to analyze PPE litter with FTIR spectroscopy, face masks were cut open and three layer was determined separately. The readings were carried out in transmittance mode at wavelengths varied from 3500 to 500 cm−1 at 8 cm−1 resolution (De-la-Torre et al., 2022b). The adsorption bands were analyzed manually to determine the presence of various functional groups and suspected polymer types. Additionally, brand-new surgical, KN95 face masks, and gloves were analyzed by FTIR for comparison.
2.4 Statistical analysis
The PPE litter density in each sampling site was followed by Okuku et al. (2020a).C=n/a
where C represents the density of PPE (PPE m−2), n denotes the number of PPE and a is the sampled area (m2). The mean density of PPE in each station was presented as a boxplot. Sample locations were combined by activity (recreational activities and fishing activity) to investigate its influence on PPE density. PPE density data were examined for normality and homoscedasticity. PPE densities were not normally distributed (Kolmogorov-Smirnov test and Levene test, p < 0.05). Therefore, nonparametric tests were employed. Significant differences in PPE density between all sites and weeks were analyzed by the Kruskal-Wallis test. The significance level was fixed at 0.05 for all statistical tests. Statistical tests were performed using SPSS software (version 20 for Windows).
3 Results and discussion
The occurrence and distribution of COVID-19-driven PPE items were monitored in the world's second longest beach (Marina beach), Tamil Nadu, Southeast coast of India. A total of 1154 COVID-19-associated PPE items were found on Marina beach. Fig. 2 shows examples of COVID-19-driven PPE items. The entire beach was predominantly polluted by face masks (99.81 %) and only 2 surgical gloves were found (0.17 %) (Fig. 3 ). Among the total face masks, 97.83 % were surgical masks, 1.21 % were cloth masks and 0.77 % were KN-95 respirators. Face shields and hazard suits were not found on the beaches. The predominance of surgical masks in the marine environment has been recorded in most studies with some exceptions (Ben Haddad et al., 2021; De-la-Torre et al., 2021; Hatami et al., 2022; Rakib et al., 2021b; Aragaw et al., 2022; Dioses-Salinas et al., 2022; Mghili et al., 2022; Ribeiro et al., 2022; Sajorne et al., 2022). This is probably due to the accessibility of surgical masks and their low cost, as well as the mandates of their use. This study showed that face masks are still abundant in the marine environment. Despite the increase in vaccination rates, the number of cases affected by the COVID-19 during the summer period is rapidly increasing. Likewise, in Tamil Nadu, the number of active cases gradually increased in May and June 2022 (10,033), whereas active cases started to reduce at end of the August 2022 (PRS, 2022). During the study period, the use of face masks is mandated by the government in public places. In addition, the WHO strongly recommends wearing masks even after vaccination to combat this virus. For this reason, visitors and tourists to Marina Beach have continued to wear masks.Fig. 2 Various types of surgical face masks, gloves found in different sampling sites on Marina beach.
Fig. 2
Fig. 3 a) Contribution of each type PPE items. b) Weekly evolution of the total number of PPE across sampling sites.
Fig. 3
A higher number of PPE items were recorded in all the stations on May 2022, followed by June 2022 and a lower number of PPE items was found on July 2022. The higher number of PPE items found in May 2022 may be due to the start of the summer holiday in India, therefore schools, colleges and the public kept visiting Marina beach. The number of tourists reached a peak during May and June which explains the large number of PPE recorded during these two months. This is consistent with the results of studies in the coastal zone, which have linked increasing numbers of beach visitors to higher PPE disposal (De-la-Torre et al., 2021; Thiel et al., 2021; Hassan et al., 2022; Sajorne et al., 2022). Following Fig. 3b, the sampling surveys with the highest densities of PPE were registered during the first four weeks. A similar temporal pattern has been recorded in Iran, Morocco, Peru, Bangladesh, Ethiopia and Brazil (Rakib et al., 2021; Aragaw et al., 2022; De-la-Torre et al., 2022a; Hatami et al., 2022; Mghili et al., 2022; Ribeiro et al., 2022). Concerning the PPE accumulation rates, we also observed a remarkable rise in the density of PPE items in weeks 2 and 3, which were marked to be Sunday and Monday. These two days of sampling coincided with the weekend. This may be attributed to the weekend effect. A total of 733 PPE litter were collected on the weekend while 421 were collected on the weekday. The number of visitors in the beach zones has increased during these weekend days, and consequently, PPE dumping increased in the coastal area. Our results were also similar to the findings of Hassan et al. (2022), where the number of PPE litter increased during weekends in Egypt and Saudi Arabia. Sajorne et al. (2022) also observed a large amount of PPE on the weekend. There was a decrease in the number of PPE during week 6 (W6). From the beginning of W6, the number of visitors suddenly decreased, which coincided with the start of the reopening of schools. In Morocco, Ben Haddad et al. (2021) recorded a low occurrence of PPE items during closures and a suddenly raised just after the beaches were reopened to the public.
The overall mean density of PPE items was 4.00 × 10−3 m−2 and ranged from 0.00 to 2.25 × 10−3 PPE m−2 (Table 2 ). These findings indicate a relatively higher abundance compared to previous studies (Table 2). The mean density of PPE in the study area was comparable to the values reported from the beaches of Tamil Nadu, India. Specifically, the abundances reported in this study are much higher than those from Morocco, Peru, Brazil, Ethiopia, Iran, and Argentina beaches. At the same time, these values are lower than the abundance recorded in Chile and Bangladesh. As the table shows, the abundance of PPE varies from one country to another. This variation may be affected by population density, COVID-19 restrictions and protocols, sampling area, weather conditions, and population density (Ben Haddad et al., 2021; Sajorne et al., 2022). The boxplot displays the mean density in each station (Fig. 4 ). PPE densities were grouped according to the principal activities performed in each sample area. A large number of PPE were recorded in the recreational activity (n = 1104; 0.0037 m−2) compared to the fishing activity (n = 50; 0.0004 m−2). PPE density differed significantly between the different activities (Kruskal Wallis test, p < 0.05). Previous research indicated a clear effect of the type of activity on the mean density of PPE. The present survey results observed that the highest number of PPE litter was found in more intensive touristic sites (S1 to S4) on Marina beach (Table 1). Particularly, the highest number of PPE items was found in S2 (n = 275), probably due to the larger number of beachgoers who occasionally gather and celebrated birthday parties, practice horse riding, and bathing. There is also a large number of commodities and snack shops located near to the S2 sampling site. Similarly, S5 to S8 are tourist sites in Marina beach where recreational and cultural activities (e.g., playing, bathing, spiritual activity) take place. Therefore, this touristic zone was also contaminated with a significant number of PPE items (n = 284). Kruskal-Wallis tests revealed that the PPE density differed significantly between the study sites (p < 0.05). Many residents and visitors were observed wearing facemasks. With a large number of visitors on recreational sites, the number of PPE litter was also reported to increase. For instance, recreational beaches in Lima, Peru (De-la-Torre et al., 2021), Cox's Bazar, Bangladesh (Rakib et al., 2021), Tetouan, Morocco (Mghili et al., 2022), and the Bushehr coast of the Persian Gulf (Akhbarizadeh et al., 2021) were considerably more polluted than the fishing beaches. Interestingly, in S9 (Marina beach fish market) and S10 (Foreshore estate beach) fishing activities primarily take place. Only 50 PPE items were found in these locations (representing 4.33 % of the total number of PPE), probably due to the reduced number of beachgoers and visitors in contrast with recreational beaches. Rakib et al. (2021) also recorded low densities in the fishing areas. Okuku et al. (2020b) indicated that beaches utilized for mixed activities had a higher density of litter than those utilized only for recreational or fishing activities.Table 2 Comparison of the mean density of PPE on different beaches in the world.
Table 2Country City PPE density (PPE m−2) Reference
Mean Range
Morocco Tetouan 1.2 × 10-3a 0.00–3.67 × 10−3 Mghili et al. (2022)
Morocco Agadir 1.13 × 10−5 0.001.21 × 10−4 Ben Haddad et al. (2021)
Kenya Kwale and Kilifi – 0.00–5.6 × 10−2 Okuku et al. (2020a)
Ethiopia Bahir Dar 1.54 × 10−4 1.22 × 10−5–2.88 × 10−4 Aragaw et al. (2022)
Peru Lima 6.42 × 10−5 0.00–7.44 × 10−4 De-la-Torre et al. (2021)
Peru Multiple 6.60 × 10−4 0.00–5.01 × 10−3 De-la-Torre et al., 2022a, De-la-Torre et al., 2022b, De-la-Torre et al., 2022c
Peru Protected areas 1.32 × 10−3 – Dioses-Salinas et al. (2022)
Argentina Multiple 7.21 × 10−4 0.00–5.60 × 10−3 De-la-Torre et al., 2022a, De-la-Torre et al., 2022b, De-la-Torre et al., 2022c
Brazil Santos 7.46 × 10−5 0.00–3.89 × 10−4 Ribeiro et al. (2022)
Chile Nationwide 6.00 × 10-3a – Thiel et al. (2021)
Bangladesh Cox’s Bazar 6.29 × 10−3 3.16 × 10−4–2.18 × 10−2 Rakib et al. (2021)
Iran Bushehr – 7.71 × 10−3 –2.70 × 10−2 Akhbarizadeh et al. (2021)
Iran Mazandaran 1.02 × 10−4 0.00–7.16 × 10−4 Hatami et al. (2022)
Iran Kish Island 2.34 × 10−4 0.00–1.18 × 10–3 Mohamadi et al. (2023)
India Tamil Nadu 1.08 × 10−3 2.80 × 10−4 –2.80 × 10−3 Gunasekaran et al. (2022)
India Marina beach 4.00 × 10−3 0.00–2.25 × 10−3 Present study
Fig. 4 Box plot diagram of the PPE number among sampling sites.
Fig. 4
The intertidal zone displayed a higher abundance of PPE (n = 664, mean = 0.0026 items/m2) compared to the supralittoral zone (n = 490, mean = 0.0019 items/m2). In contrast, Kaviarasan et al. (2022) reported a larger amount of marine litter accumulating in the supralittoral zone compared to the intertidal zone. This suggests that the vast majority of litter found may have been brought and incorrectly discarded by beachgoers instead of washed ashore. This behavior is probably due to the lack of awareness and poor environmental education (Mghili et al., 2020). The significant abundance of PPE on the supratidal zone shows that this zone is a critical litter sink. The large amounts of PPE from the intertidal zone are not only explained by recreational activities but probably due to environmental processes. Disposed masks can be carried to surface waters where they can be transported further into marine environments. In addition, PPE is discarded in landfills without appropriate management due to the lack of resources to manage this type of litter. Drainage systems are also identified as sources of PPE on beaches (Gunasekaran et al., 2022). Wind, inundation by rainwater, and rivers are the primary pathways of face masks to marine environments. Several rivers flowing from west to east transport significant quantities of plastic litter into the study area. Neelavannan et al. (2022) previously attributed a large amount of plastic litter on Poompuhar beach to the inflow of the Cauvery River, which plays an important role in the deposition of plastic on the Tamil Nadu coast. The significant presence of PPE on the intertidal zone is likely the result of this litter washing up on the beaches. This highlights that Indian waters are widely affected by this new type of pollution, which are subject to local hydrodynamics and poor environmental awareness.
Tamil Nadu is located on the South-Eastern coast of India, with a shoreline length of 1076 km (∼13 % of India's total shoreline) comprising multiple ecologically significant places, including Palk Bay (home to reefs and sea grass beds), the Gulf of Mannar Biosphere Reserve, Vedaranyam and Pichavaram (mangrove swamps), and Lake Pulicat (lagoon of ecological importance). The diversity of the environment of this region is accompanied by a great biological diversity with the presence of hundreds of endemic species of different taxa. This site is endowed with a wide range of marine biodiversity, such as fishes, bivalves, gastropods, crustaceans, corals, sea anemones, polychaete worms, echinoderms, and bryozoans (ZSI 2007; Venkatraman and Venkataraman, 2012; Tenjing et al., 2019). In addition, Marina beach supports turtle nesting sites, especially olive ridley turtles (Bhupathy et al., 2007). This region has lost its pristine and aesthetic condition due to high anthropogenic pressure, such as recreational activity, dumping untreated domestic waste, and fast-growing urbanization (Arunkumar et al., 2016; Venkatramanan et al., 2022). With the increasing number of COVID-19-related litter in Indian marine environments, the threats to marine life are becoming more and more numerous. PPE will probably become an entanglement and ingestion threat to marine fauna in Indian waters. Entanglement is the most reported impact of PPE on wildlife. Ammendolia et al. (2022) reported 114 cases of interaction between wildlife and COVID-19-related litter, most of which were affected by entanglements. Birds, mammals, and invertebrates were most affected by entanglements. These authors have documented two entanglements of animals by PPE in India. In this study, we observed the presence of a face mask near the hole of a crab (Fig. 5b). Mohamadi et al. (2023) already recorded a crab entangled in a face mask. Marine wildlife can also ingest face masks. The first case of face mask ingestion by a Magellanic penguin (Spheniscus magellanicus) was documented in Brazil (Neto et al., 2021). Also, the presence of masks in the feces of a juvenile green turtle has also been documented in Japan (Fukuoka et al., 2022). In India, many studies have documented the ingestion of microplastics by marine wildlife. Bioaccumulation of microplastics in epipelagic and mesopelagic fish, commercially important fishes, Indian white shrimp, Indian edible oysters, bivalves, and other marine animals has been reported in India (Patterson et al., 2019; Daniel et al., 2020a, Daniel et al., 2020b; Dowarah et al., 2020; James et al., 2020; Karuppasamy et al., 2020; Sathish et al., 2020). PPE can pose a hazard to many marine animals, especially those vulnerable to plastic ingestion in India. It is expected that PPE items, which are primarily composed of synthetic polymers, degrade periodically in the marine environment, leading to its fragmentation into smaller pieces (De-la-Torre et al., 2022b). PPE items may serve as a suitable substrate for the colonization of alien species. Studies have already indicated the adequacy of PPE as an artificial substrate for benthic organisms and also microorganisms, which increases the chances of biological invasion (De-la-Torre and Aragaw, 2021; De-la-Torre et al., 2021; Crisafi et al., 2022). Zhou et al. (2022) experimentally demonstrated that face masks enrich and host microbial communities, potentially acting as pathogen vectors. In addition, PPE in the marine environment acts as a carrier for the transfer of contaminants to marine organisms, which could potentially induce a range of deleterious and cytotoxic effects (Dobaradaran et al., 2018, Dobaradaran et al., 2021; Takdastan et al., 2021; Hajiouni et al., 2022). Recent studies have revealed that masks and wipes contain a large number of inorganic and organic pollutants used as UV stabilizers, plasticizers, and flame retardants in plastic production, including organophosphate esters, phthalates (di- and mono) and non-phthalates, bisphenols, antioxidants and plastic additives (Liu and Mabury, 2021; Sullivan et al., 2021; Wang et al., 2021a; Kutralam-Muniasamy et al., 2022). Face masks also contain metallic nanoparticles, such as Ag and Cu (Ardusso et al., 2021; De-la-Torre et al., 2022c). Based on previous work, the massive use of single-use plastics due to the COVID-19 clearly added additional stress to Indian marine ecosystems that are already threatened by numerous pressures.Fig. 5 a) Degraded masks found in the study area. b) a face mask near a crab hole.
Fig. 5
3.1 FTIR results
Weathered and brand-new PPE (2 surgical face masks, 1 KN95 respirator, and 1 glove) were analyzed by FTIR to determine their polymeric composition and signs of degradation (Fig. 6 ). Both surgical face masks and KN95 respirators showed typical PP absorption bands characterized by the presence of strong peaks around 2950, 2915, 2838 cm−1 (assigned to C—H stretching), 1455, and 1377 cm−1 (assigned to CH2, and CH3 bending, respectively), and weaker peaks around 1166, 997, 972, 840, and 808 cm−1 (assigned to the stretching, bending, and rocking of C—C, C—H, CH2, CH3) according to Jung et al. (2018). The spectra coincide with those from the brand-new surgical face mask and KN95 respirator with slightly weaker absorption bands. The weathered samples, regardless of the mask type, showed an increase in the strength of peak at around 1700–1780 cm−1, which is normally attributed to carboxyl groups (C <svg xmlns="http://www.w3.org/2000/svg" version="1.0" width="20.666667pt" height="16.000000pt" viewBox="0 0 20.666667 16.000000" preserveAspectRatio="xMidYMid meet"><metadata> Created by potrace 1.16, written by Peter Selinger 2001-2019 </metadata><g transform="translate(1.000000,15.000000) scale(0.019444,-0.019444)" fill="currentColor" stroke="none"><path d="M0 440 l0 -40 480 0 480 0 0 40 0 40 -480 0 -480 0 0 -40z M0 280 l0 -40 480 0 480 0 0 40 0 40 -480 0 -480 0 0 -40z"/></g></svg> O). Similar observations have been reported in face masks extracted from the environment in Peru, Argentina, Ethiopia, and The Persian Gulf (Aragaw et al., 2022; De-la-Torre et al., 2022b; Mohamadi et al., 2023). These changes are due to exposure to the sun, which induced chain scission and later oxidation of the polymer chains (Gewert et al., 2018). Highly weathered plastics tend to change their structure, mostly increasing their crystallinity in the case of polyolefins, like PE and PP (Hsu et al., 2017), ultimately becoming more brittle and subject to fragmentation under a mechanical stressor (Andrady et al., 2022).Fig. 6 Photographs of a weathered surgical face mask and their corresponding FTIR spectrum.
Fig. 6
The glove was identified as Latex, characterized by the presence of a broad peak at around 3394 cm−1 attributed to OH stretching (Fig. 7 ), as well as sharper peaks at 2957, 2918, and 2851 cm−1 attributed to C—H stretching (Agostini et al., 2008). Weaker peaks at around 1795 cm−1, and between 1650 and 1500 cm−1 are likely associated with vibrations of the CO and CC structures, respectively (Jung et al., 2018). The peak at 830 cm−1 was attributed to the CCH3 CH structure, typical of natural rubber (Rolere et al., 2015). Absorption bands at around 1412 and 872 cm−1 may be due to the presence of calcium carbonate (CO3 −2 stretching and out of plane deformation), which is a common additive in latex gloves (Baeta et al., 2009). The wavenumber and peak intensity of the weathered glove were similar to those from the brand-new one, except for a sharper peak at 3394 cm−1 (OH stretching). Wang et al. (2022) reported the occurrence of stronger peaks around 1780–1700 cm−1 assigned to CO groups in weathered latex gloves, as well as a drop in the strength of O—H absorption bands (~3000–2800 cm−1). This behavior was not followed by the IR spectra of the weathered glove in the present study. However, this is likely due to the reduced/uncontrolled time exposed to the sun, resulting in photooxidation.Fig. 7 Photographs of a weathered glove (top) and KN-95 face mask and their corresponding FTIR spectrum.
Fig. 7
According to the above data, the presence of PPE in the marine environment can generate large concentrations of microplastics and continuously pollute the environment. Several studies have documented the release of micro- and nanoplastics from disposable masks and other PPE (Aragaw, 2020; Fadare and Okoffo, 2020; Ma et al., 2021; Saliu et al., 2021; Wang et al., 2021b). Some items of the face masks showed some degree of damage or physical degradation (torn layers, Fig. 5a), similar to those reported by Akhbarizadeh et al. (2021). With a large number of PPE litter entering Indian beaches, we assume that microplastic pollution may become more extensive, especially in the areas most impacted by this type of pollution.
4 Recommendations
There is an urgent need for viable management actions to save the marine resources in Marina beach from illegal littering of PPE items. As mentioned earlier, the lack of awareness is one of the major causes of PPE pollution on Indian beaches. Awareness and education may encourage more pro-environmental behaviors and decrease incorrect disposal of plastics and PPE on beaches. One of the best methods is to involve citizens and children in the cleanup campaigns, and apply citizen science to obtain litter contamination baselines while encouraging the population to get involved in educational activities (Bouzekry et al., 2022). The reappearance of the Olive Ridley turtles on the beach of Mumbai after 20 years, following the largest cleaning operation, is one of the best examples of the importance of involving people in the cleanup efforts (AFP, 2018). Extensive media coverage of the negative impact of incorrect PPE and plastic disposal is necessary to raise awareness of better practices through promotional videos and educational campaigns. During the survey, it was observed that none of the garbage cans were placed on the beach. For this, we recommend the installation of garbage cans every 100 m on the entire beach. Periodic beach cleaning campaigns should be implemented in order to collect misplaced debris. Reusable masks should be promoted, as they can be used repeatedly and are a better choice compared to surgical masks. Recycling PPE litter provides a higher benefit to the society. Today, there are a number of initiatives to recycle PPE in India, but they are still insufficient to make up for the amount of waste generated in coastal areas. It is imperative that local and competent authorities allocate an appropriate budget to combat marine litter in India while reducing waste generation at the source by improving solid waste management systems and developing education programs.
5 Conclusions
The COVID-19 pandemic has accelerated the PPE pollution in the beach environment, negatively impacting marine biota, potentially harbour pathogens and non-native species, and release of MPs and chemical contaminants. In the present study, COVID-19-driven PPE litter was monitored for 10 continuous weeks in 10 different sampling sites on Marina Beach, India, the longest urban beach in Asia. The observed results are comparable to those from the coastal environments around the world. The overall mean density of PPE items was 4.00 × 10−3 PPE m−2 and ranged from 0 to 2.25 × 10−3 PPE m−2. The entire Marina beach was predominantly polluted by face masks (99.81 %) and only 2 surgical gloves were found (0.17 %). A higher number of PPE items were recorded in all the stations on May 2022 (summer season), likely due to the summer holiday in India, where a higher number of people visited Marina beach. The results of the present surveys demonstrate that poor solid waste management and lack of environmental awareness among beachgoers on Marina beach are the main drivers of PPE pollution. The abundance of COVID-19-driven PPE items may cause entanglement, and ingestion hazards to the intertidal biota and top predators, as well as posing a potential source of microplastics and chemical contaminants. The authorities should make the alternative mitigation routes in a durable plan with waste-to-energy recycling policies. However, further research and development are needed regarding the leaching of chemical additives and microplastic, as well as elucidating the ecotoxicological consequences in order to obtain a clear picture concerning the environmental implications of COVID-19-driven PPE pollution.
CRediT authorship contribution statement
Gunasekaran Kannan: Conceptualization, Investigation, Methodology, Writing- Original draft preparation.
Bilal Mghili: Conceptualization, Writing- Original draft preparation.
Gabriel Enrique De-la-Torre: Methodology, Writing- Reviewing and Editing.
Prabhu Kolanthasamy: Writing- Reviewing and Editing,
Mayakrishnan Machendiranathan: Writing- Reviewing and Editing,
Mayavan Veeramuthu Rajeswari: Writing- Reviewing and Editing,
Ayyappan Saravanakumar: Writing- Reviewing and Editing.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
No data was used for the research described in the article.
Acknowledgement
The First author GK is grateful to authorities of Sathyabama Institute of Science and Technology, for providing the facilities.
==== Refs
References
AFP Olive ridley turtles hatch in Mumbai after two decades From https://phys.org/news/2018-03-olive-ridley-turtles-hatch-mumbai.html 2018
Agostini D.L.S. Constantino C.J.L. Job A.E. Thermal degradation of both latex and latex cast films forming membranes J. Therm. Anal. Calorim. 91 2008 703 707 10.1007/s10973-007-8351-x
Akhbarizadeh R. Dobaradaran S. Nabipour I. Tangestani M. Abedi D. Javanfekr F. Jeddi F. Zendehboodi A. Abandoned Covid-19 personal protective equipment along the Bushehr shores, the Persian Gulf: an emerging source of secondary microplastics in coastlines Mar. Pollut. Bull. 168 2021 112386 10.1016/j.marpolbul.2021.112386
Ammendolia J. Saturno J. Brooks A.L. Jacobs S. Jambeck J.R. An emerging source of plastic pollution: environmental presence of plastic personal protective equipment (PPE) debris related to COVID-19 in a metropolitan city Environ. Pollut. 269 2021 116160 10.1016/j.envpol.2020.116160
Ammendolia J. Tony R. Walker R. Citizen science: a way forward in tackling the plastic pollution crisis during and beyond the COVID-19 pandemic Sci. Total Environ. 805 2022 149957 10.1016/j.scitotenv.2021.149957
Andrady A.L. Barnes P.W. Bornman J.F. Gouin T. Madronich S. White C.C. Zepp R.G. Jansen M.A.K. Oxidation and fragmentation of plastics in a changing environment; from UV-radiation to biological degradation Sci. Total Environ. 2022 158022 10.1016/J.SCITOTENV.2022.158022
Aragaw T.A. Surgical face masks as a potential source for microplastic pollution in the COVID-19 scenario Mar. Pollut. Bull. 159 2020 111517 10.1016/j.marpolbul.2020.111517
Aragaw T.A. De-la-Torre G.E. Teshager A.A. Personal protective equipment (PPE) pollution driven by the COVID-19 pandemic along the shoreline of Lake Tana, Bahir Dar, Ethiopia Sci. Total Environ. 820 2022 153261 10.1016/J.SCITOTENV.2022.153261
Ardusso M. Forero-Lopez A.D. Buzzi N.S. Spetter C.V. Fernandez-Severini M.D. COVID-19 pandemic repercussions on plastic and antiviral polymeric textile causing pollution on beaches and coasts of South America Sci. Total Environ. 763 2021 144365 10.1016/j.scitotenv.2020.144365
Arunkumar A. Sivakumar R. Sai Rutwik Reddy Y. Bhagya M.V. Nishanth T. Revanth V. Preliminary study on marine debris pollution along Marina beach, Chennai, India Reg. Stud. Mar. Sci. 5 2016 35 40 10.1016/j.rsma.2016.01.002
Baeta D.A. Zattera J.A. Oliveira M.G. Oliveira P.J. The use of styrene-butadiene rubber waste as a potential filler in nitrile rubber: order of addition and size of waste particles Braz. J. Chem. Eng. 26 1 2009 23 31
Banerjee P. Hazra A. Ghosh P. Ganguly A. Murmu N.C. Chatterjee P.K. Solid waste management in India: a brief review Ghosh S.K. Waste Management and Resource Efficiency 2019 Springer Singapore
Ben Haddad M. De-la-Torre G.E. Abelouah M.R. Hajji S. Alla A.A. Personal protective equipment (PPE) pollution associated with the COVID-19 pandemic along the coastline of Agadir, Morocco Sci. Total Environ. 798 2021 149282 10.1016/J.SCITOTENV.2021.149282
Benson N.U. Bassey D.E. Palanisami T. COVID pollution: impact of COVID-19 pandemic on global plastic waste footprint Heliyon 2 2021 e06343 10.1016/j.heliyon.2021.e06343
Bhupathy S. Subramanean J. Vijay M. Nesting of Lepidochelys olivacea along the southern Chennai coast, with emphasis on habitat characteristics Hamadryad 31 2 2007 274 280
Bouzekry A. Mghili B. Aksissou M. Addressing the challenge of marine plastic litter in the moroccan Mediterranean: a citizen science project with schoolchildren Mar. Pollut. Bull. 184 2022 114167 10.1016/j.marpolbul.2022.114167
Central Pollution Control Board Generation of COVID-19 related biomedical waste in states/UTs https://cpcb.nic.in/uploads/Projects/Bio-Medical-Waste/COVID19_Waste_Management_status_Jul_Dec_2021.pdf 2021
Zoological Survey of India Z.S.I. Fauna of Chennai Coast Marine Ecosystem Series 1 2007 Government of India 1 294
CPCB Central Pollution Control Board Generation of COVID-19 related biomedical waste in states/UTs 2022
Crisafi F. Smedile F. Yakimov M.M. Aulenta F. Fazi S. La Cono V. Martinelli A. Di Lisio V. Denaro R. Bacterial biofilms on medical masks disposed in the marine environment: a hotspot of biological and functional diversity Sci. Total Environ. 837 2022 155731 10.1016/j.scitotenv.2022.155731
Daniel D.B. Ashraf P.M. Thomas S.N. Microplastics in the edible and inedible tissues of pelagic fishes sold for human consumption in Kerala, India Environ. Pollut. 266 2020 115365 10.1016/j.envpol.2020.115365
Daniel D.B. Ashraf P.M. Thomas S.N. Abundance, characteristics and seasonal variation of microplastics in Indian white shrimps (Fenneropenaeus indicus) from coastal waters off Cochin, Kerala, India Sci. Total Environ. 737 2020 139839 10.1016/j.scitotenv.2020.139839
De-la-Torre G.E. Aragaw T.A. What we need to know about PP E associated with the COVID-19 pandemic in the marine environment Mar. Pollut. Bull. 163 2021 111879 10.1016/j.marpolbul.2020.111879
De-la-Torre G.E. Jahan Refat Rakib Md Pizarro-Ortega C.I. Dioses-Salinas D.C. Occurrence of personal protective equipment (PPE) associated with the COVID-19 pandemic along the coast of Lima, Peru Sci. Total Environ. 744 2021 145774 10.1016/j.scitotenv.2021.145774
De-la-Torre G.E. Dioses Salinas D.C. Pizarro Ortega C.I. Fernández Severini M.D. Forero López A.D. Mansilla R. Ayala F. Jimenez Castillo L.M. Castillo-Paico E. Torres D.A. Mendoza-Castilla L.M. Meza-Chuquizuta C. Vizcarra J.K. Mejía M. Valdivia De La Gala J.J. Sayra Ninaja E.A. Siles Calisaya D.L. Flores-Miranda W.E. Eras Rosillo J.L. Espinoza-Morriberón D. Gonzales K.N. Torres F.G. Rimondino G.N. Ben-Haddad M. Dobaradaran S. Araga T.A. Santillán L. Binational survey of personal protective equipment (PPE) pollution driven by the COVID-19 pandemic in coastal environments: abundance, distribution, and analytical characterization J. Hazard. Mater. 42 6 2022 128070 10.1016/j.jhazmat.2021.128070
De-la-Torre G.E. Dioses-Salinas D.C. Dobaradaran S. Spitz J. Keshtkar M. Akhbarizadeh R. Abedi D. Tavakolian A. Physical and chemical degradation of littered personal protective equipment (PPE) under simulated environmental conditions Mar. Pollut. 178 2022 113587 10.1016/j.marpolbul.2022.113587
De-la-Torre G.E. Dioses-Salinas D.C. Dobaradaran S. Spitz J. Keshtkar M. Akhbarizadeh R. Tangestani M. Abedi D. Javanfekr F. Release of phthalate esters (PAEs) and microplastics (MPs) from face masks and gloves during the COVID-19 pandemic Environ. Res. 215 2022 114337 10.1016/j.envres.2022.114337
Dioses-Salinas D.C. Pizarro-Ortega C.I. Dobaradaran S. Ben-Haddad M. De-la-Torre G.E. Face masks invading protected areas: risks and recommendations Sci. Total Environ. 847 2022 157636 10.1016/J.SCITOTENV.2022.157636
Dobaradaran S. Schmidt T.C. Nabipour I. Khajeahmadi N. Tajbakhsh S. Saeedi R. Javad Mohammadi M. Keshtkar M. Khorsand M. Faraji Ghasemi F. Characterization of plastic debris and association of metals with microplastics in coastline sediment along the Persian Gulf Waste Manag. 78 2018 649 658 10.1016/J.WASMAN.2018.06.037 32559956
Dobaradaran S. Soleimani F. Akhbarizadeh R. Schmidt T.C. Marzban M. BasirianJahromi R. Environmental fate of cigarette butts and their toxicity in aquatic organisms: A comprehensive systematic review Environ. Res. 195 2021 110881 10.1016/j.envres.2021.110881 33607099
Dowarah K. Patchaiyappan A. Thirunavukkarasu C. Jayakumar S. Devipriya S.P. Quantification of microplastics using Nile red in two bivalve species Perna viridis and meretrix meretrix from three estuaries in Pondicherry, India and microplastic uptake by local communities through bivalve diet Mar. Pollut. Bull. 153 2020 110982 10.1016/j.marpolbul.2020.110982
Fadare O.O. Okoffo E.D. Covid-19 face masks: a potential source of microplastic fibers in the environment Sci. Total Environ. 737 2020 140279 10.1016/j.scitotenv.2020.140279
Fukuoka T. Sakane F. Kinoshita C. Sato K. Mizukawa K. Takada H. Covid-19-derived plastic debris contaminating marine ecosystem: alert from a sea turtle Mar. Pollut. Bull. 175 2022 113389 10.1016/J.MARPOLBUL.2022.113389
Gewert B. Plassmann M. Sandblom O. Macleod M. Identification of chain scission products released to water by plastic exposed to ultraviolet light Environ. Sci. Technol. Lett. 5 5 2018 272 276 10.1021/ACS.ESTLETT.8B00119/SUPPL_FILE/EZ8B00119_SI_001.PDF
Gowri V.S. Ramachandran S. Ramesh R. Pramiladevi I.R.R. Krishnaveni K. Application of GIS in the study of mass transport of pollutants by Adyar and cooum Rivers in Chennai, tamilnadu Environ. Monit. Assess. 138 2008 41 49 10.1007/s10661-007-9789-9 17562203
Gunasekaran K. Mghili B. Saravanakumar A. Personal protective equipment (PPE) pollution driven by the COVID-19 pandemic in coastal environment, southeast coast of India Mar. Pollut. Bull. 180 2022 113769 10.1016/j.marpolbul.2022.113769
Hajiouni S. Mohammadi A. Ramavandi B. Arfaeinia H. De-la-Torre G.E. Tekle-Röttering A. Dobaradaran S. Occurrence of microplastics and phthalate esters in urban runoff: a focus on the Persian Gulf coastline Sci. Total Environ. 806 2022 150559 10.1016/J.SCITOTENV.2021.150559
Hassan A.I. Younis A. Al Ghamdi M.A. Almazroui M. Basahi J.M. El-Sheekh M.M. Abouelkhair E.K. Haiba N.S. Alhussaini M.S. Hajjar D. Abdel Wahab M.M. Maghraby D.M. Contamination of the marine environment in Egypt and Saudi Arabia with personal protective equipment during COVID-19 pandemic: a short focus Sci. Total Environ. 810 2022 152046 10.1016/j.scitotenv.2021.152046
Hatami T. Rakib M.R.J. Madadi R. De-la-Torre G.E. Idris A.M. Personal protective equipment (PPE) pollution in the Caspian Sea, the largest enclosed inland water body in the world Sci. Total Environ. 824 2022 153771 10.1016/j.scitotenv.2022.153771
Hiemstra A.F. Rambonnet L. Gravendeel B. Schilthuizen M. The effects of COVID-19 litter on animal life Anim. Biol. 71 2021 215 231 10.1163/15707563-bja10052
Hsu Y.C. Truss R.W. Laycock B. Weir M.P. Nicholson T.M. Garvey C.J. Halley P.J. The effect of comonomer concentration and distribution on the photo-oxidative degradation of linear low density polyethylene films Polymer 119 2017 66 75 10.1016/J.POLYMER.2017.05.020
James K. Vasant K. Padua S. Gopinath V. Abilash K.S. Jeyabaskaran R. Babu A. John S. An assessment of microplastics in the ecosystem and selected commercially important fishes off Kochi, south eastern Arabian Sea, India Mar. Pollut. Bull. 154 2020 111027 10.1016/j.marpolbul.2020.111027
Ji B. Zhao Y. Wei T. Kang P. Water science under the global epidemic of COVID-19: bibliometric tracking on COVID-19 publication and further research needs J. Environ. Chem. Eng. 9 2021 1 2 10.1016/j.jece.2021.105357
Jung M.R. Horgen F.D. Orski S.V. Rodriguez C.V. Beers K.L. Balazs G.H. Jones T.T. Work T.M. Brignac K.C. Royer S.J. Hyrenbach K.D. Jensen B.A. Lynch J.M. Validation of ATR FT-IR to identify polymers of plastic marine debris, including those ingested by marine organisms Mar. Pollut. Bull. 127 2018 704 716 10.1016/j.marpolbul.2017.12.061 29475714
Karthik R. Robin R.S. Purvaja R. Ganguly D. Anandavelu I. Raghuraman R. Hariharan G. Ramakrishna A. Ramesh R. Microplastics along the beaches of southeast coast of India Sci. Total Environ. 645 2018 1388 1399 10.1016/j.scitotenv.2018.07.242 30248861
Karuppasamy P.K. Ravi A. Vasudevan L. Elangovan M.P. Dyana Mary P. Vincent S.G.T. Palanisami T. Baseline survey of micro and mesoplastics in the gastrointestinal tract of commercial fish from southeast coast of the bay of Bengal Mar. Pollut. Bull. 153 2020 110974 10.1016/j.marpolbul.2020.110974
Kaviarasan T. Dhineka K. Sambandam M. Sivadas S.K. Sivyer D. Hoehn D. Pradhan U. Mishra P. Murthy M.R. Impact of multiple beach activities on litter and microplastic composition, distribution, and characterization along the southeast coast of India Ocean Coast. Manag. 223 2022 106177 10.1016/j.ocecoaman.2022.106177
Kutralam-Muniasamy G. Pérez-Guevara F. Shruti V.C. A critical synthesis of current peer-reviewed literature on the environmental and human health impacts of COVID-19 PPE litter: new findings and next steps J. Hazard. Mater. 422 2022 126945 10.1016/J.JHAZMAT.2021.126945
Liu R. Mabury S.A. Single-use face masks as a potential source of synthetic antioxidants to the environment Environ. Sci. Technol. Lett. 8 8 2021 651 655
Ma J. Chen F. Xu H. Jiang H. Liu J. Li P. Chen C.C. Pan K. Face masks as a source of nanoplastics and microplastics in the environment: quantification, characterization, and potential for bioaccumulation Environ. Pollut. 288 2021 117748 10.1016/J.ENVPOL.2021.117748
Mghili B. Analla M. Aksissou M. Aissa C. Marine debris in moroccan Mediterranean beaches: an assessment of their abundance, composition and sources Mar. Pollut. Bull. 160 2020 111692 10.1016/j.marpolbul.2020.111692
Mghili B. Analla M. Aksissou M. Face mask s relate d to COVID -19 in the beaches of the moroccan Mediterranean: an emerging source of plastic pollution Mar. Pollut. Bull. 2022 113181 10.1016/J.MARPOLBUL.2021.113181
Mohamadi S. Madadi R. Refat Jahan Rakib M. De-la-Torre G.E. Idris A.M. Abundance and characterization of personal protective equipment (PPE) polluting Kish Island, Persian Gulf Sci. Total Environ. 854 2023 158678 10.1016/j.scitotenv.2022.158678
Morgana S. Casentini B. Amalfitano S. Uncovering the release ofmicro/nanoplastics from disposable face masks at times of COVID-19 J. Hazard. Mater. 419 2021 126507 10.1016/j.jhazmat.2021.126507
Neelavannan K. Achyuthan H. Sen I.S. Krishnakumar S. Gopinath K. Dhanalakshmi R. Rajalakshmi P.R. Sajeev R. Distribution and characterization of plastic debris pollution along the Poompuhar Beach, Tamil Nadu, Southern India Mar. Pollut. Bull. 175 2022 113337 10.1016/j.marpolbul.2022.113337
Neto H.G. Bantel C.G. Browning J. Della Fina N. Ballabio T.A. Telesde Santana F. Britto M.K. Barbosa C.B. Mortality of a juvenile magellanic penguin (Spheniscus magellanicus, Spheniscidae) associated with the ingestion of a PFF-2 protective mask during the Covid-19 pandemic Mar. Pollut. Bull. 166 2021 112232 10.1016/j.marpolbul.2021.112232
OceansAsia COVID-19 facemasks & marine plastic pollution [WWW document] 2020 OceansAsia https://oceansasia.org/covid-19-facemasks/
Okuku E. Kiteresi L. Owato G. Otieno K. Mwalugha C. Mbuche M. Gwada B. Nelson A. Chepkemboi P. Achieng Q. Wanjeri V. Ndwiga J. Mulupi L. Omire J. The impacts of COVID-19 pandemic on marine litter pollution along the Kenyan coast: a synthesis after 100 days following the first reported case in Kenya Mar. Pollut. Bull. 2020 111840 10.1016/j.marpolbul.2020.111840
Okuku E. Kiteresi L. Owato G. Mwalugha C. Omire J. Otieno K. Mbuche M. Nelson A. Gwada B. Mulupi L. Marine macro-litter composition and distribution along the Kenyan coast: the first-ever documented study Mar. Pollut. Bull. 159 2020 111497 10.1016/j.marpolbul.2020.111497
Patterson J. Jeyasanta K.I. Sathish N. Booth A.M. Edward J.K.P. Profiling microplastics in the Indian edible oyster, Magallana bilineata collected from the Tuticorin coast, Gulf of Mannar, Southeastern India Sci. Total Environ. 691 2019 727 735 10.1016/j.scitotenv.2019.07.063 31325870
Peng Y. Wua P. Schartup A.T. Zhang Y. Plastic waste release caused by COVID-19 and its fate in the global ocean Proc. Natl. Acad. Sci. 118 2021 1 6 10.1073/pnas.2111530118
Prata J.C. Silva A.L.P. Walker T.R. Duarte A.C. Rocha-Santos T. COVID-19 pandemic repercussions on the use and management of plastics Environ. Sci. Technol. 54 2020 7760 7765 10.1021/acs.est.0c02178 32531154
PRS Legislative research COVID-19 number of cases https://prsindia.org/covid-19/cases 2022
Rakib M.R.J. De-la-Torre G.E. Pizarro-Ortega C.I. Dioses-Salinas D.C. Al-Nahian S. Personal protective equipment (PPE) pollution driven by the COVID-19 pandemic in Cox’s bazar, the longest natural beach in the world Mar. Pollut. Bull. 169 2021 112497 10.1016/j.marpolbul.2021.112497
Ranjani M. Veerasingam S. Venkatachalapathy R. Jinoj T.P.S. Guganathan L. Mugilarasan M. Vethamony P. Seasonal variation, polymer hazard risk and controlling factors of microplastics in beach sediments along the southeast coast of India Environ. Pollut. 305 2022 119315 10.1016/j.envpol.2022.119315
Ribeiro V.V. De-la-Torre G.E. Castro I.B. COVID-19 related personal protective equipment (PPE) contamination in the highly urbanized southeast Brazilian coast Mar. Pollut. Bull. 177 2022 113522 10.1016/2Fj.marpolbul.2022.113522
Rolere S. Liengprayoon S. Vaysse L. Sainte-Beuve J. Bonfils F. Investigating natural rubber composition with fourier transform infrared (FT-IR) spectroscopy: a rapid and non-destructive method to determine both protein and lipid contents simultaneously Polym. Test. 43 2015 83 93 10.1016/j.polymertesting.2015.02.011
Sajorne R.E. Cayabo G.D.B. Madarcos J.R.V. Madarcos K.G. Omar D.M. Jr Ardines L.B. Sabtal S.A. Mabuhay-Omar J.A. Cheung V. Creencia L.A. Bacosa H.P. Occurrence of COVID-19 personal protective equipment (PPE) litters along the eastern coast of Palawan Island, Philippines Mar. Poll. Bull. 182 2022 113934 10.1016/j.marpolbul.2022.113934
Saliu F. Veronelli M. Raguso C. Barana D. Galli P. Lasagni M. The release process of microfibers: from surgical face masks into the marine environment Environ. Adv. 4 2021 100042 10.1016/j.envadv.2021.100042
Sathish N. Jeyasanta K.I. Patterson J. Abundance, characteristics and surface degradation features of microplastics in beach sediments of five coastal areas in Tamil Nadu, India Mar. Pollut. Bull. 142 2019 112 118 10.1016/j.marpolbul.2019.03.037 31232283
Sathish M.N. Jeyasanta I. Patterson J. Occurrence of microplastics in epipelagic and mesopelagic fishes from tuticorin, southeast coast of India Sci. Total Environ. 720 2020 137614 10.1016/j.scitotenv.2020.137614
Singh E. Kumar A. Mishra R. Kumar S. Solid waste management during COVID-19 pandemic: recovery techniques and responses Chemosphere 288 2022 132451 10.1016/J.CHEMOSPHERE.2021.132451
Sullivan G.L. Delgado-Gallardo J. Watson T.M. Sarp S. An investigation into the leaching of micro and nano particles and chemical pollutants from disposable face masks - linked to the COVID-19 pandemic Water Res. 196 2021 117033 10.1016/j.watres.2021.117033
Sunitha T.G. Monisha V. Sivanesan S. Vasanthy M. Prabhakaran M. Omine K. Sivasankar V. Darchen A. Micro-plastic pollution along the Bay of Bengal coastal stretch of Tamil Nadu, south India Sci. Total Environ. 756 2021 144073 10.1016/j.scitotenv.2020.144073
Takdastan A. Niari M.H. Babaei A. Dobaradaran S. Jorfi S. Ahmadi M. Occurrence and distribution of microplastic particles and the concentration of Di 2- ethyl hexyl phthalate (DEHP) in microplastics and wastewater in the wastewater treatment plant J. Environ. Manag. 280 2021 111851 10.1016/J. JENVMAN.2020.111851
Tenjing S.Y. Rocktim Ramen D. Deepak Samuel V.K. Pandian K. Purvaja R. Ramesh R. Rapid assessment of coastal biodiversity post-2015 Chennai flood, India Environ. Asia 12 2019 91 103 10.14456/ea.2019.49
Thiel M. de Veer D. Espinoza-Fuenzalida N.L. Espinoza C. Gallardo C. Hinojosa I.A. Kiessling T. Rojas J. Sanchez A. Sotomayor F. Vasquez N. Villablanca R. COVID lessons from the global south – face masks invading tourist beaches and recommendations for the outdoor seasons Sci. Total Environ. 2021 147486 10.1016/j.scitotenv.2021.147486
Torres F.G. Dioses-Salinas D.C. Pizarro-Ortega C.I. De-la-Torre G.E. Sorption of chemical contaminants on degradable and non-degradable microplastics: recent progress and research trends Sci. Total Environ. 757 2021 143875 10.1016/j.scitotenv.2020.143875
United Nations World Population prospect Summary of Results 2022 Department of Economic and Social Affairs 1 52 file:///D:/11a.%20Marina%20PPE/000.%20PPE%20pdf%20for%20Marina%20beach%20reference/undesa_pd_2022_WPP_summary_of_results.pdf
Venkatachalapathy R. Veerasingam S. Basavaiah N. Ramkumar T. Deenadayalan K. Environmental magnetic and petroleum hydrocarbons records in sediment cores from the north east coast of Tamil Nadu, Bay of Bengal, India Mar. Pollut. Bull. 62 2011 681 690 10.1016/j.marpolbul.2011.01.030 21329949
Venkatraman C. Venkataraman K. Diversity of Molluscan Fauna Along the Chennai Coast. International Day for Biological Diversity 2012 1 7
Venkatramanan S. Chung S.Y. Selvam S. Sivakumar K. Soundhariya G.R. Elzain H.E. Bhuyan M.S. Characteristics of microplastics in the beach sediments of Marina tourist beach, Chennai, India Mar. Pollut. Bull. 176 2022 113409 10.1016/j.marpolbul.2022.113409
Wang X. Okoffo E.D. Banks A.P. Li Y. Thomas K.V. Rauert C. Aylward L.L. Mueller J.F. Phthalate esters in face masks and associated inhalation exposure risk J. Hazard. Mater. 2021 127001 10.1016/J. JHAZMAT.2021.127001
Wang Z. An C. Chen X. Lee K. Zhang B. Feng Q. Disposable masks release microplastics to the aqueous environment with exacerbation by natural weathering J. Hazard. Mater. 417 2021 126036 10.1016/j.jhazmat.2021.126036
Wang Z. An C. Lee K. Chen X. Zhang B. Yin J. Feng Q. Physicochemical change and microparticle release from disposable gloves in the aqueous environment impacted by accelerated weathering Sci. Total Environ. 832 2022 154986 10.1016/j.scitotenv.2022.154986
WHO World Health Organization Director-General’s opening remarks at the media briefing on COVID-19 – 19-25 May https://www.who.int/publications/m/item/weekly-epidemiological-update-on-covid-19---25-may-2022 2022
Zhou S.Y. Lin C. Yang K. Yang L.Y. Yang X.R. Huang F.Y. Neilson R. Su J.Q. Zhu Y.G. Discarded masks as hotspots of antibiotic resistance genes during COVID-19 pandemic J. Hazard. Mater. 425 2022 127774 10.1016/j.jhazmat.2021.127774
| 0 | PMC9726691 | NO-CC CODE | 2022-12-16 23:18:05 | no | Mar Pollut Bull. 2023 Jan 7; 186:114476 | utf-8 | Mar Pollut Bull | 2,022 | 10.1016/j.marpolbul.2022.114476 | oa_other |
==== Front
J Hum Lact
J Hum Lact
JHL
spjhl
Journal of Human Lactation
0890-3344
1552-5732
SAGE Publications Sage CA: Los Angeles, CA
36398916
10.1177/08903344221134631
10.1177_08903344221134631
Original Research
Increase in SARS-CoV-2 RBD-Specific IgA and IgG Antibodies in Human Milk From Lactating Women Following the COVID-19 Booster Vaccination
https://orcid.org/0000-0002-7383-7616
Henle Andrea M. PhD 1Conceptualization Data curation Formal analysis Funding acquisition Investigation Methodology Project administration Resources Software Supervision Validation Visualization Writing - original draft Writing - review & editing
1 Biology Department, Carthage College, Kenosha, WI, USA
Andrea M. Henle, PhD, Biology Department, Carthage College, 2001 Alford Park Drive, Kenosha, WI 53140, USA. Email: [email protected]
18 11 2022
18 11 2022
0890334422113463123 3 2022
6 10 2022
© The Author(s) 2022
2022
International Lactation Consultant Association
This article is made available via the PMC Open Access Subset for unrestricted re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the COVID-19 pandemic or until permissions are revoked in writing. Upon expiration of these permissions, PMC is granted a perpetual license to make this article available via PMC and Europe PMC, consistent with existing copyright protections.
Background:
The United States Centers for Disease Control and Prevention recommended a third dose or booster of the Pfizer-BioNTech Comirnaty (BNT162b2) COVID-19 mRNA vaccine in September 2021 for high-risk individuals. Pregnant and high-risk lactating women were encouraged to receive the booster to obtain potential prolonged protection for themselves and their infants.
Research Aim:
To investigate the ability of the booster vaccine to increase IgA and IgG antibodies specific to the receptor-binding domain of the SARS-CoV-2 spike protein in human milk compared to levels pre-booster.
Methods:
This was a prospective one-group study with a pretest-posttest design. Six of 12 participants were recruited prospectively. Participants were instructed to collect ≥ 2 ounces of milk in the morning at 30 days and 1-day pre-booster, and 7, 14, 21, 30, 45, and 60 days post-booster. Levels of IgA and IgG antibodies specific to the receptor-binding domain of the SARS-CoV-2 spike protein were quantified in human milk via an ELISA assay.
Results:
We found a significant increase in anti-receptor-binding domain-specific IgA and IgG antibodies in human milk 1–2 weeks after the Pfizer-BioNTech booster and at the study endpoint (45- and 60-days post-booster)
Conclusions:
This suggests that the booster vaccination enhances SARS-CoV-2 specific immunity in human milk, which may be protective for infants.
booster
breastfeeding
Covid-19
IgA and IgG Antibodies in Human Milk
immunology
lactation
passive immunity
SARS-CoV-2
United States
vaccination
edited-statecorrected-proof
typesetterts1
==== Body
pmcKey Messages
To date, researchers have not investigated COVID-19 immunity in human milk following the third dose (i.e., booster) of the Pfizer-BioNTech Comirnaty vaccine.
In participants (N = 12), both IgA and IgG levels specific to SARS-COV-2 RBD in human milk were significantly higher 7 days post-booster versus pre-booster.
Antibodies to SARS-COV-2 RBD were detectable in blood ≥ 60 days post-Pfizer-BioNTech booster.
This suggests that the booster vaccination enhances SARS-CoV-2 specific immunity in human milk, which may be protective for infants.
Background
As of August 2022, over 92 million cases of coronavirus disease of 2019 (COVID-19) were confirmed in the United States, and over 1 million deaths (United States Centers for Disease Prevention and Control [CDC], 2020). The American Academy of Pediatrics (2022) reported that children represented 19% of the cumulative COVID-19 cases and 3.2% of the total hospitalizations for COVID-19 in the United States. While children aged 6 months and older were eligible as of late June 2022 for the Pfizer-BioNTech Comirnaty vaccination, those under 6 months of age remain ineligible and at risk for infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes COVID-19. Vaccination of pregnant or lactating mothers can protect infants from diseases e.g., pertussis and the flu (De Schutter et al., 2015; Zaman et al., 2008). Previously, researchers have reported elevated levels of SARS-CoV-2-specific IgA and IgG in milk from lactating women following the first and second doses of the Pfizer-BioNTech vaccine (Baird et al., 2021; Gray et al., 2021; Low et al., 2021; Perl et al., 2021; Valcarce et al., 2021). Therefore, in the absence of FDA-approved COVID-19 vaccinations for infants under 6 months of age, maternal vaccination against COVID-19 may offer passive immunity to infants, similarly to other vaccines typically administered to pregnant or lactating women (e.g., Tdap and Influenza). However, to our knowledge, no one has assessed the antibody response in human milk > 2 months beyond the initial two-dose vaccination series for COVID-19.
In September 2021, the CDC (2021a) recommended a third dose or booster of the Pfizer-BioNTech Comirnaty (BNT162b2) COVID-19 mRNA vaccine for people aged 65 and older, adults with underlying health conditions, and frontline workers. Pregnant and lactating women in these groups were advised to consider the booster for possible prolonged protection for themselves and their infants. We sought to address whether the booster changed the levels of anti-SARS-CoV-2 antibodies in human milk. We investigated the ability of the booster vaccine to increase IgA and IgG antibodies specific to the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein in human milk compared to levels pre-booster. We hypothesized that the booster would increase SARS-CoV-2 RBD-specific immunoglobulins in human milk.
Methods
Research Design
This was a prospective one-group study with a pretest-posttest design. The study compared the levels of human milk antibodies after the third dose of the Pfizer-BioNTech (BNT162b2) vaccine to the levels present ≥ 6 months following the second dose of the Pfizer-BioNTech (BNT162b2) vaccine. This study design allowed us to observe the effect of the booster vaccine on antibodies in human milk, which was the outcome of interest. The study was approved by the college’s institutional review board (IRB# 1817816-1) in November 2021.
Setting and Relevant Context
The study was conducted using human milk samples from participants across the United States. Lactating mothers in academia/education and healthcare were targeted because these groups were approved earlier than the general public for the booster and would better align with the timeline of this study. All participants were eligible to receive the booster between September and November 2021.
In the United States, the American Academy of Pediatrics (AAP) recommends that infants be fed human milk exclusively for the first 6 months of life, followed by an introduction of solid foods with the continuation of human milk (Meek & Noble, 2022). In June 2022, the AAP expanded its guidelines; they now suggest that infants continue to receive human milk up to 2 years of age, which is in accordance with guidelines from the World Health Organization (WHO, 2021). Despite these recommendations, only 25.8% of infants are exclusively breastfed through 6 months in the United States (CDC, 2021b).
There are significant challenges to breastfeeding in the United States since there is no federally-mandated paid parental leave. However, some mothers may be eligible for 12 weeks of unpaid maternity leave through the federal Family and Medical Leave Act, depending on their employer and how long they have been employed. Many mothers cannot take unpaid maternity leave due to the financial burden it places on the family (Sriraman & Kellams, 2016). Additionally, only 51% of employers in the United States report having an on-site lactation room (CDC, 2021b). The lack of these facilities at a place of employment causes further challenges for those who wish to continue feeding their infant human milk while returning to the workplace.
There are also disparities in breastfeeding rates in the United States across different races, incomes, and ages (CDC, 2021b). For example, only 19.3% of mothers under age 20 are breastfeeding when the infant is 6 months of age. In comparison, 33.8% of mothers ages 20–29 and 48.5% of mothers ≥ 30 years of age are breastfeeding when the infant is 6 months of age.
Sample
Six of the 12 participants were recruited prospectively. The additional six participants were recruited approximately 1 month into the 90-day timeline for the study but were eligible because they had dated and frozen aliquots of milk from the earlier required time points in the study. Participants’ samples were represented across the majority of time points for analysis of both RBD-specific IgA and IgG antibodies, except 30 days pre-booster (Day -30) and 90 days post-booster (Day 90) (Table 1). Those two time points were considered optional milk collection dates for the participants. Two participants could not provide milk samples from 1 day pre-booster (Day -1), so their samples from 30 days pre-booster (Day -30) were used for the analysis instead. Four participants voluntarily provided milk samples at 90 days post-booster (Day 90) to further extend the study timeline. Participants were included in the study if they could provide at least one pre-booster milk sample (Day -30 or Day -1) and a post-booster milk sample on Day 60. All twelve participants satisfied the aforementioned sample criteria.
Table 1. Participants Represented at Each Time Point in the Study (N = 12).
Time
(Day) IgA Analysis
n (%) IgG Analysis
n (%)
-30 8 (67) 8 (67)
-1 10 (83) 10 (83)
7 7 (58) 9 (75)
14 10 (83) 10 (83)
21 9 (75) 9 (75)
30 11 (92) 11 (92)
45 11 (92) 11 (92)
60 12 (100) 12 (100)
90 4 (25) 4 (25)
Note. IgA = Immunoglobulin A; IgG = Immunoglobulin G.
To be eligible for this study, participants needed to be lactating, must have received the standard two-dose Pfizer-BioNTech vaccination series at least 6-months prior, and provided written informed consent. Participants were excluded if they had a known diagnosis or suspected infection for COVID-19 and were currently pregnant. Inclusion criteria were determined via an initial demographic and health questionnaire completed online by the participants at enrollment (see online Supplementary Material). Participants were not compensated for their involvement in this study. The sample size (N = 12) was adequate to detect a difference in antibody concentration of 4 units/ml with greater than 80% power for IgA and a difference in 40 units/ml with greater than 80% power for IgG.
Measurement
Anti-RBD-specific IgA and IgG levels were assessed via enzyme-linked immunosorbent assay (ELISA; Ancell Corp.). First, 96-well plates were coated for 2 hr with 100 µl/well of 4 µg/ml purified RBD-His spike protein (Ancell Corp.). The plates were then aspirated, and 300 µl/well of blocking buffer (Tris buffered saline/glycine-01% BSA-10% glycerol 0.04% sodium azide, pH = 7.45) was added for 1 hr at room temperature. Human milk specimens were diluted 1:5, and blood serum was diluted 1:100 for IgG analysis and 1:50 for IgA analysis and incubated on the plates for 1 hr at room temperature with shaking. Plates were washed twice with 300 µl/well Tris-buffered saline (TBS)/glycine, 0.1% BSA, 0.1% pluronic acid, pH = 7.49. Monoclonal mouse anti-human IgG-HRP (ICO-97; 0.8 µg/ml) and mouse anti-human IgA-HRP (Hisa43; 2 µg/ml) antibodies (100 µl/well; Ancell Corp.) were added to capture the binding signal, and the plates were incubated at room temperature with shaking for 1 hr. Plates were washed three times with 300 µl/well TBS. TMB H2O2 substrate (Ancell Corp.) was used for detection at 450 nm on a Biotek Powerwave X plate reader. All human milk and blood samples were run in triplicate. The negative control for this assay was a human milk sample from July 2019. A positive control and standard curve were generated by serially diluting blood serum obtained from a COVID-19 positive patient in May 2020 into the 2019 negative control human milk sample in duplicate. IgA and IgG levels were converted to units/ml.
Data Collection
The study milk samples were collected in 2021–2022. Twelve participants provided a total of 87 milk samples. The negative control was human milk from July 2019. Changes in participants’ health status were monitored with a post-study survey (see supplementary material). Participants were instructed to collect ≥ 2 oz of milk in the morning at 30 days and 1 day pre-booster, and 7, 14, 21, 30, 45, and 60 days post-booster.
Samples were immediately stored at -20 °C until analyzed. Upon receipt in the lab, each sample containing 2 oz or more of human milk was thawed in a 37 °C water bath, aliquoted into a 50 ml conical tube, and centrifuged for 25 min at 872 x g (2000 rpm) and 4 °C on a Thermo Scientific Sorvall Legend XTR centrifuge. After centrifugation, a 25 ml serological pipette was used to separate the aqueous layer on the bottom of the tube from the fat layer on the top. The aqueous layer was transferred to a new 15 ml conical tube and stored at -20 °C until the ELISA assay was performed.
Participants used an alcohol swab and a finger-prick lancet device to collect up to 200 µl of blood into a Becton Dickinson (BD) microtainer in January 2022 (≥ 60 days post-booster). The blood was allowed to clot for 20 min at room temperature and then promptly stored at 4 °C. Upon receipt in the lab, the blood was centrifuged at 9391 x g (10,000 rpm) for 10 min at room temperature on an Eppendorf 5424 centrifuge. The serum layer was collected and stored at -20 °C until the ELISA assay was performed. Individual data were confidentially maintained, assigned a random participant code, and stored on a secure institutional server.
Data Analysis
To describe the characteristics of the study sample, we used mean and standard deviations for the continuous variables and frequencies and percentages for the categorical variables. Data were analyzed with GraphPad Prism 7. IgA and IgG levels were reported as units/ml. Data were expressed as median values with upper and lower limits, or mean values and 95% CIs. A Kruskal-Wallis H test with Dunn’s multiple comparisons test was used to determine differences between the mean ranks of IgA and IgG pre-booster (Day -1) to all other time points. The same tests were also used to determine differences between the mean ranks of IgA and IgG levels within each time point. The significance threshold was p < .05.
Results
Characteristics of the Participants
All participants were White, not Hispanic or Latino, or of Spanish origin. Five participants had gestational diabetes during their recent pregnancy, two participants had hypertension during their recent pregnancy, and two reported using antibiotics within the past 6 months. Participants were a mean (SD) age of 35.45 (4.17) years and infants had a mean age of 3.58 (1.8) months at the time of the booster (Table 2). The mean (SD) time between the second dose of the vaccine and the booster was 7.01 (0.62) months (Table 2). Vaccine-related adverse events were reported by 91.67% of participants after the booster, with injection site soreness the most frequent event (75%; Table 3). At ≥ 60 days post-booster, 58.33% of participants had RBD-specific IgA antibodies, and 100% had RBD-specific IgG antibodies in their blood (Table 3).
Table 2. Participant Characteristics (N = 12).
Characteristic M (SD)
Maternal Characteristics
Maternal Age, years 35.45 (4.17)
Weight, kg 83.50 (23.57)
Height, m 1.654 (0.075)
Pregnancy
Gravidity 2.75 (1.42)
Birth week 39.30 (0.74)
First vaccine during pregnancy
Pregnancy week at first dose 22.02 (7.21)
Pregnancy week at second dose 25.02 (7.31)
Infant age (months) when participant received booster 3.58 (1.8)
First vaccine after pregnancy
Infant age (months) when participant received booster 10.84 (0.64)
Time between 2nd vaccine dose and booster (months) 7.01 (0.62)
Note. Missing values: First vaccine during pregnancy = 4; first vaccine after pregnancy = 8.
Table 3. Participants’ Reported or Measured Outcomes (N = 12).
Outcome n (%)
Booster vaccine adverse reactions 11 (92)
Injection site soreness 9 (75)
Injection site rash 1 (8)
Injection site swelling 3 (25)
Injection site redness 4 (33)
Headache 2 (17)
Muscle or body aches 4 (33)
Joint pain 3 (25)
Fatigue or tiredness 4 (33)
Fever 3 (25)
Chills 5 (42)
Syncope 1 (8)
IgA positive blood serum at day ≥60 7 (58)
IgG positive blood serum at day ≥60 12 (100)
Note. IgA = Immunoglobulin A; IgG = Immunoglobulin G.
RBD-Specific IgG and IgA Levels
A Kruskal-Wallis H test (p < .0001) followed by a Dunn’s multiple comparisons test indicated that the median levels of RBD-specific IgG antibodies in human milk were significantly higher than the median levels of RBD-specific IgA antibodies across all time points in this study except Day 7 (Table 4). A Kruskal-Wallis H test showed that there was a statistically significant difference (p < .0001) in RBD-specific IgG antibodies between the pre-booster Day -1 and all six post-booster time points (Table 4 and Figure 1). A post hoc Dunn’s multiple comparisons test indicated that the median for the pre-booster Day -1 (Mdn = 5.548) was significantly lower (p < .05) than post-booster Day 7 (Mdn = 9.058), Day 14 (Mdn = 22.945), Day 21 (Mdn = 15.461), Day 30 (Mdn = 14.419) Day 45 (Mdn = 12.801), and day 60 (Mdn = 28.243). RBD-specific IgG antibodies in milk significantly increased by Days 7 and 14 (Table 4 and Figure 1); On Day 7 post-booster, 33.33% of participants had an increased level compared to pre-booster (Day -1). By Day 14 post-booster, 80% of participants had an increased level compared to pre-booster (Day -1). At 21, 45, and 60 days post-booster, respectively, 66.67%, 63.64%, and 75% of participants displayed significantly increased RBD-specific IgG antibodies.
Figure 1. Levels of Anti-RBD-Specific IgA and IgG in Human Milk Following the Pfizer-BioNTech Booster Vaccine (N = 12).
Note. IgA = Immunoglobulin A; IgG = Immunoglobulin G; ELISA = enzyme-linked immunosorbent assay; CI = confidence interval. Each time point is compared to the pre-booster levels of (A) IgA and (B) IgG at Day -1. Each participant’s sample was run in triplicate in one ELISA. Table 1 indicates the number of participants represented at each time point. Data points represent means; error bars, 95% CIs. All data points have error bars present.
*p < .05.
**p < .01.
***p ≤ .001.
****p ≤ .0001.
Table 4. Comparisons of the Levels of Anti-RBD-Specific IgA and IgG in Participants’ Milk Following the Vaccine Booster (N = 12).
Time (Day) IgA Analysis IgG Analysis Group Comparison
Mdn UL, LL p Mdn UL, LL p p
-30 -0.954 36.28, -10.08 0.9926 6.281 12.85, 4.16 >0.9999 0.0394
-1 -2.433 20.74, -10.32 - 5.548 9.46, 2.68 - 0.0004
7 9.069 55.74, -1.32 <0.0001 9.058 143.02, 4.56 0.0380 >0.9999
14 -1.600 49.36, -7.00 0.9659 22.945 294.35, 3.15 <0.0001 <0.0001
21 -1.117 41.03, -11.02 0.3060 15.461 294.35, 3.93 <0.0001 <0.0001
30 -1.471 18.48, -8.80 0.7981 14.419 235.67, 3.83 <0.0001 <0.0001
45 0.795 26.91, -7.60 0.0479 12.801 296.85, 3.07 0.0003 <0.0001
60 7.915 32.81, -7.44 <0.0001 28.243 294.35, 3.31 <0.0001 0.0252
Note. Median IgA and IgG levels are reported in units/ml (U/ml). Negative values indicate that the standardized sample value was lower than the negative control milk value. The p values reported within each IgA and IgG analysis column were determined via a Dunn’s multiple comparisons test where the dataset from each time was compared to the dataset from Day -1. The p values reported in the last column are from a Dunn’s multiple comparisons test evaluating differences between IgA and IgG levels at each time point. Missing values: Day -30 = 4, Day -1 = 2, Day 7 = 5 for the IgA analysis and 3 for the IgG analysis, Day 14 = 2, Day 21 = 3, Day 30 = 1, and Day 45 = 1. IgA = Immunoglobulin A; IgG = Immunoglobulin G; Mdn = median; UL = upper limit; LL = lower limit.
A Kruskal-Wallis H test showed that there was a statistically significant difference (p < .0001) in RBD-specific IgA antibodies between the pre-booster Day -1 and three post-booster time points (Days 7, 45, and 60; Table 4 and Figure 1). A post hoc Dunn’s multiple comparisons test indicated that the RBD-specific IgA antibodies in milk were significantly increased by Day 7 (Mdn =9 .069), Day 45 (Mdn = 0.795), and Day 60 (Mdn = 7.915) compared to pre-booster (Day -1; Mdn = -2.433; Table 4 and Figure 1). On post-booster Day 7, 71.43% of participants had an increased level compared to pre-booster (Day -1). On post-booster Days 45 and 60, 45.45% and 41.67% of participants had significantly increased RBD-specific IgA antibodies, respectively.
On an individual level, 9 of the 12 participants (75%) had a significant increase (p < .05) in RBD-specific IgA levels at one or more time points post-booster compared to pre-booster, whereas all 12 (100%) had a significant increase in RBD-specific IgG levels at two or more time points post-booster compared to pre-booster (individual data not shown). Additionally, of the four participants who provided a milk sample at 90 days post-booster, 3/4 (75%) had significantly elevated RBD-specific IgA antibodies (mean 15.24 U/ml) and RBD-specific IgG antibodies (mean 55.67 U/ml) at Day 90 compared to pre-booster (Day -1; data not shown).
Discussion
We found significant increase in RBD-specific IgA and IgG antibodies in human milk following the third dose (booster) of the Pfizer-BioNTech Comirnaty vaccination was found. Our results similar to those reported previously by researchers investigating the tetanus-diphtheria-acellular pertussis vaccination (Abu Raya et al., 2014), which demonstrated that vaccine-specific antibodies were detected in human milk after intramuscular immunization. Other researchers have reported high RBD-specific IgG antibodies in human milk following the standard two-dose vaccination against SARS-CoV-2 (Baird et al., 2021; Gray et al., 2021; Low et al., 2021; Perl et al., 2021; Young et al., 2022), and we observed a similar IgG-dominant effect post-booster. This could be due to the administration method of the vaccination—intramuscular injection may induce a more robust IgG response than other routes, which would favor mucosal (IgA) immunity. While secretory IgA is the dominant immunoglobulin in human milk (Goldman & Goldblum, 1989), IgG in human milk may be important for protection against viral infections (e.g., RSV and HIV; Fouda et al., 2011; Mazur et al., 2019). These results and those of previous researchers (Baird et al., 2021; Gray et al., 2021; Low et al., 2021; Perl et al., 2021; Young et al., 2022) suggest a role for IgG in human milk for infant immunity to SARS-CoV-2.
Previously researchers (Pace et al., 2021; Young et al., 2022) reported that the IgA and IgG antibodies induced in human milk following COVID-19 infection could neutralize SARS-CoV-2 in vitro. Therefore, the results from previous neutralization assays imply that the immunoglobulins induced in human milk following the parent’s third dose of the vaccination could offer the infant enhanced protection against SARS-CoV-2. The mother would also have increased protection against SARS-CoV-2 from the systemic neutralizing antibodies induced following the Comirnaty booster vaccination (Yu et al., 2022).
Limitations
The sample size in this study was small. At the time of recruitment for this study, the CDC had authorized the third dose of the Pfizer-BioNTech vaccination for high-risk individuals or frontline workers, for example, healthcare workers, first responders, or educators. Therefore, the study population was skewed and mainly included participants who were healthcare workers or educators and were of an older average age (35.5 years) than the typical age at first birth (26.9 years). Participants were recruited via social media posts in academic or local mom groups. This recruitment strategy may have been a source of potential bias as it could have excluded participants who lacked the time and resources to engage in social media. The study population should be expanded to include participants from broader ranges of races and ethnicities and age ranges more representative of the average child-bearing population.
No functional assays were performed to assess the protective activity of the immunoglobulins from human milk following the third dose of the vaccine. However, one researcher has determined that (serum) antibodies induced following the third dose of the Comirnaty vaccine could neutralize the parental WA1/2020 strain of SARS-CoV-2 and the variant BA.1 and BA.2 Omicron strains. The neutralizing antibody titers had increased substantially after the third dose of the vaccine compared to the titers measured after the initial two doses of the vaccine (Yu et al., 2022). Future investigations should use in vitro neutralization assays to determine whether the SARS-CoV-2-specific antibodies induced in human milk post-booster are protective and capable of neutralizing the virus.
The ELISA assay in this study used mouse anti-human IgA-HRP (Hisa43) antibody to capture the binding signal of anti-RBD-specific IgA antibodies. The anti-human IgA-HRP (Hisa43) antibody is specific to the CH3 domain of the Fc region of IgA and recognizes secretory IgA, the most dominant form of IgA in human milk (Biewenga et al., 1986, 1991). A capture antibody specific to the secretory component could be utilized as a more precise method for measuring the secretory form of the anti-RBD-specific IgA antibodies.
This study did not assess immunoglobulins following a booster with the Moderna vaccine or a mix-and-match (Pfizer-BioNTech, Moderna, or Johnson & Johnson) approach to Doses 1, 2, and 3. It also did not assess the level of immunoglobulins in human milk following a fourth dose of the Comirnaty vaccine. Future research is needed to assess optimal vaccination strategies (such as the mix-and-match approach) for providing the highest levels of antibodies against SARS-CoV-2 in human milk.
Participants were not tested for SARS-CoV-2 via real-time reverse transcriptase polymerase chain reaction (RT-PCR) as part of this study; therefore, it is difficult to determine whether elevated antibody levels could also be attributed to infection (either symptomatic or asymptomatic). However, no participants reported any known exposures to COVID-19 or any positive at-home or RT-PCR test results throughout the study. Additionally, this study concluded in early January 2022, so most milk sample collection occurred before the emergence of the highly transmissible Omicron BA.1 and BA.2 variants of SARS-CoV-2 in the United States. Future researchers could quantify the levels of immunoglobulins in human milk specific to the SARS-CoV-2 nucleocapsid (N) protein as a way to detect immunity to past infections (since N antigens are not present in the currently available COVID-19 vaccinations).
Conclusions
The recent clinical trial which tested the Pfizer-BioNTech vaccination in pregnant women assessed safety and efficacy but did not directly quantify immunoglobulin levels (Dagan et al., 2021). Healthcare providers and mothers have limited information regarding the ability of the COVID-19 vaccination to induce long-term immunity in human milk. Our research suggests administering the third dose of the vaccination ≥6 months after the standard two-dose Pfizer-BioNTech vaccination increases SARS-CoV-2 RBD-specific IgA and IgG antibodies in human milk. This may be a source of passive and protective immunity for infants. As of early August 2022, less than half of the adult population (48.2%) that is booster-eligible in the United States has received their booster vaccine (CDC, 2020). Pregnant or lactating women who are booster-eligible may wish to consider the booster vaccination for potential protection for their infants.
Supplemental Material
sj-pdf-1-jhl-10.1177_08903344221134631 – Supplemental material for Increase in SARS-CoV-2 RBD-Specific IgA and IgG Antibodies in Human Milk From Lactating Women Following the COVID-19 Booster Vaccination
Click here for additional data file.
Supplemental material, sj-pdf-1-jhl-10.1177_08903344221134631 for Increase in SARS-CoV-2 RBD-Specific IgA and IgG Antibodies in Human Milk From Lactating Women Following the COVID-19 Booster Vaccination by Andrea M. Henle in Journal of Human Lactation
We thank Joan Donner and Paul Everson (Ancell Corp.) for technical assistance with the ELISA assays. We thank Qinzi Ji (Carthage College) for assistance processing the milk samples. We thank Crystal Nelson (Save the Milk) for recommendations and guidance on shipping milk samples. No one was compensated for their contributions.
Author contribution(s): Andrea M. Henle: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing.
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Andrea M. Henle https://orcid.org/0000-0002-7383-7616
Supplemental Material: Supplementary material may be found in the “Supplemental Material” tab in the online version of this article.
==== Refs
References
Abu Raya B. Srugo I. Kessel A. Peterman M. Bader D. Peri R. Ashtamker N. Gonen R. Bamberger E . (2014). The induction of breast milk pertussis specific antibodies following gestational tetanus-diphtheria-acellular pertussis vaccination. Vaccine, 32 (43 ), 5632–5637. 10.1016/j.vaccine.2014.08.006 25148774
American Academy of Pediatrics. (2022). Children and COVID-19: State-level data report. http://www.aap.org/en/pages/2019-novel-coronavirus-covid-19-infections/children-and-covid-19-state-level-data-report/
Baird J. K. Jensen S. M. Urba W. J. Fox B. A. Baird J. R. (2021). SARS-CoV-2 antibodies detected in mother’s milk post-vaccination. Journal of Human Lactation, 37 (3 ), 492–498. 10.1177/08903344211030168 34297643
Biewenga J. Faber A. Pronk J. C. Haaijman J. J. (1986). Production and characterization of pepsin fragments of human IgA1 to determine domain-specificity of monoclonal anti-IgA antibodies. Immunology, 59 (1 ), 153–158. 10.1177/000456329102800311 3093370
Biewenga J. Stoop A. E. Baker H. E. Swart S. J. Nauta J. J. P. van Kamp G. J. van der Baan S. (1991). Nasal secretions from patients with polyps and healthy individuals, collected with a new aspiration system: Evaluation of total protein and immunoglobulin concentrations. Annals of Clinical Biochemistry, 28 (3 ), 260–266. 10.1177/000456329102800311 1872572
Dagan N. Barda N. Biron-Shental T. Makov-Assif M. Key C. Kohane I. S. Hernán M. A. Lipsitch M. Hernandez-Diaz S. Reis B. Y. Balicer R. D. (2021). Effectiveness of the BNT162b2 mRNA COVID-19 vaccine in pregnancy. Nature Medicine, 27 (10 ), 1693–1695. 10.1038/s41591-021-01490-8
De Schutter S. Maertens K. Baerts L. De Meester I. Van Damme P. Leuridan E . (2015). Quantification of vaccine-induced antipertussis toxin secretory IgA antibodies in breast milk: Comparison of different vaccination strategies in women. The Pediatric Infectious Disease Journal, 34 (6 ), e149–152. 10.1097/INF.0000000000000675
Fouda G. G. Yates N. L. Pollara J. Shen X. Overman G. R. Mahlokozera T. Wilks A. B. Kang H. H. Salazar-Gonzalez J. F. Salazar M. G. Kalilani L. Meshnick S. R. Hahn B. H. Shaw G. M. Lovingood R. V. Denny T. N. Haynes B. Letvin N. L. Ferrari G. , . . . Center for HIV/AIDS Vaccine Immunology. (2011). HIV-specific functional antibody responses in breast milk mirror those in plasma and are primarily mediated by IgG antibodies. Journal of Virology, 85 (18 ), 9555–9567. 10.1128/JVI.05174-11 21734046
Goldman A. S. Goldblum R. M. (1989). Immunoglobulins in human milk. In Atkinson S. A. Lonngerdal B. , Proteins and non-protein nitrogen in human milk. 43–51. CRC Press. 10.1201/9780367812805
Gray K. J. Bordt E. A. Atyeo C. Deriso E. Akinwunmi B. Young N. Baez A. M. Shook L. L. Cvrk D. James K. De Guzman R. Brigida S. Diouf K. Goldfarb I. Bebell L. M. Yonker L. M. Fasano A. Rabi S. A. Elovitz M. A. . . . Edlow A. G. (2021). Coronavirus disease 2019 vaccine response in pregnant and lactating women: A cohort study. American Journal of Obstetrics and Gynecology, 225 (3 ), 303.e1–303.e17. 10.1016/j.ajog.2021.03.023
Low J. M. Gu Y. Ng M. S. F. Amin Z. Lee L. Y. Ng Y. P. M. Shunmuganathan B. D. Niu Y. Gupta R. Tambyah P. A. MacAry P. A. Wang L. W. Zhong Y. (2021). Codominant IgG and IgA expression with minimal vaccine mRNA in milk of BNT162b2 vaccinees. NPJ Vaccines, 6 , 105. 10.1038/s41541-021-00370-z 34413319
Mazur N. I. Horsley N. M. Englund J. A. Nederend M. Magaret A. Kumar A. Jacobino S. R. de Haan C. A. M. Khatry S. K. LeClerq S. C. Steinhoff M. C. Tielsch J. M. Katz J. Graham B. S. Bont L. J. Leusen J. H. W. Chu H. Y. (2019). Breast Milk prefusion F Immunoglobulin G as a correlate of protection against respiratory syncytial virus acute respiratory illness. The Journal of Infectious Diseases, 219 (1 ), 59–67. 10.1093/infdis/jiy477 30107412
Meek J. Y. Noble L. (2022). Breastfeeding and the use of human milk. 150 (1 ): e2022057988. 10.1542/peds.2022-057988
Pace R. M. Williams J. E. Järvinen K. M. Belfort M. B. Pace C. D. W. Lackey K. A. Gogel A. C. Nguyen-Contant P. Kanagaiah P. Fitzgerald T. Ferri R. Young B. Rosen-Carole C. Diaz N. Meehan C. L. Caffé B. Sangster M. Y. Topham D. McGuire M. A. . . . McGuire M. K. (2021). Characterization of SARS-CoV-2 RNA, antibodies, and neutralizing capacity in milk produced by women with COVID-19. MBio, 12 (1 ), e03192–20. 10.1128/mBio.03192-20
Perl S. H. Uzan-Yulzari A. Klainer H. Asiskovich L. Youngster M. Rinott E. Youngster I. (2021). SARS-CoV-2-specific antibodies in breast milk after COVID-19 vaccination of breastfeeding women. JAMA, 325 (19 ), 2013–2014. 10.1001/jama.2021.5782 33843975
Sriraman N. K. Kellams A. (2016). Breastfeeding: What are the barriers? Why women struggle to achieve their goals. Journal of Women’s Health, 25 (7 ), 714–722. 10.1089/jwh.2014.5059
United States Centers for Disease Control and Prevention. (2020). COVID data tracker. https://covid.cdc.gov/covid-data-tracker
United States Centers for Disease Control and Prevention. (2021a) Coronavirus disease 2019. https://www.cdc.gov/media/releases/2021/p0924-booster-recommendations-.html
United States Centers for Disease Control and Prevention. (2021b). Facts about nationwide breastfeeding goals. https://www.cdc.gov/breastfeeding/data/facts.html
Valcarce V. Stafford L. S. Neu J. Cacho N. Parker L. Mueller M. Burchfield D. J. Li N. Larkin J. (2021). Detection of SARS-CoV-2-specific IgA in the human milk of COVID-19 vaccinated lactating health care workers. Breastfeeding Medicine, 16 (12 ), 1004–1009. 10.1089/bfm.2021.0122 34427487
World Health Organization. (2021). Infant and young child feeding. https://www.who.int/news-room/fact-sheets/detail/infant-and-young-child-feeding
Young B. E. Seppo A. E. Diaz N. Rosen-Carole C. Nowak-Wegrzyn A. Cruz Vasquez J. M. Ferri-Huerta R. Nguyen-Contant P. Fitzgerald T. Sangster M. Y. Topham D. J. Järvinen K. M. (2022). Association of human milk antibody induction, persistence, and neutralizing capacity with SARS-CoV-2 infection vs. mRNA vaccination. JAMA Pediatrics, 176 (2 ), 159–168. 10.1001/jamapediatrics.2021.4897 34757387
Yu J. Collier A.-R. Y. Rowe M. Mardas F. Ventura J. D. Wan H. Miller J. Powers O. Chung B. Siamatu M. Hachmann N. P. Surve N. Nampanya F. Chandrashekar A. Barouch D. H. (2022). Neutralization of the SARS-CoV-2 Omicron BA.1 and BA.2 variants. The New England Journal of Medicine, 386 (16 ), 1579–1580. 10.1056/NEJMc2201849 35294809
Zaman K. Roy E. Arifeen S. E. Rahman M. Raqib R. Wilson E. Omer S. B. Shahid N. S. Breiman R. F. Breiman R. E. Steinhoff M. C. (2008). Effectiveness of maternal influenza immunization in mothers and infants. The New England Journal of Medicine, 359 (15 ), 1555–1564. 10.1056/NEJMoa0708630 18799552
| 36398916 | PMC9726888 | NO-CC CODE | 2022-12-08 23:18:20 | no | J Hum Lact. 2022 Nov 18;:08903344221134631 | utf-8 | J Hum Lact | 2,022 | 10.1177/08903344221134631 | oa_other |
==== Front
JACC Case Rep
JACC Case Rep
JACC Case Reports
2666-0849
Published by Elsevier on behalf of the American College of Cardiology Foundation.
S2666-0849(22)00866-X
10.1016/j.jaccas.2022.101706
101706
Editorial Comment
Post COVID, Ergo Propter COVID?∗
Bhave Nicole M. MD ∗
Division of Cardiovascular Medicine, Department of Internal Medicine, University of Michigan and Michigan Medicine, Ann Arbor, Michigan, USA
∗ Address for correspondence: Dr Nicole M. Bhave, 2362 Cardiovascular Center, 1500 E Medical Center Dr, SPC 5853, Ann Arbor, Michigan 48109-5853, USA.
∗ Editorials published in JACC: Case Reports reflect the views of the authors and do not necessarily represent the views of JACC: Case Reports or the American College of Cardiology.
7 12 2022
18 1 2023
7 12 2022
6 101706101706
© 2022 The Author
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Corresponding Author
Key Words
autoimmune
pericardial effusion
supraventricular arrhythmias
==== Body
pmcNearly 3 years into the COVID-19 pandemic, more than 97 million cases have occurred in the United States, with more than 1 million associated deaths.1 Beyond the acute illness, COVID often has long-term impacts on patients’ functional status, economic and social productivity, and quality of life. Patients and clinicians have struggled with uncertainty regarding the natural history of long COVID, which can be highly variable, and have experienced frustration related to the limitations of current diagnostic methods and treatments.
In this issue of JACC: Case Reports, Safronenka et al2 describe the clinical course of a patient with recrudescence of frequent symptomatic premature ventricular contractions (PVCs) accompanied by a constellation of symptoms including chest pain and shortness of breath, 3 weeks after she received a second COVID-19 vaccine. She subsequently underwent 2 PVC ablations and later tested positive for SARS-CoV-2, with symptoms including chest pain and joint pain persisting for months afterwards. The presence of a pericardial effusion on echocardiography and improvement of symptoms with nonsteroidal anti-inflammatory drugs supported a diagnosis of pericarditis. Serologic testing was notable for a positive antinuclear antibody titer, with negative titers for other autoantibodies, including double-stranded DNA.
Long COVID, or postacute sequelae of SARS-CoV-2 (PASC), is a multisystem syndrome including diverse symptoms such as fatigue, chest pain, dyspnea, palpitations, musculoskeletal pain, and headache, persisting 4 or more weeks after SARS-CoV-2 infection.3 Given the heterogeneous nature of PASC, defining an individual patient’s specific phenotype based on a thorough clinical assessment is critically important for risk stratification and management. PASC cardiovascular disease refers to cardiovascular conditions including, but not limited to, myocarditis and other myocardial involvement, pericarditis, new or worsening myocardial ischemia due to epicardial or microvascular coronary artery disease, left ventricular or right ventricular dysfunction, and arrhythmias. PASC cardiovascular syndrome encompasses cardiovascular symptoms and findings such as exercise intolerance, postexertional malaise, orthostasis, and tachycardia, in the absence of clear cardiovascular pathology based on standard diagnostic tests.3
Initial workup for a patient with persistent cardiovascular symptoms after COVID-19 should include basic laboratory testing (including serum cardiac troponin, preferably a high-sensitivity assay), electrocardiography, transthoracic echocardiography, and possibly ambulatory rhythm monitoring. The clinician should pursue further testing selectively, based on results of the initial evaluation and the patient’s clinical course. Particularly for a patient with frequent ventricular ectopy, cardiac magnetic resonance imaging is prudent to assess for myocarditis and other forms of myocardial involvement.
Myocarditis associated with COVID-19 mRNA vaccines is rare, occurs most often in male adolescents and young adults, and usually manifests with symptoms several days after the second vaccine dose.3 , 4 The association between mRNA vaccines and pericarditis is less clear.5 The patient described by Safronenka et al2 did not meet diagnostic criteria for myocarditis, and pericarditis was not diagnosed until months after vaccination.2 Therefore, it is difficult to know precisely how the vaccination affected her individual clinical course.
However, on a population level, SARS-CoV-2 infection is clearly associated with an increase in incident cardiovascular disease. In a Veterans Affairs cohort of 153,760 individuals with COVID-19, new diagnoses of multiple cardiovascular disorders were more common, from 1 to 12 months after the illness, than in control subjects.6 New diagnoses of atrial fibrillation and heart failure were particularly common in this cohort (excess burden per 1,000 persons: 10.74 and 11.61, respectively). The proinflammatory and hyperadrenergic milieu such as that seen in SARS-CoV-2 infection can trigger atrial fibrillation and heart failure as acute or subacute manifestations of underlying, previously undiagnosed, structural heart disease. On a related note, patients with historically evident arrhythmias, like the patient in this case report, could plausibly become more symptomatic as a result of cardiovascular deconditioning after SARS-CoV-2 infection. Mechanisms underlying palpitations, a frequent component of PASC, may include decreased blood volume, cardiac atrophy, and decreased stroke volume, accompanied by compensatory tachycardia.3
The associations between SARS-CoV-2 infection and autoimmunity are under active investigation. Autoantibodies, including antinuclear and antiphospholipid antibodies, have been documented in patients with SARS-CoV-2 infection, suggesting that the virus can trigger autoimmune responses via molecular mimicry.7 The long-term impact of SARS-CoV-2 infection on autoimmune serologies deserves further study. Moreover, it remains unclear whether vaccines or antiviral pharmacotherapies (namely, nirmatrelvir-ritonavir) affect the probability of autoimmune phenomena following SARS-CoV-2 infection.
As we continue to grapple with COVID-19 and its sequelae, longitudinal studies such as the National Institute of Health’s Researching COVID to Enhance Recovery (RECOVER) will help to clarify disease mechanisms, identify prognostic indicators, and evaluate efficacy of treatments. Especially as new variants emerge and as the population’s immunity waxes and wanes with infections and booster vaccines, we will need repeated population-based studies to assess attributable risks of cardiovascular disorders. Post COVID, ergo propter COVID? For an individual patient, it may be impossible to answer this question with certainty. But with high-quality epidemiologic data, we as clinicians will be better equipped to educate and care for our patients.
Funding Support and Author Disclosures
The author has reported that she has no relationships relevant to the contents of this paper to disclose.
The author attests they are in compliance with human studies committees and animal welfare regulations of the author’s institution and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
==== Refs
References
1 Centers for Disease Control and Prevention COVID data tracker https://covid.cdc.gov/covid-data-tracker
2 Safronenka A. Capcha J.M.C. Webster K.A. Autoimmune reaction associated with long COVID syndrome and cardiovascular disease: a genetic case report J Am Coll Cardiol Case Rep 5 2023 XXX
3 Gluckman T.J. Bhave N.M. Allen L.A. 2022 ACC expert consensus decision pathway on cardiovascular sequelae of COVID-19 in adults: myocarditis and other myocardial involvement, post-acute sequelae of SARS-CoV-2 infection, and return to play J Am Coll Cardiol 79 2022 1717 1756 35307156
4 Witberg G. Barda N. Hoss S. Myocarditis after COVID-19 vaccination in a large health care organization N Engl J Med 385 2021 2132 2139 34614329
5 Patone M. Mei X.W. Handunnetthi L. Risks of myocarditis, pericarditis, and cardiac arrhythmias associated with COVID-19 vaccination or SARS-CoV-2 infection Nat Med 28 2022 410 422 34907393
6 Xie Y. Xu E. Bowe B. Al-Aly Z. Long-term cardiovascular outcomes of COVID-19 Nat Med 28 2022 583 590 35132265
7 Liu Y. Sawalha A.H. Lu Q. COVID-19 and autoimmune diseases Curr Opin Rheumatol 33 2021 155 162 33332890
| 36505732 | PMC9727378 | NO-CC CODE | 2022-12-08 23:18:27 | no | JACC Case Rep. 2023 Jan 18; 6:101706 | utf-8 | JACC Case Rep | 2,022 | 10.1016/j.jaccas.2022.101706 | oa_other |
==== Front
Me´decine De Catastrophe, Urgences Collectives
1279-8479
1279-8479
Société Française de Médecine de Catastrophe. Published by Elsevier Masson SAS.
S1279-8479(22)00294-4
10.1016/j.pxur.2022.10.008
NO-FEAR project
The results of a Foresight exercise: Outcome from a NO-FEAR internal evaluation
Les résultats d’un exercice de prospective : conclusion d’une évaluation interne de NO-FEARVoicescu George Teo abc⁎
Linty Monica ac
Ler Lian-Guey e
Kaufman Stefan f
Corte Francesco Della ad
a CRIMEDIM (Center for Research and Training in Disaster Medicine, Humanitarian Aid and Global Health), Università del Piemonte Orientale, 28100 Novara, Italy
b Iuliu Hatieganu. University of Medicine and Pharmacy, 8 Babes str., 400012 Cluj-Napoca, Romania
c Department for Sustainable Development and Ecological Transition, Università del Piemonte Orientale, 13100 Vercelli, Italy
d Department of Translational Medicine, Università del Piemonte Orientale, 28100 Novara, Italy
e Université Côte d’Azur, Polytech Nice Sophia, 930, Route des Colles, 06903 Sophia, France
f Institute of Sociology, Albert-Ludwigs-Universität Freiburg, Rempartstr. 15, 79098 Freiburg, Germany
⁎ Corresponding author at: CRIMEDIM (Center for Research and Training in Disaster Medicine, Humanitarian Aid and Global Health), Università del Piemonte Orientale, 28100 Novara, Italy.
7 12 2022
12 2022
7 12 2022
6 4 247251
© 2022 Société Française de Médecine de Catastrophe. Published by Elsevier Masson SAS. All rights reserved.
2022
Société Française de Médecine de Catastrophe
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The challenges that the health systems face in the last years increased exponentially. No matter if we are talking about the impact of the COVID-19 pandemic or the Russian military action in Ukraine, the European health ecosystem is facing new problems. In the light of these uncertainties, we assessed which could be the next trends that can impact the healthcare systems, in order to better prepare and adapt to the new contexts.
Using two foresights exercises (FSE), one in 2018 and the second one in 2022, we had identified the most important trends in the political, economic, social, technological, security, environmental and medical sectors that could have an impact on health.
Fifty-three people participated in the first FSE and 40 in the second one. The respondents identified cyber security, an increased reliance on digital technologies for communications, CBRNE management of the patients, centrally coordinated attacks, demographic aging, reduced economic resources, violence against emergency medical staff and the increased need and demand for psychosocial support as the most important trends. Moreover, they considered that wars, hybrid threats, the fake news, pandemics and the influence of artificial intelligence could impact the healthcare systems.
Many of the trends identified in 2018 as having a possible impact on the health system proved to be relevant four years later. Therefore, we consider the FSE a relevant tool in foreseeing the main areas that could have an impact on health and its results could guide the preparedness for the future.
Résumé
Les défis auxquels les systèmes de santé se sont confrontés pendant ces dernières années ont augmenté de façon exponentielle. Qu’il s’agisse de l’impact de la pandémie de COVID-19 ou de l’action militaire russe en Ukraine, l’écosystème européen de la santé est confronté à de nouveaux problèmes. Au regard de ces incertitudes, nous avons évalué quelles pourraient être les prochaines tendances pouvant impacter les systèmes de santé, afin de mieux se préparer et s’adapter aux nouveaux contextes.
À l’aide de deux exercices de prospective (FSE), l’un en 2018 et le second en 2022, nous avons identifié les tendances les plus importantes dans les secteurs politique, économique, social, technologique, sécuritaire, environnemental et médical qui pourraient avoir un impact sur la santé.
Cinquante trois personnes ont participé au premier FSE et 40 au second. Le répondants ont identifié la cybersécurité, la dépendance aux technologies digitales pour les communications, la gestion des patients NRBCE, les attaques coordonnées de manière centralisée, le vieillissement démographique, la réduction des ressources économiques, la violence contre le personnel médical d’urgence et le besoin et la demande accrus de soutien psychosocial comme les tendances les plus importantes. De plus, ils ont considéré que les guerres, les menaces hybrides, les informations fallacieuses, les pandémies et l’influence de l’intelligence artificielle pourraient avoir un impact sur les systèmes de santé.
Bon nombre des tendances identifiées en 2018 comme ayant un impact possible sur le système de santé se sont avérées pertinentes quatre ans plus tard. Par conséquent, nous considérons le FSE comme un outil pertinent pour prévoir les principaux domaines qui pourraient avoir un impact sur la santé et ses résultats pourraient guider la préparation pour l’avenir.
Keywords
Disaster Medicine
Risk Assessment
Forecasting
Crisis Management
Mots clés
Médecine de catastrophe
Évaluation des risques
Prévision
Gestion de crise
==== Body
pmcIntroduction
The healthcare systems around the world are defined by constant change, striving to adapt themselves to the challenges and struggles that emerge. Therefore, regardless of their professional level, healthcare workers need to adapt to the latest society demands. An example strengthening this believe is the COVID-19 pandemic, that proved us that no healthcare system is immune to what is happening around the world and no country is safe in front of a global threat. In the context of the pandemic, the adaptation to the new situation was identified as being one of the main struggles during the response to the pandemic [1], [2].
In the light of this constant demand for change and adaptation, the assessments of how the main trends may change in the future and which scenario approaches are the most suitable to apply could prove useful. A better understanding of the future could help us prepare and adapt better to the new contexts. One method of identifying the most probable evolvement of the existing trends is a foresight exercise (FSE).
A FSE represents a participatory, systematic intelligence gathering exercise, in an effort to understand better what the future could bring. By having the goal of enabling the present decisions towards joint actions, it doesn’t aim to predict the future, but to prepare for the different possibilities that could happen. Nevertheless, it strives to develop a common understanding of what should happen if threats are to be mitigated and opportunities seized [3]. Besides thinking about the future, during a FSE, the participant also debate the future and try to find ways to shape it.
The Network Of practitioners For Emergency medical systems and critical care (NO-FEAR) project is a five-year European funded project that aims to create an active Pan-European network of practitioners, decision and policy makers, suppliers and academia in the security field, sharing knowledge, experience and necessities, thus overcoming the current state of overwhelming fragmentation.
The purpose of the present study was to challenge the members of the NO-FEAR project to identify the most important trends and their evolution in time, in the field of disaster and emergency medicine. Moreover, we analyzed how the identified trends changed over the COVID-19 pandemic.
Methods
During the NO-FEAR project, we ran two FSE. The first one was running in 2018 and the second one 4 years later, in 2022.
In November 2018, using on guided discussion, the participants identified the most important trends in the political, economic, social, technological, security, environmental and medical clusters (PEST-SEM). The PEST analysis approach originates from the business sector, where it is widely used to help managers and organizations scan and predict the external factors impacting an organization [4]. We have added to the PEST analysis the security, environmental and medical (SEM) clusters to personalize the tool fitting our project's objective.
In March 2022, for the second FSE, we recategorized the trends identified in 2018 into four clusters: technological, security, social and health (Table I ). We used an online questionnaire, distributed among the NO-FEAR Participants using the platform provided by the European Commission “EUSurvey” (https://ec.europa.eu/eusurvey/) to reanalyze the participants perception on the trend's relevance, by asking them to rank the trends on a six points scale, ranging from totally irrelevant to totally relevant.Table I The clusters and trends used in the second FSE.
Table ICluster Identified Trends
Technological Cluster Access to medical data
Artificial intelligence (AI) applications on big data
Cyber Security
Digital transformation focused on operations
Impact of social media on quantity, reliability and speed of information
Impact of technological support (5G): dangers and opportunities
Increased reliance on digital technologies for communications
Security cluster CBRNE management of patients
CBRNE management of threat
Centrally coordinated attacks
Individual coordinated attacks
Healthcare impact of global climate change
Social cluster Culture, tradition and religious factors
Demographic aging
Generation discrepancy
Instabilities of governments
Reduced economic resources related to health
Health cluster Higher frequency of mass casualty incidents
Increased need and demand for psychosocial support
Less qualified health assistance
Nursing leadership
Violence against emergency medical staff
Besides analyzing their relevance, we have asked the participants to mention the 3 most important trends in each of the four clusters. Moreover, we allowed the responders to mention any additional trend(s) that they deem relevant, that were not identified during the first FSE in 2018.
Results
First foresight exercise
A total of 53 people participated in the first FSE. The most important trends identified during the 2018 FSE are presented in Table II . When the trends were split according to the PEST-SEM cluster classification, social, technological and security clusters where highly identified in terms of primary positioning, while the political, economic, and environmental clusters were not as highly represented. Moreover, the medical cluster was not represented as much as the social, technological and security ones.Table II Trends identified during the first FSE in 2018.
Table IITrends in Consequence Trees
1 112 single digit emergency number
2 5G
3 Artificial Intelligence
4 Automated mobile universal patient monitoring
5 CBRNE
6 CBRNE management of wounded patients
7 Environmental issues: Climate change
8 Common procedures shared between different agencies
9 Demography aging
10 Development of human and material resources for mass causality
11 Digital transformation focused on human and operators
12 Education and training in emergency medicine
13 Education and training in mass casualty incidents
14 Growth of Artificial intelligence applications
15 Health economics
16 Healthy aging
17 Immersive wearable
18 Increased need and demand for psychological support
19 Individual initiative attacks (vs. centrally coordinated attacks)
20 Instabilities of governments
21 International (EU) operational organization (command and control practices)
22 Less qualified assistance: doctor/paramedic behavior
23 More frequent large-scale incidents
24 More need for operational integration
25 New technologies/big data
26 Professionalization of emergency and disaster medicine
27 Reduced economical availability
28 Remote monitoring
29 Restructuring mass casualty Incidents plans (to scoop and scooter)
30 Safety treat
31 Social media and speed of information dangers and opportunity
32 Standard operating procedures
33 Technological support
34 Technology intake/enabler
35 Training professional to mass causality incidents
36 Violence against medical personnel
37 Vulnerability of integrated system
Second foresight exercise
We received a total of 40 responses to the survey. 18 responders were emergency and crisis managers, eight were practitioners and policy makers, while 14 were suppliers and academia representatives.
When asked to reconsider the importance of the trends identified in 2018 for the technological cluster, cyber security and an increased reliance on digital technologies for communications were considered the two most relevant ones. Regarding the security cluster, the CBRNE management of the patients and the centrally coordinated attacks deemed as the most relevant trends. For the social cluster, demographic aging and the reduced economic resources related to health were the trends consider having the most relevance. Meanwhile, for the health cluster violence against emergency medical staff and the increased need and demand for psychosocial support were ranked as the most relevant trends.
When asked to rank the trends according to their importance, for the technological cluster, cyber security was considered the most important by 39% of the participants, while 29% considered the access to medical data as the most important trend. Regarding the security cluster, 29% of the responders ranked the CBRNE management of patients as being the most important one, while 26% considered the CBRNE management of the threat as the first most important.
In the social cluster, 47% of the participants ranked demographic aging as being the most important, while 32% ranked the reduced economic resources as the most important trend. Finally, in the health cluster, the increased need and demand for psychosocial support was deemed as the most important trend by 37% of the responders, while the less qualified health assistance was ranked by 20% the most important trend.
Besides the trends already identified in the first FSE, the responders identified additional ones. In the technology cluster, the impact of drones, the availability of appropriate staff and the increased number of interconnected medical devices were considered relevant. In the security cluster, wars, hybrid threats and the impact of drones were identified as relevant, while in the social cluster, the fake news, wars and cybersecurity risks in the physical world were mentioned as very important. The participants also identified the mobility of people, pandemics and the influence of artificial intelligence in decision making as relevant for the health cluster.
Discussion
We successfully used two foresight exercises to understand the future trends in emergency and disaster medicine. Surprisingly, infectious outbreaks were not considered an important trend in 2018, although the COVID-19 pandemic proved us wrong.
During the first FSE, healthcare workers safety was identified as very relevant for future: “safety threat”, “violence against medical personnel”, “security” trends being identified as very important. The high number of attacks on healthcare workers led to the development of a number of initiatives e.g. healthcare in danger, attacks on health care initiative etc., emphasizing the high international attention given to these trends [5], [6], [7].
New technologies are arising fast, impacting the medical sector without consideration for professional level or specialty [8], [9]. During the first FSE, high importance was given to “5G”, “e-health”, “social media, digital transformation” and “artificial intelligence”, showcasing that the impact of technology development on the medical field was one of the main interests of the participants. This is in accordance with the existing literature, emphasizing people's natural interest towards novelty, but also people's skepticism regarding emerging technologies [10], [11].
The overwhelming benefits of the new technologies in medicine comes with additional problems. Among them, the “cyber security” and “big data management” were tendencies identified as important during the first FSE.
The different healthcare systems structural problems, e.g. “reduced economical availability”, “less qualified assistance: doctor/paramedic behavior” were rated as important trends during the first FSE. These tendencies were confirmed, the personnel shortage forcing less qualified staff to care for patients [12].
During the second FSE, performed three years later, compared with the first FSE, people's perception about trend's relevance remained similar. Cyber security was once again considered the most important trend in the technology cluster, following the increasing number of cyber security attacks that happened in the last years [13]. Compared with the first FSE, during the second one, “violence against emergency healthcare workers” was still considered very relevant, together with the “increased need and demand for psychosocial support”. One of the reasons of ranking the psychosocial support as very important could be the high impact that the COVID-19 pandemic had on mental health. Considering that both patient's and healthcare worker's mental wellbeing were severely affected by the pandemic, a further increase in psychosocial support is expected in the future [14].
Interestingly, during the second FSE, “demographic aging” was rated as the most important trend in the social cluster, together with reduced economic resources. It is known that the proportion of elderly is increasing, with longer life expectancy and lower birth rates. This population shift already has an impact on the healthcare systems, and this impact is expected to increase, considering that the proportion of the world's population over 60 years is expected to double until 2050 [15].
In the security cluster, the “CBRNE management of patients” and threats were identified as the main trends. This finding could be linked with the Russian military action in Ukraine, since this European conflict reignited the fear of CBRNE threats [16].
Additional to the trends identified in 2018, during the second FSE the participants considered “the impact of drones” as very important in the field of emergency and disaster medicine. In the last years, drones have been used for the delivery of medical products, for biological hazards surveillance, for telemedicine etc. [17]. Their versatility and reliance make them important tools for the future, their usage being expected to increase.
Following the COVID-19 pandemic both governments and scholars emphasized the negative impact that fake news had on the public health management of the outbreak [18]. Therefore, “fake news” was considered very relevant by the participants to the second FSE, confirming the high impact of false news on the medical field, perceived by the participants to our study.
Naturally, after the Russian military action in Ukraine, the responders identified wars as being a relevant trend for the future. Interestingly, while infectious outbreaks and the high mobility of people were not mentioned as important trends in 2018, during the second FSE the participants rated both as very important for the future.
Conclusion
Using the network created by the NO-FEAR project, we have run two FSE to identify the trends that will impact the future of emergency and disaster medicine. Although some important trends were not foreseen, e.g. pandemic, the vast majority of those identified by the first FSE proved relevant 4 years later, while the second FSE brought forward additional tendencies.
Disclosure of interest
The authors declare that they have no competing interest.
Funding
The NO-FEAR project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 786670
==== Refs
References
1 Pearson C. Britt H. Schapiro L. Adapting Health Care Delivery in Response to COVID-19: International Lessons for the United States 2021 Commonwealth Fund 10.26099/nr66-tn77
2 Yaffee A.Q. Peacock E. Seitz R. Hughes G. Haun P. Ross M. Preparedness, Adaptation, and Innovation: Approach to the COVID-19 Pandemic at a Decentralized, Quaternary Care Department of Emergency Medicine West J Emerg Med 21 6 2020 63 70 33052812
3 Masum H. Ranck J. Singer P.A. Five promising methods for health foresight Foresight 12 1 2010 54 66 10.1108/14636681011020182
4 AHRQ. Political, Economic, Social, and Technological Forces Analysis | Digital Healthcare Research. https://digital.ahrq.gov/health-it-tools-and-resources/evaluation-resources/workflow-assessment-health-it-toolkit/all-workflow-tools/pest-analysis.
5 SHCC. More Than 4000 Attacks Against Health Workers, Facilities, and Transports Since 2016 Underscore Need for Action to Protect Health Care in Conflict - World | ReliefWeb. https://reliefweb.int/report/world/more-4000-attacks-against-health-workers-facilities-and-transports-2016-underscore-need.
6 WHO. Stopping attacks on health care. https://www.who.int/activities/stopping-attacks-on-health-care.
7 Healthcare in Danger Initiative. ICRC Health Care In Danger 2015 https://healthcareindanger.org/
8 Ahuja A.S. The impact of artificial intelligence in medicine on the future role of the physician PeerJ 7 2019 e7702 10.7717/peerj.7702 31592346
9 The future of healthcare technology: How is medicine changing? 2021 FutureLearn https://www.futurelearn.com/info/blog/future-of-healthcare-technology
10 Ball W. Holland S. The fear of new technology: a naturally occurring phenomenon Am J Bioeth 9 1 2009 14 16 10.1080/15265160802617977
11 Why many doctors & patients are scared of new health technologies 2022 TFOT https://thefutureofthings.com/16753-why-many-doctors-patients-are-scared-of-new-health-technologies/
12 Allkins S. ‘Cause for concern’ as nursing shortages force less qualified staff to step up for patient care 2019 IndependentNurse https://www.independentnurse.co.uk/news/cause-for-concern-as-nursing-shortages-force-less-qualified-staff-to-step-up-for-patient-care/221956
13 Landi H. Healthcare data breaches hit all-time high in 2021, impacting 45M people 2022 Fierce Healthcare https://www.fiercehealthcare.com/health-tech/healthcare-data-breaches-hit-all-time-high-2021-impacting-45m-people
14 Talevi D. Socci V. Carai M. Carnaghi G. Faleri S. Trebbi E. Mental health outcomes of the CoViD-19 pandemic Rivista di Psichiatria 55 3 2020 137 144 10.1708/3382.33569 32489190
15 WHO. Ageing and health [Internet]. WHO October 2021. https://www.who.int/news-room/fact-sheets/detail/ageing-and-health.
16 Putin puts Russia's nuclear deterrent forces on alert 2022 Aljazeera https://www.aljazeera.com/news/2022/2/27/putin-puts-russias-nuclear-deterrent-forces-on-alert
17 Johnson A.M. Cunningham C.J. Arnold E. Rosamond W.D. Zègre-Hemsey J.K. Impact of using drones in emergency medicine: what does the future hold? Open Access Emerg Med 13 2021 487 498 10.2147/OAEM.S247020 34815722
18 van der Linden S. Roozenbeek J. Compton J. Inoculating against fake news about COVID-19 Front Psychol 11 2020 566790 10.3389/fpsyg.2020.566790 33192844
| 0 | PMC9727550 | NO-CC CODE | 2022-12-08 23:18:52 | no | 2022 Dec 7; 6(4):247-251 | utf-8 | null | null | null | oa_other |
==== Front
Rev Neurol (Paris)
Rev Neurol (Paris)
Revue Neurologique
0035-3787
0035-3787
Elsevier Masson SAS.
S0035-3787(22)00824-4
10.1016/j.neurol.2022.11.003
Original Article
Serology results after COVID vaccine in multiple sclerosis patients treated with fingolimod
Ciccone A. a
Mathey G. abc
Prunis C. a
Debouverie M. abc*
a Service de Neurologie, CHRU-Nancy, Université de Lorraine, 54000 Nancy, France
b Inserm, CIC-1433 Épidemiologie Clinique, CHRU-Nancy, Université de Lorraine, 54000 Nancy, France
c EA 4360 APEMAC, Université de Lorraine, 54000 Nancy, France
⁎ Corresponding author.
7 12 2022
7 12 2022
12 6 2022
4 11 2022
17 11 2022
© 2022 Elsevier Masson SAS. All rights reserved.
2022
Elsevier Masson SAS
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Introduction
While it is recommended that patients with multiple sclerosis (MS) be vaccinated against COVID-19, it is unknown what the vaccine response is in MS patients treated with fingolimod, an agent which modulates the humoral response. We aimed to characterize the immune response to the COVID-19 vaccine in MS patients treated with fingolimod and to explore which factors influenced response.
Method
We collected the following data from 59 MS patients treated with fingolimod and vaccinated against COVID-19: age, sex, duration of treatment, number of vaccine doses, date of last vaccination, type of vaccine, lymphocyte count, history of COVID-19, and serology to measure the vaccine response. We used Student's t-test and Chi2 test to see whether there was a relationship between these variables and seropositivity. A multivariate logistic regression model was used to identify factors influencing the serology result. A multivariate linear regression model was used to identify factors influencing the antibody titer.
Results
Twenty-eight participants (47%) developed a positive serology. Age (P < 0.001) and the duration of treatment (P = 0.002) were significantly related to seropositivity. Gender (P = 0.73), number of vaccinations (P = 0.78), lymphocyte count (P = 0.46), and the time between the last vaccine dose and blood sampling (P = 0.84) were not significant variables. Multivariate analysis using logistic regression (n = 59) showed that age (P = 0.003, RR = 2.28, 95%CI = 1.28, 4.07) and duration of treatment (P = 0.04, RR = 1.91, 95%CI = 1.04, 3.50) were significantly and independently correlated with COVID serology. Multivariate linear regression analysis of the antibody titer (n = 59) found the duration of treatment to be significant (P = 0.015), but not age (P = 0.53). After removing three outliers, age (P = 0.005, RR = 6.82, 95%CI = 1.66, 27.98) and duration of treatment (P = 0.008, RR = 5.12, 95%CI = 1.24, 21.03) were significantly correlated with the antibody titer.
Conclusion
COVID-19 seropositivity was present in 47% of our sample of 59 MS patients on fingolimod. A strong relationship was found between antibody development, age, and duration of treatment, as well as between antibody titer and age and duration of treatment.
Keywords
Multiple sclerosis
Fingolimod
COVID-19
Serology
==== Body
pmc1 Introduction
Since December 2019, the world has experienced a pandemic following the outbreak of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1]. Several types of vaccines are currently available including mRNA vaccines (Moderna and Pfizer), and viral vector vaccines (Johnson&Johnson and Astrazeneca) [2]. These vaccines have already proven their efficacy and safety in the general population [3], [4], [5].
Several studies have already shown that multiple sclerosis (MS) is not a contraindication to vaccination (except for live attenuated vaccines which are contraindicated for patients on immunosuppressive therapies), and that patients vaccinated against COVID-19 do not experience increased disease activity. Consequently, MS patients are advised to be vaccinated early as they are often immunocompromised and therefore constitute a population at risk [6], [7].
MS is a chronic, autoimmune, inflammatory disease of the central nervous system in which the immune system attacks the patient's myelin sheath. Patients are administered immunosuppressive treatments to reduce this autoimmunity and limit relapses.
Fingolimod, an orally administered sphingosine 1 phosphate (S1P) receptor inhibitor, is one of the background treatments for patients with MS. This class of drugs works by blocking the migratory signal of S1P thus preventing the lymphocytes in lymphoid tissues from moving to peripheral tissues where they destroy the myelin sheath. The lymphocytes remain sequestered in the lymph nodes and are not altered [8], [9], [10], [11].
It has already been shown that, unlike other immunosuppressive background treatments, fingolimod does not increase the risk of developing severe COVID (requiring hospitalization) compared with the general population [12], [13], [14], [15], [16], [17]. Furthermore, those patients who do develop severe disease have similar risk factors to those of the general population of patients at risk [18]. This means that patients currently on fingolimod can continue their treatment during the pandemic which is reassuring as there is a risk of rebound after stopping fingolimod [19]. Background immunosuppressive treatment should always be introduced on a case-by-case basis, weighing up the benefits of each treatment on MS and the potential risk of severe COVID that the treatment could cause [20]. In terms of vaccine efficacy, disease-modifying therapies for MS can modulate the vaccine response. S1P modulators, in particular, can produce attenuated vaccine responses [21]. A randomised trial of vaccination in fingolimod-treated patients with MS showed that the immune response was lower in patients on fingolimod than in the control groups (placebo). It is therefore advisable to check the humoral response with serology [22].
The aim of the current study was to describe the qualitative (development of antibodies) and quantitative (antibody titer) immune response of the COVID-19 vaccine in MS patients treated with fingolimod, and to identify factors influencing response.
2 Materials and methods
2.1 Participants and study design
We included all patients with MS who consulted at the Neurology Department of Nancy University Hospital between 15/08/2021 and 09/12/2021 and who met the following criteria: patients vaccinated against COVID-19; patients treated by fingolimod.
For each patient, we collected the following data: age, sex, duration of fingolimod treatment, number of vaccine doses, date of last vaccination, type of vaccine, lymphocyte count, history of COVID-19, and the results and date of the serology showing the patient's vaccine response.
Antibodies were detected by DiaSorin's liaison SARS-COV2 trimericS IgG control set in a dry tube with gel separator. Lymphocytes were measured by complete blood count (CBC). Patients with BAU/mL values over 33.80 (threshold) were considered seropositive. The LIAISON® SARS-CoV-2 TrimericS IgG assay uses chemiluminescence immunoassay technology (CLIA) for the quantitative determination of specific IgG antibodies to the SARS-CoV-2 trimeric Spike protein in human serum and plasma samples. The main components of the test are magnetic particles (solid phase) coated with recombinant SARS-CoV-2 trimeric Spike protein and a conjugate reagent containing a mouse monoclonal anti-human IgG antibody linked to an isoluminol derivative (isoluminol-antibody conjugate). The amount of isoluminol-antibody conjugate is translated into a light signal and is measured by a photomultiplier as relative light units (RLU), which indicates the presence of anti-SARS-CoV-2 antibodies in standards, samples and controls.
2.2 Statistical analysis
Quantitative variables (age, duration of treatment, lymphocyte count, time between last vaccination and blood sampling, antibody titer) are described as means and standard deviations after checking the normality of the variables. Categorical variables (gender, number of doses) are described as numbers and proportions (Table 1 ).Table 1 Descriptive analyses (n = 59).
Table 1 Min Q1 Median Mean Q3 Max Standard deviation
Age (years) 19 34 44 43.29 54 65 11.84
Length of treatment (month) 6 20 44 50.63 82 126 33.13
Lymphocyte count (G/L) 0.2 0.38 0.479 0.51 0.61 1.419 0.19
Time between serology and last dose (month) 0 1 3 2.73 4 8 2.07
Antibodies titer (BAU/mL) 4.81 5.68 30.10 214.44 218 2080 472.58
A Student's t-test was used after checking the normality of the variables to compare age, duration of treatment, lymphocyte count, and time between the last vaccination and blood sampling between the group that developed antibodies and the group that did not. A Chi2 test was used to check whether there was a relationship between seropositivity and age, gender, or the number of doses received. A multivariate logistic regression model was used to identify factors influencing the serology result. The variable to be explained was antibody development and the explanatory variables were age and duration of treatment. A multivariate linear regression model was used to identify factors influencing the antibody titer. The variable to be explained was the antibody titer, the explanatory variables were the duration of treatment and age. We defined age (40 years) and treatment duration (4 years), thresholds, according to literature data, that correspond to the median. We only kept variables with a P < 0.10 in the multivariate model.
Statistical analyses were performed using Excel and Rstudio software.
2.3 Ethics
Patients were identified through the “Registre Lorrain de la Sclérose en Plaques” (RelSEP). All the patients gave their informed consent. Data collection was approved by the French National Commission for Data Protection and Liberties (CNIL No. 913001-2014.01.06).
3 Results
Of the 368 patients with MS who consulted at our Neurology Department during the study period, 59 patients met the inclusion criteria (43 women (73%) and 16 men (27%)) with an average age of 43 years (SD = 12). Thirty-three people (56%) had received two doses of vaccine, 23 (39%) three doses, and three (5%) one dose, two of whom had active COVID-19 infection at the time of sampling. Eight participants (13.6%) had a history of COVID infection. None of these eight patients required hospitalisation. Overall, 28 participants (47%) were seropositive.
Age was significantly the main explanatory factor for seropositivity (P < 0.001) (Table 2 ). We modelled the proportion of COVID seropositivity in four age groups: under 30, between 30 and 39, between 40 and 50 and over 50. Seropositivity was found in eight of the nine patients under 30; eight of the 13 patients between 30 and 39; eight of the 17 patients between 40 and 50; and four of the 20 patients over 50 (Fig. 1 ).Table 2 Characteristics of patients with a positive and a negative serology.
Table 2 Positive serology Negative serology P-value
N n = 28 n = 31 /
Sex (W/M) 21/7 22/9 P = 0.73
Age, years mean (SD) 37.07 (10.83) 48.90 (9.72) P < 0.001
Duration of treatment, month mean (SD) 37.04 (27.56) 62.90 (32.94) P = 0.002
Lymphocyte count, G/L mean (SD°) 0.486 (0.148) 0.524 (0.223) P = 0.46
Time between last dose and blood sampling, month mean (SD) 3 (2.01) 2.68 (2.12) P = 0.84
Fig. 1 Proportion by age group of COVID seropositivity. Global P-value = 0.004.
Gender (P = 0.73), number of vaccinations (P = 0.78), lymphocyte count (P = 0.46), and the time between the last vaccine dose and blood sampling (P = 0.84) were not significant variables (Table 2). We added a univariate analysis to see if there was an association between any of these variables and the serology results, for example, gender, but we did not find a significant relationship (P = 0.728).
The duration of fingolimod treatment was significantly related to COVID seropositivity (P = 0.002) (Table 2). We modelled the proportion of COVID seropositivity in four treatment duration groups: less than 22 months, between 22 and 44 months, between 45 and 80 months, more than 80 months. Each group contained, respectively: 15 patients (of whom 11 were seropositive), 15 patients (of whom eight were seropositive), 14 patients (of whom six were seropositive), and 15 patients (of whom three were seropositive) (Fig. 2 ).Fig. 2 Proportion of COVID seropositivity for a given treatment du ration. Global P-value = 0.03.
In the multivariate logistic regression analysis (n = 59), age (P = 0.003) and duration of treatment (P = 0.04) emerged as being significantly and independently correlated with COVID seropositivity (Table 3 ). Thus, the higher the age and the longer the duration of treatment, the greater the risk of having a negative serology.Table 3 Multivariate analysis (logistic regression): development of antibodies according to age and duration of treatment.
Table 3 RR (95% CI) P-value
Age > 40 years 2.28 (1.28, 4.07) 0.003
Duration of Fingolimod treatment > 48 months 1.91 (1.04, 3.50) 0.04
The risk of being seronegative was 2.28 times greater after 40 years of age and 1.91 times greater after 48 months of treatment.
Multivariate linear regression analysis (n = 59) showed a positive significance of the duration of treatment (P = 0.015) but not age (P = 0.53). We repeated this analysis by excluding three patients aged 46, 50 and 32 years (two of whom had had COVID) with an antibody titer greater than 2080 BAU/mL, which could bias the result. In the multivariate linear regression analysis with the remaining 56 patients, age (P = 0.005) and duration of treatment (P = 0.008) emerged as being significantly correlated with the antibody titer (Table 4, Table 5 ).Table 4 Descriptive analyses (n = 56).
Table 4 Min Q1 Median Mean Q3 Max Standard deviation
Age (years) 19 34.5 43 43.32 54 65 12.02
Duration of treatment (months) 6 25.5 44.5 52.14 82 126 33.04
Antibodies titer (BAU/mL) 4.81 5.41 23.35 114,53 110 838 197.16
Table 5 Multivariate analysis (linear regression): antibody titer according to age and duration of treatment.
Table 5 RR (95% CI) P-value
Age > 40 years 6.82 (1.66, 27.98) 0.005
Duration of Fingolimod treatment > 48 months 5.12 (1.24, 21.03) 0.008
The risk of having an antibody level < 114 BAU/mL was 6.82 times greater after 40 years and 5.12 times greater after 48 months of treatment.
Concerning COVID history, we added a univariate analysis to see if there was a possible association between covid history and serology results and found no significant result (P = 0.112). However, 2 of the 3 patients had a very high antibody titer with a history of covid, so the antibody titer is significantly increased in patients with a history of COVID (P = 0.002) but the small number of patients does not allow us to retain this result.
4 Discussion
In this study investigating the immune response in 59 MS patients treated with fingolimod and vaccinated against COVID-19, nearly 50% were seropositive. We also showed a strong relationship between antibody development and age and duration of treatment as well as between antibody titer and age and duration of treatment.
Around 2.8 million people have MS and most are treated by a background immunotherapy such as fingolimod [23]. While the anti-COVID-19 vaccination is recommended for people with MS [24], more data are required to quantify the immune response in MS patients and to better understand which factors influence the response. This study provides additional data about MS patients treated with fingolimod and could contribute to supporting the current recommendations to vaccinate all MS patients.
Overall, 28 participants (47%) in our sample were seropositive, which is lower than the rate in the general population which is close to 100% for several types of vaccine [3], [4], [5]. Our results are in line with those of the randomised study by Kappos et al., which found that patients on fingolimod developed a diminished immune response compared with the placebo group [22]. Other studies have also found a diminished immune response in patients on fingolimod compared with patients treated with different disease-modifying therapies [25]. Finally, one study also showed an attenuated antibody response under immunosuppressive therapy (including fingolimod) [26] and a more recent paper shows that the vaccine response is attenuated in patients treated with fingolimod compared to other patients without background treatment [27]. We found that the development of antibodies was correlated to age: the proportion of people with a positive COVID serology decreases as age increases. This relationship between age and the development of antibodies was confirmed by logistic regression.
Linear regression did not find a positive significance of age, which was inconsistent with our results in terms of antibody development. We noticed that three of our patients–aged 46, 50 and 32 years (two of whom had had COVID) had an antibody titer greater than 2080 BAU/mL. Given that no other patient exceeded 1000 BAU/mL, we hypothesized that these outliers could distort the results. Once they were removed from analysis, age was found to be significantly correlated with antibody titer: the higher the age, the higher the risk of having a lower antibody titer.
The fact that our results showed an influence of age on vaccine response is in line with studies showing that vaccine responses in the elderly are weaker for several types of vaccine (excluding COVID) [28]. Although a few studies have shown that the elderly can develop robust vaccine responses against COVID [29], [30], many show that the vaccine response in this population is attenuated both qualitatively (antibody development) and quantitatively (antibody titer). A study of the COVID vaccine response in 45,965 adults from the general population in United Kingdom showed that seropositivity varied with age at the first vaccination, and that in patients with and without evidence of previous infection, younger subjects were more likely to be seropositive. The antibody titer followed the pattern of binary representation (older people are found to have lower antibody titer) [31]. Other studies on vaccine response tend to show similar results [32], [33], [34], [35], [36].
We demonstrate a significant association with the duration of treatment and the development of antibodies: the shorter the duration of treatment, the higher the probability of having a positive serology. Similarly, linear regression showed a positive significance of the duration of treatment with respect to antibody titer: the longer the duration of treatment, the greater the risk of having a lower antibody titer. The same was true for the patient sample once the outliers had been excluded.
Gender, the number of vaccinations, lymphocyte count, and time between last vaccine dose and blood sampling were not significant variables. According to the literature, gender may play a role in vaccine response [37], [38], [39], [40]. In addition, some studies show that the number of circulation lymphocytes could predict the vaccine response, which we do not find in our study [41], [42].
Our results are somewhat discrepant with previously published data on COVID seropositivity in MS patients treated by fingolimod [43]. For example, in the study conducted by Achiron et al., only one patient (out of 26) treated with fingolimod developed a positive serology to COVID, whereas in our sample almost half of the patients developed a humoral response. Furthermore, age did not seem to affect the vaccine response in their study whereas in our sample, age was significantly related to the development of antibodies. However, our results are similar to those published by Guerrieri et al. [44], which showed a positive COVID serological response in 10 of their 16 patients. They found no relationship between serological response and duration of treatment, time between vaccination and last dose of treatment, and white blood cell count. This is broadly consistent with the results of our study except for duration of treatment which was found to be significant in our sample.
The first limitation of our study is the relatively small sample size. Another limitation is that we did not measure cellular immunity: it is feasible that even if a patient did not develop humoral immunity, they might have cellular immunity and thus be protected. Tallantyre et al. studied the COVID-19 vaccine response in MS patients and, while they showed that seroconversion under fingolimod was lower, they also found that one of the six patients who had a negative humoral response had developed a cellular response [45]. An investigation of humoral, and T-cell-specific, immune responses to COVID vaccination in MS patients on different therapies by Tortorella et al. showed firstly a lower antibody response rate and antibody titer in patients on fingolimod compared with other disease-modifying therapies, and also that the lowest frequency of T-cell response was observed in patients treated with fingolimod [46]. The T-cell response in MS patients treated with fingolimod has been measured for other viruses, such as the varicella zoster virus, and the antiviral T-cell response was also lower [47]. Finally, one study recommends measuring cellular immunity to check vaccine efficacy in patients with treatments that have an action on B cells [48].
5 Conclusion
Our results show a COVID seropositivity of 47% in our sample of 59 MS patients on fingolimod. We also showed a strong relationship between antibody development and age and duration of treatment, as well as between antibody titer and age and duration of treatment. Therefore, we cannot afford not to recommend vaccination in MS patients treated with fingolimod, especially as research has shown that the third booster dose (mRNA vaccine in the study) revives the humoral response independently of any clinical variables [49]. Further studies on this topic are needed to support our results, especially regarding the influence of the duration of treatment on antibody development and titer (scarce data at present), although one study supports our point and also shows a relationship between treatment duration and vaccine response [50], ideally with a larger sample size and a measurement of cellular immunity.
Disclosure of interest
The authors declare that they have no competing interest.
==== Refs
References
1 Coronavirus disease (COVID-19)–World Health Organization [Internet]. [cited 2022 Jan 9]. Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019.s.
2 The different types of COVID-19 vaccines [Internet]. [cited 2022 Jan 9]. Available from: https://www.who.int/news-room/feature-stories/detail/the-race-for-a-covid-19-vaccine-explained.
3 Polack F.P. Thomas S.J. Kitchin N. Absalon J. Gurtman A. Lockhart S. Safety and efficacy of the BNT162b2 mRNA COVID-19 vaccine N Engl J Med. 383 27 2020 2603 2615 33301246
4 Voysey M. Clemens S.A.C. Madhi S.A. Weckx L.Y. Folegatti P.M. Aley P.K. Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK Lancet 397 10269 2021 99 111 33306989
5 Baden L.R. El Sahly H.M. Essink B. Kotloff K. Frey S. Novak R. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine N Engl J Med. 384 5 2021 403 416 33378609
6 Kelly H. Sokola B. Abboud H. Safety and efficacy of COVID-19 vaccines in multiple sclerosis patients J Neuroimmunol. 356 2021 577599 34000472
7 Achiron A. Dolev M. Menascu S. Zohar D.-N. Dreyer-Alster S. Miron S. COVID-19 vaccination in patients with multiple sclerosis: what we have learnt by February 2021 Mult Scler Houndmills Basingstoke Engl 27 6 2021 864 870
8 Lublin F. Miller D.H. Freedman M.S. Cree B.A.C. Wolinsky J.S. Weiner H. Oral fingolimod in primary progressive multiple sclerosis (INFORMS): a phase 3, randomised, double-blind, placebo-controlled trial Lancet. 387 10023 2016 1075 1084 26827074
9 Kappos L. Radue E.-W. O’Connor P. Polman C. Hohlfeld R. Calabresi P. A placebo-controlled trial of oral fingolimod in relapsing multiple sclerosis N Engl J Med. 362 5 2010 387 401 20089952
10 Cohen J.A. Barkhof F. Comi G. Hartung H.-P. Khatri B.O. Montalban X. Oral fingolimod or intramuscular interferon for relapsing multiple sclerosis N Engl J Med. 362 5 2010 402 415 20089954
11 Ingwersen J. Aktas O. Kuery P. Kieseier B. Boyko A. Hartung H.-P. Fingolimod in multiple sclerosis: mechanisms of action and clinical efficacy Clin Immunol. 142 1 2012 15 24 21669553
12 Simpson-Yap S. Brouwer E.D. Kalincik T. Rijke N. Hillert J.A. Walton C. Associations of disease-modifying therapies with COVID-19 severity in multiple sclerosis Neurology. 97 19 2021 e1870 e1885 34610987
13 Laroni A. Schiavetti I. Sormani M.P. Uccelli A. COVID-19 in patients with multiple sclerosis undergoing disease-modifying treatments Mult Scler Houndmills Basingstoke Engl. 27 14 2021 2126 2136
14 Sormani M.P. Schiavetti I. Carmisciano L. Cordioli C. Filippi M. Radaelli M. COVID-19 severity in multiple sclerosis: putting data into context Neurol Neuroimmunol Neuroinflammation. 9 1 2021 e1105
15 Zheng C. Kar I. Chen C.K. Sau C. Woodson S. Serra A. Multiple sclerosis disease-modifying therapy and the COVID-19 pandemic: implications on the risk of infection and future vaccination CNS Drugs. 34 9 2020 879 896 32780300
16 Francis G. Kappos L. O’Connor P. Collins W. Tang D. Mercier F. Temporal profile of lymphocyte counts and relationship with infections with fingolimod therapy Mult Scler J. 20 4 2014 471 480
17 Sormani M.P. De Rossi N. Schiavetti I. Carmisciano L. Cordioli C. Moiola L. Disease-modifying therapies and coronavirus disease 2019 severity in multiple sclerosis Ann Neurol. 89 4 2021 780 789 33480077
18 Parrotta E. Kister I. Charvet L. Sammarco C. Saha V. Charlson R.E. COVID-19 outcomes in MS: observational study of early experience from NYU Multiple Sclerosis Comprehensive Care Center Neurol Neuroimmunol Neuroinflammation. 7 5 2020 e835
19 Hatcher S.E. Waubant E. Nourbakhsh B. Crabtree-Hartman E. Graves J.S. Rebound syndrome in patients with multiple sclerosis after cessation of fingolimod treatment JAMA Neurol. 73 7 2016 790 794 27135594
20 Sastre-Garriga J. Tintoré M. Montalban X. Keeping standards of multiple sclerosis care through the COVID-19 pandemic Mult Scler J. 26 10 2020 1153 1156
21 Ciotti J.R. Valtcheva M.V. Cross A.H. Effects of MS disease-modifying therapies on responses to vaccinations: a review Mult Scler Relat Disord. 45 2020 102439 32769063
22 Kappos L. Mehling M. Arroyo R. Izquierdo G. Selmaj K. Curovic-Perisic V. Randomized trial of vaccination in fingolimod-treated patients with multiple sclerosis Neurology. 84 9 2015 872 879 25636714
23 Number of people with MS | Atlas of MS [Internet]. [cited 2022 Jan 9]. Available from: https://www.atlasofms.org/map/global/epidemiology/number-of-people-with-ms.
24 The coronavirus and MS – updated global advice [Internet] 2020 MS International Federation [cited 2022 Jan 9. Available from: https://www.msif.org/news/2020/02/10/the-coronavirus-and-ms-what-you-need-to-know/]
25 Bigaut K. Kremer L. Fabacher T. Lanotte L. Fleury M.-C. Collongues N. Impact of disease-modifying treatments of multiple sclerosis on anti-SARS-CoV-2 antibodies: an observational study Neurol Neuroimmunol Neuroinflammation. 8 5 2021 e1055
26 Wagner A. Jasinska J. Tomosel E. Zielinski C.C. Wiedermann U. Absent antibody production following COVID19 vaccination with mRNA in patients under immunosuppressive treatments Vaccine. 39 51 2021 7375 7378 34785100
27 Wu X. Wang L. Shen L. Tang K. Response of COVID-19 vaccination in multiple sclerosis patients following disease-modifying therapies: a meta-analysis eBioMedicine 81 2022 104102 35759920
28 Zimmermann P. Curtis N. Factors that influence the immune response to vaccination Clin Microbiol Rev 2019 [cited 2021 Dec 27. Available from: https://journals.asm.org/doi/abs/10.1128/CMR.00084-18]
29 Parry H.M. Tut G. Faustini S. Stephens C. Saunders P. Bentley C. BNT162b2 vaccination in people over 80 years of age induces strong humoral immune responses with cross neutralisation of P.1 Brazilian variant. Rochester, NY Soc Sci Res Network 10 2021 e69375
30 Subbarao S. Warrener L.A. Hoschler K. Perry K.R. Shute J. Whitaker H. Robust antibody responses in 70–80-year-olds 3 weeks after the first or second doses of Pfizer/BioNTech COVID-19 vaccine, United Kingdom, January to February 2021 Eurosurveillance. 26 12 2021 2100329 33769252
31 Wei J. Stoesser N. Matthews P.C. Ayoubkhani D. Studley R. Bell I. Antibody responses to SARS-CoV-2 vaccines in 45,965 adults from the general population of the United Kingdom Nat Microbiol. 6 9 2021 1140 1149 34290390
32 Ward H. Cooke G. Whitaker M. Redd R. Eales O. Brown J.C. REACT-2 Round 5: increasing prevalence of SARS-CoV-2 antibodies demonstrate impact of the second wave and of vaccine roll-out in England [Internet] 2021 [cited 2022 Jan 6. p. 2021.02.26.21252512. Available from: https://www.medrxiv.org/content/10.1101/2021.02.26.21252512v1]
33 Jabal K.A. Ben-Amram H. Beiruti K. Batheesh Y. Sussan C. Zarka S. Impact of age, ethnicity, sex and prior infection status on immunogenicity following a single dose of the BNT162b2 mRNA COVID-19 vaccine: real-world evidence from healthcare workers, Israel, December 2020 to January 2021 Eurosurveillance. 26 6 2021 2100096 33573712
34 Müller L. Andrée M. Moskorz W. Drexler I. Walotka L. Grothmann R. Age-dependent immune response to the Biontech/Pfizer BNT162b2 COVID-19 vaccination [Internet] 2021 [cited 2022 Jan 6 p. 2021.03.03.21251066. Available from: https://www.medrxiv.org/content/10.1101/2021.03.03.21251066v1]
35 Naaber P. Tserel L. Kangro K. Sepp E. Jürjenson V. Adamson A. Dynamics of antibody response to BNT162b2 vaccine after six months: a longitudinal prospective study Lancet Reg Health Eur. 10 2021 100208 34514454
36 Eyre D.W. Lumley S.F. Wei J. Cox S. James T. Justice A. Quantitative SARS-CoV-2 anti-spike responses to Pfizer–BioNTech and Oxford–AstraZeneca vaccines by previous infection status Clin Microbiol Infect. 27 10 2021 1516e7 1516e14
37 Scully E.P. Haverfield J. Ursin R.L. Tannenbaum C. Klein S.L. Considering how biological sex impacts immune responses and COVID-19 outcomes Nat Rev Immunol. 20 7 2020 442 447 32528136
38 Klein S.L. Flanagan K.L. Sex differences in immune responses Nat Rev Immunol. 16 10 2016 626 638 27546235
39 Park M.D. Sex differences in immune responses in COVID-19 Nat Rev Immunol. 20 8 2020 [461-461]
40 Fischinger S. Boudreau C.M. Butler A.L. Streeck H. Alter G. Sex differences in vaccine-induced humoral immunity Semin Immunopathol. 41 2 2019 239 249 30547182
41 Achiron A. Mandel M. Gurevich M. Dreyer-Alster S. Magalashvili D. Sonis P. Immune response to the third COVID-19 vaccine dose is related to lymphocyte count in multiple sclerosis patients treated with fingolimod J Neurol. 269 5 2022 2286 2292 35235002
42 Schiavetti I. Barcellini L. Lapucci C. Tazza F. Cellerino M. Capello E. CD19+ B cell numbers predict the increase of anti-SARS-CoV-2 antibodies in fingolimod-treated and COVID-19-vaccinated patients with multiple sclerosis [Internet] medRxiv 2022 [cited 2022 Oct 25. p. 2022.07.02.22277178. Available from: https://www.medrxiv.org/content/10.1101/2022.07.02.22277178v1]
43 Achiron A. Mandel M. Dreyer-Alster S. Harari G. Magalashvili D. Sonis P. Humoral immune response to COVID-19 mRNA vaccine in patients with multiple sclerosis treated with high-efficacy disease-modifying therapies Ther Adv Neurol Disord 14 2021 [17562864211012836]
44 S G S L C Z A N M F L M Serological response to SARS-CoV-2 vaccination in multiple sclerosis patients treated with fingolimod or ocrelizumab: an initial real-life experience J Neurol. 269 1 2021 39 43 34189719
45 Tallantyre E.C. Vickaryous N. Anderson V. Asardag A.N. Baker D. Bestwick J. COVID-19 vaccine response in people with multiple sclerosis Ann Neurol. 91 1 2022 89 100 34687063
46 Tortorella C. Aiello A. Gasperini C. Agrati C. Castilletti C. Ruggieri S. Humoral- and T-cell – specific immune responses to SARS-CoV-2 mRNA vaccination in patients with MS using different disease-modifying therapies Neurology. 98(5) 2021 e541 e554 34810244
47 Ricklin M.E. Lorscheider J. Waschbisch A. Paroz C. Mehta S.K. Pierson D.L. T-cell response against varicella-zoster virus in fingolimod-treated MS patients Neurology. 81 2 2013 174 181 23700335
48 Signoriello E. Bonavita S. Sinisi L. Russo C.V. Maniscalco G.T. Casertano S. Is antibody titer useful to verify the immunization after VZV vaccine in MS patients treated with Fingolimod? A case series Mult Scler Relat Disord. 40 2020 101963 31986424
49 Capuano R. Altieri M. Conte M. Bisecco A. d’Ambrosio A. Donnarumma G. Humoral response and safety of the third booster dose of BNT162b2 mRNA COVID-19 vaccine in patients with multiple sclerosis treated with ocrelizumab or fingolimod J Neurol 269(12) 2022 6185 6192 35879563
50 Capuano R. Bisecco A. Conte M. Donnarumma G. Altieri M. Grimaldi E. Six-month humoral response to mRNA SARS-CoV-2 vaccination in patients with multiple sclerosis treated with ocrelizumab and fingolimod Mult Scler Relat Disord 60 2022 [Available from: https://www.msard-journal.com/article/S2211-0348(22)00239-5/fulltext]
| 36496270 | PMC9727589 | NO-CC CODE | 2022-12-08 23:18:52 | no | Rev Neurol (Paris). 2022 Dec 7; doi: 10.1016/j.neurol.2022.11.003 | utf-8 | Rev Neurol (Paris) | 2,022 | 10.1016/j.neurol.2022.11.003 | oa_other |
==== Front
Archives Des Maladies Professionnelles et De L'Environnement
1775-8785
1775-8785
Published by Elsevier Masson SAS
S1775-8785(22)00374-5
10.1016/j.admp.2022.10.041
Article
Facteurs associés à la détresse psychologique des étudiants en médecine lors de la crise sanitaire de la COVID-19 : une étude transversale
Factors associated with psychological distress in medical students during the COVID-19 health crisis: A cross-sectional studyPelissier C. ab⁎
Viale M. b
Berthelot P. c
Poizat B. d
Massoubre C. e
Tiffet T. f
Fontana L. ab
a Université Lyon, université Lyon 1, université Saint-Étienne, université Gustave-Eiffel, UMRESTTE, UMR_T9405, 42005 Saint-Étienne, France
b Service de santé au travail, centre hospitalier universitaire de Saint-Étienne, 42005 Saint-Étienne, France
c Services de maladies infectieuses, centre hospitalier universitaire de Saint-Étienne, faculté de médecine, université Jean-Monnet, 42000 Saint-Étienne, France
d Service de médecine préventive, université Jean-Monnet, 42000 Saint-Étienne, France
e Département de psychiatrie, centre hospitalier universitaire de Saint-Étienne, 42005 Saint-Étienne, France
f Service de Santé Publique, centre hospitalier universitaire de Saint-Étienne, 42005 Saint-Étienne, France
⁎ Auteur correspondant.
7 12 2022
12 2022
7 12 2022
83 6 618618
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objectifs
Les modifications organisationnelles liées à la crise sanitaire COVID-19 (cours en distanciel et stage en service d’unités COVID pour certains externes) ont pu favoriser la détresse psychologique des étudiants inscrits en médecine habituellement exposés à une charge de travail intense. L’objectif de cette étude était d’évaluer la prévalence de la détresse psychologique chez les étudiants en médecine pendant la crise sanitaire du COVID-19, et d’identifier les facteurs associés à la détresse psychologique.
Méthodes
Une étude observationnelle transversale a été proposée à 1814 étudiants en médecine (de la première à la sixième année) dans un centre hospitalier universitaire français du 10 mars au 25 mars 2021. Un avis favorable du comité d’éthique a été obtenu (IRBN272021/CHUSTE). Des informations sociodémographiques, professionnelles et médicales (détresse psychologique mesurée sur l’échelle française GHQ12) ont été recueillies par le biais d’un questionnaire anonyme auto-administré en ligne. Les variables associées à la détresse psychologique ont été étudiées en utilisant une analyse univariée (test du Chi2 et Fisher). Les variables avec une valeur p < 0,2 ont été inclues dans un modèle de régression de Poisson modifiée et les variables avec une valeur de p < 0,05 ont été conservées dans le modèle.
Résultats
Au total, 832 étudiants en médecine ont répondu (46 %) et 699 ont rempli le questionnaire en entier (39 %) : 625 (75 %) ont montré des signes de détresse psychologique et 109 (15 %) ont rapporté des idées suicidaires. Le sexe féminin, un traumatisme psychologique lors de la crise sanitaire COVID-19, une modification de la consommation d’alcool et des difficultés liées à l’enseignement en distanciel étaient positivement associés à la détresse psychologique, alors qu’un sentiment d’entraide et de coopération dans le cadre des études, la reconnaissance du travail effectué étaient négativement associés à la détresse psychologique. Un stage dans une unité COVID n’apparaît pas significativement associé à la détresse psychologique.
Conclusions
Des soins de santé mentale intégrant la prévention du suicide devraient être proposés aux étudiants à risque dans ce contexte de crise sanitaire COVID-19. Connaître les facteurs éducatifs et médicaux associés à la détresse psychologique permet d’identifier des axes de prévention.
==== Body
pmcDéclaration de liens d’intérêts
Les auteurs déclarent ne pas avoir de liens d’intérêts.
| 0 | PMC9727626 | NO-CC CODE | 2022-12-08 23:18:52 | no | 2022 Dec 7; 83(6):618 | utf-8 | null | null | null | oa_other |
==== Front
Archives Des Maladies Professionnelles et De L'Environnement
1775-8785
1775-8785
Published by Elsevier Masson SAS
S1775-8785(22)00373-3
10.1016/j.admp.2022.10.040
Article
Le risque d’infection par la COVID-19 chez les professionnels de santé. Revue de la littérature
The risk of COVID-19 infection among healthcare workers. A reviewTelle-Lamberton M.
Observatoire régional de santé d’Île-de-France, 15, rue Falguière, 75015 Paris
7 12 2022
12 2022
7 12 2022
83 6 617618
Copyright © 2022 Published by Elsevier Masson SAS.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objectifs
Par leurs multiples contacts avec les patients, leurs collègues ou le public, les professionnels de santé ont un risque d’exposition au SARS-CoV-2 élevé et leur contamination a des conséquences tant pour eux-mêmes que pour les patients. Cette revue examine leur risque de contamination avant l’arrivée massive de la vaccination et les facteurs de risque associés.
Méthodes
La recherche bibliographique porte sur les documents publiés jusqu’au 24/11/2021 (principalement avant la vaccination). Elle a consisté à : rechercher par mots-clés les articles au sein des bases de données usuelles : sélectionner les études pertinentes : ajouter les études issues des bibliographies des articles sélectionnés et enfin analyser systématiquement les articles retenus (lieu de l’étude, période, protocole, type de population étudiée, principaux résultats).
Résultats
Ce recensement a conduit à plus de 150 articles. Les résultats de prévalence obtenus sont très hétérogènes (0,2 % à 36 %). Les valeurs les plus élevées concernent des zones à forte prévalence en population générale. Les valeurs les plus faibles concernent des populations avec des pratiques très strictes de confinement ou des régions où l’épidémie n’était pas encore forte. Les valeurs intermédiaires ne laissent pas apparaître de logique particulière. Les données de séroprévalence apportent des conclusions analogues. La profession apparaît comme facteur de risque : infirmiers et aides-soignants ont un risque plus élevé que les médecins dans la plupart des études. Les autres facteurs de risque sont : le travail en unité dédiée à la COVID-19 (seules quelques études font exception) : l’activité en contact avec des patients contaminés (12 études sur 19, 2 en limite de significativité et 5 non conclusives) : le contact avec des collègues contaminés (une seule étude fait exception) et le contact avec une personne contaminée hors travail (une seule étude fait exception). Les études sur les moyens de protection confirment l’apport des masques chirurgicaux, FFP2 et respiratoires renforcés et la prudence à apporter sur l’utilisation de gants et de lunettes.
Conclusions
L’hétérogénéité des niveaux de contamination publiés dans les études analysées est à rapprocher des protocoles utilisés : enquêtes transversales à des périodes diverses et souvent sur de petits effectifs. Les résultats sur les facteurs de risque ont été établis par les études les plus robustes. Les protocoles associés à ces dernières sont à encourager.
==== Body
pmcDéclaration de liens d’intérêts
L’auteure déclare ne pas avoir de liens d’intérêts.
| 0 | PMC9727627 | NO-CC CODE | 2022-12-08 23:18:53 | no | 2022 Dec 7; 83(6):617-618 | utf-8 | null | null | null | oa_other |
==== Front
Aten Primaria
Aten Primaria
Atencion Primaria
0212-6567
1578-1275
Published by Elsevier España, S.L.U.
S0212-6567(22)00237-2
10.1016/j.aprim.2022.102517
102517
Editorial
¿Está justificada la cuarta dosis para el SARS-CoV-2? Entre la necesidad y la evidencia
Is the fourth dose justified for SARS-CoV-2? Between necessity and evidenceJavierre Miranda Ana Pilar a⁎
Herce Pablo Aldaz b
Marco José Javier Gómez c
a Especialista en Medicina Familiar y Comunitaria, Centro de Salud Goya, Servicio Madrileño de Salud (SERMAS), Madrid. Miembro del Grupo de Prevención de Enfermedades Infecciosas del PAPPS de semFYC
b Especialista en Medicina Familiar y Comunitaria, Centro de Salud San Juan, Servicio Navarro de Salud (SNS), Pamplona. Miembro de Grupo de Prevención de Enfermedades Infecciosas del PAPPS
c Especialista en Medicina Familiar y Comunitaria, Centro de Salud Las Calesas, Servicio Madrileño de Salud (SERMAS), Madrid. Miembro del Grupo de Prevención de Enfermedades Infecciosas del PAPPS de semFYC
⁎ Autor para correspondencia.
7 12 2022
12 2022
7 12 2022
54 12 102517102517
© 2022 Published by Elsevier España, S.L.U.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmcSituación epidemiológica actual, las vacunas disponiblesy las coberturas vacunales actuales en nuestro país
Desde marzo del 2020 que la Organización Mundial de la Salud declaró el estado de pandemia por el virus SARS-CoV-2, ésta ha evolucionado en forma de siete periodos de diferente duración y repercusión en términos de morbimortalidad. Desde el 10 de marzo de 2020 se han estimado 124.099 excesos de defunciones por todas las causas 1.
Uno de los condicionantes de la morbimortalidad ha sido el comportamiento del virus durante la pandemia desde la cepa original (Wuhan-alfa) hasta la actualmente predominante Omicrón. Dicho comportamiento del virus, además de las medidas sanitarias implementadas explica las diferencias entre los periodos pandémicos respecto a la infección (sintomática o asintomática), morbilidad, ingresos y defunciones. Actualmente, la variante predominante en España es la Omicrón que tiene una mayor capacidad de escape inmune que variantes previas, es más transmisible y tiene menor gravedad. Comprende cinco linajes- mutaciones: BA.1, BA.2, BA.3, BA.4 y BA.5. Hasta el momento no se han observado diferencias en la gravedad de los casos para ninguna de ellas. Actualmente, la variante BA.5 es la dominante 2.
Desde el inicio de la vacunación hasta el 14 de septiembre del 2022, el 92,8% de los ciudadanos mayores de 12 años tienen la pauta completa de vacunación.3
Además de las vacunas monovalentes utilizadas hasta la fecha, se han desarrollado vacunas bivalentes llamadas así porque contienen material de RNA de la cepa original añadida a linajes de la variante Omicrón actualmente predominante: vacuna bivalente Comirnaty Original/Omicron BA.1 y Spikevax bivalente Original/Omicron BA.1, aprobadas por la Agencia Europea de Medicamentos (EMA)4 para uso como refuerzo (y no como vacunación primaria), en personas mayores de 12 años y las vacunas bivalentes Comirnaty Original/ Omicron BA.4-BA.5 y Spikevax /Original BA.4-BA.5 aprobadas en EEUU y en Europa.5
Las Autoridades Sanitarias han decidido iniciar la vacunación de la cuarta dosis a personas mayores de 80 años e institucionalizados con una vacuna bivalente disponible. En función de la disponibilidad de vacunas se ampliará a mayores de 60 años, inmunodeprimidos, sociosanitarios y gestantes.6
Evidencias sobre la protección conferida por cada vacunay cada pauta vacunal
Desde que tenemos vacunas disponibles frente al SARS-CoV-2 se han utilizado varias pautas vacunales: primovacunación con dos dosis y un refuerzo monovalente, primovacunación y dos refuerzos monovalentes. Actualmente las vacunas que vamos a tener disponibles para el segundo refuerzo son las vacunas bivalentes.
Los refuerzos monovalentes han conseguido recuperar la efectividad vacunal hasta el 85-90% frente a hospitalización y fallecimiento, que había disminuido con el paso de los meses y coincidiendo con la circulación de la variante Omicron y su evasión inmune7. Estudios observacionales en Israel describieron una reducción de 18 veces del riesgo de enfermedad grave en mayores de 60 años.
Los refuerzos con vacunas bivalentes que incluían la proteína espiga BA.1 produjeron niveles más altos de anticuerpos neutralizantes contra las subvariantes de Omicron, así como otras variantes preocupantes, en comparación con los refuerzos monovalentes. Con la adaptación de las vacunas se pretende ampliar la protección frente a diferentes variantes.
Estudios predictivos de eficacia de refuerzos bivalentes en población vacunada concluyen que la protección frente a infección grave pasaría del 50% al 86,6% con refuerzo monovalente, mientras que llegaría al 98% al usar refuerzo con bivalente. Además, cuanto menor sea la inmunidad pre-refuerzo mayor será el beneficio relativo de un refuerzo bivalente 8.
Se hace la similitud de las vacunas bivalentes con la vacuna de la gripe que cada año se actualiza para coincidir con los virus circulantes y disponer de ellas sin precisar repetir evaluaciones clínicas7.
Vacunas bivalentes disponibles para dosis de refuerzo en la Unión Europea.
Desde el pasado 1 de septiembre la Comisión Europea y la EMA autorizaron el uso de las vacunas COVID-19 adaptadas Comirnaty Original/Omicron BA.1 y Spikevax bivalente Original/Omicron BA.1, como vacunas para segunda dosis de refuerzo en personas mayores de 12 años con la pauta de primovacunación completa y al menos 3 meses después de la última dosis de vacuna administrada 4. Se basan en estudios que demuestran que pueden desencadenar fuertes respuestas inmunitarias frente a Omicron BA.1 y la cepa SARS-CoV-2 original en personas previamente vacunadas. Y mayores respuestas frente a subvariante BA.1 que las vacunas originales.
Igualmente el 12 de septiembre la EMA aprobó la vacuna bivalente COVID-19 adaptada Comirnaty Original/Omicron BA.4- BA.5 para uso en personas mayores de 12 años que han recibido al menos un ciclo primario de vacunación contra COVID-19 9. Su aprobación se ha basado en los datos clínicos disponibles con Comirnaty Original/Omicron BA.1 y en estudios de inmunogenicidad en laboratorio. Los estudios clínicos están en curso.
Los datos presentados sobre eficacia se resumen en la siguiente tabla 1 , adaptada de referencias 10 y 11.Tabla 1 Eficacia comparada de vacunas mono y bivalentes. Elaboración propia adaptada de referencias 10, 11
Tabla 1% participantes con respuesta serológica Neutralización frente a cepa original SARS-CoV-2 Neutralización frente a Ómicron
Comirnaty original
49,2% 57%
Comirnaty bivalente (original/Omicrom BA.1)
50% 71,6%
Vacuna Spikevax original
42,7% 53,1%
Vacuna Spikevax bivalente (original/omicron BA.1) 53,9% 74,9%
Los efectos secundarios observados con las vacunas adaptadas fueron comparables a los observados con las originales y, por lo general, fueron leves y de corta duración.
La pauta recomendada de administración del segundo refuerzo con vacuna bivalente en mayores de 60 años es al menos tras 5 meses de la infección por SARS-CoV-2 o de la dosis previa de vacuna frente al SARS-CoV-2. En mayores de 80 años, institucionalizados o en inmunodeprimidos el tiempo en ambos casos se reduce a 3 meses12.
Las vacunas originales, Comirnaty y Spikevax, siguen siendo eficaces para prevenir enfermedades graves, hospitalizaciones y muertes asociadas con la COVID-19 y se seguirán utilizando en las campañas de vacunación en la UE, en particular para la vacunación primaria. Se recomienda completar lo antes posible la primovacunación.
Aunque las vacunas adaptadas están autorizadas para su uso en personas de 12 años o más que hayan recibido al menos la vacunación primaria contra el COVID-19, el ECDC y la EMA aconsejan que estos refuerzos se dirijan de forma prioritaria a las personas que corren un mayor riesgo de enfermedad grave con factores de riesgo. Es por ello importante administrar este segundo refuerzo a personas inmunodeprimidas, embarazadas o en los primeros 6 meses tras el parto, convivientes de pacientes inmunodeprimidos y personal sanitario y sociosanitario.
Argumentos a favor y en contra de la cuarta dosis
Los grupos más vulnerables (inmunosenescentes, inmunodeprimidos…) tienen una protección limitada en el tiempo ya sea derivada de la vacunación o de la infección natural. Son los que más se han beneficiado de la vacunación y los que sin duda deben recibir nuevas dosis de refuerzo.
Las personas no vacunadas y que no se han infectado con el virus se han mostrado como un grupo de riesgo para presentar las formas graves de la enfermedad, y por lo tanto serían candidatas cuanto antes a iniciar su vacunación.
En el resto de población adolecemos de falta de estudios, especialmente en lo referente a la inmunidad celular y a la duración de la inmunidad.
Aplicar una cuarta dosis de manera general tiene como inconvenientes que nos puede abocar a una baja cobertura vacunal por fatiga pandémica, y que la reiteración con nuevas dosis nos puede provocar un fenómeno de tolerancia inmune.
El hecho de que aparecieran nuevas variantes del virus nos obligaría a evaluar nuevamente la protección conseguida con las vacunas disponibles en la actualidad.
En un escenario tan cambiante se debe reforzar la vigilancia epidemiológica y promover nuevos estudios de inmunidad en toda la población.
Conflicto de Intereses
Los tres autores (Ana Pilar Javierre Miranda, Pablo Aldaz Herce y Javier Gómez Marco) declaran no tener conflicto de intereses.
==== Refs
Bibliografia
1 Instituto de Salud Carlos III. ISCI. RENAVE. Informe n° 145. Situación de COVID-19 en España. Informe COVID-19. 6 de septiembre de 2022. Disponible en: Informe n° 145 Situación de COVID-19 en España a 6 de septiembre de 2022.pdf (isciii.es).(consultado 15-9-2022).
2 Ministerio de Sanidad. Actualización de la situación epidemiológica de las variantes de SARS-CoV-2 en España. 6 de septiembre del 2022. Disponible en: COVID19_Actualizacion_variantes_20220913.pdf (sanidad.gob.es).(consultado el 18-9-2022).
3 Ministerio de Sanidad. GIV. Gestión integral de la vacunación COVID-19. Disponible en: https://www.sanidad.gob.es/profesionales/saludPublica/ccayes/alertasActual/nCov/documentos/Informe_GIV_comunicacion_20220916.pdf.(consultado el 22-9-2022).
4 EMA. Agencia Europea del Medicamento. Primeras vacunas de refuerzo adaptadas contra la COVID-19 recomendadas para su aprobación en la UE.01/09/2022. Disponible en: Primeras vacunas de refuerzo adaptadas contra la COVID-19 recomendadas para su aprobación en la UE | Agencia Europea de Medicamentos (europa.eu).(consultado el 14-9-2022).
5 FDA. U. S Food and Drug Administration. COVID-19 Bivalent Vaccine Boosters. 31-8-2022. Disponible en: COVID-19 Bivalent Vaccine Boosters | FDA.(consultado el 16-9-2022).
6 Ministerio de Sanidad, Consejo Interterritorial. Comisión de Salud Pública. Recomendaciones de vacunación frente a COVID-19 para el otoño en España. Aprobado por la Comisión de Salud Pública el 22 de septiembre de 2022. Elaborado por la Ponencia de Programa y Registro de Vacunaciones. Disponible en: https://www.sanidad.gob.es/profesionales/saludPublica/prevPromocion/vacunaciones/covid19/docs/Recomendaciones_vacunacion_Otono_Covid_VF.pdf.(consultado 25-9-2022).
7 UpToDate. Vacunas para Covid 19. https://www-uptodate-com.bvcscm.a17.csinet.es/contents/covid-19-vaccines?search=vacuna%20bibalentes%20para%20el%20covid-19&source=search_result&selectedTitle=1∼150&usage_type=default&display_rank=1#H1355847968.
8 David S. Khoury, Steffen S. Docken, Kanta Subbarao, Stephen J. Kent, Miles P. Davenport, and Deborah Cromer. Predicting the efficacy of variant-modified COVID-19 vaccine boosters. Disponible en: https://www.medrxiv.org/content/10.1101/2022.08.25.22279237v1.full.pdf.(Consultado el 18-9-2022).
9 EMA. Agencia Europea del Medicamento. Adapted vaccine targeting BA.4 and BA.5 Omicron variants and original SARS-CoV-2 recommended for approval. 12/9/2022. Disponible en: https://www.ema.europa.eu/en/news/adapted-vaccine-targeting-ba4-ba5-omicron-variants-original-sars-cov-2-recommended-approval.
10 AUTORIZACIÓN DE USO DE EMERGENCIA (EUA) para la VACUNA PFIZER-BIONTECH COVID-19, BIVALENTE (ORIGINAL Y OMICRON BA.4/BA.5). https://www.fda.gov/media/161327/download.(Consultado el 25 de septiembre de 2022).
11 AUTORIZACIÓN DE USO DE EMERGENCIA (EUA) para la VACUNA MODERNA COVID-19, BIVALENTE (ORIGINAL Y OMICRON BA.4/BA.5). https://www.fda.gov/media/161318/download.(Consultado el 25 de septiembre de 2022).
12 Documento técnico de vacunación frente a COVID-19 en la Comunidad de Madrid en el otoño de 2022. Dosis de refuerzo con vacunas adaptadas a las nuevas variantes. 23/9/2022. Disponible en: https://www.comunidad.madrid/sites/default/files/doc/sanidad/prev/23.09.2022_doc_tecnico_vacunacion_covid-19_recuerdo.pdf.
| 36496254 | PMC9727668 | NO-CC CODE | 2022-12-08 23:18:53 | no | Aten Primaria. 2022 Dec 7; 54(12):102517 | utf-8 | Aten Primaria | 2,022 | 10.1016/j.aprim.2022.102517 | oa_other |
==== Front
Journal of Pipeline Science and Engineering
2667-1433
2667-1433
The Authors. Publishing Services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.
S2667-1433(21)00079-2
10.1016/j.jpse.2021.12.004
Article
Planning of a flexible refined products transportation network in response to emergencies
Wang Bohong a⁎
Klemeš Jiří Jaromír a
Yu Xiao b
Qiu Rui b
Zheng Jianqin b
Lin Yuming c
Zhu Baikang d
a Sustainable Process Integration Laboratory – SPIL, NETME Centre, Faculty of Mechanical Engineering, Brno University of Technology - VUT BRNO, Technická 2896/2, 616 69 Brno, Czech Republic
b National Engineering Laboratory for Pipeline Safety/Beijing Key Laboratory of Urban Oil and Gas Distribution Technology, China University of Petroleum-Beijing, Beijing, 102249, PR China
c PIPECHINA Oil & Gas Control Center, Beijing, 100007, PR China
d School of Petrochemical Engineering and Environment, Zhejiang Ocean University, Zhoushan, 316022, PR China
⁎ Corresponding author.
2 1 2022
12 2021
2 1 2022
1 4 468475
26 10 2021
28 12 2021
30 12 2021
© 2021 The Authors. Publishing Services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The stable supply of refined products is a vital measure to maintain national security and economic development. However, unexpected accidents and incidents, such as pipeline failures, natural hazards, and even this Covid-19 period can disrupt the supply of refined products to the end-users, leading to shortages on the demand side. This issue should be considered in advance in the planning process to increase transportation flexibility. In this paper, a MILP model-based method is proposed to increase the operational flexibility of refined products transportation. This model takes into account several failure conditions of routes during the transportation process and determines the backup transport route plans. The illustration case study demonstrates how to build a more flexible refined products transportation process through expanding transport and sending capacities to satisfy the demand for refined products in market depots. The results show that the capacity expansion plans are different when disruptions are individually and simultaneously considered, a 33.60% difference in the expansion cost. The proposed method could be helpful to ensure the supply of retail markets when facing emergencies.
Keywords
Transportation
Refined products
Flexibility
Oil industry
Disruption
==== Body
pmc1 Introduction
The gasoline consumption of China is due to reach 125.1 Mt in 2021 (Xu et al., 2021). A well-running supply chain is required to ensure that such a huge amount of refined products can be transported smoothly from the refineries to the markets.
The primary level of refined products transportation network studied in this paper is part of the refined products supply chain. The transportation network is operated from refineries to market depots. The distribution of refined products is the focus but without the consideration of other supply chain activities such as production, inventory management, and vendition (Zhang, 2006). The goal is to meet the needs of the downstream market for different types of refined products. An operable and money-saving distribution plan should be determined to achieve this purpose.
In the distribution plan formulated by the operator of the refined products transportation, various transportation modes should be rationally arranged based on the production of refineries, the inventory of transit depots and market depots. The maximum delivery capacity of each transportation mode and the sending capacity of refineries and transit depots should be considered. The goals include minimising the transportation cost and maximising the satisfaction rate of the market depots.
Guarantee the supply of refined products is the priority of sales and transport companies. However, the end-user market shortage and fears of shortage exist. Such as the coronavirus epidemic, which has severely disrupted the input supply of some supply chains, e.g., the poultry supply chain (Palouj et al., 2021).
There are occasions that the operation of refined products transportation is affected by some unexpected factors, such as the sudden increase in the demand of market depots, the totally or partly shutdown of upstream refineries, the outage of transportation facilities. For example, the offline of the Colonial pipeline in 2021 caused fear of refined products shortage in the US East Coast (Navarro, 2021), the 2008 Wenchuan earthquake, which caused the transport routes of refined products to shutdown (Xinhua News Agency, 2008), and in the COVID-19 pandemic period when the energy supply-demand balance (Jiang et al., 2021) and by a worldwide vaccination campaign (Klemeš et al., 2021), as well as the transport and sale schemes, are significantly influenced (Norouzi et al., 2020). Many factors in a transport network of refined products are affecting the satisfaction of the end-user market.
In the daily operation of the supply of the refined products chain, there are always some occasions that unexpected events happen. Some accidents may occur during the operation of the refined products pipeline, such as corrosion (Shafeek et al., 2021), equipment failure (Alves and Lima, 2021), pipeline leakage (Zheng et al., 2021), etc. If the accidents are severe, partial sections or the whole pipeline will be shutdown, affecting the normal distribution of refined products, which may result in a shortage in the downstream market.
Refinery accidents such as explosions (Isimite and Rubini, 2016), fire incidents (Shaluf et al., 2003), failure of refinery equipment (Etminanfar et al., 2020), etc., will affect the normal production of the refinery, causing disruptions of equipment or suspension the delivery of the refined product. In addition, other accidents, as, e.g. depot explosions (Xie et al., 2021), would also have a significant impact on the product of the refined oil, resulting in the shortage of the downstream market.
These occasions are barriers for transportation to achieve the main goals of reducing shortage and increasing uptime (Emenike and Falcone, 2020). These two goals are vital as the shortage will affect peoples’ lives the industrial production and even cause economic recessions. Meanwhile, with the intelligent trends in the oil and gas industry (Wang et al., 2019), a quick response becomes possible. It is important to design a flexible transportation network that can handle unnormal issues, strengthen the emergency measures when the accident happens, and ensure the efficient and precise configuration of refined products, and reduce transportation costs.
There are some existing contributions focused on the optimisation of refined products transportation network, or on the supply chain, to reduce total cost in strategical, tactical, and operation levels. Lima et al. (2018) developed a stochastic programming model for tactical planning of downstream oil supply chains. The scenario tree approach was applied to tackle uncertainties, including oil price and demand. A real-world case of Portugal was studied. Wang et al. (2019) proposed a MILP model which considers both refined products distribution planning and new pipeline routes planning. The demand uncertainty of the refined products was considered, and a three-level supply chain network was modelled. Zhou et al. (2019) developed a stochastic model for a coal-to-liquid supply chain to determine facility location, coal purchase plan, transport route and amount, and the number of vehicles. Pudasaini (2021) proposed a multi-objective mixed-integer linear programming model for strategic and tactical planning of downstream petroleum supply chain (DPSC). The demand uncertainty was modelled using the scenario trees, and both the transport cost and product loss cost were considered. Lima et al. (2021) considered the uncertainties of network costs and refined products demands in the design and plan of a downstream oil supply chain. A mixed-integer linear programming model was developed, and fuzzy mathematical programming was applied to deal with the uncertainty. A case of the Brazilian oil network was studied with the objective function of the total cost. Wang et al. (2021) studied the fair profit allocation of a refined products supply chain and optimised the transport plan of refined products distributed from refineries to transit depots and customers to get the Nash bargaining solution.
Some studies integrated the environmental cost in the refined products transport and supply chain optimisation. Yuan et al. (2019b) quantified the impacts of pipeline network reform on the downstream oil supply chain of China. A mathematical programming model was developed to determine the transport plan of refined products. The performance of reformation in five scenarios on energy, economy, and environment was assessed. Wang et al. (2020a) proposed a method for simultaneously considering the economic and environmental costs for refined products supply chain optimisation at the operational level. GHG and SOx emissions were calculated together, with the total cost was minimised.
The policy can also have an impact on the efficiency of supply chains (Zheng et al., 2020). Studies showed that applying advanced energy-saving technical measures for pipelines (Yuan et al., 2019a) and the interconnectivity of infrastructures of different companies (Yuan et al., 2020b) can help to improve the efficiency of refined products supply chains.
There are a few pieces of the research study the disruptions in the refined products transportations. One aspect is reliability. Wang et al. (2020b) proposed a framework to assess the reliability of refined products supply chains. A MILP model was developed to simulate the operation of the supply chain, and indices were proposed to evaluate the performances under several scenarios. Yuan et al. (2020a) studied two levels, i.e., provincial level and regional level, to evaluate the downstream oil supply security in China. The vulnerability of the oil supply network was studied by calculating the oil shortage amount in scenarios of different levels of oil import disruption, duration times, and initial inventory levels. A downstream oil supply chain planning model was developed for the calculation. Zhou et al. (2020) developed a supply reliability evaluation method for multi-product pipeline systems. The pump failures were considered, the stochastic failure characteristic was described by the discrete-time Markov process, and the system states transition was simulated by the Monte Carlo method.
However, the reliability study aims to locate and analyse the weak points in the supply chain and strengthen these weak points to increase the satisfaction of the market depots. While in the flexibility study, the main aim is to find an alternative plan to compensate for the loss caused by the disruption. The strategic, tactical, and operational decision-making of the supply chain when responses to dramatic disruptions were studied by Kumar and Sharma (2021) recently to help provide guidelines.
The contributions of this work are as follows:(i) A method is proposed for refined products transportation network optimisation under disruption to tackle unexpected situations and to satisfy the market depots.
(ii) The proposed operating level model for refined products transportation network optimisation integrates the transport and sending capacity expansion.
(iii) This model can be used to predetermine the emergency measures before the disruption happen. The facility sending capacity expansion for some transport modes can be determined as the backup. When the disruption happens, it would be easier to switch to alternative plans as reserved capacities have been determined.
(iv) The expansion plan of individual disruption and the case that simultaneously considers several disruptions in a model are compared.
The rest of this work is as follows. In Section 2, the mathematical model and procedure of how to determine the alternative plans are introduced. In Section 3, a refined products transportation network in a region of China is studied to verify the effects of the proposed method. Section 4 summarises the work and provides the outlook.
2 Model and procedure
The refined products transportation network studied in this paper has three levels of facilities, i.e., refineries, transit depots, and market depots. Refined products are produced in the refineries, transported via transit depots, or directly to the market depots as sinks. There are transport routes between these facilities. The transport mode is related to the existing infrastructure, refined products sending equipment and receiving equipment. The transit depots can receive refined products from upstream refineries, store the refined products, and distribute them to the market depots. Some refineries and transit depots have pipeline connections to reduce the transport cost.
2.1 Modelling assumptions
The following assumptions have been made for model formulation:(i) The planning of the refined products transportation network is operated by a company, and the refined products can be transported through railway, pipeline, waterway, and road.
(ii) The inventory in market depots is not sufficient to meet the demand of the market depot when disruption occurs.
(iii) The production capacities of refineries and the demand for market depots are fixed.
(iv) The current sending capacity and transport capacity are known, and they can be extended.
2.2 Model development
In the developed model, givens are:(i) Locations of refineries, transit depots, and market depots.
(ii) Production capacity of refineries and demand of market depots.
(iii) The existing routes between facilities, their distances, and their current maximum transport capacities by four types of transport mode.
(iv) Unit prices of four types of transport mode, and the prices for extending the transport and sending capacities.
It is necessary to estimate:(i) The distribution plan of the refined products transportation network.
(ii) The routes that should be expanded and the sending capacity that should be increased when disruption of current routes happens.
2.2.1 Objective function
(1) fmin=CT+CSN+CTN
The objective function (1) in this model is the sum of transport cost, sending capacity expansion cost, and transportation capacity expansion cost.
2.2.2 Transport cost constraints
(2) CT=∑i∑j∑r∑t∑sXi,j,r,t,sALi,jCi,j,rU+∑i∑k∑r∑t∑sXi,k,r,t,sALi,kCi,k,rU+∑j∑k∑r∑t∑sXj,k,r,t,sALj,kCj,k,rU
Where Xi,j,r,t,sA, Xi,k,r,t,sA, and Xj,k,r,t,sA are volumes for transporting refined products from 1) refinery i to transit depot j, 2) from refinery i to market depot k, and 3) from transit depot j to market depot k by transport mode r at time slice t in scenario s. Ci,j,rU, Ci,k,rU, and Cj,k,rU are corresponding unit transport prices. Li,j, Li,k, and Lj,k are distances for transportation.
The total transport cost is the sum of transport cost of three types of routes, i.e., from refinery to transit depot, from refinery to a market depot, and from transit depot to market depot.
2.2.3 Material balance constraints
(3) VLi,t,s=VLi,t−1,s+Vi,t,sP−∑k∑rXi,k,r,t,sA−∑j∑rXi,j,r,t,sA
(4) VSj,t,s=VSj,t−1,s+∑i∑rXi,j,r,t,sA−∑k∑rXj,k,r,t,sA
(5) VZk,t,s=VZk,t−1,s+∑i∑rXi,k,r,t,sA+∑j∑rXj,k,r,t,sA−Dk,t,s
Where VLi,t,s, VSj,t,s, and VZk,t,s are the volume of refined products storage at refinery depot, transit depot, and market depot at time slice t; Vi,t,sP is the production volume at time slice t for refinery i; and Dk,t,s is the demand of refined products for market depot k at time slice t in scenario s.
Constraints (3–5) are material balance constraints for refineries, transit depots, and market depots. For a refinery, the volume of refined products storage at time slice t equals the volume at time slice t-1 plus the production and minus the volume transported to the storage and market depots through all transport modes. For a transit depot, its volume at time slice t equals the volume at the previous time slice plus the amount transported from refineries and minus the amount transported to market depots. While for a market depot, its volume at time slice t is the sum of the volume at the previous time slice and the volume from refineries and transit depots, then minus the demand.
2.2.4 Capacity constraints
(6) Xi,j,rAmin≤Xi,j,r,t,sA≤Xi,j,r,sAmax
(7) Xi,k,rAmin≤Xi,k,r,t,sA≤Xi,k,r,sAmax
(8) Xj,k,rAmin≤Xj,k,r,t,sA≤Xj,k,r,sAmax
Where Xi,j,r,sAmax, Xi,k,r,sAmax, and Xj,k,r,sAmax are the maximum transport capacity of the refined products transportation network in scenario s after the new route expansion. For those transport routes that are disrupted in scenario s, they can be set as 0. Xi,j,rAmin, Xi,k,rAmin, and Xj,k,rAmin are the minimum transport capacity of the refined products in routes.
Constraints (6–8) ensure that the transport volume of each route is in a feasible transport volume range. Each transport route has its minimum and maximum transport capacities at each time slice. For railway and road, the amount is restricted by the number of tanks or vehicles, while for the pipeline, to ensure safe transport, the amount cannot be lower than the minimum transport restriction and cannot exceed the highest transport capacity.(9) ∑jXi,j,r,t,sA+∑kXi,k,r,t,sA≤Si,r,sAMax
(10) ∑kXj,k,r,t,sA≤Sj,r,sAMax
Where Si,r,sMax is the maximum sending capacity of refineries i for transport mode r in scenario s, and Sj,r,sMax is the maximum sending capacity of transit depot j for transport mode r in scenario s.
Constraints (9–10) restrict the sending capacity of refineries and transit depots at each time slice. The number of equipment or facilities for sending refined products through different ways are limited. For example, the number of crane tubes which is used for injecting refined products into the tanker truck is fixed, and the amount of refined products which is sent to road transport has its maximum volume.
2.2.5 Capacity expansion constraints
(11) Xi,j,r,sAMax≤Xi,j,rN+Xi,j,rCMax
(12) Xi,j,rN=∑gBi,j,r,gE×PgE
(13) ∑gBi,j,r,gE≤1
Where Xi,j,rN is the expanded capacity for the transport route from refinery i to transit depot j by transport mode r of the current refined products transportation network, which is a decision variable in the model; Bi,j,r,gE is a binary variable that represents the expansion level of the transport mode.
For the transport route from refinery to transit depot, constraint (11) ensures that the maximum transport capacity of route i to j by transport mode r in scenario s should be lower than the previous capacity plus the expanded capacity. The maximum transport capacity equals the current transport capacity plus the new or expanded transport capacity. One of the main objectives of this study is to determine how much capacity should be expanded or newly developed to ensure the flexible operation of the transportation network under unnormal conditions. Constraint (12) calculates the expanded capacity of route i to j by transport mode r. Constraint (13) ensures that at most of 1 level transport expansion should be selected.(14) Xi,k,r,sAMax≤Xi,k,rN+Xi,k,rCMax
(15) Xi,k,rN=∑gBi,k,r,gEPgE
(16) ∑gBi,k,r,gE≤1
For other routes, such as from refineries to market depots and from transit depots to market depots, similar constraints can be formulated. Constraints (14–16) are formulated for the routes from refineries to market depots.(17) Xj,k,r,sAMax≤Xj,k,rN+Xj,k,rCMax
(18) Xj,k,rN=∑gBj,k,r,gE×PgE
(19) ∑gBj,k,r,gE≤1
Constraints (17–19) are used for the routes from transit depots to market depots.(20) Si,r,sAMax≤Si,rN+Si,rCMax
(21) Si,rN=∑hBi,r,hSPhS
(22) ∑hBi,r,hS≤1
Where Bi,r,hS is a binary variable that represents the expansion level of the sending capacity of facilities.
For refineries, Constraint (20) ensures that the maximum sending capacity of the refinery i for transport mode r in scenario s should be lower than the previous sending capacity plus the expanded sending capacity. Constraint (21) calculates the expanded capacity of refinery i for transport mode r. Constraint (22) ensures that at most 1 level of facility expansion should be selected.(23) Sj,r,sAMax≤Sj,rN+Sj,rCMax
(24) Sj,rN=∑hBj,r,hSPhS
(25) ∑hBj,r,hS≤1
For transit depot, similar constraints as Constraints (23–25) can be formulated to determine the expansion capacity.
2.2.6 Expansion cost constraints
(26) CSN=∑i∑r∑hBi,r,hSCr,hS+∑j∑r∑hBj,r,hSCr,hS
(27) CTN=∑i∑j∑r∑gBi,j,r,gECr,gE+∑i∑k∑r∑gBi,k,r,gECr,gE+∑j∑k∑r∑gBj,k,r,gECr,gE
Where Cr,hS and Cr,gE are costs for expanding sending and transport capacities for transport mode r and in their levels, which are indexed by h and g.
Eq. (26) calculates the sending capacity expansion cost of facilities, including refineries and transit depots, and Eq. (27) calculates the transport capacity expansion cost of every type of transport route.
2.3 Scenarios
There are several types of route disruptions in the supply of the refined products chain, including railway route failure, pipeline failure, waterway route failure, and road route failure. These routes all have the possibility to be disrupted by natural disasters or human errors. When these disruptions happen, it is important to introduce new routes or substituted routes to revive the operation of the refined products transportation network. The basic model in the normal condition plus the constraints of route disruption should be solved to propose the alternative plan. The method can also be used to determine an alternative distribution plan when several disruptions have happened. The scenarios in the next section Case study are based on these route disruption types.
3 Case study
A case is studied to test the proposed method in this section. There are three refineries, two transit depots, and 30 market depots in the studied region. They are distributed as shown in Fig. 1 . The case was solved by Gurobi (Gurobi Optimization LLC, 2021), and the optimality gap was set to 0.Fig. 1 Distribution of facilities in the studied refined products transportation network.
Fig. 1
Transit depot 1 is affiliated with refinery 2, and a pipeline connects these two stations. Another pipeline starts at refinery 1, connecting market depot numbered 1, 4, 5, 11, 12, 16, 19, 23, 30, and transit depot 2. Transit depot 2 can transport refined oil by a waterway, and market depot numbered 24, 25, 26, 27, and 28 have the unloading capacity of the waterway. The transportation costs for the railway, pipeline, waterway, and road are set as: 0.31 CNY/(km•t), 0.143 CNY/(km•t), 0.1 CNY/(km•t), and 0.68 CNY/(km•t).
First, the original distribution plan without transport disruption is optimised. The results are shown in Fig. 2 . Refinery 1 transports refined products to several market depots through pipelines, also to the transit depot 2. Then it further distributes refined products via waterway to these market depots, which can receive from docks. Refinery 2 and transit depot 1 are mainly used to distribute refined products in the southwest part of the studied region. The upper part is distributed by the railway from the refinery, and the lower part is distributed by the transit depot 1, which receives refined products through the pipeline links these two stations first. The northeast part of the region receives refined products from refinery 3 mainly through railways.Fig. 2 Distribution plan of the refined products transportation network under normal condition.
Fig. 2
Then several scenarios for transport routes disruption are set as the following:(i) The railway transport route from refinery 2 to market depot 9 is disrupted.
(ii) Market depot 22 and 29 cannot receive refined products through the pipeline.
(iii) The waterway transport route from transit depot to market depot 23 is disrupted.
(iv) Consider the above three scenarios simultaneously.
In scenario (i), two new transport routes are used to satisfy the demand of market depot 9, the first is a roadway from transit depot 1 to market depot 9, and the second is a roadway from transit depot 2 to market depot 9. The new distribution plan for this scenario is shown in Fig. 3 . The influence on the transport cost is minor, only 0.01%.Fig. 3 Distribution plan of the refined products transportation network under scenario (i).
Fig. 3
In scenario (ii), these two market depots need a relatively large volume of refined products, and mainly through pipeline transportation when the whole system operates well. When the pipeline that transports refined products to these two market depots is disrupted, some changes happen to continually satisfy the demand. Refinery 1 starts to transport refined products through railways directly. However, the amount still cannot meet all their requirement. The refined products received from refinery 2 via railway to market depot 22 is increased to make up for the loss brought by the pipeline disruption, which also influences the way that other market depots receive the refined products. The market depot 9 changes its source from refinery 2 to refinery 3, and market depots 14 and 17 change their sources from transit depot 1 to refinery 3, still by railway transport. Market depot 29 receives the refined products from both refineries 1 and 3 through railways. The scheme is shown in Fig. 4 . In this way, the transport cost increases by 2.29%. The change of sources is due to the production capacities of refineries. To compensate for the loss of those two market depots influenced by the pipeline disruption, other refineries and transit depots also have to adjust their distribution plans to ensure the demand of the whole system can be satisfied.Fig. 4 Distribution plan of the refined products transportation network under scenario (ii).
Fig. 4
In scenario (iii), the disruption of the waterway stops the market depot from receiving refined products from transit depot 2. As shown in Fig. 5 , alternative ways from refinery 3 by road and from transit depot 2 by road should be used to meet the requirement of the influenced market depot, which increases the transport cosy by 3.03%. Because of the insufficient sending capacity of refinery 3 and transit depot 2, the cost also happens on the expansion of the sending capacities. In both stations, the road sending mode is selected, which means the current sending capacities of these two stations are limited and should be expanded to better tackle future disruptions.Fig. 5 Distribution plan of the refined products transportation network under scenario (iii).
Fig. 5
In scenario (iv), when all three scenarios are considered, the plans for expanding transport and sending capacities are different from the previous individual considerations. The road sending capacity for refinery 3 is extended to satisfy the demand from market depot 8, 9, 22, and 29. The road sending capacity for refinery 1 is also extended to satisfy the requirement from market depots 10, 22, and 29. Unlike the case in scenario (ii), that railway is mainly used, as it requires a little cost to expand the sending capacity. While in scenario (iv), it tends to use more road transport due to the relatively low cost of increasing the road sending capacities of refineries and transit depots. Although the transport cost increases 22.01% compared to the normal condition, the expansion cost reduces 33.6% compared to the sum of the expansion costs of scenarios (i) to (iii). This is because some expansion of routes can be together used for several scenarios. When only one scenario is considered, the model gives the best result only for that scenario, but when several scenarios are simultaneously considered, the model can comprehensively consider these scenarios together and try to maximise the usage of expanded route transport capacities and sending capacities. In this way, the total expansion cost can be reduced. The increase of transport cost is because although some transport modes such as road transport are relatively expensive compared to other routes, it can save capacity expansion cost, which in total, reduces the overall cost. This suggests that to increase the flexibility of the refined products transportation network, it is a good option to expand the sending capacities of depots for road transport as it can adapt to more unexpected situations.
4 Conclusions
This paper optimises refined products transportation network under unexpected disruptions to determine the distribution plan, transport capacity expanding plan, and sending capacity expanding plan. The transportation cost and capacity expansion cost are set as the objective functions. The results show that considering disruptions individually and simultaneously would lead to different expanding plans. In the simultaneous consideration scenario, it tends to expand the sending capacity of refineries and transit depots on the road transportation as it is easy to operate and costs less for increasing the sending capacity. As a result, the simultaneous consideration would reduce the total cost for expansion compared to the sum of the expansion cost individually considered; in this case, it achieves a 33.60% reduction.
The future work should consider the expansion cost in detail – e.g., setting parameters for each route and each facility. In terms of methodological development, the current model could be extended to a stochastic programming model to consider the uncertainties such as demand and production in the distribution and expansion planning.
Nomenclature
Sets and Indices
I Set of refineries, denoted by index i
J Set of transit depots, denoted by index j
K Set of market depots, denoted by index k
R Set of transport modes, denoted by index r
T Set of time slices, denoted by index t
S Set of scenarios, denoted by index s
G Set of expanding capacities of transport, denoted by index g
H Set of expanding capacities of sending, denoted by index h
Continuous Variables
CT Transportation cost, CNY
CSN Sending capacity expansion cost, CNY
CTN Transportation capacity expansion cost, CNY
VLi,t,s Volume of the product stored at refinery i in the time period t in scenario s, t
VSj,t,s Volume of the product stored at the transit depot j in the time period t in scenario s, t
VZk,t,s Volume of the product stored at the market depot k in the time period t in scenario s, t
Xi,j,r,t,sA Volume of product transported from the refinery i to the transit depot j by transport mode r at time slice t in scenario s, t
Xi,k,r,t,sA Volume of product c transported from the refinery i to the market depot k by transport mode r at time slice t in scenario s, t
Xj,k,r,t,sA Volume of product c transported from the transit depot j to the market depot k by transport mode r at time slice t in scenario s, t
Si,rN Expanded sending capacity for the refinery i by transport mode r, t
Sj,rN Expanded sending capacity for the transit depot j by transport mode r, t
Xi,j,rN Expanded capacity for the transport route from refinery i to transit depot j by transport mode r, t
Xi,k,rN Expanded capacity for the transport route from refinery i to market depot k by transport mode r, t
Xj,k,rN Expanded capacity for the transport route from transit depot j to market depot k by transport mode r, t
Xi,j,r,sAmax Maximum transport capacity of the refined products from refinery i to transit depot j by transport mode r in scenario s after new route development and expansion, t
Xi,k,r,sAmax Maximum transport capacity of the refined products from refinery i to market depot k by transport mode r in scenario s after new route development and expansion, t
Xj,k,r,sAmax Maximum transport capacity of the refined products from transit depot j to the market depot k by transport mode r in scenario s after new route development and expansion, t
Si,r,sAMax Maximum sending capacity of refineries i for transport mode r in scenario s, t
Sk,r,sAMax Maximum sending capacity of transit depot k for transport mode r in scenario s, t
Binary Variables
Bi,j,r,gE Binary variable which represents the expansion level g of the transport mode r from refinery i to transit depot j
Bi,k,r,gE Binary variable which represents the expansion level g of the transport mode r from refinery i to market depot k
Bj,k,r,gE Binary variable which represents the expansion level g of the transport mode r from transit depot j to market depot k
Bi,r,hS Binary variable which represents the expansion level h of the sending capacity of refinery i
Bj,r,hS Binary variable which represents the expansion level h of the sending capacity of transit depot j
Parameters
Li,j Distance from the refinery i to the transit depot j, km
Li,k Distance from the refinery i to the market depot k, km
Lj,k Distance from the transit depot j to the market depot k, km
Ci,j,rU Unit transport price for transporting refined products from refinery i to transit depot j through transport mode r, CNY/(km•t)
Ci,k,rU Unit transport price for transporting refined products from refinery i to market depot k through transport mode r, CNY/(km•t)
Cj,k,rU Unit transport price for transporting refined products from transit depot j to market depot k through transport mode r, CNY/(km•t)
Vi,t,sP Production volume at time slice t for refinery i, t
VLi,0,s Initial storage volume for refinery i at scenario s, t
VSj,0,s Initial storage volume for transit depot j at scenario s, t
VZk,0,s Initial storage volume for market depot k at scenario s, t
Dk,t,s Demand of refined products for market depot k at time slice t in scenario s, t
Xi,j,rAmin Minimum transport capacity of the refined products from refinery i to transit depot j by transport mode r, t
Xi,k,rAmin Minimum transport capacity of the refined products from refinery i to market depot k by transport mode r, t
Xj,k,rAmin Minimum transport capacity of the refined products from transit depot j to market depot k by transport mode r, t
Xi,j,rCMax Previous maximum transport capacity from refinery i to transit depot j by transport mode r before expansion, t
Xi,k,rCMax Previous maximum transport capacity from refinery i to market depot k by transport mode r before expansion, t
Xj,k,rCMax Previous maximum transport capacity from transit depot j to market depot k by transport mode r, t
PgE The expanded transport capacity for level g, t
Si,rCMax Previous maximum sending capacity of refinery i by transport mode r before expansion, t
Sj,rCMax Previous maximum sending capacity of transit depot j by transport mode r before expansion, t
PhS The expanded sending capacity for level h, t
Cr,hS Cost for expanding sending capacity for transport mode r in h level, CNY
Cr,gE Cost for expanding transport capacity for transport mode r in g level, CNY
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
EU project “Sustainable Process Integration Laboratory – SPIL”, project No. CZ.02.1.01/0.0/0.0/15_003/0000456 funded by EU “CZ Operational Programme Research, Development and Education”, Priority 1: Strengthening capacity for quality research.
==== Refs
References
Alves D.T.S. Lima G.B.A. Establishing an onshore pipeline incident database to support operational risk management in Brazil - Part 2: bowtie proposition and statistics of failure Process Saf. Environ. Protect. 155 2021 80 97
Emenike S.N. Falcone G. A review on energy supply chain resilience through optimisation Renew. Sustain. Energy Rev. 134 2020 110088
Etminanfar M.R. Safavi M.S. Abbasian-Vardin N. An analysis of magnetic pumps failure used in the oil refinery industry: metallurgical point of view Eng. Fail. Anal. 112 2020 104533
Gurobi Optimization LLC, 2021. Gurobi Optimizer Reference Manual. < https://www.gurobi.com >(accessed 25 October 2021)
Isimite J. Rubini P. A dynamic HAZOP case study using the Texas City refinery explosion J. Loss Prev. Process Ind. 40 2016 496 501
Jiang P. Fan Y.V. Klemeš J.J. Impacts of COVID-19 on energy demand and consumption: challenges, lessons and emerging opportunities Appl. Energy 285 2021 116441
Klemeš J.J. Jiang P. Fan Y.V. Bokhari A. Wang X.C. COVID-19 pandemics Stage II – Energy and environmental impacts of vaccination Renew. Sustain. Energy Rev. 150 2021 111400 10.1016/j.rser.2021.111400
Kumar B. Sharma A. Managing the supply chain during disruptions: developing a framework for decision-making Ind. Mark. Manag. 97 2021 159 172
Lima C. Relvas S. Barbosa-Póvoa A. Designing and planning the downstream oil supply chain under uncertainty using a fuzzy programming approach Comput. Chem. Eng. 151 2021 107373
Lima C. Relvas S. Barbosa-Póvoa A. Stochastic programming approach for the optimal tactical planning of the downstream oil supply chain Comput. Chem. Eng. 108 2018 314 336
Navarro, M., 2021. Gas shortage: what to know about the Colonial Pipeline for Wednesday Gas shortage: what to know about the Colonial Pipeline for Wednesday. <https://www.greenvilleonline.com/story/news/local/south-carolina/2021/05/12/colonial-fuel-pipelines-offline-after-cyber-attack-what-know-update/5042759001/> (accessed 7 October 2021).
Norouzi N. Zarazua de Rubens G. Choupanpiesheh S. Enevoldsen P. When pandemics impact economies and climate change: exploring the impacts of COVID-19 on oil and electricity demand in China Energy Res. Soc. Sci. 68 2020 101654
Palouj M. Lavaei Adaryani R. Alambeigi A. Movarej M. Safi Sis Y. Surveying the impact of the coronavirus (COVID-19) on the poultry supply chain: a mixed methods study Food Control 126 2021 108084
Pudasaini P. Integrated planning of downstream petroleum supply chain: a multi-objective stochastic approach Oper. Res. Perspect. 8 2021 100189
Shafeek H. Soltan H.A. Abdel-Aziz M.H. Corrosion monitoring in pipelines with a computerised system Alex. Eng. J. 60 2021 5771 5778
Shaluf I.M. Ahmadun F.-.R. Said A.M. Fire incident at a refinery in West Malaysia: the causes and lessons learned J. Loss Prev. Process. Ind. 16 2003 297 303
Wang B. Fan Y.V. Chin H.H. Klemeš J.J. Liang Y. Emission-cost nexus optimisation and performance analysis of downstream oil supply chains J. Clean. Prod. 266 2020 121831
Wang B. Klemeš J.J. Zheng T. Liang Y. A fair profit allocation model for the distribution plan optimisation of refined products supply chains Comput. Aid. Chem. Eng. 50 2021 1847 1852 2021
Wang B. Liang Y. Zheng T. Yuan M. Zhang H. Optimisation of a downstream oil supply chain with new pipeline route planning Chem. Eng. Res. Des. 145 2019 300 313
Wang B. Zhang H. Yuan M. Guo Z. Liang Y. Sustainable refined products supply chain: a reliability assessment for demand-side management in primary distribution processes Energy Sci. & Eng. 8 2020 1029 1049
Wang S. Peng X. Lü Q. Long F. Jiang Y. Meng Y. Design and implementation of the overall architecture of the Puguang intelligent gas-field project Nat. Gas Ind. B 6 2019 262 271
Xie S. Dong S. Chen Y. Peng Y. Li X. A novel risk evaluation method for fire and explosion accidents in oil depots using bow-tie analysis and risk matrix analysis method based on cloud model theory Reliab. Eng. Syst. Saf. 215 2021 107791
Xinhua News Agency Major Oil Pipeline in Quake Region Back in Service 2008 <www.china.org.cn/business/2008-05/15/content_15243226.htm> (accessed 7 October 2021)
Xu, M., Kelly, S., Sharafedin, B., 2021. Weekend driving in U.S., China pave way for gasoline market recovery. Reuters, 27 April 2021, <https://www.reuters.com/world/china/weekend-driving-us-china-pave-way-gasoline-market-recovery-2021-04-27/> (accessed 25 October 2021)
Yuan M. Zhang H. Long Y. Shen R. Wang B. Liang Y. Economic, energy-saving and carbon-abatement potential forecast of multi-product pipelines: a case study in China J. Clean. Prod. 211 2019 1209 1227
Yuan M. Zhang H. Wang B. Huang L. Fang K. Liang Y. Downstream oil supply security in China: policy implications from quantifying the impact of oil import disruption Energy Policy 136 2020 111077
Yuan M. Zhang H. Wang B. Shen R. Long Y. Liang Y. Future scenario of China's downstream oil supply chain: an energy, economy and environment analysis for impacts of pipeline network reform J. Clean. Prod. 232 2019 1513 1528
Yuan M. Zhang H. Wang B. Zhang Y. Zhou X. Liang Y. Future scenario of China's downstream oil reform: improving the energy-environmental efficiency of the pipeline networks through interconnectivity Energy Policy 140 2020 111403
Zhang D. A network economic model for supply chain versus supply chain competition Omega (Westport) 34 3 2006 283 295
Zheng J. Liang Y. Xu N. Wang B. Zheng T. Li Z. Deeppipe: a customised generative model for estimations of liquid pipeline leakage parameters Comput. Chem. Eng. 149 2021 107290
Zheng T. Wang B. Rajaeifar M.A. Heidrich O. Zheng J. Liang Y. How government policies can make waste cooking oil-to-biodiesel supply chains more efficient and sustainable J. Clean. Prod. 263 2020 121494
Zhou X. van Gelder P.H.A.J.M. Liang Y. Zhang H. An integrated methodology for the supply reliability analysis of multi-product pipeline systems under pumps failure Reliab. Eng. Syst. Saf. 204 2020 107185
Zhou X. Zhang H. Qiu R. Lv M. Xiang C. Long Y. A two-stage stochastic programming model for the optimal planning of a coal-to-liquids supply chain under demand uncertainty J. Clean. Prod. 228 2019 10 28
| 0 | PMC9727701 | NO-CC CODE | 2022-12-08 23:18:53 | no | 2021 Dec 2; 1(4):468-475 | utf-8 | null | null | null | oa_other |
==== Front
J Am Acad Dermatol
J Am Acad Dermatol
Journal of the American Academy of Dermatology
0190-9622
1097-6787
Published by Elsevier on behalf of the American Academy of Dermatology, Inc.
S0190-9622(22)03172-3
10.1016/j.jaad.2022.10.062
Article
Demographics and Disease Associations of Patients with Monkeypox and Recipients of Monkeypox Vaccine from Safety Net Hospitals in New York City: A Cross-Sectional Study
Cline Abigail MD, PhD abc
Marmon Shoshana MD, PhD acd∗
a New York Medical College, Valhalla, New York
b Metropolitan Medical Center, Department of Dermatology, New York, New York
c Coney Island Hospital, Department of Dermatology, Brooklyn, New York
d Cumberland Diagnostic and Treatment Center, Department of Dermatology, Brooklyn, New York
∗ Corresponding Author: Shoshana Marmon MD, PhD, Coney Island Hospital Dermatology Department, 2601 Ocean Pkwy, Brooklyn, NY 11235
5 12 2022
5 12 2022
23 8 2022
13 10 2022
19 10 2022
© 2022 Published by Elsevier on behalf of the American Academy of Dermatology, Inc.
2022
Elsevier has created a Monkeypox Information Center (https://www.elsevier.com/connect/monkeypox-information-center) in response to the declared public health emergency of international concern, with free information in English on the monkeypox virus. The Monkeypox Information Center is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its monkeypox related research that is available on the Monkeypox Information Center - including this research content - immediately available in publicly funded repositories, with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the Monkeypox Information Center remains active.
Key words
Orthopox
Monkeypox
HIV
Hepatitis C
Syphilis
Vaccination
Skin of Color
Safety net
==== Body
pmcFundingsources: None
IRB status: Exempt
Conflict of Interests and Financial Disclosures: There are no conflicts or competing interests to disclose.
| 36481377 | PMC9727738 | NO-CC CODE | 2022-12-08 23:18:53 | no | J Am Acad Dermatol. 2022 Dec 5; doi: 10.1016/j.jaad.2022.10.062 | utf-8 | J Am Acad Dermatol | 2,022 | 10.1016/j.jaad.2022.10.062 | oa_other |
==== Front
Cell Rep Med
Cell Rep Med
Cell Reports Medicine
2666-3791
The Authors.
S2666-3791(22)00407-4
10.1016/j.xcrm.2022.100843
100843
Article
A genetically engineered, stem-cell-derived cellular vaccine
Cooper Amanda 2
Sidaway Adam 2
Chandrashekar Abishek 3
Latta Elizabeth 2
Chakraborty Krishnendu 2
Yu Jingyou 3
McMahan Katherine 3
Giffin Victoria 3
Manickam Cordelia 3
Kroll Kyle 3
Mosher Matthew 3
Reeves R. Keith 3
Gam Rihab 2
Arthofer Elisa 2
Choudhry Modassir 12
Henley Tom 12∗
Barouch Dan H. 345∗∗
1 Praesidium Bioscience, Inc., New York, NY, USA
2 Intima Bioscience, Inc., New York, NY, USA
3 Center for Virology and Vaccine Research, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA
4 Ragon Institute of Massachusetts General Hospital, Massachusetts Institute of Technology, and Harvard University, Cambridge, MA, USA
∗ Corresponding author
∗∗ Corresponding author
5 Lead contact
7 12 2022
7 12 2022
1008434 7 2022
19 10 2022
10 11 2022
© 2022 The Authors
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Despite rapid clinical translation of COVID-19 vaccines in response to the global pandemic, an opportunity remains for vaccine technology innovation to address current limitations and meet challenges of inevitable future pandemics. We describe a universal vaccine cell (UVC) genetically engineered to mimic natural physiological immunity induced upon viral infection of host cells. Cells engineered to express the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike as a representative viral antigen induce robust neutralizing antibodies in immunized non-human primates. Similar titers generated in this established non-human primate (NHP) model have translated into protective human neutralizing antibody levels in SARS-CoV-2-vaccinated individuals. Animals vaccinated with ancestral spike antigens and subsequently challenged with SARS-CoV-2 Delta variant in a heterologous challenge have an approximately 3 log decrease in viral subgenomic RNA in the lungs. This cellular vaccine is designed as a scalable cell line with a modular poly-antigenic payload, allowing for rapid, large-scale clinical manufacturing and use in an evolving viral variant environment.
Graphical abstract
Cooper et al. describe a genetically engineered cellular vaccine designed to recapitulate natural physiological immunity induced upon viral infection of host cells. The scalable and poly-antigenic vaccine technology can induce robust and durable antibody responses to SARS-CoV-2 virus as a representative model of viral infection.
Key words
COVID-19
SARS-CoV-2
CRISPR
stem cell
iPSC
vaccine
vaccinology
immuno-vaccinology
universal cell
Published: December 7, 2022
==== Body
pmcIntroduction
The COVID-19 pandemic has demonstrated the urgent need for new innovations in vaccinology to enable the rapid development of novel vaccines against emerging viral variants that engender robust and long-lasting immune protection. The unprecedented success of both mRNA and adenoviral vaccines has established the capability of a rapid global vaccination program.1 , 2 , 3 However, the waning antibody responses seen with these emergency-use-authorized vaccine technologies, and the need for vaccine boosters, has highlighted the requirement for further improvements in vaccine approaches to drive higher, longer-lasting protective immunity.4 , 5 , 6 , 7 , 8 , 9 The newly emerging viral variants of SARS-CoV-2, and the evident reduced efficacy of the existing vaccines to protect against transmissible and symptomatic infection of these variants, also highlights the need for vaccines that can ideally deliver multiple variant antigens (polyvalency) and be rapidly manufactured at scale as soon as new viral variants are discovered.10 , 11 , 12 , 13
Theoretically, an ideal vaccine technology would have four core attributes, namely robust immunity, self-adjuvancy, polyvalency, and scalability. Immunity is self-evident and speaks to the requirement of generating robust humoral neutralizing antibody and ideally T cell responses that are durable. Self-adjuvancy, or conversely the absence of the need for exogenous excipients to elicit a robust immune response, may prove to be a meaningful innovation in that the immune side effects of current vaccines may be mediated by the non-target-antigen-specific adjuncts.14 Thirdly, polyvalency, or the ability to protect against multiple immunodominant epitopes, is a core feature of overlapping and orthogonal mechanisms of protection.15 , 16 Lastly, scalability or the ability to deliver preventative doses of vaccines in an economic, large-scale, and clinically relevant fashion in both the developed and developing worlds is a sine qua non of any human vaccine.
Current mRNA-, protein-, and viral-vector-based vaccines have certain limitations, such as their requirement for excipient adjuvants to activate the recipient immune system or to deliver the viral antigenic payload.17 , 18 These include the artificial lipid nanoparticles delivering the mRNA, or MF59, AS03, Alum, ISCOMATRIX, and Matrix-M chemical emulsions, for example, or the adenoviral protein antigens themselves that stimulate innate immune cell activation.18 , 19 , 20 , 21 , 22 , 23 , 24 Adjuvants are required to increase the effectiveness of vaccines, and their use can cause side effects including local reactions (redness, swelling, and pain at the injection site) and systemic reactions (fever, chills, and body aches).25 , 26 , 27
The size constraint of the adenoviral vector genome, and the limited length of stable mRNA that can be produced and packaged into nanoparticles, restricts the number and size of nucleic-acid-encoded antigens and epitopes that can be delivered in these vaccines.28 Thus, these vaccines are constrained in their ability to provide multiple immunodominant proteins to address emerging pandemic variants or to easily combine multiple pathogens into one vaccine.
To address some of the current limitations of vaccine technologies, we have developed a vaccine platform based on a CRISPR genetically engineered human stem cell, termed the universal vaccine cell (UVC). The principal feature of this vaccine platform is to attempt to reproduce physiologic immunity that is engendered naturally through lytic viral infection and the resulting apoptosis of primary human cells. The platform is designed to deliver an antigenic payload within the context of a physiological apoptotic environment to both release antigen and simultaneously stimulate the host immune response. Here, we use the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus as a rigorous test platform to demonstrate that this self-adjuvanting, polyvalent UVC can generate a robust and antigen-specific humoral immune response in vaccinated macaques. This vaccine resulted in reduced viral loads in animals challenged with heterologous SARS-CoV-2 Delta variant, which is consistent with current clinical experience with vaccines encoding the WA1/2020 spike against SARS-CoV-2 variants.29 , 30 , 31
Results
Genetic engineering of iPSCs to create a cellular vaccine to deliver the SARS-CoV-2 spike antigen
To create a cellular vaccine platform to deliver viral antigens and simultaneously engage host innate immune cells to present these antigens to lymphocytes, we attempted to create a cell with a robust immunogenic phenotype. We selected human induced pluripotent stem cells (iPSCs) as the UVC cell line due to their stable genetics, non-transformed phenotype, ease of genetic engineering, and capacity for rapid, scalable propagation.32 , 33 iPSCs also retained the unique ability for programmable differentiation into any cell lineage, thus retaining the future opportunity to explore differentiation of the UVC into different cell types that may have enhanced immunogenic properties.34
We first genetically engineered iPSCs to create an immunogenic phenotype by stable integration of the SARS-CoV-2 full-length spike antigen into the AAVS1 safe-harbor locus using CRISPR-Cas9 gene editing (Figure 1A). We selected the original and well-characterized WA1/2020 variant of the SARS-CoV-2 spike antigen sequence with a mutation of the furin cleavage site and proline-stabilizing mutations that are identical to those in the current emergency-use-authorized vaccines being deployed globally to vaccinate against COVID-1935 , 36 , 37 (Figure S1). By including the spike transmembrane domain sequence in the gene encoding this antigen, we were able to detect high levels of the viral spike on the cell surface of the engineered iPSCs (Figure 1B). This high level of surface viral spike expression was maintained throughout the expansion of the cells and cryopreservation several weeks after CRISPR engineering (data not shown). Spike protein was also readily observed in engineered cell lysates when measured by western blotting (Figure 1C). The yield of antigen released upon lysis was quantified using a spike-specific ELISA assay, and we observed an abundant and dose-dependent release of protein from the cells, which would equate to approximately ∼20 μg spike antigen protein delivered in a 108 UVC dose (Figure 1D).Figure 1 CRISPR genetic engineering of an iPSC line to create an immunogenic, self-adjuvanting cellular vaccine
(A) Universal vaccine cell CRISPR genetic engineering strategy to create an apoptotic cellular vehicle for antigen delivery.
(B and C) Representative flow cytometric analysis showing expression of SARS-CoV-2 WA1/2020 spike protein on the cell surface (B) and by western blot showing spike protein within UVC whole-cell lysates (C).
(D) ELISA quantification of spike protein released upon UVC lysis for 2 independent UVC cultures.
(E) Proportion of apoptotic cells at 24 and 72 h post-irradiation as measured by 7-AAD staining and flow cytometry.
(F) Morphology, observed by light microscopy, of engineered UVC during expansion culture and, when reseeded into culture 24 h after irradiation, showing apoptosis and cell death.
(G) Absence of detectable proliferation of irradiated UVC as determined by CellTrace yellow proliferation dye staining and measuring the dilution of the dye by flow cytometry over 72 h.
(H) Representative flow cytometric analysis showing deletion of MHC class I and overexpression of MICA on the UVC surface by CRISPR engineering.
(I) Cell counts of 3 independent cultures showing exponential expansion of live engineered UVCs over 21 days in culture.
(J) Relative expression of pluripotency and self-renewal genes by UVC and the control iPSCs from which they were derived, as measured by quantitative PCR, showing maintenance of an iPSC gene expression profile after genetic engineering and expansion. Error bars represent mean ± SEM of 3 technical replicas. All experiments were repeated at least three times.
To ensure robust delivery of this immunodominant antigen to the recipient immune system, we incorporate an apoptosis-inducing lethal irradiation step during vaccine manufacture by exposing the UVCs to a 10 Gy dose of gamma radiation prior to cryopreservation and vaccination. Thus, when subjects are immunized with the UVC, we reasoned that the cells would undergo apoptosis and release the SARS-CoV-2 spike antigen into the immune microenvironment via production of apoptotic bodies (Figure 1A). In theory, these apoptotic bodies would be phagocytosed by innate immune cells and antigen-presenting cells and be presented to T and B lymphocytes to generate a spike-antigen-specific immune response.
In addition to creating a mechanism for delivery of immunogenic antigens via apoptotic bodies, the irradiation of the UVC is also a safety feature, as it renders the cells unable to proliferate or persist in vivo upon vaccination. In support of this, we observed a robust elevation in the proportion of apoptotic cells after 24 and 72 h of culture of irradiated UVC using both apoptotic dyes and flow cytometry (Figure 1E) and by observation of cell morphology under the microscope (Figure 1F). Furthermore, unlike non-irradiated UVCs, irradiation prevented any detectible proliferation of the cells over 72 h in culture as measured by proliferation dyes using flow cytometry (Figure 1G) and by colony-formation assays (Figure S2). To confirm the absence of proliferation post-irradiation in vivo, we transplanted irradiated and non-irradiated UVCs into immunocompetent mice, monitored for teratoma formation over 6 weeks, and showed that neither irradiated nor non-irradiated cells formed tumors in any of the mice evaluated (Figure S3).
Incorporation of NK cell activation signals by genetic engineering to create a self-adjuvanting vaccine cell
In addition to the proposed immunogenicity expected from apoptosis and release of immunogenic antigens upon vaccination, we attempted to increase the immunogenic potential by incorporating a self-adjuvanting phenotype into the UVC. As a form of physiological cell death, apoptosis is generally non-inflammatory.38 Therefore, to promote effective local inflammation and engage the innate immune system that can mobilize effector cells, we engineered the UVC to mimic a virally infected cell to be recognized and rapidly lysed by host innate immune cells, principally natural killer (NK) cells.39 , 40 Many viruses attempt to evade immune recognition by limiting major histocompatibility complex (MHC) class I cell surface expression to reduce the presentation of viral antigens to CD8+ T cells.41 , 42 This “missing-self’ signal can aid in the activation of NK cells and promote cytolysis, and therefore the iPSCs were engineered to completely remove MHC class I molecules from the cell surface via CRISPR knockout of the β2 microglobulin (B2M) gene, a critical component of MHC class I molecules (Figure 1H).
In vivo, lack of MHC class I on the target cell is not sufficient to trigger full NK cell activation alone and a further hallmark of cells undergoing stress or viral infection, is the expression of NK cell activating natural killer group 2 member D (NKG2D) ligands on their cell surface.43,44 Therefore, the UVC was further engineered to integrate a gene expression cassette in a safe-harbor locus to drive constitutive expression of the human MICA gene (MHC class I polypeptide-related sequence A), a potent activator of NK cells. Using flow cytometry, abundant levels of MICA could be detected on the surface of the engineered UVC (Figure 1H).
Rapid growth kinetics of engineered UVC
Prior to irradiation and cryopreservation of the UVC ready for immunization, we evaluated the growth kinetics of the cells to confirm the capacity for rapid, scalable proliferation that would be needed for a vaccine technology to address the needs of a pandemic. iPSCs are known to have relatively short doubling times in the range of 18–20 h,45 , 46 and we observed similar kinetics with an average exponential growth of >50-fold over a 7 day culture period (Figure 1I). Thus, from a starting UVC number of 106 cells, the vaccine can be theoretically expanded to provide millions of doses in under 8 weeks and even more quickly if adapted to bioreactor manufacturing.
The consistent rapid cell growth of the UVC and the morphological similarity to unmodified iPSCs suggested the UVC exhibited characteristics of the iPSCs from which they are derived. We thus assessed the stem cell characteristics of the UVC after genetic engineering and rapid expansion to confirm that the cells have retained their original stemness gene expression signatures without acquiring any detectible or obvious changes in phenotype beyond those introduced by genetic engineering. The expanded UVC expressing the SARS-CoV-2 spike antigen, human MICA ligand, and CRISPR knockout of B2M showed a similar level of expression of three important pluripotent transcription factors, NANOG, OCT4, and SOX2, suggesting that they have retained a stem-cell-like transcriptional profile (Figure 1J). Engineered UVCs also showed similar expression to control iPSCs for genes (DCN, vimentin, HES5, and GATA6) that are known to increase in expression as iPSCs differentiate into mesoderm, endoderm, and ectoderm lineages, confirming that the UVCs have a consistent undifferentiated iPSC gene expression profile, morphology, and growth characteristics.47 , 48
Human and primate NK cell cytolysis of UVCs
To further explore the impact of MHC class I loss and overexpression of NK cell ligands on recognition and killing of the UVCs by NK cells, we performed a series of in vitro NK cell activation and cytolysis assays. When MHC class I was removed via B2M knockout alone, the UVCs were robustly killed by human NK cells, which increased in an E:T (effector:target) ratio-dependent manner (Figure 2A). We compared the level of UVC cytolysis with that observed with the MHC-deficient K562 leukemia cell line, known to be potent targets for NK cell killing, and found a similar level of cytolysis confirming that the MHC class I-deficient UVCs are readily targeted by NK cells. We extended this analysis to macaque NK cells and found that while control iPSCs (expressing MHC class I) show low levels of killing, the MHC class I knockout UVCs were lysed more readily by the NK cells (Figure 2B).Figure 2 Self-adjuvancy: Enhanced cytolysis of genetically engineered UVC iPSCs via engineered MHC class I deletion and NK apoptotic ligand expression
(A) CRISPR knockout of B2M and loss of MHC class I enhances the killing of UVC cells by human primary NK cells, showing equivalent levels of cytolysis seen with the MHC class I-deficient K562 cell line in 3 independent assays.
(B) A similar elevated cytolysis of MHC-deficient UVC cells is observed with macaque NK cells.
(C) When overexpressed transiently on the UVC, NKG2D family ligands show no elevation in markers of NK cell activation by macaque NK cells, except MICA, which significantly elevates levels of macrophage inflammatory protein-1β (MIP-1β).
(D) When stably overexpressed on the UVC by CRISPR editing, MICA enhanced the NK cell functional responses as measured by ICS of NK cells from 7 macaques. ∗p < 0.05, ∗∗p < 0.01, error bars represent mean ± SEM of 3 biological replicates.
To assess the relative contribution of overexpressing NK activating ligands on UVC cytolysis by macaque NK cells, we performed an analysis of UVCs transiently overexpressing different NKG2D ligands, including MICA, MICB, and UL16 binding protein 1 (ULBP1). While the level of macrophage inflammatory protein-1β (MIP-1β) was significantly elevated when MICA was overexpressed, proinflammatory and activation markers for NK cells were generally the same regardless of ligand overexpression (Figure 2C). With stable overexpression of MICA by CRISPR engineering, we confirmed a significant increase in total responding macaque NK cells and a significant elevation in MIP-1β (Figure 2D).
Immunogenicity of UVCs in macaques
To evaluate the immunogenicity of the UVCs, we immunized cynomolgus macaques and followed neutralizing and spike-specific antibodies for 6 months. We immunized 9 macaques, aged 6–12 years old, with either 107 (n = 3) or 108 UVCs (n = 3) expressing the WA1/2020 SARS-COV-2 spike and sham controls (n = 3). Macaques were primed by the intramuscular route without adjuvant at week 0 and were boosted at week 6 (Figure 3A). Neutralizing antibody responses were assessed using a pseudovirus neutralization assay.49 , 50 , 51 , 52 We observed neutralizing antibodies in all UVC vaccinated macaques at week 2 and higher levels at week 4 (Figure 3B). The higher dose of 108 UVCs resulted in the most robust titers of neutralizing antibodies at all time points tested. Following the boost immunization at week 6, neutralizing antibody titers increased further, reaching titers close to 1,000 with the 108 cell dose. Six months after the initial UVC immunization, neutralizing antibodies showed a durable response, particularly for macaques immunized with the 108 UVC dose. We also observed robust spike-specific and receptor-binding domain (RBD)-specific antibody titers as measured by enzyme-linked immunosorbent assay (ELISA) in vaccinated animals (Figures 3C and 3D). At 6 months after immunization, detectible levels of neutralizing antibodies against Beta and Delta variants were also observed, albeit lower than seen with the immunizing antigen variant WA1/2020 spike, suggesting that humoral immunity is also generated against SARS-CoV-2 variants (Figure 3E).Figure 3 Humoral immune responses in UVC-vaccinated macaques
(A) Macaques received a high WA1/2020 spike expressing UVC prime dose (108) or low UVC prime dose (107) at week 0 and a boost dose matched to that of the prime dose at week 6.
(B–D) Humoral immune responses were assessed at 2 week intervals up to week 10 and then again at weeks 24 and 26 by (B) pseudovirus neutralization assays and (C) RBD-specific and (D) spike-binding antibody ELISA.
(E) In addition to the WA1/2020 SARS-CoV-2 variant, detectible neutralizing antibodies against the B.1.351 (Beta) and B.1.617.2 (Delta) variants were observed in immunized macaques at weeks 24 and 26. Red bars reflect median responses. Dotted lines reflect assay limit of quantification. Data points represent individual primates, 3 per group. NAb, neutralizing antibody.
Protective efficacy against heterologous SARS-CoV-2 challenge
In a second macaque study, we immunized 12 rhesus macaques, aged 6–12 years old, with 108 UVCs (n = 6) expressing the SARS-CoV-2 WA1/2020 spike antigen and sham controls (n = 6) (Figure 4A). At week 8, the macaques were challenged with 105 50% tissue culture infectious dose (TCID50) of heterologous SARS-CoV-2 B.1.617.2 (Delta) by intranasal and intratracheal routes.51 , 52 In addition to measuring neutralizing antibodies (Figure S4), viral loads in bronchoalveolar lavage (BAL) and nasal swabs were assessed by reverse transcription PCR (RT-PCR) specific for subgenomic mRNA (sgRNA), which is thought to measure replicating virus.52 , 53 Sham controls showed a median peak of 5.39 (range 4.60–5.88) log10 sgRNA copies/mL in BAL samples (Figures 4B and 4D). In macaques immunized with the UVC, a significantly lower level of virus was detected in BAL samples, with a median peak of 2.78 (range 1.70–4.63) log10 sgRNA copies/mL, representing a 2.81 log reduction in virus in UVC-vaccinated animals (p = 0.0152). A significant 0.96 log reduction of virus was also observed in nasal swabs from UVC-immunized macaques compared with sham controls (Figures 4C and 4E; p = 0.0260). These data demonstrate that a two-dose regimen of UVC promoted antigen-specific antibody responses with levels of neutralizing antibodies and durability similar to current approved COVID-19 vaccines, and this can lead to partial protection against a heterologous SARS-CoV-2 Delta challenge.49 , 54 , 55 Figure 4 Viral loads in UVC-vaccinated macaques after heterologous SARS-CoV-2 challenge
(A) Rhesus macaques were immunized with 108 WA1/2020 spike expressing UVCs at week 0 and received a boost dose of 108 matched UVCs at week 4. Macaques were then challenged at week 6 by intranasal and intratracheal routes with 1.0 × 105 TCID50 of SARS-CoV-2 B.1.617.2 (Delta).
(B and C) Log10(sgRNA [copies per mL]) (limit of quantification 50 copies per mL) were assessed, and peak viral loads are shown in (B) bronchoalveolar lavage (BAL) samples and (C) nasal swabs (NSs) in sham controls and vaccinated macaques after challenge.
(D and E) Viral loads were assessed every 2 days. Dotted lines reflect assay limit of quantification. Data points represent individual primates, 6 per group. NAb, neutralizing antibody.
Discussion
The global need for improved vaccine technologies to meet future pandemics is driving a renaissance of innovation in vaccinology. COVID-19 has demonstrated the rapid pace at which viral mutations can accumulate and new variants emerge that can escape the protective efficacy of existing vaccines designed against earlier viral antigen sequences.10 , 11 , 29 , 30 , 31 , 56 , 57 To address the need for novel vaccine technologies, we developed a UVC platform technology to generate immunity via self-adjuvancy through apoptosis and NK-cell-mediated cytolysis within the immune microenvironment.
Our data demonstrate that the UVC vaccine platform can induce robust neutralizing antibody responses in macaques when delivering the SARS-CoV-2 WA1/202058 spike that contains a mutation of the furin cleavage site and two proline-stabilizing mutations.35 , 36 , 37 The neutralizing antibody titers around 1,000 are similar to those reported for the current COVID-19 mRNA vaccines.49 , 54 , 55 Following a high dose, heterologous SARS-CoV-2 Delta challenge, the UVC vaccine reduced viral loads 2.81 logs in the BAL and 0.96 logs in nasal swabs. The more robust protection in the lower respiratory tract compared with the upper respiratory tract is consistent with clinical data showing that all current COVID-19 vaccines are better at protecting against severe disease than infection with emerging SARS-CoV-2 variants.29 , 31 A prior study showed that the Moderna mRNA-1273 vaccine resulted in an approximate 3 log reduction in viral loads in BAL and a 1 log reduction in nasal swabs against a heterologous SARS-CoV-2 Delta challenge in macaques, which appears similar to our data with UVC.
Regarding safety, the UVC undergoes lethal irradiation during manufacture and rapid apoptosis in the immune microenvironment upon vaccination. This is the principal mechanism of efficacy of the UVC and provides an important safety feature with no detectable persistence or teratogenicity. The irradiation-induced apoptosis is further enhanced by CRISPR genetic engineering to remove MHC class I expression and introduce cell surface expression of the NKG2D ligand MICA, making the UVCs potent targets for host NK cells. Recruited NK cells will likely recognize the UVCs as virally infected cells through MHC class I absence and MICA activation of NKG2D signaling to mediate a direct killing effect and release of protein antigen.59 The apoptosis- and NK-mediated cytolysis enables the UVC to be a self-adjuvanting vaccine vector without the need for additional chemicals adjuvants or additional foreign antigens. Thus, the UVC may mimic the physiological engagement of the immune system typical of virally infected cells within the tissues of an individual suffering with the disease.
The CRISPR genetic engineering to render the UVC highly immunogenic and self-adjuvanting also presents a unique opportunity to address antigen polyvalency. Unlike mRNA or DNA vaccines or recombinant viral vector vaccines, which have limits on the size or number of independent encoded antigens they can deliver, the UVC can be engineered to deliver a higher number of full-length protein antigens. Thus, there is the ability to create polyvalency against multiple epitopes in a rapid modular gene cassette fashion through CRISPR engineering of the iPSC genome.
A perceived limitation of the UVC technology may be the seemingly complex and costly nature of developing and manufacturing a human cell as a vector for vaccination at scale. However, the UVC is a cell line, not a complex cell therapy, and can thus be scaled within appropriate parameters for such a biologic agent. The UVC cell line can be expanded rapidly to scale with predictable growth kinetics and quality assurance/quality control (QA/QC) controls. The modular nature of the UVC and the ability to integrate emerging viral antigens into the cellular genome using CRISPR can allow scalable manufacture of new polyvalent vaccines to address emerging variants. In fact, the genetic engineering of the UVCs can be accomplished in a matter of weeks prior to exponential cell culture expansion to create millions of clinical doses. Moreover, quantification of viral spike protein released from lysed UVCs suggests that a 108 UVC number can deliver an antigen dose comparable to, and in excess of, that administered by other approved COVID-19 protein vaccines. At the 108 UVC dose, billions of doses could theoretically be generated under good manufacturing practice (GMP) conditions in under 8 weeks.
In summary, our data establish a cellular vaccine platform and demonstrate that immunization with UVC expressing the WA1/2020 SARS-CoV-2 spike elicits robust neutralizing antibody responses that provide partial protection against heterologous Delta SARS-CoV-2 challenge in rhesus macaques. This platform offers a unique class of gene and cell therapy prophylaxis for potential future viral pandemics.
Limitations of the study
One limitation of this study is that we have yet to observe the generation of a robust T cell response in animals vaccinated with the UVC. The measurable, albeit modest, CD8+ T cell responses seen with the adenoviral and mRNA vaccines for COVID-19 have not resulted in neutralizing antibody (nAb) titers and duration of protection longer than 6–9 months.8 , 60 One potential hypothesis to begin to establish a clinically meaningful cellular response is to explore in future studies non-spike antigens, leveraging the simultaneous polyvalency of the UVC platform, including immunodominant T cell epitopes such as those in the nucleocapsid and viral accessory proteins.61 , 62 , 63
Another limitation of this study is that investigation into persistence of irradiated UVCs in animal models was limited to immunocompetent mice, selected because they represent the target population of this vaccine technology: healthy individuals with an intact immune system. Immunocompromised animals lacking cells of the adaptive immune system may not clear injected cells as effectively, and although lethal irradiation of the UVC designed to induce apoptosis and prevent cell survival is likely to prevent any persistence and proliferation of cells in recipients, this has yet to be addressed experimentally.
STAR★Methods
Key resources table
REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
Rabbit Anti-SARS-CoV-2 Spike Glycoprotein S1 antibody Abcam Cat# ab275759; RRID:AB_2892127
Goat Anti-Rabbit IgG H&L Alexa 488 Abcam Cat# ab150077; RRID:AB_2630356
Mouse anti-human MICA/MICB PE Biolegend Cat# 320906; RRID:AB_493193
Mouse anti-human HLA-A,B,C Alexa 647 Biolegend Cat# 311416; RRID:AB_493136
Rabbit anti-SARS-Cov2 Sino Biological Cat# 40591-T62; RRID:AB_2893171
Mouse anti-human b-actin Abcam Cat# ab8226; RRID:AB_306371
Goat anti-rabbit HRP Sino Biological Cat# SSA003; RRID:AB_2814815
Mouse anti-human ULBP-1 R & D Systems Cat# MAB1380; RRID:AB_2214683
Mouse anti-human CD107a BD Biosciences Cat# 561343; RRID:AB_10644020
Mouse anti-human CD3 BV421 BD Biosciences Cat# 562877; RRID:AB_2737860
Mouse anti-human CD14 BV650 BD Biosciences Cat# 563420; RRID:AB_2744286
Mouse anti-human CD16 BV496 BD Biosciences Cat# 612945; RRID: AB_2870224
Mouse anti-human CD20 BV570 BD Biosciences Cat# 741210; RRID:AB_2870766
Mouse anti-human CD56 BV605 BD Biosciences Cat# 742659; RRID:AB_2740950
Mouse anti-human HLA-DR APC-H7 BD Biosciences Cat# 561358; RRID:AB_10611876
Mouse anti-human NKG2A PE-Cy7 Beckman Coulter Cat# B10246; RRID:AB_2687887
Mouse anti-human MIP1β FITC BD Biosciences Cat# 560565; RRID:AB_1645489
Mouse anti-human interferon-ɣ BV395 BD Biosciences Cat# 563563; RRID:AB_2738277
Mouse anti-human TNF-α BV650 BD Biosciences Cat# 563418; RRID:AB_2738194
anti-macaque IgG HRP NIH NHP Reagent Program Cat# 1b3-HRP: 0320K235/070920SC
Goat anti-Mouse HRP Abcam Cat# ab205719; RRID:AB_2755049
Bacterial and virus strains
SARS-CoV-2 B.1.617.2 (Delta variant) BEI Resource N/A
Biological samples
NK cells (Macaca fascicularis) BIDMC N/A
Bronchoalveolar lavage from Non-Human Primates Bioqual, Inc. N/A
Nasal swabs from Non-Human Primates Bioqual, Inc. N/A
EDTA, SST, Paxgene collection tubes with whole blood, from Non-Human Primates Bioqual, Inc. N/A
Formalin fixed paraffin embedded skin sections excised from the left flank of C57BL/6J mice Propath UK N/A
Chemicals, peptides, and recombinant proteins
Cas9 protein IDT Cat# 1081066
VEG-F protein Peprotech Cat# 100-20
Leukocyte activation cocktail BD Biosciences Cat# 550583
Vitronectin StemCell Technologies Cat# 100-0763
G418 Sigma-Aldrich Cat# 4727878001
Puromycin Sigma-Aldrich Cat# P9620
Accutase StemCell Technologies Cat# 07920_C
CryoStor-CS10 StemCell Technologies Cat# 07930_C
Proteinase K Promega Cat# MC5005
GoTaq G2 PCR master mix Promega Cat# M7422
Human Heat Inactivated AB Serum Sigma Cat# H3667
RIPA buffer ThermoFisher Cat# 89900
LDS Sample Buffer ThermoFisher Cat# NP0007
Mini-PROTEAN TGX Gel 4–15% Bio-Rad Cat# 4561083
Cell Extraction Buffer Invitrogen Cat# FNN0011
Protease inhibitor cocktail Sigma Cat# P8340
Crystal Violet Thermo Scientific Cat# 212121000
Calcein acetoxymethyl ester CAM Invitrogen Cat# C3099
Critical commercial assays
QIAquick PCR Purification Kit Qiagen Cat# 28104
GoTaq G2 PCR mastermix Promega Cat# M7823
BD Cytofix/Cytoperm Fixation Kit ThermoFisher Cat# AB_2869008
Live/Dead Fixable Dead Cell Stains Invitrogen Cat# L23101
ReliaPrep RNA miniprep Promega Cat# Z6010
High-Capacity cDNA Reverse Transcription Kit Applied Biosystems Cat# 4368814
Brilliant III Ultra-Fast SYBR green qPCR mix Agilent Technologies Cat# 600882
Covid-19 S-protein ELISA kit Abcam Cat# ab284402
FITC Annexin V Apoptosis Kit with 7-AAD Biolegend Cat# 640922
Cell Trace Yellow Molecular Probes Cat#
CD3 cell depletion kit Miltenyi Biotech Cat# 130-050-101
Primary cell 4D nucleofector kit Lonza Cat# V4XP-3024
Superscript III VILO Invitrogen Cat# 11754050
KPL TMB SureBlue Start solution SeraCare Cat# 5120-0075
KPL TMB Stop solution SeraCare Cat# 5150-0022
Steady-Glo Luciferase Assay Promega Cat# E2510
Pierce BCA Assay ThermoFisher Cat# 23225
AmpliCap-Max T7 High Yield Message Maker Kit Cellscript Cat# C-ACM04037
Super-Signal West Femto kit ThermoFisher Cat# 34094
Experimental models: Cell lines
Human induced pluripotent stem cells Thermo Fisher Cat# A18945
HEK293T ATCC Cat# CRL-1573
HEK293T-hACE2 Chandrashekar et al.64 N/A
K562 ATCC Cat# CCL-243
Experimental models: Organisms/strains
C57BL/6J immunocompetent wildtype mice Charles River Laboratories N/A
Macaca mulatta Bioqual, Rockville, MD N/A
Macaca fascicularis Bioqual, Rockville, MD N/A
Oligonucleotides
Human PPP1R12C (AAVS1) sgRNA: GTCACCAATCCTGTCCCTAG This paper N/A
Human ROSAβgeo26 sgRNA: AAGTAATTAGGACTCACTCA This paper N/A
Human B2M sgRNA: AAGTCAACTTCAATGTCGGA This paper N/A
q-PCR primers for measurement of human stemness genes See Table S1 for oligonucleotide sequence N/A
Recombinant DNA
human MICA expressing plasmid This paper N/A
human MICB expressing plasmid This paper N/A
human ULBP1 expressing plasmid This paper N/A
SARS-CoV-2 WA1/2020 spike gene targeting vector This paper N/A
psPAX2 AIDS Resource and Reagent Program Cat# 11348
pLenti-CMV Puro-Luc Addgene Cat# 17477
pcDNA3.1-SARS CoV-2 SΔCT Chandrashekar et al.64 N/A
Software and algorithms
FlowJo 10 BD Biosciences https://www.flowjo.com/solutions/flowjo
Synthego ICE tool Synthego https://ice.synthego.com/
GraphPad Prism GraphPad Software https://www.graphpad.com/scientific-software/prism/
BioRender BioRender https://biorender.com/
Other
RNA Standard: SARS-CoV-2 E gene subgenomic RNA (sgRNA) Chandrashekar et al.64 N/A
Resource availability
Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the corresponding author, Dr. Dan H. Barouch ([email protected]).
Materials availability
The genetically modified iPS cells (UVC) generated in this study will be made available on request, but we may require a payment and/or a completed Materials Transfer Agreement if there is potential for commercial application.
Experimental model and subject details
Cell lines
Human iPS cells (Thermo Fisher) were cultured on vitronectin-coated T225cm2 flasks using complete mTesSR Plus medium (StemCell Technologies) supplemented with 1% penicillin/streptomycin, Rock inhibitor (StemCell Technologies) at 1:1000 dilution. For drug selection, G148 was used at 500ug/mL and puromycin at 5ug/mL (Sigma-Aldrich). Cultures were maintained at 37°C, 5% CO2 in a humidified incubator. NK cell effectors were enriched from normal cynomolgus macaque (Macaca fascicularis, male and female) blood samples using a CD3 depletion Kit (Miltenyi Biotec). NK cells were maintained in RPMI 1640 with Glutamax (Life Technologies) supplemented with 10% heat-inactivated FCS, 2 mM, l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin and 100 IU/mL of IL-2 and 10 ng/mL of IL-15.
Animals and study design
8- to 9-week-old female C57BL/6J immunocompetent wildtype mice (Charles River Laboratories) were randomly allocated to groups, housed at Crown Bioscience UK and acclimatized for 7-day. Irradiated or non-irradiated UVC were resuspended at 1 × 106 in 200 μL of PBS and injected subcutaneously into the left hind flank (10 animals per group). Animals were checked and weighed daily and measured for tumor growth 3 times a week in 2-dimensions using electronic calipers. Animals were sacrificed at 6 weeks and tissue surrounding the injection site was excised and prepared for histological analysis. Animal welfare for this study complies with the UK Animals Scientific Procedures Act 1986 (ASPA).
Outbred adult male and female rhesus macaques (M. mulatta) and cynomolgus macaques (M. fascicularis), 6–12 years old, were randomly allocated to groups. All macaques were housed at Bioqual. Macaques were treated with irradiated UVC at doses of either 1 × 107 or 1 × 108 cells (n = 3–6), and sham controls (n = 3–6). Prior to immunization, the cryopreserved doses of irradiated UVC were thawed at 37°C, then 900 μL of 1xPBS was added to each vial of 100 μL UVC in CryoStore freezing media. Macaques received a prime immunization of 1mL of UVC by the intramuscular route without adjuvant at week 0. At weeks 4 or 6, macaques received a boost immunization of either 1 × 107 or 1 × 108 UVC. At week 10 all macaques were challenged with 1.0 × 105 TCID50 (1.2 × 108 RNA copies, 1.1 × 104 PFU) SARS-CoV-2, which was derived from B.1.617.2 (Delta). Viral particle titers were assessed by RT–PCR. Virus was administered as 1 mL by the intranasal route (0.5 mL in each nare) and 1 mL by the intratracheal route. All immunological and virological assays were performed blinded. All animal studies were conducted in compliance with all relevant local, state, and federal regulations and were approved by the Bioqual Institutional Animal Care and Use Committee (IACUC).
Method details
iPS cell irradiation and cryopreservation
Harvesting of engineered UVC was performed using accutase (StemCell Technologies) and cells were counted using a CellDrop cell counter (De-Novix). Cells were irradiated at a total single dose of 10 Gy, before centrifugation at 300 xg for 10 min followed by resuspension in 100 μL of CryoStor-CS10 freezing media (StemCell Technologies). The UVC preparations for use in non-human primate studies were analyzed for endotoxin levels (Wickham Laboratories Ltd) and absence of mycoplasma (Mycoplasma Experience Ltd).
CRISPR genetic engineering
CRISPR sgRNAs targeting the human B2M gene, PPP1R12C (AAVS1), and the ROSAβgeo26 locus were designed and validated for indel formation at the selected genomic site. Up to 6 sgRNAs per target gene were tested and the most efficient sgRNA was selected containing 2′-O-methyl and 3′ phosphorothioate modifications to the first three 5′ and the last three 3′ nucleotides (Synthego). 2 × 106 UVC cells were electroporated using a Neon Nucleofector (Lonza) in Buffer P3 (Lonza) with Cas9 protein (IDT) precomplexed with sgRNA, in a total volume of 100 μL using electroporation program CM138. Gene targeting vectors carrying an expression cassette for expression of human MICA or the SARS-CoV-2 WA1/2020 spike gene, targeting the Rosa26 and AAVS1 locus respectively, were co-electroporated at 4 μg. Indels introduced by CRISPR editing were detected by PCR and Sanger sequence using DNA primers designed to amplify a 600–900 base pair region surrounding the sgRNA target site. A minimum of 24 h after electroporation, genomic DNA was extracted using the DirectPCR Lysis solution (Viagen Biotech) containing Proteinase K and target regions were amplified by PCR using the GoTaq G2 PCR mastermix (Promega). Correct and unique amplification of the target regions was verified by agarose gel electrophoresis before purifying PCR products using the QIAquick PCR Purification Kit (Qiagen). For analysis by TIDE, PCR amplicons were Sanger sequenced (Eurofins or Genewiz) and paired.ab1 files of control versus edited samples were analyzed using Synthego’s ICE tool (https://ice.synthego.com/).
Intracellular spike protein staining
Engineered UVC were harvested and then fixed and permeabilized using Cytofix/Cytoperm Fixation/Permeabilization Solution (ThermoFisher). Cells were then stained for intracellular spike protein using an Anti-SARS-CoV-2 Spike Glycoprotein S1 antibody (Abcam, ab275759, 1:50) followed by Goat Anti-Rabbit IgG H&L (Alexa Fluor 488) (Abcam, ab150077, 1:500). Flow analysis was carried out on a Fortessa flow cytometer (BD Bioscience), and data analyzed, and flow cytometry figures generated using FlowJo 10 software (BD Biosciences).
Flow cytometry analysis of cell surface antigen expression
For flow cytometric analysis of cell surface expression of MHC-I, MICA and SARS-CoV-2 spike protein, cells were harvested from culture plates and washed using PBS with 1% Bovine Serum Albumen (Thermo Scientific) and were then stained with PE anti-human MICA/MICB Antibody (6D4, Biolegend), Alexa Fluor 647 anti-human HLA-A,B,C (W6/32, Biolegend), and anti-SARS-CoV-2 Spike Glycoprotein S1 antibody (Abcam, ab275759, 1:50) followed by Goat Anti-Rabbit IgG H&L (Alexa Fluor 488) (Abcam, ab150077, 1:500). Live/Dead Fixable Dead Cell Stains (Invitrogen) were included in all experiments to exclude dead cells. After staining, cells were resuspended in PBS with 2% Human Heat Inactivated AB Serum (Sigma) and 0.1 M EDTA pH 8.0 (Invitrogen) before analysis on a Fortessa flow cytometer (BD Bioscience) and data analyzed using FlowJo 10 software (BD Biosciences).
Western blot
The SARS-CoV-2 spike glycoprotein was detected in UVC lysates by western blotting. Briefly, cells were lysed by RIPA buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.1% SDS, 1% NP40, 1x protease inhibitor cocktail). Samples were spun at 4°C for 10 min at 12,000 xg and the pellet discarded. Protein content was measured using BCA Assay (ThermoFisher) using a PHERAstar plate reader (BMG Labtech) at 560 nm. LDS Sample Buffer was added to 30 ng of protein sample to make a 1x solution, with 0.5 μL of b-mercaptoethanol per well and heated at 70°C for 10 min before separation on a polyacrylamide gel (Bio-Rad Mini-PROTEAN TGX Gel 4–15%) and transferred to a PVDF membrane. Membranes were blocked in blocking buffer (5% non-fat powdered milk in TBST), before incubation with primary antibodies in blocking buffer (Rabbit polyclonal anti-SARS-Cov2, Sino Biological 40,591-T62, 1:6000 dilution or Mouse b-actin, Abcam 8226, 1 μg/mL), detected with HRP conjugated secondaries in blocking buffer (Goat anti-Rabbit HRP, Sino Biological SSA003, 0.5 μg/mL or Goat anti-Mouse HRP, Abcam ab205719, 1: 4000 dilution) and visualised using the Super-Signal West Femto kit (ThermoFisher) as per kit instructions.
qPCR measurement of stem cell factors
Total RNA was extracted from UVC cells using the ReliaPrep RNA miniprep (Promega) according to the manufacturer’s instructions (a DNase treatment was included for all samples), and RNA concentration and absorbance ratios were measured using a Nanodrop One Spectrophotometer (ThermoFisher). cDNA was synthesized using a High-Capacity cDNA Reverse Transcription Kit (the Applied Biosystems) in a total volume of 20 μL to produce DNA that was subsequently assessed by spectrophotometric analysis and diluted to 100 ng/μL. Individual master mixes with each of the DNA-primer combinations for detection of human SOX2, NANOG, OCT4, DCN, Vimentin, HES5 and GATA6 genes were made for 3 replicates using the Brilliant III Ultra-Fast SYBR green qPCR master mix (Agilent Technologies) and analyzed on a CFX Opus Real-Time PCR system (BioRad) using the following program: 95°C for 15 min for 1 cycle; 95°C for 15 s for 40 cycles; 60°C for 30 s.
SARS-CoV-2 spike protein ELISA
Cell pellets were harvested and lysed in 20 μL Cell Extraction Buffer (Invitrogen) containing protease inhibitors (Sigma) on ice for 30 min, with 3 brief vortexing every 10 min. Samples were centrifuged at 13,000 rpm for 10 min at 4°C to pellet insoluble contents. S1 Spike protein was detected using a Covid-19 S-protein ELISA kit (Abcam) specific to S1RBD. Samples were diluted to a range determined to be within the working range of the ELISA kit used and the assay procedure was followed as per manufacturer’s instructions. The resulting colorimetric signal was detected at 450 nm using a PHERAstar (BMG LABTECH) plate reader. GraphPad Prism was used to plot a standard curve and interpolate the sample values using a 4-parameter logistic fit.
UVC proliferation and apoptosis assays
To quantify apoptosis of UVC post-irradiation, cells were stained using a FITC Annexin V Apoptosis Detection Kit with 7-AAD (Biolegend). Proliferation of cells was measured by staining of control and Irradiated UVC with 2 μM Cell Trace Yellow (Invitrogen) according to kit protocol and analyzing the dilution of the dye at 24-h periods over 3-days and measuring fluorescence intensity. Flow analysis was carried out on a Fortessa flow cytometer (BD Bioscience), and data analyzed, and flow cytometry figures generated using FlowJo 10 software (BD Biosciences).
Colony formation assay
Cells were seeded in vitronectin coated 60 mm diameter plastic culture dishes in triplicate, at densities of between 100 and 3000 cells per dish, and cultured in mTeSR-Plus with Rock Inhibitor (StemCell Technologies) for 24 h to allow cells to adhere. Media was then replaced with mTeSR-Plus without Rock Inhibitor for an additional 9 days. Media was replaced every 2 days before finally being removed and plates dried and stained with 0.25% Crystal Violet (Thermo Scientific) in 20% methanol to visualise colonies prior to counting.
CAM cytotoxicity assay
Both MHC-I expressing and MHC-I deficient (B2M knockout) UVC were used as target cells for NK cell cytotoxicity assay. Trypsinized cells were stained with calcein acetoxymethyl ester (CAM, Invitrogen) at a 10 μM concentration for 1 h at 37°C and then washed to remove excess dye. NK cells highly enriched from normal cynomolgus macaque (Macaca fascicularis) blood samples using a CD3 depletion kit (Miltenyi Biotec), were used as effector cells. NK cell effectors and stained target cells were co-cultured in 96 well round bottom plates at effector: target (E:T) ratios of 1:1 and 5:1. Control wells included – only target cells for spontaneous release of CAM and target cells treated with Triton X-100 for maximum release of CAM. At the end of 4-h incubation, supernatant was collected for CAM measurement in a fluorescent plate reader at 530 nm. Percent-specific lysis = (test release - spontaneous release)/(maximum release - spontaneous release).
Nucleofection of NKG2D ligands in iPS cells
UVC were cultured in EGM2 (Lonza) media supplemented with 20 ng/mL VEG-F (Peprotech) until 70–90% confluent, in tissue culture flasks pre-coated with sterile 0.1% gelatin in PBS for 1 h at 37°C. The cells were removed from culture flasks using trypsin, washed, and transfected with plasmid DNA containing either MICA, MICB or ULBP-1 genes after optimizing nucleofection conditions using primary cell 4D nucleofector kit and 4D nucleofector system (Lonza). After 48 h of culture, transfected cells were stained with aqua dye for live/dead discrimination and corresponding antibodies- MICA/MICB (Clone 6D4, PE, BioLegend) or ULBP-1 (clone 170,818, PE, R & D Systems). Stained cells were fixed with 2% paraformaldehyde and acquired on LSRII flow cytometer. Transfection efficiency was calculated as % live cells expressing transfected protein.
NK cell intracellular cytokine staining assay
UVC target and NK effector cells were plated at E:T ratio of 2:1 in a 96 well round bottom plate. Anti-CD107a antibody (clone H4A3, ECD conjugate, BD Biosciences), brefeldin A and monensin (BD Biosciences) were added to all the samples prior to incubation. After 6 h of incubation at 37°C, the cells were washed and stained with aqua dye used for live and dead cell discrimination for 20 min at room temperature. The cells were then washed and stained for surface markers that included CD3 (SP34.2, BV421, BD Biosciences), CD14 (M5E2, BV650, BD Biosciences), CD16 (3G8, BUV496, BD Biosciences), CD20 (L27, BV570, BD Biosciences), CD56 (NCAM1.2, BV605, BD Biosciences), HLA-DR (G46-6, APC-H7, BD Biosciences) and NKG2A (Z199, PE-Cy7, Beckman Coulter) to delineate NK effector cells. Following incubation for 20 min, cells were washed and permeabilized using fix & perm reagent (Thermofisher Scientific) as per manufacturer’s recommendation. Intracellular cytokine staining was performed for macrophage inflammatory protein 1β (MIP-1β; D21-1351, FITC, BD Biosciences) interferon-ɣ (IFN-ɣ; B27, BUV395, BD Biosciences), tumor necrosis factor alpha (TNF-α; Mab11, BV650, BD Biosciences) at 4°C for 15 min. Cells were washed, fixed, and acquired on LSRII flow cytometer. Unstimulated NK cells were used for background subtraction of percent positive cells. NK cells stimulated with leukocyte activation cocktail (BD Biosciences) were used as positive control for the assay.
Immunohistochemical analysis
Following fixation in 10% NBF, tissue samples excised from mice injected with UVC were dehydrated in graded alcohols and embedded side-on in paraffin wax. FFPE blocks were trimmed until at full-face before placing on slides for H&E staining. Following heat fixation to the slide, the tissue sections were deparaffinised in xylene and rehydrated through graded alcohol before staining with Haematoxylin and Eosin. Whole slide scans were imaged using a Hamamatsu slide scanner.
Subgenomic viral mRNA assay
SARS-CoV-2 E gene sgRNA was assessed by RT–PCR using primers and probes as previously described.49 , 50 , 51 , 52 In brief, to generate a standard curve, the SARS-CoV-2 E gene sgRNA was cloned into a pcDNA3.1 expression plasmid; this insert was transcribed using an AmpliCap-Max T7 High Yield Message Maker Kit (Cellscript) to obtain RNA for standards. Before RT–PCR, samples collected from challenged macaques or standards were reverse-transcribed using Superscript III VILO (Invitrogen) according to the manufacturer’s instructions. A Taqman custom gene expression assay (ThermoFisher Scientific) was designed using the sequences targeting the E gene sgRNA. Reactions were carried out on a QuantStudio 6 and 7 Flex Real-Time PCR System (Applied Biosystems) according to the manufacturer’s specifications. Standard curves were used to calculate sgRNA in copies per mL or per swab; the quantitative assay sensitivity was 50 copies per mL or per swab.
Serum antibody ELISA
RBD-specific binding antibodies were assessed by ELISA as previously described.9 , 10 In brief, 96-well plates were coated with 1 μg mL−1 SARS-CoV-2 RBD protein (A. Schmidt, MassCPR) in 1× DPBS and incubated at 4°C overnight. After incubation, plates were washed once with wash buffer (0.05% Tween 20 in 1× DPBS) and blocked with 350 μL casein block per well for 2–3 h at room temperature. After incubation, block solution was discarded, and plates were blotted dry. Serial dilutions of heat-inactivated serum diluted in casein block were added to wells and plates were incubated for 1 h at room temperature, before three further washes and a 1-h incubation with a 1:1,000 dilution of anti-macaque IgG HRP (NIH NHP Reagent Program) at room temperature in the dark. Plates were then washed three times, and 100 μL of SeraCare KPL TMB SureBlue Start solution was added to each well; plate development was halted by the addition of 100 μL SeraCare KPL TMB Stop solution per well. The absorbance at 450 nm was recorded using a VersaMax or Omega microplate reader. ELISA endpoint titers were defined as the highest reciprocal serum dilution that yielded an absorbance >0.2. The log10(endpoint titers) are reported.
Pseudovirus neutralization assay
The SARS-CoV-2 pseudovirus expressing a luciferase reporter gene were generated in a similar approach to that previously described.9 , 10 , 16 In brief, the packaging construct psPAX2 (AIDS Resource and Reagent Program), luciferase reporter plasmid pLenti-CMV Puro-Luc (Addgene), and spike protein expressing pcDNA3.1-SARS-CoV-2 SΔCT were co-transfected into HEK293T cells with calcium phosphate. The supernatants containing the pseudotype viruses were collected 48 h after transfection; pseudotype viruses were purified by filtration with 0.45-μm filter. To determine the neutralization activity of the antisera from vaccinated macaques, HEK293T-hACE2 cells were seeded in 96-well tissue culture plates at a density of 1.75 × 104 cells per well overnight. 2-fold serial dilutions of heat-inactivated serum samples were prepared and mixed with 50 μL of pseudovirus. The mixture was incubated at 37°C for 1 h before adding to HEK293T-hACE2 cells. After 48 h, cells were lysed in Steady-Glo Luciferase Assay (Promega) according to the manufacturer’s instructions. SARS-CoV-2 neutralization titers were defined as the sample dilution at which a 50% reduction in relative light units was observed relative to the average of the virus control wells.
Quantification and statistical analysis
Statistical differences between two sample groups, where appropriate, were analyzed by a standard Student’s two-tailed, non-paired, t-test and between three or more sample groups using two-way or three-way ANOVA using GraphPad Prism 9. Analysis of virological data was performed using two-sided Mann–Whitney tests. Correlations were assessed by two-sided Spearman rank-correlation tests. p values are included in the figures or referred to in the legends where statistical analyses have been carried out. p values of less than 0.05 were considered significant.
Supplemental information
Document S1. Figures S1–S4 and Table S1
Document S2. Article plus supplemental information
Data and code availability
• All data reported in this paper will be shared by the lead contact upon request.
• This paper does not report original code.
• Any additional information required to reanalyze the data reported in this work paper is available from the lead contact upon request.
Acknowledgments
This work is funded by Intima Bioscience.
Author contributions
Genetic engineering and in vitro characterization of the UVC were led by T.H., with A.C., A.S., E.L., K.C., R.G., and E.A. generating and interpreting experimental data and performing statistical analysis. R.K.R. designed and led the NK cell killing assays, with C.M., K.K., and M.M. generating and interpreting experimental data. The vaccination study was designed and led by D.H.B. Immunologic and virologic assays were led by A.C., J.Y., K.M., and V.G. Humoral immune responses were assessed by K.M. and J.Y. The paper was written by T.H., D.H.B., and M.C., with the involvement of all co-authors.
Declaration of interests
D.H.B. has a sponsored research collaboration funded by Intima Bioscience. Praesidium Bioscience has patents filed based on the findings described herein (application WO2021216729A1).
Supplemental information can be found online at https://doi.org/10.1016/j.xcrm.2022.100843.
==== Refs
References
1 Polack F.P. Thomas S.J. Kitchin N. Absalon J. Gurtman A. Lockhart S. Perez J.L. Pérez Marc G. Moreira E.D. Zerbini C. Safety and efficacy of the BNT162b2 mRNA covid-19 vaccine N. Engl. J. Med. 383 2020 2603 2615 10.1056/NEJMoa2034577 33301246
2 Baden L.R. El Sahly H.M. Essink B. Kotloff K. Frey S. Novak R. Diemert D. Spector S.A. Rouphael N. Creech C.B. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine N. Engl. J. Med. 384 2021 403 416 10.1056/NEJMoa2035389 33378609
3 Sadoff J. Gray G. Vandebosch A. Cárdenas V. Shukarev G. Grinsztejn B. Goepfert P.A. Truyers C. Fennema H. Spiessens B. Safety and efficacy of single-dose Ad26.COV2.S vaccine against covid-19 N. Engl. J. Med. 384 2021 2187 2201 10.1056/NEJMoa2101544 33882225
4 Levin E.G. Lustig Y. Cohen C. Fluss R. Indenbaum V. Amit S. Doolman R. Asraf K. Mendelson E. Ziv A. Waning immune humoral response to BNT162b2 covid-19 vaccine over 6 months N. Engl. J. Med. 385 2021 e84 10.1056/NEJMoa2114583 34614326
5 Naaber P. Tserel L. Kangro K. Sepp E. Jürjenson V. Adamson A. Haljasmägi L. Rumm A.P. Maruste R. Kärner J. Dynamics of antibody response to BNT162b2 vaccine after six months: a longitudinal prospective study Lancet Reg. Health. Eur. 10 2021 100208 10.1016/j.lanepe.2021.100208 34514454
6 Thomas S.J. Moreira E.D. Jr. Kitchin N. Absalon J. Gurtman A. Lockhart S. Perez J.L. Pérez Marc G. Polack F.P. Zerbini C. Safety and efficacy of the BNT162b2 mRNA covid-19 vaccine through 6 months N. Engl. J. Med. 385 2021 1761 1773 10.1056/NEJMoa2110345 34525277
7 Cohn B.A. Cirillo P.M. Murphy C.C. Krigbaum N.Y. Wallace A.W. SARS-CoV-2 vaccine protection and deaths among US veterans during 2021 Science 375 2022 331 336 10.1126/science.abm0620 34735261
8 Collier A.R.Y. Yu J. McMahan K. Liu J. Chandrashekar A. Maron J.S. Atyeo C. Martinez D.R. Ansel J.L. Aguayo R. Differential kinetics of immune responses elicited by covid-19 vaccines N. Engl. J. Med. 385 2021 2010 2012 10.1056/NEJMc2115596 34648703
9 Barouch D.H. Stephenson K.E. Sadoff J. Yu J. Chang A. Gebre M. McMahan K. Liu J. Chandrashekar A. Patel S. Durable humoral and cellular immune responses 8 Months after Ad26.COV2.S vaccination N. Engl. J. Med. 385 2021 951 953 10.1056/NEJMc2108829 34260834
10 Planas D. Veyer D. Baidaliuk A. Staropoli I. Guivel-Benhassine F. Rajah M.M. Planchais C. Porrot F. Robillard N. Puech J. Reduced sensitivity of SARS-CoV-2 variant Delta to antibody neutralization Nature 596 2021 276 280 10.1038/s41586-021-03777-9 34237773
11 Callaway E. Heavily mutated Omicron variant puts scientists on alert Nature 600 2021 21 10.1038/d41586-021-03552-w 34824381
12 Lefèvre B. Tondeur L. Madec Y. Grant R. Lina B. van der Werf S. Rabaud C. Fontanet A. Beta SARS-CoV-2 variant and BNT162b2 vaccine effectiveness in long-term care facilities in France Lancet. Healthy Longev. 2 2021 e685 e687 10.1016/S2666-7568(21)00230-0 34580665
13 Abu-Raddad L.J. Chemaitelly H. Butt A.A. National Study Group for COVID-19 Vaccination Effectiveness of the BNT162b2 covid-19 vaccine against the B.1.1.7 and B.1.351 variants N. Engl. J. Med. 385 2021 187 189 10.1056/NEJMc2104974 33951357
14 Villarreal R. Casale T.B. Commonly used adjuvant human vaccines: advantages and side effects J. Allergy Clin. Immunol. Pract. 8 2020 2953 2957 10.1016/j.jaip.2020.04.045 32360184
15 Lauer K.B. Borrow R. Blanchard T.J. Multivalent and multipathogen viral vector vaccines Clin. Vaccine Immunol. 24 2017 10.1128/CVI.00298-16 e00298-16
16 Schlingmann B. Castiglia K.R. Stobart C.C. Moore M.L. Polyvalent vaccines: high-maintenance heroes PLoS Pathog. 14 2018 10.1371/journal.ppat.1006904 e1006904
17 Pardi N. Hogan M.J. Porter F.W. Weissman D. mRNA vaccines - a new era in vaccinology Nat. Rev. Drug Discov. 17 2018 261 279 10.1038/nrd.2017.243 29326426
18 Pulendran B. S Arunachalam P. O'Hagan D.T. Emerging concepts in the science of vaccine adjuvants Nat. Rev. Drug Discov. 20 2021 454 475 10.1038/s41573-021-00163-y 33824489
19 Yousefi Avarvand A. Meshkat Z. Khademi F. Tafaghodi M. Immunogenicity of HspX/EsxS fusion protein of Mycobacterium tuberculosis along with ISCOMATRIX and PLUSCOM nano-adjuvants after subcutaneous administration in animal model Microb. Pathog. 154 2021 104842 10.1016/j.micpath.2021.104842 33762199
20 Cebon J.S. Gore M. Thompson J.F. Davis I.D. McArthur G.A. Walpole E. Smithers M. Cerundolo V. Dunbar P.R. MacGregor D. Results of a randomized, double-blind phase II clinical trial of NY-ESO-1 vaccine with ISCOMATRIX adjuvant versus ISCOMATRIX alone in participants with high-risk resected melanoma J. Immunother. Cancer 8 2020 10.1136/jitc-2019-000410 e000410
21 Baz Morelli A. Becher D. Koernig S. Silva A. Drane D. Maraskovsky E. ISCOMATRIX: a novel adjuvant for use in prophylactic and therapeutic vaccines against infectious diseases J. Med. Microbiol. 61 2012 935 943 10.1099/jmm.0.040857-0 22442293
22 O'Hagan D.T. Lodaya R.N. Lofano G. The continued advance of vaccine adjuvants - 'we can work it out Semin. Immunol. 50 2020 101426 10.1016/j.smim.2020.101426 33257234
23 Zhou F. Hansen L. Pedersen G. Grødeland G. Cox R. Matrix M adjuvanted H5N1 vaccine elicits broadly neutralizing antibodies and neuraminidase inhibiting antibodies in humans that correlate with in vivo protection Front. Immunol. 12 2021 747774 10.3389/fimmu.2021.747774 34887855
24 Bengtsson K.L. Song H. Stertman L. Liu Y. Flyer D.C. Massare M.J. Xu R.H. Zhou B. Lu H. Kwilas S.A. Matrix-M adjuvant enhances antibody, cellular and protective immune responses of a Zaire Ebola/Makona virus glycoprotein (GP) nanoparticle vaccine in mice Vaccine 34 2016 1927 1935 10.1016/j.vaccine.2016.02.033 26921779
25 Pavord S. Scully M. Hunt B.J. Lester W. Bagot C. Craven B. Rampotas A. Ambler G. Makris M. Clinical features of vaccine-induced immune thrombocytopenia and thrombosis N. Engl. J. Med. 385 2021 1680 1689 10.1056/NEJMoa2109908 34379914
26 Long B. Bridwell R. Gottlieb M. Thrombosis with thrombocytopenia syndrome associated with COVID-19 vaccines Am. J. Emerg. Med. 49 2021 58 61 10.1016/j.ajem.2021.05.054 34062319
27 Cattaneo M. Thrombosis with Thrombocytopenia Syndrome associated with viral vector COVID-19 vaccines Eur. J. Intern. Med. 89 2021 22 24 10.1016/j.ejim.2021.05.031 34092488
28 Afkhami S. Yao Y. Xing Z. Methods and clinical development of adenovirus-vectored vaccines against mucosal pathogens Mol. Ther. Methods Clin. Dev. 3 2016 16030 10.1038/mtm.2016.30 27162933
29 Lopez Bernal J. Gower C. Andrews N. Public Health England Delta Variant Vaccine Effectiveness Study Group Effectiveness of covid-19 vaccines against the B.1.617.2 (Delta) variant N. Engl. J. Med. 385 2021 e92 10.1056/NEJMc2113090
30 Grant R. Charmet T. Schaeffer L. Galmiche S. Madec Y. Von Platen C. Chény O. Omar F. David C. Rogoff A. Impact of SARS-CoV-2 Delta variant on incubation, transmission settings and vaccine effectiveness: results from a nationwide case-control study in France Lancet Reg. Health. Eur. 13 2022 100278 10.1016/j.lanepe.2021.100278 34849500
31 Fowlkes A. Gaglani M. Groover K. Thiese M.S. Tyner H. Ellingson K. HEROES-RECOVER Cohorts Effectiveness of COVID-19 vaccines in preventing SARS-CoV-2 infection among frontline workers before and during B.1.617.2 (Delta) variant predominance - eight U.S. Locations, December 2020-August 2021 MMWR Morb. Mortal. Wkly. Rep. 70 2021 1167 1169 10.15585/mmwr.mm7034e4 34437521
32 Abu-Dawud R. Graffmann N. Ferber S. Wruck W. Adjaye J. Pluripotent stem cells: induction and self-renewal Philos. Trans. R. Soc. Lond. B Biol. Sci. 373 2018 10.1098/rstb.2017.0213 20170213
33 Deinsberger J. Reisinger D. Weber B. Global trends in clinical trials involving pluripotent stem cells: a systematic multi-database analysis NPJ Regen. Med. 5 2020 15 10.1038/s41536-020-00100-4 32983575
34 Shi Y. Inoue H. Wu J.C. Yamanaka S. Induced pluripotent stem cell technology: a decade of progress Nat. Rev. Drug Discov. 16 2017 115 130 10.1038/nrd.2016.245 27980341
35 Kirchdoerfer R.N. Cottrell C.A. Wang N. Pallesen J. Yassine H.M. Turner H.L. Corbett K.S. Graham B.S. McLellan J.S. Ward A.B. Pre-fusion structure of a human coronavirus spike protein Nature 531 2016 118 121 10.1038/nature17200 26935699
36 Pallesen J. Wang N. Corbett K.S. Wrapp D. Kirchdoerfer R.N. Turner H.L. Cottrell C.A. Becker M.M. Wang L. Shi W. Immunogenicity and structures of a rationally designed prefusion MERS-CoV spike antigen Proc. Natl. Acad. Sci. USA 114 2017 E7348 E7357 10.1073/pnas.1707304114 28807998
37 Wrapp D. Wang N. Corbett K.S. Goldsmith J.A. Hsieh C.L. Abiona O. Graham B.S. McLellan J.S. Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation Preprint at bioRxiv 2020 10.1101/2020.02.11.944462
38 Szondy Z. Sarang Z. Kiss B. Garabuczi É. Köröskényi K. Anti-inflammatory mechanisms triggered by apoptotic cells during their clearance Front. Immunol. 8 2017 909 10.3389/fimmu.2017.00909 28824635
39 Brandstadter J.D. Yang Y. Natural killer cell responses to viral infection J. Innate Immun. 3 2011 274 279 10.1159/000324176 21411975
40 Waggoner S.N. Reighard S.D. Gyurova I.E. Cranert S.A. Mahl S.E. Karmele E.P. McNally J.P. Moran M.T. Brooks T.R. Yaqoob F. Rydyznski C.E. Roles of natural killer cells in antiviral immunity Curr. Opin. Virol. 16 2016 15 23 10.1016/j.coviro.2015.10.008 26590692
41 Koutsakos M. McWilliam H.E.G. Aktepe T.E. Fritzlar S. Illing P.T. Mifsud N.A. Purcell A.W. Rockman S. Reading P.C. Vivian J.P. Downregulation of MHC class I expression by influenza A and B viruses Front. Immunol. 10 2019 1158 10.3389/fimmu.2019.01158 31191533
42 Schuren A.B. Costa A.I. Wiertz E.J. Recent advances in viral evasion of the MHC Class I processing pathway Curr. Opin. Immunol. 40 2016 43 50 10.1016/j.coi.2016.02.007 27065088
43 Paul S. Lal G. The molecular mechanism of natural killer cells function and its importance in cancer immunotherapy Front. Immunol. 8 2017 1124 10.3389/fimmu.2017.01124 28955340
44 Wensveen F.M. Jelenčić V. Polić B. NKG2D: a master regulator of immune cell responsiveness Front. Immunol. 9 2018 441 10.3389/fimmu.2018.00441 29568297
45 Chatterjee I. Li F. Kohler E.E. Rehman J. Malik A.B. Wary K.K. Induced pluripotent stem (iPS) cell culture methods and induction of differentiation into endothelial cells Methods Mol. Biol. 1357 2016 311 327 10.1007/7651_2015_203 25687301
46 Rivera T. Zhao Y. Ni Y. Wang J. Human-induced pluripotent stem cell culture methods under cGMP conditions Curr. Protoc. Stem Cell Biol. 54 2020 e117 10.1002/cpsc.117 32649060
47 Swaidan N.T. Salloum-Asfar S. Palangi F. Errafii K. Soliman N.H. Aboughalia A.T. Wali A.H.S. Abdulla S.A. Emara M.M. Identification of potential transcription factors that enhance human iPSC generation Sci. Rep. 10 2020 21950 10.1038/s41598-020-78932-9 33319795
48 Shi G. Jin Y. Role of Oct4 in maintaining and regaining stem cell pluripotency Stem Cell Res. Ther. 1 2010 39 10.1186/scrt39 21156086
49 Mercado N.B. Zahn R. Wegmann F. Loos C. Chandrashekar A. Yu J. Liu J. Peter L. McMahan K. Tostanoski L.H. Single-shot Ad26 vaccine protects against SARS-CoV-2 in rhesus macaques Nature 586 2020 583 588 10.1038/s41586-020-2607-z 32731257
50 Yang Z.Y. Kong W.P. Huang Y. Roberts A. Murphy B.R. Subbarao K. Nabel G.J. A DNA vaccine induces SARS coronavirus neutralization and protective immunity in mice Nature 428 2004 561 564 10.1038/nature02463 15024391
51 Yu J. Tostanoski L.H. Peter L. Mercado N.B. McMahan K. Mahrokhian S.H. Nkolola J.P. Liu J. Li Z. Chandrashekar A. DNA vaccine protection against SARS-CoV-2 in rhesus macaques Science 369 2020 806 811 10.1126/science.abc6284 32434945
52 Chandrashekar A. Liu J. Martinot A.J. McMahan K. Mercado N.B. Peter L. Tostanoski L.H. Yu J. Maliga Z. Nekorchuk M. SARS-CoV-2 infection protects against rechallenge in rhesus macaques Science 369 2020 812 817 10.1126/science.abc4776 32434946
53 Wölfel R. Corman V.M. Guggemos W. Seilmaier M. Zange S. Müller M.A. Niemeyer D. Jones T.C. Vollmar P. Rothe C. Virological assessment of hospitalized patients with COVID-2019 Nature 581 2020 465 469 10.1038/s41586-020-2196-x 32235945
54 Vogel A.B. Kanevsky I. Che Y. Swanson K.A. Muik A. Vormehr M. Kranz L.M. Walzer K.C. Hein S. Güler A. BNT162b vaccines protect rhesus macaques from SARS-CoV-2 Nature 592 2021 283 289 10.1038/s41586-021-03275-y 33524990
55 Corbett K.S. Flynn B. Foulds K.E. Francica J.R. Boyoglu-Barnum S. Werner A.P. Flach B. O'Connell S. Bock K.W. Minai M. Evaluation of the mRNA-1273 vaccine against SARS-CoV-2 in nonhuman primates N. Engl. J. Med. 383 2020 1544 1555 10.1056/NEJMoa2024671 32722908
56 Wang L. Cheng G. Sequence analysis of the emerging sars-CoV-2 variant omicron in South Africa J. Med. Virol. 94 2022 1728 1733 10.1002/jmv.27516 34897752
57 CDC COVID-19 Response Team SARS-CoV-2 B.1.1.529 (omicron) variant - United States, December 1-8, 2021 MMWR Morb. Mortal. Wkly. Rep. 70 2021 1731 1734 10.15585/mmwr.mm7050e1 34914670
58 Huang C. Wang Y. Li X. Ren L. Zhao J. Hu Y. Zhang L. Fan G. Xu J. Gu X. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China Lancet 395 2020 497 506 10.1016/S0140-6736(20)30183-5 31986264
59 Björkström N.K. Strunz B. Ljunggren H.G. Natural killer cells in antiviral immunity Nat. Rev. Immunol. 22 2022 112 123 10.1038/s41577-021-00558-3 34117484
60 Tartof S.Y. Slezak J.M. Fischer H. Hong V. Ackerson B.K. Ranasinghe O.N. Frankland T.B. Ogun O.A. Zamparo J.M. Gray S. Effectiveness of mRNA BNT162b2 COVID-19 vaccine up to 6 months in a large integrated health system in the USA: a retrospective cohort study Lancet 398 2021 1407 1416 10.1016/S0140-6736(21)02183-8 34619098
61 Smits V.A.J. Hernández-Carralero E. Paz-Cabrera M.C. Cabrera E. Hernández-Reyes Y. Hernández-Fernaud J.R. Gillespie D.A. Salido E. Hernández-Porto M. Freire R. The Nucleocapsid protein triggers the main humoral immune response in COVID-19 patients Biochem. Biophys. Res. Commun. 543 2021 45 49 10.1016/j.bbrc.2021.01.073 33515911
62 Vidal S.J. Collier A.R.Y. Yu J. McMahan K. Tostanoski L.H. Ventura J.D. Aid M. Peter L. Jacob-Dolan C. Anioke T. Correlates of neutralization against SARS-CoV-2 variants of concern by early pandemic sera J. Virol. 95 2021 10.1128/JVI.00404-21 e0040421
63 Lu S. Xie X.X. Zhao L. Wang B. Zhu J. Yang T.R. Yang G.W. Ji M. Lv C.P. Xue J. The immunodominant and neutralization linear epitopes for SARS-CoV-2 Cell Rep. 34 2021 10.1016/j.celrep.2020.108666 108666
64 Chandrashekar A. Yu J. McMahan K. Jacob-Dolan C. Liu J. He X. Hope D. Anioke T. Barrett J. Chung B. Vaccine protection against the SARS-CoV-2 Omicron variant in macaques Cell 185 2022 1549 1555.e11 10.1016/j.cell.2022.03.024 35427477
| 36480934 | PMC9727836 | NO-CC CODE | 2022-12-08 23:18:54 | no | Cell Rep Med. 2022 Dec 7;:100843 | utf-8 | Cell Rep Med | 2,022 | 10.1016/j.xcrm.2022.100843 | oa_other |
==== Front
Sci Total Environ
Sci Total Environ
The Science of the Total Environment
0048-9697
1879-1026
Elsevier B.V.
S0048-9697(22)07814-7
10.1016/j.scitotenv.2022.160711
160711
Review
Ozone based inactivation and disinfection in the pandemic time and beyond: Taking forward what has been learned and best practice
Cai Yamei ab
Zhao Yaqian ab⁎
Yadav Asheesh Kumar c
Ji Bin ad
Kang Peiying ab
Wei Ting ae
a State Key Laboratory of Eco-Hydraulics in Northwest Arid Region, Xi'an University of Technology, Xi'an 710048, PR China
b Department of Municipal and Environmental Engineering, School of Water Resources and Hydroelectric Engineering, Xi'an University of Technology, Xi'an 710048, PR China
c Department of Chemical and Environmental Technology, Rey Juan Carlos University, Madrid, Spain
d School of Civil Engineering, Yantai University, Yantai 264005, PR China
e Department of Analytical Chemistry, Physical Chemistry and Chemical Engineering, University of Alcalá, Madrid, Spain
⁎ Corresponding author at: State Key Laboratory of Eco-Hydraulics in Northwest Arid Region, Xi'an University of Technology, Xi'an 710048, PR China.
7 12 2022
7 12 2022
16071111 9 2022
27 11 2022
2 12 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The large-scale global COVID-19 has a profound impact on human society. Timely and effectively blocking the virus spread is the key to controlling the pandemic growth. Ozone-based inactivation and disinfection techniques have been shown to effectively kill SARS-CoV-2 in water, aerosols and on solid surface. However, the lack of an unified information and discussion on ozone-based inactivation and disinfection in current and previous pandemics and the absence of consensus on the main mechanisms by which ozone-based inactivation of pandemic causing viruses have hindered the possibility of establishing a common basis for identifying best practices in the utilization of ozone technology. This article reviews the research status of ozone (O3) disinfection on pandemic viruses (especially SARS-CoV-2). Taking sterilization kinetics as the starting point while followed by distinguishing the pandemic viruses by enveloped and non-enveloped viruses, this review focuses on analyzing the scope of application of the sterilization model and the influencing factors from the experimental studies and data induction. It is expected that the review could provide an useful reference for the safe and effective O3 utilization of SARS-CoV-2 inactivation in the post-pandemic era.
Graphical abstract
Unlabelled Image
Keywords
Ozone
Pandemic
SARS-CoV-2
Inactivation
Kinetic model
Editor: Pavlos Kassomenos
==== Body
pmcData availability
Data will be made available on request.
| 36496014 | PMC9727960 | NO-CC CODE | 2022-12-11 23:16:13 | no | Sci Total Environ. 2023 Mar 1; 862:160711 | utf-8 | Sci Total Environ | 2,022 | 10.1016/j.scitotenv.2022.160711 | oa_other |
==== Front
J Hosp Infect
J Hosp Infect
The Journal of Hospital Infection
0195-6701
1532-2939
Published by Elsevier Ltd on behalf of The Healthcare Infection Society.
S0195-6701(22)00377-2
10.1016/j.jhin.2022.11.022
Practice Points
(Mis-)Judgment of infection risks is associated with additional workload among healthcare workers when treating isolated patients
Gaube Susanne 12∗
Däumling Sara 13
Biebl Isabell 23
Rath Anca 1
Caplunik-Pratsch Aila 1
Schneider-Brachert Wulf 1
1 Department of Infection Prevention and Infectious Diseases, University Hospital Regensburg, Regensburg, Germany
2 LMU Center for Leadership and People Management, LMU Munich, Munich, Germany
3 Department of Experimental Psychology, University of Regensburg, Regensburg, Germany
∗ Corresponding author. , Franz-Josef-Strauß-Allee 11, 93053 Regensburg, Germany. Tel.: +49 941-944-8960.
7 12 2022
7 12 2022
27 7 2022
2 11 2022
10 11 2022
© 2022 Published by Elsevier Ltd on behalf of The Healthcare Infection Society.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
==== Body
pmcThe COVID-19 pandemic has an immense negative impact on healthcare workers' (HCWs) mental health leading to more burnout, depression, anxiety, and insomnia [1,2]. According to a meta-analysis, high workloads, elevated infection risk perception, and lacking specialised training are especially detrimental [1]. However, COVID-19 is not the only infectious disease that HCWs face. Multidrug-resistant organisms (MDROs) are ubiquitous in hospitals and lead to increased patient mortality [3,4]. To prevent their transmission, similar protective measures to those for COVID-19 are taken, such as contact isolation and wearing personal protective equipment (PPE) [5]. Therefore, caring for patients with MDROs might likewise cause higher workloads and perception of infection risk. The present study is the first to compare HCWs’ self-reported workload and task-related colonisation risk perception after performing care tasks on isolated and non-isolated patients. Moreover, we investigated whether the perception of task-related and task-independent colonisation risks, as well as knowledge about appropriate infection prevention behaviour were associated with experiencing additional workload when caring for isolated patients.
We conducted a repeated-measures study with N = 45 HCWs (71.1% female, 95.6% nurses) at a tertiary care hospital. The National Aeronautics and Space Administration Task Load Index (NASA-TLX) was used to measure self-reported workload [6,7]. HCWs rated all NASA-TLX dimensions twice, directly after a care task on an isolated patient and after the same task on a non-isolated patient. Participants evaluated their task-related risk of becoming colonised while performing the task both times. Moreover, HCWs rated their task-independent risk once for the following pathogens: Vancomycin-resistant enterococci (VRE), Methicillin-resistant Staphylococcus aureus (MRSA), multidrug-resistant gram-negative bacteria (3MRGN, 4MRGN according to the German classification [8]), and COVID-19. Finally, we assessed participants’ perception of additional workload and their self-reported knowledge about appropriate infection prevention behaviour when caring for isolated patients. Data were collected between October 2021 and February 2022. The study was approved by the Research Ethics Committee at the University Hospital Regensburg (# 21-2428-101), and participants gave informed consent. The questionnaire, de-identified data, analysis script, and supplements are available in the Supplementary Information.
First, we compared the NASA-TLX dimensions and the overall task workload after completing the care task on an isolated and a non-isolated patient (Figure 1 a). We found that the task on an isolated patient was more physically demanding, effortful, and frustrating. Moreover, the overall task workload was significantly higher when caring for an isolated patient. Next, we saw that caring for isolated patients resulted in significantly higher perception of task-related risk of becoming colonised while performing the task (Figure 1b). Furthermore, Figure 1c shows that HCWs, on average, rate their task-independent risk for various pathogens between 3 = medium and 4 = high. Finally, we tested if knowledge about appropriate infection prevention behaviour, task-related and overall task-independent colonisation risk perception predicts HCWs’ evaluation of additional workload when caring for isolated patients. Higher task-related risk perception was associated with experiencing additional workload (t = 2.39, p = 0.022) while self-reported knowledge (t = 1.28, p = 0.209) and task-independent risk perception (t = -1.04, p = 0.306) were not. A secondary analysis including only nurses yielded nearly identical results (see OSF-supplements).Figure 1 Means of a) NASA-TLX dimensions and overall task workload; b) task-related colonisation risk perception; c) task-independent colonisation risk perception. *p ≤ 0.05, **p ≤ 0.001, ns = not significant.
Figure 1
This is the first study using the NASA-TXL to empirically show that HCWs’ self-reported workload is higher after performing a care task on isolated than non-isolated patients. It is plausible that caring for an isolated patient is physically demanding, requires effort, and leads to frustration because PPE can limit mobility and be prohibitively warm. Surprisingly, HCWs rated their task-related risk of becoming colonised with a pathogen while caring for an isolated patient higher. This finding may indicate misconceptions about the extent and effectiveness of existing infection prevention measures. Hospitals in Germany generally do not screen every patient for MDROs and other infectious diseases, only at-risk patients. Therefore, the colonisation risk from treating an unscreened patient without PPE should be perceived as higher than treating a positively tested patient with PPE. This effect might also be concerning when considering the significant association between task-related colonisation risk ratings and the perception that caring for isolated patients increases the workload. The present study has some limitations: a) the small sample size, b) it was conducted at a single centre, and c) data were collected during the COVID-19 pandemic, which might have biased the results.
Our findings contribute to a better understanding of how the COVID-19 pandemic and the occurrence of MDROs might impact HCWs’ mental health. More patients are treated in contact isolation resulting in higher workloads and infection risk perceptions. Considering the increasing prevalence of MDROs and risk of emerging infectious pathogens, ideally the ratio between HCWs and isolated patients should be improved, and staff should be reassured about the effectiveness of infection prevention measures to reduce the workload and risk perception to some degree.
Author contributions
S.G.: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Project administration, Visualization, Writing – original draft; Writing - Review & Editing.
S.D.: Investigation (data collection), Writing – original draft.
I.B.: Investigation (data collection), Writing - Review & Editing.
A.R.: Methodology, Writing - Review & Editing.
A.C.P.: Methodology, Writing - Review & Editing.
W.S.B.: Resources, Writing - Review & Editing, Supervision.
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
None.
==== Refs
References
1 Galanis P. Vraka I. Fragkou D. Bilali A. Kaitelidou D. Nurses’ burnout and associated risk factors during the COVID‐19 pandemic: A systematic review and meta‐analysis J Adv Nurs 77 2021 3286 3302 10.1111/jan.14839 33764561
2 Dragioti E. Tsartsalis D. Mentis M. Mantzoukas S. Gouva M. Impact of the COVID-19 pandemic on the mental health of hospital staff: An umbrella review of 44 meta-analyses Int J Nurs Stud 131 2022 104272 10.1016/j.ijnurstu.2022.104272
3 Tanwar J. Das S. Fatima Z. Hameed S. Multidrug resistance: An emerging crisis Interdiscip Perspect Infect Dis 2014 2014 1 7 10.1155/2014/541340
4 WHO. Global antimicrobial resistance and use surveillance system (GLASS) report 2021. Geneva: World Health Organization; 2021.
5 Siegel J.D. Rhinehart E. Jackson M. Chiarello L. Management of multidrug-resistant organisms in health care settings 2006 Am J Infect Control 35 2007 165 193 10.1016/j.ajic.2007.10.006
6 Hart SG. NASA-Task Load Index (NASA-TLX): 20 years later. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 50, 2006, p. 904–908. 10.1177/154193120605000909.
7 Hart S.G. Staveland L.E. Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research Elsevier Advances in Psychology 52 1988 139 183 10.1016/S0166-4115(08)62386-9
8 Kommission für Krankenhaushygiene und Infektionsprävention Hygienemaßnahmen bei Infektionen oder Besiedlung mit multiresistenten gramnegativen Stäbchen: Empfehlung der Kommission für Krankenhaushygiene und Infektionsprävention (KRINKO) beim Robert Koch-Institut Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 55 2012 1311 1354 10.1007/s00103-012-1549-5 23011096
| 36496091 | PMC9727961 | NO-CC CODE | 2022-12-08 23:19:01 | no | J Hosp Infect. 2022 Dec 7; doi: 10.1016/j.jhin.2022.11.022 | utf-8 | J Hosp Infect | 2,022 | 10.1016/j.jhin.2022.11.022 | oa_other |
==== Front
Clin Nutr ESPEN
Clin Nutr ESPEN
Clinical Nutrition Espen
2405-4577
Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
S2405-4577(22)01410-3
10.1016/j.clnesp.2022.12.001
Letter to the Editor
Letter to the editor: “Clinical Significance of Micronutrient Supplements in Patients with Coronavirus Disease 2019: A Comprehensive Systematic Review and Meta-Analysis” by Azizulla Beran et al., Clinical Nutrition ESPEN, https://doi.org/10.1016/j.clnesp.2021.12.033
Graaff Reindert MSc. PhD a∗
Eggersdorfer Manfred L. PhD b
a former unpaid associate Professor at University Medical Centre Groningen and University of Groningen, the Netherlands. Fultsemaheerd 16, 9736CN Groningen, the Netherlands
b Professor at University Medical Centre Groningen, the Netherlands, Department of Internal Medicine, Münchwilerstrasse 20, 4332 Stein, Switzerland
∗ Corresponding author. ..
7 12 2022
7 12 2022
5 5 2022
2 12 2022
© 2022 Published by Elsevier Ltd on behalf of European Society for Clinical Nutrition and Metabolism.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Keywords
Micronutritions
vitamin C
vitamin D
zinc
clinical significance
==== Body
pmc
| 0 | PMC9727962 | NO-CC CODE | 2022-12-13 23:17:23 | no | Clin Nutr ESPEN. 2022 Dec 7; doi: 10.1016/j.clnesp.2022.12.001 | utf-8 | Clin Nutr ESPEN | 2,022 | 10.1016/j.clnesp.2022.12.001 | oa_other |
==== Front
Gen Hosp Psychiatry
Gen Hosp Psychiatry
General Hospital Psychiatry
0163-8343
1873-7714
Elsevier Inc.
S0163-8343(22)00143-8
10.1016/j.genhosppsych.2022.12.002
Review Article
Prevalence of depression in SARS-CoV-2 infected patients: An umbrella review of meta-analyses
Mazza Mario Gennaro abc⁎
Palladini Mariagrazia abc
Villa Gaia a
Agnoletto Elena a
Harrington Yasmine abc
Vai Benedetta ab
Benedetti Francesco ab
a Psychiatry & Clinical Psychobiology, Division of Neuroscience, IRCCS Scientific Institute Ospedale San Raffaele, Via Stamira d'Ancona 20, Milano 20127, Italy
b Vita-Salute San Raffaele University, Via Olgettina 58, Milano 20132, Italy
c Cognitive Neuroscience, Via Olgettina 58, Milano 20132, Italy
⁎ Corresponding author at: Istituto Scientifico IRCCS Ospedale San Raffaele, Dipartimento di Neuroscienze Cliniche, San Raffaele Turro, Via Stamira d'Ancona 20, Milano 20127, Italy.
7 12 2022
7 12 2022
13 7 2022
12 11 2022
5 12 2022
© 2022 Elsevier Inc. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Objective
The COVID-19 pandemic is still spreading worldwide two years after its outbreak. Depression has been reported in around 30% of SARS-CoV-2 infected patients. We aim to synthesize the available meta-analytical evidence in an umbrella review exploring the prevalence of depression during and after SARS-CoV-2 infection.
Methods
First, we performed a narrative umbrella review including only meta-analyses providing a quantitative summary of the prevalence of depression during or after SARS-CoV-2 infection. Then we extracted the prevalence and sample size from the original studies included in each meta-analysis, and after removing duplicate studies, we performed a random-effects model meta-analysis based on single original study estimates. Heterogeneity, publication bias, leave-one-out sensitivity, and subgroup analyses were performed.
Results
14 meta-analyses were included in the umbrella review. The prevalence of depression ranged from 12% to 55% in the presence of high heterogeneity. The meta-analysis based on 85 original studies derived from the included 14 meta-analyses showed a pooled prevalence of depression of 31% (95% CI:25–38%) in the presence of high and significant heterogeneity (Q = 8988; p < 10−6; I2 = 99%) and publication bias (p < 0.001).
Conclusion
The burden of post-COVID depression substantially exceeds the pre-pandemic prevalence. Health care services for COVID-19 survivors should monitor and treat emergent depression, reducing its potential detrimental long-term effects.
Keywords
COVID-19
SARS-CoV-2
Depression
Mental health
Systematic review
Meta-analysis
==== Body
pmc1 Introduction
The Coronavirus Disease 2019 (COVID-19) has affected >600 million people and resulted in 6 million deaths two years from its outbreak (WHO) [1].
After Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) infection, approximately 20% of patients develop severe illness [2]. Furthermore, persistent and prolonged symptoms have been observed after the acute infection regardless of the COVID-19 severity [2]. The post-acute sequelae of COVID-19 (PASC) are now recognized and defined as a condition pertaining to those who still have signs and symptoms for weeks after the onset of the illness [3].
Since the pandemic spread, acute and post-acute psychopathological consequences have been reported in COVID-19 survivors [4,5]. Available evidence indicates depression as a major adverse psychiatric outcome COVID-19 survivors struggle with during infection and for several months after the infection itself [6]. Clinically significant depression during and after COVID-19 was reported in one out of three patients, with a higher risk of presenting post-COVID depressive symptomatology in females and patients with preexisting psychiatric disorders [[7], [8], [9]]. COVID-19-related depression, characterized by depressed mood and decreased interest and pleasure, displays similar psychopathological brain imaging correlates [10] and negative thinking styles [11] as major depression and affects fatigue syndrome [12], neurocognitive functioning [13], and quality of life [14] in PACS. The mechanisms underlying post-COVID depression have been associated with infection-related neuroinflammation as well as persistent psychological distress in the peri-infection period [15,16]. Furthermore, pre-existing depression during SARS-CoV-2 infection was found to detrimentally affect COVID-19 outcome, being associated with hospitalization, intensive care unit admission, and mortality [17].
Notwithstanding the impact of depression on SARS-CoV-2 infected patients' quality of life, a more precise estimate of its prevalence and associated risk factors is still lacking. Considering the overall burden of COVID-19 triggered depressive psychopathology and the previous efforts to quantitatively meta-analyze data about its epidemiology, we aim to systematically synthesize the available meta-analytical evidence of the prevalence of depression in COVID-19 patients during acute and post SARS-CoV-2 infection by providing an umbrella review.
2 Materials & methods
We performed an umbrella review exploring the prevalence of depression in COVID-19 infected patients according to the state-of-the-art methodological guidance, namely the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [18] and specific empirical recommendations for conducting umbrella reviews [19]. PRISMA checklist for meta-analysis and systematic reviews is available in eTable1. In accordance with the guidelines, we registered the protocol in the International Prospective Register of Systematic Reviews (PROSPERO) under the code CRD42022310693. Any critical protocol amendments were noted in eMethod 1.
2.1 Search strategy and study eligibility criteria
We conducted a systematic literature search of all eligible articles updated to April 5, 2022, on PubMed, Ovid/PsycINFO, Web of Science (Clarivate Analytics), and the Cochrane Library databases (eMethods 2 for the search strategy in each database). Furthermore, the reference lists of included papers were manually screened for further eligible articles.
Medical databases were queried using relevant keywords and MeSH terms with Boolean Logic operators (OR and AND). The following keywords were adapted to the database's rules: (depress* OR mood OR affective OR psychiatric OR mental health OR neuropsychiatric OR psychopathol*) AND (COVID-19 OR SARS-CoV-2 OR coronavirus) AND (prevalence OR incidence OR epidemiology OR rate OR occurrence OR frequency OR percentage OR burden) AND (meta-analysis).
In this umbrella review, meta-analyses were included if they provided a quantitative summary of the prevalence of depression in patients who contracted the SARS-CoV-2, both considering patients in acute and post-acute infection. Only systematic reviews with a meta-analytical approach including at least three studies were considered. We only included studies reporting a categorical definition of depression according to questionnaire cut-off scores or clinical diagnosis. Studies with only mean scores of depressive symptoms were excluded. Meta-analyses utilizing multiple populations or outcomes were included only if they provided separate data for depression prevalence in COVID-19 patients. Original studies, clinical case reports, abstracts, conference proceedings, preprints, or comprehensive studies that did not undergo a peer-review process were excluded. The language was restricted to English.
2.2 Study selection and data extraction
After removing duplicates, two independent researchers (MP and EA) completed the preliminary screening based on titles, abstracts, and full text according to the eligibility criteria. Any disagreement between the researchers was resolved through discussion with a third reviewer (MGM). A PRISMA flowchart describes the selection process (Fig. 1 ).Fig. 1 PRISMA 2020 flow diagram for the selection process.
Fig. 1
Two authors (GV and EA) independently extracted the data using a standardized pre-defined template discussed among the authors. A third author (MGM) cross-checked the data extraction. From each of the eligible meta-analyses, the following information was extracted: first author, year of publication, type of assessment of depressive symptoms, the number of studies included in the meta-analysis, sample size, the pooled prevalence and 95% confidence interval of depression in SARS-CoV-2 infected patients, heterogeneity, publication bias, and list and findings of subgroup analyses. When sample sizes of the primary studies were not available in the meta-analyses, the information was retrieved and calculated from the original papers.
2.3 Narrative review and statistical analysis
All included meta-analyses were summarized and a narrative data synthesis was reported, discussing the findings from subgroup analyses of each meta-analysis to investigate potential risk and protective factors (Table 1 and Table 2 ).Table 1 The characteristics of included studies.
Table 1Study Number of included original studies Total sample size Sources searched Data collection period Instruments used to measure depressive symptoms Overall prevalence and 95% CI Heterogeneity AMSTAR-2 JBI Class of evidence
Deng et al., 2020 [23] 23 4028 PubMed, Embase, Web of Science, Medline CINAHL, Wangfang Data, WangFang Med Online, CNKI, and CQVIP From 2019 to August 18, 2020 ZSDS, PHQ-9, HADS, SCL-90, or unvalidated custom questionnaires or interviews 45% (37–54%) 96% Moderate High NS
Dong et al., 2021 [24] 27 6002 PubMed, Embase, PsycInfo, Wanfang Data, CNKI, CQVIP, and Sinomed From January 1, 2020 to October 7, 2020 ZSDS, PHQ-9, HADS, SCL-90, HAMD, or PHQ-2 38% (29–46%) 98% Critically low High III
Dorri et al., 2021 [25] 7 2393a PubMed, Embase, Google Scholar Up to January 16, 2021 ZSDS, PHQ-9, HADS, BDI-13, clinical diagnosis, or a single question on depression 12% (8–17%) 92% Critically low High II
Dragioti et al., 2021 [30] 9 2088a PubMed, PsycInfo, Who COVID-19 Up to September 29, 2020 ZSDS, PHQ-9, DASS-21, GADS 28% (21–36%) 96% Low High NS
Iqbal et al., 2021 [31] 3 621a MEDLINE, Embase, PsycInfo, HMIC Up to March 6, 2021 ZSDS, BDI-13, or self-reported symptom questionnaire 20% (9–33%) 92% Critically low High IV
Khraisat et al., 2021 [6] 20 8478a PubMed, Google Scholar, MedRxiv, ScienceDirect Up to February 2021 na 21% (16–28%) 97% Critically low High II
Krishnamoorty et al., 2020 [32] 3 398a MEDLINE, CNKI, Cochrane Library, ScienceDirect, Google Scholar Up to April 22, 2020 ZSDS, PHQ-9, or HADS 42% (28–57%) 88%b Moderate High NS
Lao et al., 2020 [27] 8 2206 PubMed, Embase, Chocrane Library Up to July 30, 2020 ZSDS, PHQ-9, or HADS 44% (30–57%) 98% Critically low High NS
Liu C. et al., 2021 [26] 20 3716a PubMed, Embase, Web of Science, PsycInfo From January 1, 2020, to December 26, 2020 ZSDS, PHQ-9, HADS, SCL-90, BDI, DASS-21 38% (25–51%) 98% Moderate High NS
Liu X. et al., 2021 [33] 4 444a PubMed, Embase, Web of Science, Cochrane Library, EBSCO, Wangfang Data, CNKI, and Chinese biomedical literature service system From January 1, 2020, to July 1, 2020 ZSDS or SCL-90 55% (33–76%) 95%b Moderate High NS
Premraj et al., 2022 [29] 8 3104 PubMed, Embase, Web of Science, Google Scholar, Scopus From January 1, 2020 to August 1, 2021 na 12% (7–22%) 98% Low High II
Rogers et al., 2021 [34] 10 43,128 MEDLINE, Embase, PsycInfo, CINAHL From January 1, 2020, to July 18, 2020 na 23% (12–40%) 99% Low High IV
Wu et al., 2021 [28] 4 480 PubMed, Embase, Web of Science, Ovid, CNKI, Wanfang Data, SSRN, bioRxiv, MedRxiv From January 1, 2020, to March 16, 2020 ZSDS, HAMD, GHQ, or self-made questionnaire 42% (26–58%) 90% Moderate High IV
Yan et al., 2022 [35] 5 137a PubMed, Embase, Web of Science, PsycINFO, Scopus, CNKI, Academic Search Premier, PsycARTICLES, Psychology and Behavioral Sciences Collection, Wanfang Standards Database, CQVIP, MedRxiv Up to March 2021 ZSDS, PHQ-9, HADS 27% (14–48%) 75% Moderate High IV
Beck Depression Inventory-13 (BDI-13); Cumulative Index of Nursing and Allied Health Literature (CINAHL); China National Knowledge Infrastructure (CNKI); Chongqing VIP Information (CQVIP); Depression Anxiety Stress Scales-21 (DASS-21); General Health Questionnaire (GHQ); Goldberg Anxiety and Depression Scale (GADS); Hospital Anxiety and Depression Scale (HADS); Hamilton Rating Scale for Depression (HAMD); Healthcare Management Information Consortium (HMIC); Patient Health Questionnaire-2 (PHQ-2); Patient Health Questionnaire-9 (PHQ-9); Social Science Research Network (SSRN); Symptom Checklist-90 (SCL-90); Zung Self-Rating Depression Scale (ZSDS).
a Sample size calculated.
b I2 Calculated.
Table 2 Subgroup analyses and relative findings performed in the included meta-analyses.
Table 2Subgroup analysis Study Findings from subgroup analysis
Country Deng et al., 2020 [23] The prevalence in 20 studies conducted in China was 45% (36–55%). The prevalence for Italy, Ecuador, and Iran was 38% (29–47%), 60% (55–66%), and 38% (28–49%), respectively, as reported by one study in each subgroup
Dragioti et al., 20210 Prevalence was higher in low/middle income countries
Liu C. et al., 2021 [26] The prevalence reported in 13 studies conducted in China was 39% (25–54%), the prevalence reported in 2 studies conducted in Iran was 65% (0–100%, 95% CI). The prevalence for South Korea, Ecuador, Jordan, Turkey, and Italy was 39% (23–57%), 23% (18–28%), 44% (32–56%), 32% (17–50%), and 11% (6–18%), respectively, as reported by one study in each subgroup
Sex Deng et al., 2020 [23] Stratified data for gender were available in 9 studies, prevalence was higher in females 50% (38–62%) than in males 39% (26–53%)
Dorri et al., 2021 [25] Stratified data for gender were available in 2 studies, prevalence was higher in females 19% (15–22%) than in males 12% (9–15%)
Liu C. et al., 2021 [26] Stratified data for gender were available in 11 studies, prevalence was higher in females 46% (32–60%) than in males 32% (17–47%)
Study design Deng et al., 2020 [23] The prevalence of one cohort-study (74%, 62–83%) was higher than the prevalence of 22 cross-sectional studies (44%, 36–53%)
Dorri et al., 2021 [25] The prevalence was similar between 4 retrospective (12%, 5–18%) and 3 prospective (12%, 9–15%) cohort studies
Liu C. et al., 2021 [26] The prevalence of 3 cohort-studies (88%, 44–100%) was higher than the prevalence of 17 cross-sectional studies (34%, 21–46%)
Severity of depression Deng et al., 2020 [23] The prevalence of mild, moderate, and severe depression was 33% (26–39%, 11 studies), 14% (11–16%, 11 studies), and 7% (4–10%, 12 studies), respectively.
Lao et al., 2020 [27] The prevalence of mild, moderate, and severe depression was 31% (19–43%, 5 studies), 13% (11–15%, 4 studies), and 5% (2–8%, 4 studies), respectively.
Liu C. et al., 2021 [26] The prevalence of mild, moderate, and severe depression was 29% (24–34% 9 studies), 17% (11–22% 9 studies), and 10% (2–20% 11 studies), respectively.
Depression screening tool Deng et al., 2020 [23] The prevalence according to different depression screening tools were PHQ-9 (n = 9, 52% (45–59%)), HADS (n = 2, 20% (16–23%)), ZSDS (n = 6, 53% (42–65%)), and SCL-90 (n = 2, 19% (17–22%)). The remaining studies used unvalidated custom questionnaires or interviews (n = 4, 47% (15–80%)).
Dong et al., 2021 [4] The prevalence according to different depression screening tools were PHQ-9 (n = 9, 33%), ZSDS (n = 6, 22%), HADS (n = 4, 15%), SCL-90 (n = 4, 15%), HAMD (n = 2, 7%), and PHQ-2 (n = 2, 7%).
Dorri et al., 2021 [25] The prevalence according to different depression screening tools were PHQ-9 (n = 2, 16% (13–18%)), HADS (n = 2, 16% (13–20%)), ZSDS (n = 1, 31% (26–36%)), Self-reported questionnaire (n = 1, 4% (3–6%)), DSM-IV (n = 1, 10% (7–13%)), BDI-13 (n = 1, 11% (8–15%))
Disease stage Liu C. et al., 2021 [26] 17 studies had reported the prevalence of 42% (29–56%) in patients who were experiencing SARS-CoV-2 infection, and 3 studies conducted reported a prevalence of 14% (0–48%, 95% CI) in patients who were at the recovery stage
Lao et al., 2020 [27] The prevalence of depressive symptoms in discharged patients was higher (55%, 34–77%, 2 studies) than the prevalence in hospitalized patients was (40%, 28–52%, 6 studies)
COVID-19 severity Dong et al., 2021 [4] The prevalence of depression in patients with severe COVID-19 was 66% (16–117%, 2 studies), higher than 31% (7–55%, 4 studies) with clinically stable COVID-19. prevalence for discharged patients was 52% (25–79%, 2 studies).
Dorri et al., 2021 [25] The prevalence of depression in patients with severe COVID-19 was 22% (16–28%, 2 studies), higher than 15% (11–18%, 2 studies) with moderate and 13% (8–18%, 2 studies) mild forms.
Setting of care for COVID-19 Deng et al., 2020 [23] The prevalence was higher in 12 study reporting prevalence for inpatients (48%, 35–61%) than for the single study reporting prevalence for outpatients (35%, 22–48%)
Premraj et al., 2022 [29] The prevalence was higher in studies reporting prevalence for outpatients (25%) than for the studies reporting prevalence for inpatients (14%)
Quality Liu C. et al., 2021 [26] 7 high-quality studies had reported a prevalence of 38% (17–60%) and the remaining 13 low-quality studies reported a prevalence of 34% (16–54%)
Follow-up duration Dorri et al., 2021 [25] The prevalence was similar between studies with mean/median follow-up duration ≤31 days and those with longer follow-up time (>31): 16% (13–18%, 3 studies) vs. 15% (2–28%, 4 studies) respectively
Additionally, to better summarize the reported prevalence of depression in SARS-CoV-2 infected patients, we extracted the prevalence and sample size from the original studies included in each meta-analysis. We then performed our own meta-analysis based on all single original study estimates, removing duplicates. In doing so, we amplified the power of the analysis considering a large number of single studies and patients included. We applied random-effects models, considering the high heterogeneity observed in included meta-analyses. Between-study heterogeneity was assessed using Cochran's Q and I2 statistics [20]. Publication bias was assessed using visual inspection of funnel plots and Egger linear regression tests [20]. To test the prevalence of depression at different stages of COVID-19, when data were available in the original studies, a subgroup analysis was performed according to the time of depression assessment (during SARS-CoV-2 infection, one to three months after infection, and more than three months after infection). Finally, we did leave-one-out sensitivity analyses to investigate the effect of single studies on the overall estimate. All analyses were two-sided and were done using Comprehensive Meta-Analysis (version 3.3.070).
2.4 Quality assessment and credibility criteria
The methodological quality of included meta-analyses was independently assessed by two investigators (MP and EA) using A Measurement Tool to Assess Systematic Reviews version 2 (AMSTAR-2) [21] and the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Systematic Reviews and Research Syntheses [22].
The credibility of associations was classified into five levels according to the strength of the meta-analytical evidence: i) convincing (when the number of cases>1000, p < 10−6, I2 < 50%, 95% prediction interval excluding the null, no small-study effects, and no excess significance bias); ii) highly suggestive (when the number of cases>1000, p < 10−6, largest study with a statistically significant effect, and class I criteria were not met); iii) suggestive (when the number of cases>1000, p < 10−3 and class I–II criteria were not met); iv) weak (when p < 0.05 and class I–III criteria were not met); v) non-significant (when p > 0.05) [19]. When data to assess the strength of evidence were unavailable in the published meta-analysis, we reran the meta-analysis with the reported sample size and prevalence to calculate, when necessary, overall significance, I2, publication bias, and small-study effects. Regardless of methodological quality, all studies were included in the review.
3 Results
3.1 Umbrella narrative review of meta-analyses
The literature search identified 683 studies. After excluding duplicate, non-relevant titles, and abstracts, 56 full-text articles were screened. 14 meta-analyses [6,[23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35]] met our inclusion criteria and were included in this umbrella review (Fig. 1, Table 1, eTable 2 for excluded meta-analyses after full-text assessment). The median number of original studies included in the meta-analyses was 8 (ranging from 3 [32] to 27 [24] studies), while the median sample size of each meta-analysis was 2299 (ranging from 137 [35] to 43,128 [34] patients). Among the included meta-analyses, the prevalence of depression in SARS-CoV-2 infected patients ranged from 12% (95% CI 7–22%) [29] to 55% (95% CI 33–76%) [33]. All the included meta-analyses showed high heterogeneity (I [2] ranging from 88 to 99%). The complete list of original studies for each included meta-analysis is available in eTable 3.
Some meta-analyses explored the effect of covariates known to affect depression prevalence in subgroup analyses (Table 2). Three meta-analyses consistently found that females showed a higher prevalence of depressive symptoms after SARS-CoV-2 infection compared to males [23,25,26]. Stratifying by depression severity, three meta-analyses consistently reported a higher prevalence of mild depression (prevalence from 29% to 33%), followed by moderate depression (prevalence from 13% to 17%) and, ultimately, severe depression (prevalence from 5% and 10%) [23,26,27]. When subgrouping the prevalence of screening tools for depression, the highest occurrence of depressive psychopathology was recorded through the Patient-Health Questionnaire-9 (PHQ-9) and Zung Self-Rating Depression Scale (ZSDS), while a lower prevalence was found using the Hospital Anxiety and Depression Scale (HADS-depression subscale), Symptom Checklist-90 (SCL-90), and Hamilton Rating Scale for Depression (HAMD) [[23], [24], [25]]. Two meta-analyses consistently revealed an impressive higher prevalence of depression in cohort studies (74%–88%) compared to cross-sectional investigations (34%–44%) [23,26]. The prevalence of depression in patients with severe COVID-19 was consistently higher than the prevalence in patients with milder forms in two meta-analyses [24,25]. Inconsistent findings were reported on the effect of setting of care for acute COVID-19 in two studies, with depression prevalence being alternately higher or lower in inpatients compared to outpatients [23,29]. Stratifying by the stage of COVID-19 inflection also revealed mixed evidence, highlighting higher depression burden during the ongoing infection or, conversely, after hospital discharge [26,27]. One meta-analysis investigating the effect of quality of study [26] and another exploring the follow-up duration [25] found no differences between groups.
3.2 Meta-analysis of the original studies
Based on 85 single original studies derived from the included 14 meta-analyses (eTable3), the estimated pooled prevalence of depression in 62,318 COVID-19 infected patients was 31% (95% CI:25–38%) (Fig. 2 ). Significantly high heterogeneity was observed among included studies (Q = 8988; p < 0.001; I2 = 99%).Fig. 2 Forest plot of depression prevalence in SARS-CoV-2 infected patients.
Fig. 2
Subgroup analysis showed a significantly lower prevalence of depression at longer post-COVID follow-ups (T = 14.20, p < 0.001). Specifically, the prevalence was 44% during infection (39 studies; 95%CI 32–55%; I2 = 99.47), 20% one to three months after infection (14 studies; 95%CI 10–35%; I2 = 91.34), and 15% more than three months after infection (18 studies; 95%CI 8–26%; I2 = 94.11) (eFigure 1). Sensitivity analysis leaving out single studies revealed that no single study had a significant impact on the pooled prevalence that lay between the value of 30% and 32% always remaining statistically significant (p ranging from <0. 0000001 to 0.0000007) (eTable 4). Visual examination of the funnel plot (eFigure 2) and Egger's and Begg's tests revealed evidence of publication bias (p < 0.001).
3.3 Quality assessment and credibility of evaluation
Based on the AMSTAR-2 assessment, six meta-analyses (43%) met the moderate quality level, three (21%) were of low quality, and finally, five (36%) were rated as critically low quality (Table 1, eTable 5). On the other hand, when considering JBI Checklist for Systematic Reviews and Research Syntheses, six meta-analyses met all 11 criteria for quality assessment, and all the included meta-analyses were classified as high quality, satisfying at least eight items (Table 1, eTable 6).
The credibility of associations was found to be highly suggestive in three (21%) meta-analyses, suggestive in one (7%) meta-analysis, and weak in four (29%) meta-analyses. Moreover, six (43%) meta-analyses were classified as non-significant reporting an overall p-value >0.05 (Table 1, eTable 7). Due to the high rate of heterogeneity in all the included meta-analyses, none of them was classified as convincing.
4 Discussion
To our knowledge, this is the first umbrella review that systematically provides a comprehensive synthesis of the prevalence of depression during and after SARS-CoV-2 infection. Fourteen existing meta-analyses met the inclusion criteria, and the narrative review showed that the prevalence of depression in SARS-CoV-2 infected patients is relatively high, ranging from 12% up to 55%. Moreover, after removing duplicate studies, the overall meta-analysis of 85 single original studies derived from the 14 included meta-analyses showed a prevalence of 31% (95% CI:25–38%) in 62,318 patients in the presence of high heterogeneity.
The available prevalence estimates indicate a remarkable boost in the burden of COVID-19-related depression [36]. Globally, an increase of 27% in the cases of major depressive disorder was estimated due to the COVID-19 pandemic, potentially causing almost 50 million disability-adjusted life-years in 2020 [37]. Two COVID-19 pandemic severity indicators, daily SARS-CoV-2 infection rates and reductions in human mobility, were significantly associated with an increased prevalence of major depression [37]. Interesting, depressive episodes were significantly more common in patients who had COVID-19 than in those who had influenza (hazard ratio [HR] 1.47, 95% CI 1.42–1.53) and those who had other respiratory tract infections (HR 1.23, 1.20–1.26) [4]. Moreover, a significant difference in depression prevalence was found between SARS-CoV-2 infected and non-infected patients during the COVID-19 pandemic. Infected individuals were at higher risk of developing depressive symptoms than both healthcare workers and the general population during the COVID-19 outbreak [28].
Current insight into immunopsychiatry suggests that SARS-CoV-2 infection-triggered perturbation of the immune-inflammatory system can foster depressive psychopathology, serving as a biological risk factor alongside COVID-19 pandemic related psychological stressors for depression [38]. In particular, from a biological perspective, the dysregulation of the innate and adaptive immune systems induced by SARS-CoV-2 infection, which leads to neurotransmitter dysregulation, is a proposed mechanism underpinning depressive psychopathology [39]. In addition to the immunological mechanisms, social isolation, quarantine, uncertainty of the future, massive media exposure, and survivor's guilt experienced by patients during the COVID-19 pandemic are all significant psychological stressors that may define psychopathological outcome [40,41].
We observed a decreasing depression prevalence at post-COVID follow-ups. The depression prevalence decreased from 44% during acute COVID-19 infection to 15% at three or more months follow-up after infection. These findings are consistent with prevalence patterns of depression during previous outbreaks of other coronaviruses [42]. Sparse longitudinal studies have directly investigated the trajectory of depression after SARS-CoV-2 infection, reporting inconsistent findings suggesting both persistent [8,9,43] and decreasing [44] depressive psychopathology over follow-ups. The subgroups meta-analysis however did not explain the heterogeneity that remained high even stratifying the studies according to the time of depression assessment. To better understand the course of depression over time long-term longitudinal follow-up studies on large cohorts of patients are needed. However, given the global burden of COVID-19 that has affected >500 million patients, the reported prevalence of depression represents an urgent clinical need that merits the attention of mental health services. Preliminary small studies suggest the potential efficacy and tolerability of pharmacological [45] and psychological [46] treatment of depressive symptomatology in SARS-CoV-2 infected patients.
Notably, in our findings, the reported high prevalence of depression was systematically accompanied by high heterogeneity thus indicating a potential effect of several clinical variables on the prevalence in the included single studies. The included studies examined very different populations in terms of COVID-19 severity, site of care, premorbid mental health history, also using different assessment instruments. When data were available in the included meta-analyses, we investigated the effect of these factors to better dealing with the source of heterogeneity. Accordingly, various meta-analyses suggested that sex, the severity of depressive symptoms, and screening tools for depressive psychopathology assessment consistently affected the prevalence estimate (Table 2). However, other risk factors potentially affecting depressive psychopathology in COVID-19 survivors where not investigated in previous meta-analyses and need to be addressed. In this context, social isolation, pandemic related psychological stressor, previous psychiatric history, pre-infection psychopharmacological treatment, medical comorbidity, and COVID-19 severity could influence the risk of presenting post-COVID depressive psychopathology directly or by interacting with the neuro-immune pathway. Although it is reasonable that all these factors could contribute to a single point estimate for depression prevalence in infected patients, no clear answers can be concluded from meta-analytical evidence and further studies are needed to investigate the reported high heterogeneity.
To check for the risk of methodological biases, we rated the included meta-analyses according both the AMSTAR-2 and JBI checklist. Discrepancies emerged between the two quality assessment tools as all included meta-analyses were rated as high quality according to the JBI checklist. In contrast, the lack of justification for excluding individual studies (item 7) in 13 of the 14 included meta-analyses prevented the studies from being classified as high quality according with AMSTAR-2. The JBI critical appraisal checklist seems to be one of the most accurate assessment tools for descriptive studies reporting prevalence data [47]. On the other hand, AMSTAR-2, one of the most commonly used tools, was designed and recommended for randomized or non-randomized studies on healthcare interventions [21]. The differing purposes of the study quality rating methods could partially explain inconsistencies in our results. Moreover, we graded the strength of associations by using well-recognized credibility criteria [19]. The high heterogeneity prevented all the meta-analyses from being classified as convincing. Moreover, six out of 14 meta-analyses had an overall p-value >0.05 and thus were classified as non-significant. However, these methodological weaknesses do not significantly bias the main findings that the burden of COVID related depression substantially exceeds the pre-pandemic prevalence of depression [37] in the general population using the same rating scales [48], and are similar to the prevalence observed in patients with immune-inflammatory diseases [49,50].
Despite the rigour in which this umbrella review was conducted, the emerging framework must be interpreted in light of some limitations. First, all the included meta-analyses as well our meta-analysis of the original studies were characterized by high heterogeneity, thus limiting the generalizability of our findings. Second, the prevalence estimates were based on observational studies, which do not imply causality but only association. Finally, the majority of reviewed studies, used self-rated questionnaires for depressive psychopathology without a psychiatric interview, thus preventing the clinical diagnosis of a major depressive episode.
5 Conclusion
In conclusion, considering the COVID-19 pandemic is still spreading worldwide two years from its outbreak and given the alarming prevalence of potentially persistent depression in infected patients, there is an urgent need for clinical intervention. According to current literature [3], mental-health follow-up services for COVID-19 survivors should be implemented to monitor emergent symptoms and provide early treatment in order to reduce the impact of psychopathology on global functioning and quality of life. The worrying rise of clinically relevant depression prevalence in acute and post-COVID stages places a new challenge on mental care services that needs to be addressed. Finally, given the reported significant heterogeneity further studies are needed to investigate the relevance of different risk and protective factors able to affect the COVID-related depression prevalence.
Funding statement
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
None.
Appendix A Supplementary data
Supplementary material
Image 1
Data availability
Data will be made available on request.
Acknowledgments
None.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.genhosppsych.2022.12.002.
==== Refs
References
1 Organization WH WHO Coronavirus (COVID-19) Dashboard In https://covid19.who.int/ 2022
2 Huang C. Wang Y. Li X. Clinical features of patients infected with 2019 novel coronavirus in Wuhan China Lancet 395 10223 2020 497 506 31986264
3 Nalbandian A. Sehgal K. Gupta A. Post-acute COVID-19 syndrome Nat Med 27 4 2021 601 615 33753937
4 Taquet M. Geddes J.R. Husain M. Luciano S. Harrison P.J. 6-month neurological and psychiatric outcomes in 236 379 survivors of COVID-19: A retrospective cohort study using electronic health records Lancet Psychiatry 8 5 2021 416 427 33836148
5 Taquet M. Luciano S. Geddes J.R. Harrison P.J. Bidirectional associations between COVID-19 and psychiatric disorder: Retrospective cohort studies of 62 354 COVID-19 cases in the USA Lancet Psychiatry 8 2 2021 130 140 33181098
6 Khraisat B. Toubasi A. AlZoubi L. Al-Sayegh T. Mansour A. Meta-analysis of prevalence: The psychological sequelae among COVID-19 survivors Int J Psychiatry Clin Pract 1-10 2021
7 Mazza M.G. De Lorenzo R. Conte C. Anxiety and depression in COVID-19 survivors: Role of inflammatory and clinical predictors Brain Behav Immun 89 2020 594 600 32738287
8 Mazza M.G. Palladini M. De Lorenzo R. One-year mental health outcomes in a cohort of COVID-19 survivors J Psychiatr Res 145 2021 118 124 34894521
9 Mazza M.G. Palladini M. De Lorenzo R. Persistent psychopathology and neurocognitive impairment in COVID-19 survivors: Effect of inflammatory biomarkers at three-month follow-up Brain Behav Immun 94 2021 138 147 33639239
10 Benedetti F. Palladini M. Paolini M. Brain correlates of depression, post-traumatic distress, and inflammatory biomarkers in COVID-19 survivors: A multimodal magnetic resonance imaging study Brain Behav Immun Health 18 2021 100387
11 Benedetti F. Palladini M. D'Orsi G. Mood-congruent negative thinking styles and cognitive vulnerability in depressed COVID-19 survivors: A comparison with major depressive disorder J Affect Disord 308 2022 554 561 35460737
12 Townsend L. Dyer A.H. Jones K. Persistent fatigue following SARS-CoV-2 infection is common and independent of severity of initial infection PLoS One 15 11 2020 e0240784
13 Poletti S. Palladini M. Mazza M.G. Long-term consequences of COVID-19 on cognitive functioning up to 6 months after discharge: role of depression and impact on quality of life Eur Arch Psychiatry Clin Neurosci 2021
14 Babicki M. Bogudzinska B. Kowalski K. Mastalerz-Migas A. Anxiety and depressive disorders and quality of life assessment of poles-a study covering two waves of the COVID-19 pandemic Front Psychol 12 2021 704248
15 Troyer E.A. Kohn J.N. Hong S. Are we facing a crashing wave of neuropsychiatric sequelae of COVID-19? Neuropsychiatric symptoms and potential immunologic mechanisms Brain Behav Immun 87 2020 34 39 32298803
16 Passavanti M. Argentieri A. Barbieri D.M. The psychological impact of COVID-19 and restrictive measures in the world J Affect Disord 283 2021 36 51 33516085
17 Vai B. Mazza M.G. Delli Colli C. Mental disorders and risk of COVID-19-related mortality, hospitalisation, and intensive care unit admission: A systematic review and meta-analysis Lancet Psychiatry 8 9 2021 797 812 34274033
18 Moher D. Liberati A. Tetzlaff J. Altman D.G. Group P Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement BMJ. 339 2009 b2535
19 Fusar-Poli P. Radua J. Ten simple rules for conducting umbrella reviews Evid Based Ment Health 21 3 2018 95 100 30006442
20 Cumpston M. Li T. Page M.J. Updated guidance for trusted systematic reviews: A new edition of the Cochrane handbook for systematic reviews of interventions Cochrane Database Syst Rev 10 2019 ED000142
21 Shea B.J. Reeves B.C. Wells G. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both BMJ. 358 2017 j4008
22 Aromataris E. Fernandez R. Godfrey C.M. Holly C. Khalil H. Tungpunkom P. Summarizing systematic reviews: Methodological development, conduct and reporting of an umbrella review approach Int J Evid Based Health C 13 3 2015 132 140
23 Deng J. Zhou F. Hou W. The prevalence of depression, anxiety, and sleep disturbances in COVID-19 patients: a meta-analysis Ann N Y Acad Sci 1486 1 2021 90 111 33009668
24 Dong F. Liu H.L. Dai N. Yang M. Liu J.P. A living systematic review of the psychological problems in people suffering from COVID-19 J Affect Disord 292 2021 172 188 34126309
25 Dorri M. Mozafari Bazargany M.H. Khodaparast Z. Psychological problems and reduced health-related quality of life in the COVID-19 survivors J Affect Disord Rep 6 2021 100248
26 Liu C. Pan W. Li L. Li B. Ren Y. Ma X. Prevalence of depression, anxiety, and insomnia symptoms among patients with COVID-19: a meta-analysis of quality effects model J Psychosom Res 147 2021 110516
27 Lao Y. Jiang Y. Luo X. Liu X. Focus on the depressive symptoms in COVID-19 patients: Perspective based on a rapid meta-analysis Asian J Psychiatr 54 2020 102421
28 Wu T. Jia X. Shi H. Prevalence of mental health problems during the COVID-19 pandemic: A systematic review and meta-analysis J Affect Disord 281 2021 91 98 33310451
29 Premraj L. Kannapadi N.V. Briggs J. Mid and long-term neurological and neuropsychiatric manifestations of post-COVID-19 syndrome: A meta-analysis J Neurol Sci 434 2022 120162
30 Dragioti E. Li H. Tsitsas G. A large-scale meta-analytic atlas of mental health problems prevalence during the COVID-19 early pandemic J Med Virol 94 5 2022 1935 1949 34958144
31 Iqbal F.M. Lam K. Sounderajah V. Clarke J.M. Ashrafian H. Darzi A. Characteristics and predictors of acute and chronic post-COVID syndrome: a systematic review and meta-analysis EClinicalMedicine. 36 2021 100899
32 Krishnamoorthy Y. Nagarajan R. Saya G.K. Menon V. Prevalence of psychological morbidities among general population, healthcare workers and COVID-19 patients amidst the COVID-19 pandemic: A systematic review and meta-analysis Psychiatry Res 293 2020 113382
33 Liu X. Zhu M. Zhang R. Public mental health problems during COVID-19 pandemic: A large-scale meta-analysis of the evidence Transl Psychiatry 11 1 2021 384 34244469
34 Rogers J.P. Watson C.J. Badenoch J. Neurology and neuropsychiatry of COVID-19: A systematic review and meta-analysis of the early literature reveals frequent CNS manifestations and key emerging narratives J Neurol Neurosurg Psychiatry 92 9 2021 932 941 34083395
35 Yan Y. Du X. Lai L. Ren Z. Li H. Prevalence of depressive and anxiety symptoms among Chinese older adults during the COVID-19 pandemic: A systematic review and meta-analysis J Geriatr Psychiatry Neurol 35 2 2022 182 195 35245999
36 Ettman C.K. Abdalla S.M. Cohen G.H. Sampson L. Vivier P.M. Galea S. Prevalence of depression symptoms in US adults before and during the COVID-19 pandemic JAMA Netw Open 3 9 2020 e2019686
37 Collaborators C.-M.D. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic Lancet. 398 10312 2021 1700 1712 34634250
38 Miller A.H. Raison C.L. The role of inflammation in depression: From evolutionary imperative to modern treatment target Nat Rev Immunol 16 1 2016 22 34 26711676
39 Najjar S. Pearlman D.M. Alper K. Najjar A. Devinsky O. Neuroinflammation and psychiatric illness J Neuroinflammation 10 2013 43 23547920
40 Singh S. Roy D. Sinha K. Parveen S. Sharma G. Joshi G. Impact of COVID-19 and lockdown on mental health of children and adolescents: A narrative review with recommendations Psychiatry Res 293 2020 113429
41 Guo Q. Zheng Y. Shi J. Immediate psychological distress in quarantined patients with COVID-19 and its association with peripheral inflammation: A mixed-method study Brain Behav Immun 88 2020 17 27 32416290
42 Rogers J.P. Chesney E. Oliver D. Psychiatric and neuropsychiatric presentations associated with severe coronavirus infections: A systematic review and meta-analysis with comparison to the COVID-19 pandemic Lancet Psychiatry 7 7 2020 611 627 32437679
43 Huang L. Yao Q. Gu X. 1-year outcomes in hospital survivors with COVID-19: A longitudinal cohort study Lancet. 398 10302 2021 747 758 34454673
44 Fernandez-de-Las-Penas C. Martin-Guerrero J.D. Cancela-Cilleruelo I. Moro-Lopez-Menchero P. Rodriguez-Jimenez J. Pellicer-Valero O.J. Trajectory curves of post-COVID anxiety/depressive symptoms and sleep quality in previously hospitalized COVID-19 survivors: The LONG-COVID-EXP-CM multicenter study Psychol Med 1-2 2022
45 Mazza M.G. Zanardi R. Palladini M. Rovere-Querini P. Benedetti F. Rapid response to selective serotonin reuptake inhibitors in post-COVID depression Eur Neuropsychopharmacol 54 2022 1 6 34634679
46 Abas M.A. Combining active ingredients to treat depression in the wake of COVID-19 Lancet Psychiatry 9 3 2022 190 191 34895478
47 Munn Z. Moola S. Lisy K. Riitano D. Tufanaru C. Methodological guidance for systematic reviews of observational epidemiological studies reporting prevalence and cumulative incidence data Int J Evid Based Health C 13 3 2015 147 153
48 Serrano D. Marti-Lluch R. Cardenas M. Gender analysis of the frequency and course of depressive disorders and relationship with personality traits in general population: A prospective cohort study J Affect Disord 302 2022 241 248 35085673
49 Zhang L. Wu Y. Liu S. Zhu W. Prevalence of depression in ankylosing spondylitis: A systematic review and Meta-analysis Psychiatry Investig 16 8 2019 565 574
50 Zhao S. Thong D. Miller N. The prevalence of depression in axial spondyloarthritis and its association with disease activity: A systematic review and meta-analysis Arthritis Res Ther 20 1 2018 140 29996916
| 0 | PMC9727963 | NO-CC CODE | 2022-12-15 23:15:13 | no | Gen Hosp Psychiatry. 2023 Dec 7 January-February; 80:17-25 | utf-8 | Gen Hosp Psychiatry | 2,022 | 10.1016/j.genhosppsych.2022.12.002 | oa_other |
==== Front
Epidemics
Epidemics
Epidemics
1755-4365
1878-0067
The Authors. Published by Elsevier B.V.
S1755-4365(22)00100-1
10.1016/j.epidem.2022.100660
100660
Article
The risk of SARS-CoV-2 Omicron variant emergence in low and middle-income countries (LMICs)
Bi Kaiming a1
Herrera-Diestra Jose Luis a1
Bai Yuan bc1
Du Zhanwei bc
Wang Lin d
Gibson Graham a
Johnson-Leon Maureen a
Fox Spencer J. a
Meyers Lauren Ancel ae⁎
a The University of Texas at Austin, Austin, Texas 78712, The United States of America
b WHO Collaborating Center for Infectious Disease Epidemiology and Control, School of Public Health, LKS Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
c Laboratory of Data Discovery for Health, Hong Kong Science and Technology Park, Hong Kong Special Administrative Region, China
d University of Cambridge, Cambridge CB2 3EH, UK
e Santa Fe Institute, Santa Fe, New Mexico, The United States of America
⁎ Corresponding author at: The University of Texas at Austin, Austin, Texas 78712, The United States of America,
1 These first authors contributed equally to this article
7 12 2022
7 12 2022
10066024 1 2022
13 11 2022
6 12 2022
© 2022 The Authors. Published by Elsevier B.V.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
We estimated the probability of undetected emergence of the SARS-CoV-2 Omicron variant in 25 low and middle-income countries (LMICs) prior to December 5, 2021. In nine countries, the risk exceeds 50%; in Turkey, Pakistan and the Philippines, it exceeds 99%. Risks are generally lower in the Americas than Europe or Asia.
Keywords
Omicron
LMICs
Risk
Emergence
Covid Importation
==== Body
pmc1 Main text
The B.1.1.529 SARS-COV-2 variant was first detected and reported in South Africa on November 26th, 2021 (WHO, 2021). By December 5th, 2021, more than 40 countries reported Omicron variant cases. Most of these countries are developed and high-income countries, including the United Kingdom, the United States, and Netherlands (GISAID,). However, low and middle income countries (LMICs) may be less likely to detect a new variant (Sisa et al., 2021) and more vulnerable to catastrophic public health outcomes than high-income countries, because of lower capacity for COVID-19 testing, vaccination, and medical treatment (Helmy et al., 2016, Siow et al., 2020). Only a few LMICs have direct flights from the countries in Southern Africa where Omicron was initially detected (Flightradar24,) and many enacted border policies to reduce Omicron importation risks from these countries (Mallapaty.). However, LMICs were at risk for Omicron importations from large international destinations outside of Southern Africa in which Omicron emerged in late 2021.
We analyzed the risks of the Omicron variant importation in 25 LMICs in which Omicron was not reported as of December 5, 2021: Bangladesh, Nepal, Philippines, Colombia, Egypt, Pakistan, Paraguay, Turkey, Serbia, Bolivia, Argentina, Uruguay, Bhutan, Indonesia, Albania, Jordan, Panama, Dominican Republic, Ecuador, Peru, Jamaica, Honduras, Guatemala, Costa Rica, and El Salvador. We first estimated the daily travel volume to each country from 13 large countries in which Omicron had already been detected, based on data from Facebook Data for Good (Figure A) (Data for Good Tools and Data.). We estimated the prevalence of Omicron in each of the 13 Omicron detected countries (ODCs) assuming that only 2.5% of early cases were identified and reported (GISAID, n.d, Davis et al., 2021), and then estimated the probability of travel-based introductions into each LMIC by December 5, 2021 (see Supporting Information). The European LMICs (Serbia and Turkey), which are highly connected to Western European countries that reported Omicron cases by November 2021, have the highest estimated risks, followed by the Asian LMICs (Pakistan, Bangladesh, and Nepal), with high inflows of travelers from South Africa (via connecting flights), the UK, and India. LMICs in the Americas (Colombia, Dominican Republic, and Paraguay) are primarily at risk for importations from the US and Brazil. We estimate that 6 of the 25 studied LMICs had over a 50% chance of having received at least one travel-based Omicron importation from ODCs by December 5, 2021 (Figure B).
To assess the risk of Omicron transmission following importation, we estimate the immunity-based effective reproduction number (Re) in each LMIC as of December 5, 2021, based on reported vaccination levels (Mathieu et al., 2021), estimates of infection-acquired immunity (World Health Organization, 2021) (Figure C), and recent estimates for the transmissibility and immune-evasiveness of the Omicron variant (Miller et al., 2012) (see Supporting Information). Recent studies have estimated that Omicron has double the reproduction number of the Delta variant, suggesting a basic reproduction number (R0) of 11.88 (95%CI: 9.16-14.61) (Figure SI.3) (Chen et al., 2021). The same study also suggests that SARS-CoV-2 vaccines are 50% less effective (VE) against Omicron compared to Delta (Figure SI.5) (Chen et al., 2021). Given the low vaccination coverage and high infectivity of Omicron, we estimate that the immunity-based Re of Omicron on December 5, 2021 ranges from 7.0 to 9.4 across the 25 studied LMICs, without additional public health interventions (Figure D).
Combining our estimates of Omicron importation and transmission risks, we estimate the probability of undetected Omicron transmission in LMICs by early December (Figure E- details see Supporting Information). Among the studied LMICs, Turkey (99.99%), Pakistan (99.95%), and Serbia (99.81%) have the highest estimated risk, followed by Nepal (87.98%), Bangladesh (84.86%), and the Dominican Republic (82.21%). If these countries implement non-pharmaceutical interventions that reduce transmission by 80%, the probability of undetected emergence declines by 12.02% to 80.77%% across the 25 LMICs (Figure E). Given the high socioeconomic costs of travel restrictions and some non-pharmaceutical interventions, many of these LMICs did not take measures to prevent introductions or slow spread (Torres-Rueda et al., 2021). Our analyses suggest that SARS-CoV-2 variants like Omicron can rapidly emerge in LMICs and spread for weeks before detection.( Fig. 1)Fig. 1 : Risks of Omicron emergence in 25 low- and middle-income countries by December 5, 2021. A. Cumulative number of travelers to each of the 25 LMICs reported by Facebook Data for Good from November 16 to December 5, 2021 from 13 countries in which Omicron had already been detected by December 5 (ODCs): Belgium, Italy, United Kingdom, Germany, Canada, Austria, Japan, Brazil, United State, United Arab Emirates, Saudi Arabia, France, India (Data for Good Tools and Data.). B. The probability of receiving at least one Omicron importation via travelers from the 13 ODCs between November 16 and December 5, 2021. C. Estimated vaccination and prior infection rates for each LMIC, assuming a 40% infection reporting rate for all countries (Irons and Raftery, 2021). D. Estimated effective reproduction number for the Omicron variant in each LMIC, assuming that Omicron is twice as transmissible as the Delta variant and that vaccines have 50% lower efficacy against Omicron in comparison to Delta. E. Comparisons of the estimated risks of the undetected Omicron transmission in LMICs with and without intervention that reduces transmission by 80%.
Fig. 1
In November 2022, we retrospectively compared our early estimates of Omicron importation and transmission risks with reported data and found a significant correlation between our estimated risks of importation and the timing the first case was detected (p < 0.05) and a non-significant but positive correlation between our estimates of transmission risks and the time at which incidence reached 25% of the winter Omicron peak ( Fig. 2 and Table SI.10). Across all 25 LMICs, the average time between the first detection and incidence reaching the 25% mark was 19.88 days, with a standard deviation of 1.51 days. Given the short window between detection and wide transmission, LMIC countries with limited surveillance capabilities should consider initiating control measures when threatening variants are first identified in other countries.Fig. 2 Retrospective comparison of estimated to observed Omicron importation and transmission events. (A) The probability of receiving at least one Omicron importation by December 5, 2021 compared to the first Omicron detection dates in LMICs. (B) Estimated risks of undetected Omicron transmission compared to the date on which reported incidence reached 25% of the eventual peak incidence in LMICs (Ritchie et al., 2020). Solid lines and shading represent the fitted regression lines and 95% confidence intervals, respectively.
Fig. 2
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary material
Supplementary material
Data availability
The authors do not have permission to share data.
Appendix A Supplementary data associated with this article can be found in the online version at doi:10.1016/j.epidem.2022.100660.
==== Refs
References
WHO. 2021. “Update on Omicron.” Www.who.int. November 28, 2021. https://www.who.int/news/item/28-11-2021-update-on-omicron.
GISAID - HCov19 Variants.” n.d. Www.gisaid.org. https://www.gisaid.org/hcov19-variants/.
Sisa Ivan Fornasini Marco Teran Enrique "COVID-19 research in LMICs." The Lancet 398 no. 10307 2021 1212 1213
Helmy Mohamed Awad Mohamed Mosa Kareem A. "Limited resources of genome sequencing in developing countries: challenges and solutions." Applied & translational genomics 9 2016 15 19 27354935
Siow Wen Ting Mei Fong Liew Babu Raja Shrestha Faisal Muchtar Kay Choong See. "Managing COVID-19 in resource-limited settings: critical care considerations Critical Care 24 no. 1 2020 1 5
Flightradar24. n.d. “Live Flight Tracker - Real-Time Flight Tracker Map.” Flightradar24. Accessed December 12, 2021. https://www.flightradar24.com/data/airports.
Mallapaty, Smriti. "Omicron-variant border bans ignore the evidence, say scientists." Nature.
Data for Good Tools and Data.” n.d. Dataforgood.facebook.com. https://dataforgood.facebook.com/dfg/tools.
Davis Jessica T. Matteo Chinazzi Nicola Perra Kunpeng Mu Marco Ajelli Dean Natalie E. Corrado Gioannini Cryptic transmission of SARS-CoV-2 and the first COVID-19 wave Nature 2021 1 9
Mathieu E. Ritchie H. Ortiz-Ospina E. A global database of COVID-19 vaccinations Nat Hum Behav 2021 10.1038/s41562-021-01122-8
World Health Organization. 2021. “WHO COVID-19 Dashboard.” Covid19.Who.int. World Health Organization. 2021. https://covid19.who.int/.
Miller Joel C. Anja C.Slim Erik M.Volz "Edge-based compartmental modelling for infectious disease spread." Journal of the Royal Society Interface 9 no. 70 2012 890 906
Chen, Jiahui, Rui Wang, Nancy Benovich Gilby, and Guo-Wei Wei. "Omicron (B. 1.1. 529): Infectivity, vaccine breakthrough, and antibody resistance." arXiv preprint arXiv:2112.01318 (2021).
Torres-Rueda Sergio Sweeney Sedona Bozzani Fiammetta Naylor Nichola R. Baker Tim Pearson Carl Rosalind Eggo "Stark choices: exploring health sector costs of policy responses to COVID-19 in low-income and middle-income countries BMJ global health 6 12 2021 e005759
Irons Nicholas J. Raftery Adrian E. "Estimating SARS-CoV-2 Infections from Deaths, Confirmed Cases, Tests, and Random Surveys arXiv preprint arXiv 2102 2021 10741
Hannah Ritchie, Edouard Mathieu, Lucas Rodés-Guirao, Cameron Appel, Charlie Giattino, Esteban Ortiz-Ospina, Joe Hasell, Bobbie Macdonald, Diana Beltekian and Max Roser (2020) - "Coronavirus Pandemic (COVID-19)". Published online at OurWorldInData.org. Ret
| 0 | PMC9727964 | NO-CC CODE | 2022-12-08 23:18:58 | no | Epidemics. 2022 Dec 7;:100660 | utf-8 | Epidemics | 2,022 | 10.1016/j.epidem.2022.100660 | oa_other |
Subsets and Splits