title
stringlengths 1
1.19k
| keywords
stringlengths 0
668
| concept
stringlengths 0
909
| paragraph
stringlengths 0
61.8k
| PMID
stringlengths 10
11
|
|---|---|---|---|---|
Participants and recruitment | RECRUITMENT, UPPER GASTROINTESTINAL BLEEDING | Participants who fulfil the following criteria are eligible:Inclusion criteriaAged ≥ 18 yearsAim to undergo screening, surveillance, and diagnosisUndergo sedated EGDAble to read, understand, and sign informed consentExclusion criteriaEGD contraindicationsNot suitable for sedated endoscopy after anaesthesia evaluationBiopsy contraindicationsActive upper gastrointestinal bleeding or emergency oesophagogastroduodenoscopy (EGD)PregnancyUpper gastrointestinal surgery or residual stomachNot suitable for recruitment after investigator evaluation because of other high-risk conditions | PMC10176798 |
|
Interventions | PMC10176798 |
|||
Procedures | nodularity, depression, upper gastrointestinal tract landmarks, upper gastrointestinal lesion | GASTRIC NEOPLASM, BLIND | The ENDOANGEL-GC is an AI system designed for assisting in EGD and possesses three functions in real time: (1) to time the entire procedure (from the endoscope intubating into the mouth to it being drawn out), (2) to record observed upper gastrointestinal tract landmarks and blind spots, and (3) to mark the upper gastrointestinal lesion with a blue rectangular box (box colour turns red if the lesion is predicted to have a high risk of gastric neoplasm) [The ENDOANGEL-GC is connected to the endoscopy processor, receives digital images as input, and outputs the predictions of landmark monitoring and lesions. The ENDOANGEL-GC is installed on a separate computer system, and the output of the system appears on the same monitor as that of the endoscope. A button on the system is set to switch between “assistance” and “no assistance” modes. Before examination in the control group, the ENDOANGEL-GC is shifted to the “no assistance” mode, so the predictions will be concealed, as they are not displayed on the monitor. Before the examination in the experimental group, the ENDOANGEL-GC is shifted to the “assistance” mode, so the predictions and prompts are displayed on the screen.The endoscopists who participate in the study are required to have EGD experience with more than 500 cases. All participating endoscopists will be trained in the AI’s functions, including monitoring anatomical landmarks, detecting suspicious lesions, and becoming familiar with the operation requirements during endoscopy. Each endoscopist will perform at least 10 examinations with AI assistance. In both groups, endoscopists are required to screen the entire stomach under white light mode. Only when the screening process is completed and suspicious lesions are found, the image-enhanced mode will be allowed for further judgement. In the experimental group, the endoscopists will operate the endoscope with the assistance of ENDOANGEL-GC. They are required to adhere to the following criteria: (1) screen anatomical landmarks according to the feedback of ENDOANGEL-GC and (2) lesions marked in red boxes by ENDOANGEL-GC are required to undergo biopsy sampling. Lesions that are not marked in red boxes but meet any of the following criteria are also recommended for biopsy: differences in colour, loss of vascularity, slight elevation or depression, nodularity, thickening, abnormal convergence or flattening of folds, irregular margins, irregular discoloration, or irregular surfaces. In the control group, endoscopists will routinely operate on the endoscope. Lesions that meet the above criteria are also recommended for biopsy. | PMC10176798 |
Adherence and protocol deviations | PATHOLOGY, GASTRIC RETENTION, OESOPHAGEAL OBSTRUCTION | To enhance the validity of the data, face-to-face adherence reminder sessions and a pilot study will be conducted before enrolment at each study site. Additionally, a key method will be followed for assessing adherence. All raw videos of the examinations, with and without the ENDOANGEL-GC dashboard, will be recorded, stored, and reviewed.The following conditions will be defined as protocol deviations in the experimental group and will be excluded from the per-protocol (PP) population: (1) more than two anatomical landmarks not observed after video review and (2) biopsy sampling or further treatment not performed for red-box lesions with no pathology results for such lesions.The following conditions were defined as protocol deviations in both the experimental and control groups and were excluded from the PP population: (1) the inclusion and exclusion criteria are determined to be unfulfilled after the participant is randomised; (2) inability to fully observe the gastric anatomical landmarks due to gastric retention, residual stomach, or oesophageal obstruction; and (3) incorrect mode of ENDOANGEL-GC. | PMC10176798 |
|
Recruited cases | PMC10176798 |
|||
Consent | Patients eligible to participate in this study will be provided with further discussions and informed consent. Discussions may be completed by a local study coordinator or a staff member. Written informed consent will be obtained from all participants who agree and wish to take part in the study. | PMC10176798 |
||
Eligibility and baseline assessment | Once written informed consent is obtained, an eligibility assessment will be performed by the local study coordinator or staff, according to the inclusion/exclusion criteria. For each eligible participant, the following baseline characteristics will be collected: gender; age; weight; height; indications; education; nationality; registered residence (rural or urban); source (inpatient or outpatient); permanent address; | PMC10176798 |
||
Recruited case management | All recruited participants will be assigned identifications (IDs) comprising eight codes in the EDC system. The first four codes represent different hospitals, whereas the last four codes represent the recruiting sequences. This ID will not be used repeatedly. | PMC10176798 |
||
Specimen management and pathological traceability | gastric cancers, neoplasia, gastric cancer, carcinoma, inflammation, atrophy, diganosis | GASTRIC CANCERS, NEOPLASIA, INFLAMMATION, CARCINOMA, INTESTINAL METAPLASIA, ATROPHY, PATHOLOGY, GASTRIC CANCER | After tissue sampling, the specimens that arrive in the pathology department will be dehydrated, embedded, sliced, and stained. Pathological results will be obtained from local study centres by expert pathologists at each centre. Regarding patients diagnosed with gastric cancers by biopsy and those highly suspected with gastric cancer without biopsy who will possibly take further elective operations, their pathological results of resected specimens will be traced for up to 60 days from their endoscopy date to determine the invasion depth of gastric cancer.After the training program, a group of five research assistants will watch the raw video of each participant and capture clean, high-quality frames of the lesions with pathological results under the guidance of an expert endoscopist. All images of the lesions and corresponding pathological results will be sent to an independent group of three expert endoscopists for further endoscopic evaluation. Randomisation and personal information will be concealed, and the experts will be unaware of the allocation of the lesions. A group of three experts will evaluate the endoscopic and pathological diganosis of a lesion and determine whether they are unmatched. (Here was a clerical error of describing the determination of the unmatched lesions, and we have corrected it). De-identified digital pathological slides of unmatched lesions will be sent to an independent group of expert pathologists for consultation. Only unmatched lesions will be reviewed by an independent group of expert pathologists for the final diagnosis.Each expert pathologist is required to determine whether the slide contains acute inflammation, chronic inflammation, atrophy, or intestinal metaplasia and to rate the severity (mild, moderate, or severe) of these pathological changes. In addition, they are required to determine the presence of low-grade intraepithelial neoplasia, high-grade intraepithelial neoplasia, or carcinoma. Multiple pathological changes may occur in the same lesion, the most severe being the primary diagnosis. The severity of the pathological changes increases from inflammation to carcinoma. A lesion is diagnosed only when two or more experts reach a consensus. If the three expert pathologists did not reach a consensus, the lesion was re-evaluated until a consensus was reached. | PMC10176798 |
Randomisation | Local study coordinators or staff accomplished randomisation using an EDC system. Participants were randomised into either the experimental or control group at a 1:1 ratio before the examinations. The randomisation results, time points, and case IDs will be stored online in the EDC system.Participants will be randomised into one of the two arms by block randomisation stratified by centre. The randomisation sequence will be developed using a computer-generated random numerical series, with one encoding for the AI-assisted group and zero encoding for the control group. The original sequence will be stored in an online central randomisation system database. The online sequence will not accessible to investigators or study coordinators. If a subject fulfils the enrolment criteria, the authorised study coordinators or staff will log in to the password-protected account to obtain the assignment for him/her. | PMC10176798 |
||
Blinding | ADVERSE EVENT | Patients, pathologists, and data analysts will be blinded to the randomisation. Masking of study group allocations will not attempted by the endoscopists. Randomisation results will be concealed in information brochures or other documents for the participants.The interventions in this study will not add additional risks to participants, compared with routine sedated EGD. However, if a patient has an unexpected adverse event unrelated to the intervention and requires disclosure of study assignment information, then unblinding can be performed. | PMC10176798 |
|
Outcomes | premalignant gastric lesions, neoplasms, neoplasia, gastric neoplasm, gastric cancer | BLIND, NEOPLASMS, NEOPLASIA, GASTRIC NEOPLASM, GASTRIC NEOPLASMS, BLIND, GASTRIC CANCER | The primary outcome measures are detection rate of gastric neoplasms and EGC detection rate. The gastric neoplasm detection rate is defined as the ratio of patients with neoplasms to the recruited population. The EGC detection rate is defined as the ratio of patients with EGC to all patients with gastric cancer. EGC includes pathologically proven high-grade intraepithelial neoplasia and gastric cancer restricted to the mucosa and submucosa.Secondary outcome measures include the detection rate of premalignant gastric lesions, biopsy rate, observation time, and number of blind spots on EGD examination. The detection rate of premalignant gastric lesions was defined as the ratio of patients with premalignant gastric lesions to the entire recruited population. The biopsy rate is defined as the ratio of patients who undergo biopsy in the recruited population. The observation time is defined as the overall examination time under WLE minus the biopsy operation time. Blind spots are defined as unobserved landmarks. | PMC10176798 |
Study withdrawal | haemorrhage, allergy | HAEMORRHAGE, ALLERGY, GASTRIC RETENTION | Eligible participants will be included in this study after providing informed consent and undergoing randomisation. The participants can withdraw at any time during the study. Data collected prior to withdrawal can be used in this study if informed consent is obtained. Participants should be withdrawn in the following situations: (1) perforation, (2) massive haemorrhage, (3) allergy to sedative medication, (4) poor preparation, and (5) gastric retention. Withdrawn participants will not be replaced by other recruited participants after they sign the informed consent form. | PMC10176798 |
Safety evaluation | ADVERSE EVENT, ADVERSE EVENT | Adverse events are evaluated according to the Common Terminology Criteria for Adverse Events [ | PMC10176798 |
|
Data analysis | PMC10176798 |
|||
Sample size calculation | GASTRIC NEOPLASMS | The sample size is calculated on the basis of the primary outcomes.The detection rate of gastric neoplasms and EGC detection rate without AI assistance were determined by literature research. Zhang et al. reported an EGC detection rate of 20.0 to 20.9% [According to our previous study, the detection rate of gastric neoplasms increased by 1.35% when assisted by an AI system; therefore, the detection rate of gastric neoplasms in the experimental group is estimated to be 4.0%. There are no published data evaluating the improvement in EGC detection rates with AI assistance. Thus, we referred to two studies reporting improvements of > 80% in EGC detection sensitivity and proportion [The sample size was calculated, using the To achieve adequate participant enrolment, we first evaluated the annual volume of sedated EGD examinations at all participating centres. The sum of the examination volumes of all centres (> 20,000 per year) is sufficient to provide support for the completion of the study. In each centre, a study coordinator, group of nurses, or endoscopists will be pretrained for informed consent before the examination. Consecutive patients will be informed about the study and assessed for eligibility. | PMC10176798 |
|
Data collection | Data will be collected in a standard case-report form through the EDC system and anonymised for further analysis. Data include baseline information, endoscopic reports, and pathological results. Data will be de-identified before being entered into the database. Regular quality monitoring and database checking will be performed at each centre to ensure data accuracy.In addition, the computer on which the ENDOANGEL-GC runs will be equipped with video-recording software. It is used to capture and store video signals from the endoscope device and predictions from the ENDOANGEL-GC system. For each examination, an unprocessed raw video from the endoscope and a video with ENDOANGEL-GC predictions (“AI” videos) are recorded and stored. | PMC10176798 |
||
Data analysis plan | GASTRIC NEOPLASMS | The analysis will use intention-to-treat (ITT) and PP approaches. The ITT population will include all patients who are randomised, whereas the PP population will include patients who undergo EGD in accordance with the assigned intervention. The null hypothesis is that the detection rates of gastric neoplasms and EGC in the experimental group will not differ from those in the control group. An alternative hypothesis is that the detection rates of gastric neoplasms and EGC in the experimental group will differ from those in the control group. The experimental group (with ENDOANGEL-GC assistance) will be compared with the control group (without ENDOANGEL-GC assistance) for the two primary outcomes using the Continuous variables will be expressed as means (standard deviations [SDs]) or medians (interquartile ranges [IQRs]), according to their distributions, and categorical variables will be presented as numbers (percentages). Baseline characteristics between the two groups will be compared using the In this study, the potential missing data include the following: (1) characteristic information (the missed data will be marked with “NA” and will not be included in the analysis); (2) pathological results, the gold standards for analysis (participants with missing pathological results are excluded from the analysis); and (3) original videos (the data contained information on whether the endoscopic procedure adhered to the protocol requirements). Participants without this information will be excluded from the per-protocol analyses. | PMC10176798 |
|
Dissemination of results | The data in this study are the properties of the chief investigator and the other co-investigators. This publication is the responsibility of the chief investigator. All co-investigators will have access to anonymised trial data for further analysis and publication of peer-reviewed journal articles. | PMC10176798 |
||
Study monitoring | Research assistants of chief investigating centre are responsible for regular study monitoring. | PMC10176798 |
||
Discussion | GASTRIC NEOPLASMS | This study will explore the effectiveness of the AI system ENDOANGEL-GC in improving the detection rate of gastric neoplasms and EGC detection rate. We plan to enrol 30,000 participants from > 20 large-scale primary digestive centres in China. Enrolment began in December 2021. At the time of manuscript preparation, more than 10,000 patients had been enrolled. The results of this large, multi-centre RCT will provide high-level evidence for the application of AI systems in clinical settings. | PMC10176798 |
|
Trial status | RECRUITMENT | The enrolment of this study is ongoing at the time of manuscript submission, adhering to the protocol with version 3.0 (July 5, 2021). Recruitment began on December 21, 2021, and is estimated to be completed on December 20, 2024. | PMC10176798 |
|
Roles and responsibilities | The principal investigator and research physician have contributed the following: designing and conducting of the trial, preparation of the protocol and revisions, and publication of study reports.There were no trial steering committee or data monitoring committee in this trial. | PMC10176798 |
||
Acknowledgements | Not applicable. | PMC10176798 |
||
Authors’ contributions | HGY is the chief investigator; he conceived the study and led the proposal and protocol development. ZHD and LLW contributed to the study design and to the development of the proposal. YJZ, HLD, JXW, XQZ, XT, TY, JMW, MD, and JL contributed to the data collection and system construction. All authors read and approved the final manuscript. | PMC10176798 |
||
Funding | 1. College-enterprise Deepening Reform Project of Wuhan University (to Honggang Yu) (Here the name of the funding was misrefered, we have corrected it to the correct name, which still refers to the same project.2. Artificial Intelligence Application Demonstration Scenario Project Wuhan (grant no.2022YYCJ01) (to Honggang Yu).3. National Natural Science Foundation of China-Youth Science Fund (grant no.82202257) (To Lianlian Wu).4. Special projects for knowledge innovation of Wuhan (grant no.2022020801020482) (To Lianlian Wu). | PMC10176798 |
||
Availability of data and materials | All principal investigators will be given access to the cleaned datasets. All datasets will be password protected. Project principal investigators will have direct access to their own site’s datasets and will have access to other sites data by request. The data in this study are available from the corresponding author on reasonable request. The study protocol is accessible in this manuscript and on the registration website. The statistical software is publicly available. | PMC10176798 |
||
Declarations | PMC10176798 |
|||
Ethics approval and consent to participate | Ethics approval has been sought from the Ethical Review Board of Renmin Hospital of Wuhan University (2019K-K068) and all participating centres. Written informed consent to participate will be obtained from all participants. | PMC10176798 |
||
Consent for publication | Not applicable. | PMC10176798 |
||
Competing interests | The authors declare that they have no competing interests. | PMC10176798 |
||
References | PMC10176798 |
|||
Introduction | There is growing evidence of a mental health crisis in graduate student populations [One of the contributing factors is the level of stress graduate students experience and the lack of coping skills, resulting in adverse emotional, academic and health outcomes [The impact of stress on the mental health is compounded by the trend of low treatment rates among engineering graduate students, with only a quarter of those with apparent mental health issues seeking help [We present an investigation of the impact of a mindfulness-based training program as a practical, cost-effective method for promoting well-being in the context of graduate engineering education. We recruited subjects to participate in an eight-week training designed to provide them with both an understanding of the scientific underpinnings of mindfulness practice and the experience of engaging with these practices firsthand. We hypothesized that training in mindfulness-based techniques would have a positive impact on graduate student well-being and their capacity to engage productively in research. We conducted survey-based assessments that were primarily quantitative in nature, augmented with qualitative data in the form of open-ended responses. Our investigation spanned multiple years involving training cohorts (each including an intervention group and a wait-list control group) at two research doctoral universities with very high research activity. | PMC10032494 |
||
Background | anxiety, eating disorders, psychosomatic, psychosis, psychiatric, depression | DISORDERS | “Mindfulness” refers to a process of relating to the present-moment experience in an open, non-judgmental, curious, and accepting manner [There is an accumulating body of scientific evidence that mindfulness training can provide numerous benefits associated with mental health and general well-being. In clinical populations, mindfulness-based interventions have been found to be effective in improving mental health and alleviating suffering associated with physical, psychosomatic and psychiatric disorders such as anxiety, depression, eating disorders, and psychosis [Prior studies have also explored the impact of mindfulness on a variety of student populations, and educators at the pre-kindergarten level through high school are introducing mindfulness-based practices as a means of improving student academic performance and emotional well-being [ | PMC10032494 |
Methods | PMC10032494 |
|||
Participants | bipolar disorders, schizophrenia, psychotic disorders | Our investigation comprises two distinct phases of studies: Phase 1 was conducted over the course of one academic year at a single large public university (University of Wisconsin-Madison, denoted as University A) with one intervention group and one wait-list control group. Phase 2 was conducted over the course of multiple years at University A with a larger participant population (in comparison to Phase 1) and at a second large public university (University of Virginia, denoted as University B), with an intervention group and a wait-list control group in each of two cohort years (Year 1 and Year 2). University A is classified by the Carnegie Commission on Higher Education as a research doctoral university with very high research activity, located in the Midwest of the United States; University B is in the same Carnegie classification, but located in the Mid-Atlantic.The human subject participation was approved and overseen by the Education and Social/Behavioral Science Institutional Review Board (IRB) at the University of Wisconsin-Madison (ID#2016–0917 for Phase 1 and ID#2018–1055 for Phase 2). All research was performed in accordance with relevant guidelines/regulations and informed consent was obtained from all participants. The IRB granted a waiver of signed consent so that the Research Participation Information and Consent Form could be obtained online.The Phase 1 participants were recruited from the general population of M.S. and Ph.D. graduate students in engineering departments at University A who identified themselves as being engaged in engineering research. People with a history of schizophrenia spectrum, bipolar disorders, or other psychotic disorders were excluded from participation. Participants were recruited through paper and electronic advertisements citing the general benefits of mindfulness training. The participants were randomly assigned to either an intervention group or a wait-list control group. This strategy enabled all participants to eventually receive the training while allowing for the control of many variables during data collection [ | PMC10032494 |
|
Timing of recruitment, training, and surveys for intervention and control groups. | RECRUITMENT, RECRUITMENT | After recruiting participants and assigning them to either the Intervention Group or Control Group, all participants completed the Pre-Test Survey. After training of the Intervention Group all participants completed the Post-test Survey. The Summative Survey was completed by all participants after the Control Group completed training.Out of the 58 students who completed Phase 1 (i.e., who engaged in the intervention and completed both the pre-test and post-test surveys), 23 self-identified as female and 35 as male. They had a mean age of 27.12 (We recruited participants for Phase 2 from the general population of M.S. and Ph.D. graduate students in the engineering departments as well as computer sciences departments at University A and University B. Computer science students were added to the recruitment pool at University A in this phase to match the majors associated with both institutions. All participants identified themselves as being engaged in research. Recruitment, intervention/control group assignment, and all other major study aspects were conducted following the protocols and timing established for Phase 1. The one exception was that the Year 1 cohort participated in not only the summative survey as shown in Of the 157 participants who successfully completed Phase 2 (i.e., who engaged in the training and completed both the pre-test and post-test surveys and passed the attention checks), 73 self-identified as female, 83 as male, and 1 as transgender. They had a mean age of 27.53 ( | PMC10032494 |
|
Training | cognitive skills | SESSION | The “Cultivating Transformative Research Through Mindfulness” (or An introduction to and exploration of the six dimensions of emotional style–An overview of the neuroplasticity of the brain and how the brain can be trained to change responses to emotions;Training in mindfulness meditation and other contemplative practices, as well as cognitive skills and techniques; andStrategies for creating and maintaining healthy mental and emotional habits.The curriculum was developed in a culturally conscious way to be sensitive to the diverse and varied backgrounds and experiences of the graduate students, given the diversity represented at the institutions where the studies took place. The course content drew from the challenges of graduate school as a common denominator. The focus was to examine the underpinnings of how beliefs, opinions, biases, and behaviors are formed. Based on neuroscientific and psychological research, the participants were taught methods for developing insight into their biases, mental reactions, and habits in a non-judgmental manner. The cognitive and meditative tools provided throughout the course were designed to assist participants in managing emotional reactions and developing healthier perspectives, responses, and habits for improved resilience and emotional well-being.The curriculum was delivered over a total of six semesters in Phase 1 and Phase 2, and in parallel at University A and University B during Phase 2. Instruction for both phases at University A was delivered by two instructors (BH and SAM) with over 20 years of personal meditation practice and over 10 years of meditation teaching experience. These instructors were also responsible for customizing the curriculum for this research investigation. Training for University B in Phase 2 was led by an instructor with over 20 years of personal meditation/mindfulness practice and 17 years of mindfulness/meditation instruction. This instructor received comprehensive in-person curriculum training and a detailed training manual in order to ensure content and delivery consistency across sites.The curriculum consisted of eight weekly sessions of 50–60 minutes in duration. Our decision to offer eight instructor-led training sessions with limited expectations for practice outside of the formal sessions aligns with the recommendations of a recent meta-analysis [The training provided numerous opportunities for cognitive and contemplative skill-building in and outside the “classroom.” Each session began with a ten-minute check-in for sharing, discussion, and answering questions regarding the previous week’s home practice experience. In addition, time was allotted for reflection and discussion following each new meditation and cognitive exercise. Time spent on meditation and cognitive exercises, including reflection and discussion, was between 20–25 minutes per class session. The remainder of the time was spent introducing the emotional style being discussed and related concepts and content. The course emphasis was not only on learning the meditation techniques, but also on helping participants build insights, skills and strategies to support integrating mindfulness practices into their daily lives.Attention and self-awareness were considered key foundational concepts in developing a mindfulness practice and were covered in greater depth–two sessions each. Sessions 1 and 2 were devoted to the science of attention and the benefits of strengthening mental focus using techniques such as breath-focused meditations. Sessions 3 and 4 were devoted to the concept of meta-awareness and cultivating self-awareness of the mind-body connection as experienced through a body scan meditation. Session 5 (resilience) focused on the impact of stress on well-being and personal effectiveness, and cognitive reframing practices to enhance resilience in stressful times. Participants were also introduced to an open awareness meditation for maintaining mindfulness and staying present in challenging situations. Session 6 (positive outlook) focused on exploring the concepts of balance and equanimity, seeing and savoring positive experiences, practicing gratitude daily, and learning a meditation practice to open one to seeing potential benefits in challenging situations. Session 7 (social intuition) focused on cultivating the ability to read social cues, exploring connection as a basic human need, recognizing personal biases and deterrents to connecting, and practicing exercises in reading social cues and mindful listening. Session 8 (sensitivity to context) focused on recognizing and regulating emotions in a context-sensitive fashion, learning to stabilize reactions through balanced breath exercise, and using loving-kindness meditation to strengthen compassion and non-judgmental acceptance of self and others.In addition to the guided meditations and exercises conducted in the weekly sessions, participants were encouraged to practice formal mindfulness meditation (e.g., breath meditation, body scan meditation) daily for at least 10 minutes per day and to integrate brief mindfulness practices daily through the duration of the study. Each week the participants were reminded of the guided meditations and exercises they had learned so far and were asked to focus some of their weekly practice time on recently acquired skills. Additionally, on-line resources, including audio recordings of all meditations as taught during the course, were provided weekly to support participants in developing their practice and incorporating mindfulness into their daily life. Open-ended responses to the question “How much time did you devote to mindfulness activities in an average week?”, which was asked of the intervention group in the post-test immediately following the completion of training, indicated that the average practice time was approximately an hour-and-a-quarter per week in Phase 1 and approximately an hour per week in Phase 2. This was consistent with the practice time requested of the participants outside the “classroom.” | PMC10032494 |
Measures | Pre-test, post-test, and summative surveys were administered following standard practices of the University of Wisconsin-Madison and with approval of the IRB of the University of Virginia. Participants completed all the measures on the online survey platform Qualtrics. We employed a variety of well-being-related measures, as well as mindfulness-related measures and custom measures related to research satisfaction. These are described in detail below. | PMC10032494 |
||
Emotional Style Questionnaire (ESQ) | We used the 48-item ESQ to investigate the changes in participants’ emotional lives as a result of the training. This was a revised version of the questionnaire found in | PMC10032494 |
||
Ten Item Personality Inventory (TIPI) | To capture participants’ personality, we used the TIPI, which has been validated as a brief measure of Big Five personality [ | PMC10032494 |
||
Positive and Negative Affect Schedule (PANAS) | Participants indicated to what extent they have been experiencing certain negative and positive emotions “during the past two weeks including today.” The scale consisted of 10 positive (e.g., attentive, enthusiastic) and 10 negative emotion words (e.g., upset, guilty). Responses could range from 1 ( | PMC10032494 |
||
Cohen-Hoberman Inventory of Physical Symptoms (CHIPS) | cough, nosebleed, acne, pain | NOSEBLEED, COLD, ACNE | CHIPS is a list of 39 common physical symptoms (e.g., back pain, cold or cough, acne, nosebleed) [ | PMC10032494 |
Mindful Attention and Awareness Scale (MAAS) | On a scale ranging from 1 ( | PMC10032494 |
||
Five Facet Mindfulness Questionnaire—Short Form (FFMQ-SF) | We employed the 24-item short version of the FFMQ to assess different aspects of mindfulness [ | PMC10032494 |
||
Research Satisfaction Scale | We developed eight face-valid items to measure participants’ satisfaction with their research and their ability to make progress in their research. These items tapped into participant perceptions of making progress (e.g., “I am satisfied with how my research is progressing”), ability to do creative research (e.g., “I feel creative and innovative in my research”), and ability to persist in the face of obstacles (e.g., “I feel like I’m able to overcome obstacles in my research”). Participants indicated their agreement with these statements on a scale from 1 ( | PMC10032494 |
||
Contributive desire scale | ’ desire | We developed five statements to evaluate participants’ desire to contribute to the well-being of other people and the betterment of society, particularly through conducting impactful research. Scale items included “I am motivated to use my knowledge and skills to make a difference in people’s lives” and “I feel a responsibility to do things to improve the well-being of people.” The complete list of items is provided in | PMC10032494 |
|
Summative and Final Summative Surveys | The summative survey included open-ended questions regarding the impact and value of the training related to participants’ professional work and personal life. Questions included “In what ways has the training impacted your research and other professional work?”, “In what ways has the training impacted your personal life?”, and “What was most valuable to you about the training?”. Participants were also asked to report on the amount of time, frequency, and type of mindfulness activities that they were currently doing. All participants completed the Summative Survey at the end of the study year (within 12 months of the completion of their training). Phase 2 Year 1 participants were asked to complete the assessment again as a Final Summative Survey 18–24 months after the completion of their training. | PMC10032494 |
||
Results | PMC10032494 |
|||
Phase 1: Quantitative measures | The data described below comes from participants who completed at least 75% of the training. Some attrition occurred prior to the start of and over the course of the study, and not all participants completed the eight-week training. Participants who did not attend the training or had poor attendance chose not to complete the survey instruments. The initially enrolled sample was 73 (38 were assigned to the intervention group and 35 to the control group), composed of individuals who consented to the study and completed the pre-test. 24 participants in the intervention group and 34 participants in the control group completed both the pre- and post-test surveys and passed all built-in attention checks. The summative survey, conducted after both the intervention and control groups had completed the training, was completed by 14 participants assigned to the intervention group and 21 to the control group.The majority of the participants (48 out of 57 respondents) had little to some familiarity with mindfulness prior to the study, as indicated by their responses to a question in the pre-test. Six reported no familiarity at all and three reported “a great deal” of familiarity.Independent t-tests were conducted to compare groups at baseline (i.e., at pre-test) for demographic variables and outcome measures. Despite the use of random assignment to groups, analyses revealed that those assigned to the intervention group were significantly lower on the To analyze the data, Repeated Measures ANOVA’s (RMANOVA’s) were conducted for each outcome variable, with Time (Time 1 and Time 2) specified as the within-subject factor and Group (intervention vs. control) specified as the between-group factor. Pre- and post-test scores for each group are presented in | PMC10032494 |
||
Phase 1 significant effects. | Graphical representations of significant effects observed between pre- and post-test means in four measures for intervention and control groups. The vertical bars represent standard error of the mean. | PMC10032494 |
||
Phase 1 pre- and post-test Means ( | POSITIVE | ESQ = Emotional Style Questionnaire. TIPI = Ten Item Personality Inventory. PANAS = Positive and Negative Affect Schedule. CHIPS = Cohen-Hoberman Inventory of Physical Symptoms. MAAS = Mindful Attention and Awareness Scale. FFMQ-SF = Five Facet Mindfulness Questionnaire-Short Form. For each of the measures, the sample size varied between 23–24 for the intervention group and 33–34 for the control group.* P < 0.05.** P < 0.01.We also calculated Cohen’s The intercorrelations between all the measures used in the study, at pre-test and post-test separately, are provided in Even though only a small number of our outcome variables yielded significant Group by Time interactions, overall we had a very consistent pattern of change favoring the intervention group. | PMC10032494 |
|
Phase 1: Qualitative measures | anger, anxiety | The qualitative summative survey was conducted approximately five months after the intervention group and one month after the control group had completed the training. The assessment was completed by 35 participants. It posed several open-ended questions, including “In what ways has the training impacted your research and other professional work?” Although not all respondents indicated that they had noticed changes, many provided exceptionally compelling responses indicating how the mindfulness practices were helping them better manage stress, regulate their emotions, ease anxiety, lessen anger, be more focused and efficient in their work, better handle interpersonal interactions, and deal with research setbacks in a positive manner (see | PMC10032494 |
|
Phase 2: Quantitative measures | In Year 1 of Phase 2, a combined total of 90 participants completed both the pre- and post-test surveys and 89 passed built-in attention checks. The summative survey, conducted after the training had concluded for both the intervention and control groups, was completed by a combined total of 60 participants, and 44 went on to complete the final summative survey after Year 2. In Year 2, a combined total of 69 participants completed both the pre- and post-test surveys and 68 passed built-in attention checks. The summative survey was completed by a combined total of 43 participants.The majority of the participants in Phase 2 (129 out of 156 respondents) had little to some familiarity with mindfulness prior to the study. 14 reported no familiarity at all and 13 reported “a great deal” of familiarity.The data analytic procedures for Phase 2 were identical to those of Phase 1. We conducted Repeated Measures ANOVA’s (RMANOVA’s) for each outcome variable, with Time (Time 1 and Time 2) specified as the within-subject factor and Group (intervention vs. control) specified as the between-group factor. Given our relatively low sample sizes per university, we combined the samples from both universities and analyzed them together. This decision was justified by the absence of any systematic differences between the results of the two samples, as evinced by a lack of significant three-way interactions (i.e., Group x Time x University) when University was included as a between-group factor in our RMANOVA’s.Pre- and post-test scores for Phase 2 are presented in | PMC10032494 |
||
Phase 2 significant effects. | Graphical representations of eight significant effects observed between pre- and post-test means for intervention and control groups. The vertical bars represent standard error of the mean. Phase 2 Year 1 and Year 2 are combined. | PMC10032494 |
||
Phase 2 pre- and post-test Means ( | Neuroticism | POSITIVE | ESQ = Emotional Style Questionnaire. TIPI = Ten Item Personality Inventory. PANAS = Positive and Negative Affect Schedule. CHIPS = Cohen-Hoberman Inventory of Physical Symptoms. MAAS = Mindful Attention and Awareness Scale. FFMQ-SF = Five Facet Mindfulness Questionnaire-Short Form For each of the measures, the sample size varied between 72–73 for the intervention groups and 83–84 for the control groups.* ** We break down the presentation of Phase 2 results by Year 1 and Year 2. As the effects we observed were not qualified by year (i.e., no significant Group x Time x Year interactions), we also report the results of the analyses in the combined sample (N = 157) to provide a more comprehensive picture for the reader.Before running the analyses, independent t-tests were conducted to compare groups at baseline (i.e., at pre-test) for demographic variables and outcome measures. In Year 1, a significant difference was observed for the Neuroticism variable, such that those in the intervention group scored significantly higher in Neuroticism at the pre-test than those in the waitlist control group (There were no significant differences for any of the other variables for Phase 2 (Overall, we had a very consistent pattern of change favoring the intervention group, particularly when it comes to variables related to emotional well-being and mindfulness. The effects were more apparent in the combined, larger sample.Details on these results can be found in the | PMC10032494 |
Phase 2: Qualitative measures | The qualitative summative survey was conducted approximately 5 months after the intervention group and 1 month after the control group had completed the training in Year 1 of Phase 2. The Summative Survey for Year 2 cohorts was impacted by the pandemic in two ways: the control group received a hybrid training experience with the first five weeks in person, a pause of two weeks, then the remaining three weeks delivered online. Both the intervention and control groups completed the Summative Survey in the midst of the COVID-19 pandemic and the administration of this was conducted approximately twelve months after the intervention group and six months after the control group had completed the training.The assessment was completed by 60 participants in Year 1 and 43 participants in Year 2. It included the same open-ended questions as those asked in Phase 1. The majority of participants noted positive changes echoing those reported in Phase 1 (see Combining Year 1 and Year 2 summative data, 68% of respondents reported that they were currently practicing mindfulness activities weekly or more frequently (see | PMC10032494 |
||
Discussion | anxiety | POSITIVE, REGRESSION | The purpose of this research was to explore the impact of a mindfulness training on engineering graduate students, with a particular focus on how this training affected their well-being and capacity for research. During the course of the two studies comprising three years, where each group engaged in training for eight weeks in the middle of a semester, the intervention groups reported significantly improved emotional well-being at the end of the training. In contrast, the control groups consistently held steady or dropped on their well-being during that same time frame. While statistical significance was not observed across each cohort (e.g., intervention/control pair) for all measures, the combined data across both studies, two institutions, and multiple years of Phase 2 produced consistent and robust findings. Together these findings provide compelling evidence of the efficacy of this type of training with engineering graduate students.More specifically, both studies showed higher Emotional Health over time as measured by the ESQ, with significant effects observed for the Resilience and Outlook subscales in Phase 1 and the Attention, Self-Awareness, and Outlook subscales in Phase 2. Additionally, across both studies, participants in the intervention group perceived themselves as less neurotic over time. Following the trend seen in Phase 1, there were significantly higher levels of Positive Affect and significantly lower levels of Negative Affect as well as significant gains in mindful attention and awareness in the Phase 2 intervention groups. A significant increase in mindfulness was observed in the intervention groups of Phase 2 with all the subscales of the FFMQ-SF. This too is consistent with trends observed in Phase 1. Although the Research Satisfaction Scale did not reach significance in Phase 1, the intervention groups in Phase 2 reported statistically significant improvement in their satisfaction with their research across time. More extensive study of the impact on research satisfaction is warranted, especially given its links to persistence and achievement [The improvements in emotional health and wellbeing observed are consistent with those reported in prior studies of similar interventions with postbaccalaureate students. Guided mindfulness practices in 8-week interventions in a similar age group of medical school students and psychology graduate students have been shown to reduce stress and anxiety [Another notable finding of our study was that engineering graduate students were open to this type of training and ultimately expressed high levels of contentment with it. An inspection of qualitative responses to the summative survey revealed that participants appreciated the opportunity to devote time weekly to the in-class training where they would practice the meditation skills being learned. These weekly training sessions also provided them with an opportunity to connect with other graduate students. The small groups formed within each training group added an opportunity for regular check-ins with peers, which provided additional support and accountability. This social support aspect of the training could have presumably contributed to the positive outcomes observed in the intervention group. Overall, we consider the training’s positive reception among engineering graduate students a valuable learning from our study, given that this population might not necessarily be thought of as naturally drawn to this type of contemplative practice.Although we did not select participants, we found post facto that women were substantially overrepresented (40% and 46% for Phase 1 and Phase 2) and domestic minority students slightly overrepresented (10% for both studies) among the participants in this study compared to graduate student enrollment in engineering departments at University A (22–25% female and 6–8% domestic minority, over the time frames of the two studies). International students were modestly underrepresented (36% and 39% for Phase 1 and Phase 2, respectively, compared to 41–42% over the time frames of the two studies). Although the overrepresentation of female and minority students allows for more robust conclusions over the breadth of engineering graduate student backgrounds, it does indicate that voluntary participation in such activities may be more challenging to achieve among other student populations. This may have ties to the differences seen in treatment seeking for mental health and related services seen across disciplines in undergraduate students [Further analyses conducted with Phase 2 data revealed that for a few of the variables, the three-way Time x Group x Gender interaction was significant, suggesting that the significant Time x Group interactions we observed were qualified by Gender. Looking at those more closely, it appeared that in those cases although both men and women benefited from the intervention, women benefited slightly more. Whereas a recent meta-analysis (Bamber and Morpeth 2019) did not find evidence that gender moderated the effects regarding anxiety in the studies of mindfulness meditation on college student anxiety, there is other evidence in the literature that women may be more likely to benefit from mindfulness interventions [We note some limitations of our study. One limitation is that despite the use of random assignment, we observed some baseline differences between the intervention and the control groups, although not consistently recurring between studies and years. These pre-existing differences invite caution when interpreting the results related to these measures in an individual study or year. Regression to the mean is a possibility that we cannot rule out with our study design. We also acknowledge the possibility of demand characteristic effects in the intervention group, especially with measures of mindfulness and the Emotional Style Questionnaire, which closely align with the training content. Yet another limitation is methodological: while we had a wait-list control group that allowed us to control for the passage of time, we did not have an active control group that could control for other non-specific training effects [Mounting evidence points to the increasing prevalence of problematic mental health and well-being issues in the graduate student populations of our higher education institutions [ | PMC10032494 |
Conclusion | anxiety | The eight-week mindfulness-based training program provided to engineering graduate student participants was well-received and produced measurable positive impacts on emotional well-being. The improvements in emotional health were highly meaningful for a graduate student sample, where high levels of stress and anxiety are commonly reported. Furthermore, the positive implications for mindfulness and research satisfaction are highly encouraging. These findings motivate broader utilization of mindfulness-based contemplative training to enhance both the personal and professional lives of engineering graduate students. | PMC10032494 |
|
Supporting information | PMC10032494 |
|||
Supplementary methods. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Phase 1 supplementary results. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Additional supplementary results. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Additional summative survey findings. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Phase 2 supplementary results. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Supplementary references. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Outline of the training curriculum developed by Healthy Minds Innovations. | BRAIN | Page numbers refer to the relevant sections of The Emotional Life of Your Brain for each week of training (Davidson and Begley 2012).(DOCX)Click here for additional data file. | PMC10032494 |
|
Intercorrelations between measures at pre-test for Phase 1. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Intercorrelations between measures at post-test for Phase 1. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Summative survey results and representative responses for Phase 1 (n = 35). | (DOCX)Click here for additional data file. | PMC10032494 |
||
Summary of participant and completion statistics for Phase 2—Year 1. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Summary of participant and completion statistics for Phase 2—Year 2. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Phase 2 Year 1 pre- and post-test Means (M), Standard Deviations (SD), between group effect sizes (d), and RMANOVA results (F-value, p-value) for intervention and control groups. | POSITIVE | ESQ = Emotional Style Questionnaire. TIPI = Ten Item Personality Inventory. PANAS = Positive and Negative Affect Schedule. CHIPS = Cohen-Hoberman Inventory of Physical Symptoms. MAAS = Mindful Attention and Awareness Scale. FFMQ-SF = Five Facet Mindfulness Questionnaire-Short Form. AUT = Alternate Uses Task. For each of the measures, the sample size varied between 37–38 for the intervention group and 50–51 for the control group. * p < 0.05. ** p < 0.01.(DOCX)Click here for additional data file. | PMC10032494 |
|
Phase 2 Year 2 pre- and post-test Means (M), Standard Deviations (SD), between group effect sizes (d), and RMANOVA results (F-value, p-value) for intervention and control groups. | POSITIVE | ESQ = Emotional Style Questionnaire. TIPI = Ten Item Personality Inventory. PANAS = Positive and Negative Affect Schedule. CHIPS = Cohen-Hoberman Inventory of Physical Symptoms. MAAS = Mindful Attention and Awareness Scale. FFMQ-SF = Five Facet Mindfulness Questionnaire-Short Form. AUT = Alternate Uses Task. For each of the measures, the sample size was 35 for the intervention group and 33 for the control group. * p < 0.05. ** p < 0.01.(DOCX)Click here for additional data file. | PMC10032494 |
|
Intercorrelations between measures at pre-test for Phase 2. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Intercorrelations between measures at post-test for Phase 2. | (DOCX)Click here for additional data file. | PMC10032494 |
||
Summative survey results and representative responses for Phase 2 Year 1 (n = 49). | (DOCX)Click here for additional data file. | PMC10032494 |
||
Summative survey results and representative responses for Phase 2 Year 2 (n = 43). | (DOCX)Click here for additional data file. | PMC10032494 |
||
Summative survey results and representative responses for phase 2 Year 1 final summative (n = 44). | (DOCX)Click here for additional data file. | PMC10032494 |
||
Impact on research. | POSITIVE | Summative Data Comparison Across Years for Positive, Negative and Neutral Responses to “In what ways has the training impacted your research and other professional work?”.(DOCX)Click here for additional data file. | PMC10032494 |
|
Impact on personal life. | POSITIVE | Summative Data Comparison Across Years for Positive, Negative and Neutral Responses to “In what ways has the training impacted your personal life?”.(DOCX)Click here for additional data file. | PMC10032494 |
|
Recommend training to others. | Summative Data Comparison across Years for Yes, Maybe, and No Responses to “Would you recommend this training to other engineering graduate students?”.(DOCX)Click here for additional data file.We would like to recognize and thank the following colleagues at the University of Virginia (UVA) for partnering with us to implement the UVA component of Phase 2: Professor David Germano, Executive Director of the Contemplative Sciences Center; Professor Pamela Norris, Executive Dean of the School of Engineering and Applied Science; and Leslie Hubbard, Program Director in the Contemplative Sciences Center. We thank them for their roles in facilitating the participation of 95 UVA graduate students (Germano, Norris) and serving as the on-site mindfulness training instructor (Hubbard).This material is based upon work supported by the National Science Foundation under Grant No. EFMA-1628916, EFMA-1745774, and while serving at the National Science Foundation (Crone). Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. | PMC10032494 |
||
References | PMC10032494 |
|||
ABSTRACT | METABOLIC DISEASES, MICROBIAL COLONIZATION | The authors declare no conflict of interest.Children delivered by elective, prelabor Cesarean section (C-section) are not exposed to the birth canal microbiota and, in relation to vaginally delivered children, show altered microbiota development. Perturbed microbial colonization during critical early-life windows of development alters metabolic and immune programming and is associated with an increased risk of immune and metabolic diseases. In nonrandomized studies, vaginal seeding of C-section-born neonates partially restores their microbiota colonization to that of their vaginally delivered counterparts, but without randomization, confounding factors cannot be excluded. In a double-blind, randomized, placebo-controlled trial, we determined the effect of vaginal seeding versus placebo seeding (control arm) on the skin and stool microbiota of elective, prelabor C-section-born neonates ( | PMC10294643 |
|
KEYWORDS | PMC10294643 |
|||
INTRODUCTION | UTERUS | The microbes that first colonize a neonate at birth play a key role in metabolic programming and immune system development (Cesarean section (C-section)—31.8% of all births in the United States in 2020 (Neonates delivered by C-section bypass the vaginal canal and therefore miss out on the exposure to the first live microbes and pioneer microbial colonizers that babies receive during vaginal birth, namely, the vaginal-perineal maternal microbes. Numerous studies confirm substantial differences in the microbiota acquisition and maturation in neonates delivered by C-section compared with vaginal delivery (C-section is correlated with other factors that could perturb the microbiota of the baby, including perinatal antibiotics, shorter gestational time, and lack of labor. No previous study has determined in a randomized, placebo-controlled trial, the effect of transplanting microbes via vaginal seeding to neonates that are extracted from the uterus into the air of operating rooms. In this double-blind, randomized, placebo-controlled study, we assess causation in the relationship between vaginal seeding of C-section-delivered babies and changes in the microbiome. We compare the fecal and skin microbiota of neonates randomly assigned to being seeded by their mother’s vaginal microbiome versus placebo seeding (control). We further assessed engraftment of maternal microbes in the infant microbiota by comparing the proportion of microbes shared between mothers and their children. | PMC10294643 |
|
RESULTS | PMC10294643 |
|||
Subjects. | Twenty mother-child dyads were randomized to either vaginal seeding (Baseline characteristics of randomized infants and mothers. Download Vaginal seeding was performed immediately after delivery and before skin-to-skin contact with their mother, following the procedure performed in our previous observational studies of vaginal seeding ( | PMC10294643 |
||
The maternal microbiota. | As expected, there were no differences in the maternal vaginal microbiota (prior to randomization) in terms of bacterial DNA load (measured by quantitative PCR [qPCR]) or alpha diversity, when comparing mothers from the vaginal-seeding versus control groups (Differences in bacterial DNA load and composition between treatment groups at differing body sites, collection types, and times. (A) Bacterial DNA load in maternal vaginal swabs (both inoculated with vaginal fluids), gauze (control gauze not inoculated with vaginal fluids), infant skin (forearm), and infant stool. (B) Bacterial load of the 20 most abundant genera in gauzes from the two groups. Samples and taxa are ordered by unsupervised hierarchical clustering based on bacteria load. Bacterial load was estimated by 16S rRNA gene copy number based on qPCR using 16S rRNA gene universal primers. Wilcoxon signed-rank tests were performed for intergroup comparisons, with statistical significance indicated as follows: *, Shannon diversity index by treatment group at differing body sites, collection types, and times. Shannon diversity index values were estimated using the R package DivNet and assessed for significance using the function “betta” from the R package breakaway. Intergroup comparisons were considered statistically significant and were indicated as follows: *, | PMC10294643 |
||
Effect of vaginal seeding on the infant stool and skin microbiota. | Vaginal seeding significantly increased bacterial load in the skin (forearm; average, 32 copies/μL versus 14 copies/μL, in seeded and control babies, respectively; Compared with the control, vaginal seeding caused a significant reduction in alpha diversity (Shannon index) in the skin at day 1 (Treatment (seeding versus control) explained considerable variance in the beta diversity of the skin microbiota of day 1 neonates (Principal-coordinate Analysis of infant microbiota beta diversity (weighted UniFrac) by treatment group. (A) Transitional stool. (B) Day 30 stool. (C) Day 1 skin. A permutational multivariate analysis of variance (PERMANOVA) test was performed, and statistical results were noted on the top of each panel. Download Relative to the control, vaginal seeding altered the relative abundance of multiple bacterial amplicon sequence variants (ASVs), causing a decrease in the relative abundance of 5 ASVs in the transitional stool (detected by the analysis of compositions of microbiomes [ANCOM] method) (Log | PMC10294643 |
||
Maternal sources of the infant microbiota. | VAGINA | Source tracking analyses revealed a significantly higher proportion of maternal vaginal microbiota in the skin of 1-day old neonates, relative to control infants (adjusted Proportion of maternal bacterial sources in infant and maternal body sites. (A) Contribution of maternal bacterial sources from different body sites to the microbiota of the infant skin (i), infant transitional stool (ii), and infant day 30 stool (iii). (B) Contribution of maternal vaginal bacteria (from vaginal swab) to the microbiota of the infant skin (i), infant transitional stool (ii), and infant day 30 stool (iii). (C) Bacterial sharing between vagina and other maternal body sites. Wilcoxon signed-rank tests were performed to compare source differences between groups, with statistical significance indicated as follows: *, With respect to composition of maternal microbiota sources, the perinatal vaginal microbiota substantially overlapped with the microbiota in other maternal body sites (Shared ASVs between vaginal swabs and maternal stool. Download | PMC10294643 |
|
DISCUSSION | Our study is the first double-blind, randomized, placebo-controlled trial to determine whether vaginal seeding causes differential engraftment of maternal bacteria in the skin and stool of neonates. The study has important limitations, such as a small sample size and only 2 time points until 30 days of life. Despite these limitations, we observed significant effects of vaginal seeding on the neonatal microbiota. Vaginal seeding of C-section-delivered neonates increased mother-to-neonate microbial transfer and changed the neonatal microbiota in different body sites, urging further studies to determine whether these changes provide health benefits to the more than 1 million babies annually born by C-section in the United States (The maturation of the stool microbiota in mammals progresses first with a postpartum reduction of the stool diversity—demonstrated in mice (C-section-born neonates have, in relation to their vaginally born counterparts, a higher proportion of potential pathobionts, such as from the Consistent with infants born vaginally showing stool bacterial communities resembling their mother’s vaginal microbiota and C-section-delivered infants showing skin-like microbiota (Our findings also support the growing recognition that microbiota from different maternal body sites are important for infant microbiome maturation (In conclusion, vaginal seeding caused changes in the neonatal skin and stool microbiota, leading to a pattern of reduced bacterial diversity that is characteristic of the microbiota of vaginally delivered and breastfed infants. There is now a critical need to evaluate the health benefits and safety of vaginal seeding in large randomized controlled trials. | PMC10294643 |
||
MATERIALS AND METHODS | PMC10294643 |
|||
Study subjects. | infection | INFECTION | We performed this institutional review board (IRB)-approved study (WCG IRB number 1300043) at the Inova Health System in Northern Virginia under the US Food and Drug Administration (FDA) Investigational New Drug Application (IND) number 18076 (with an IND required by the FDA). We recruited pregnant women who were scheduled for an elective C-section at ≥37 weeks gestation and obtained written informed consent. We applied stringent inclusion and exclusion criteria to ensure the least risk of maternal-to-infant transmission of infection as discussed with the FDA and the exclusion of maternal health conditions associated with dysbiosis of the vaginal microbiome. For full details of inclusion and exclusion criteria, see supplemental data and | PMC10294643 |
Study procedure: randomization and vaginal seeding. | ’s mouth | STERILE, OBESE, VAGINA, BLIND | On the day of the scheduled C-section, a team member blind to treatments inserted a gauze moistened with sterile saline into the mother’s vagina. The gauze was incubated in the vagina for approximately 1 h and then removed and placed into a lidded sterile container prior to the mother receiving perioperative antibiotics (with the exception of 2 mothers who were penicillin allergic and received clindamycin and gentamicin when the gauze was still in the vagina due to the longer antibiotic infusion time) and kept at room temperature. Shortly after the insertion of the vaginal gauze, a control gauze was also made by identical preparation of the gauze (i.e., folding of the gauze and moistening with sterile saline following the same protocol as the vaginal gauze) but without insertion into the mother’s vagina. The control gauze was also placed in a separate lidded sterile container and kept at room temperature. A team member not blind to the treatments then randomized the mother to either the vaginal seeding arm or the control arm. Randomization occurred in a 1:1 ratio and was also stratified to the following 3 prepregnancy BMI strata: normal weight, overweight, and obese.In the operating room, after C-section delivery of the infant and prior to the infant having skin-skin contact with its mother, the infant was wiped down with the vaginal seeding gauze or the control gauze by a team member who was blind to the treatments. The gauze was wiped first over the infant’s mouth, followed by the face and the rest of the body in a standardized manner as described previously, typically just after the 1-min Apgar score was taken ( | PMC10294643 |
Data and samples. | ADVERSE EVENTS, STERILE | We collected detailed data on mothers, including demographics, medical history, anthropometrics, and medication use, including antibiotics during pregnancy. Data collected for the infants included demographics, method of first feeding and subsequent feedings, medication use, and adverse events. We collected and managed study data using the REDCap electronic data capture tools hosted at the Inova Health System.For this study, maternal samples collected and analyzed were as follows: (i) prenatal maternal stool from after 35 weeks gestation, (ii) vaginal swab collected just prior to the vaginal gauze insertion before randomization, (iii) gauze used for swabbing the neonate, (iv) skin swab from the forearm obtained the day after delivery, and (iv) buccal swab obtained the day after delivery. Infant samples collected and analyzed were as follows: (i) transitional stool sample at day of life 2 to 3, (ii) stool sample around day 30 of life (mean, 34.9 days [SD 3.73]), and (iii) skin swab from the forearm (using a dry sterile swab rolled over forearm skin area of approximately 2 by 2 inches) obtained the day after delivery before the infant’s first bath. We stored all samples at −80°C until analysis. | PMC10294643 |
|
DNA extraction and sequencing. | We extracted microbial DNA using the Qiagen DNeasy PowerSoil HTP 96 kit (Hilden, Germany) following the manufacturer’s instructions. For 16S rRNA gene sequencing, we amplified the V4 region of 16S rRNA genes using the 515F/806R primers according to the Earth Microbiome Project protocol ( | PMC10294643 |
||
Determination of bacterial DNA load. | We determined bacterial copy number by qPCR using a Quantstudio 3 system (Thermo Fisher, Waltham, MA) with universal 16S gene primers (338F, | PMC10294643 |
||
Metadata analysis. | BLIND | All analyses were conducted by the statistical team in a blind manner. Select members of the statistical team were unblinded only after the final analysis of the data. Descriptive statistics (median with interquartile range and percentages) were presented on demographic and clinical characteristics for mothers and infants according to treatment arm. | PMC10294643 |
|
Microbiota analysis. | BLIND | The bioinformatics team conducted all microbiota analyses in a blind manner, and select members were unblinded only after the final analysis. | PMC10294643 |
Subsets and Splits