The performance of artificial intelligence large language model-linked chatbots in surgical decision-making for gastroesophageal reflux disease.

BACKGROUND: Large language model (LLM)-linked chatbots may be an efficient source of clinical recommendations for healthcare providers and patients. This study evaluated the performance of LLM-linked chatbots in providing recommendations for the surgical management of gastroesophageal reflux disease (GERD). METHODS: Nine patient cases were created based on key questions addressed by the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) guidelines for the surgical treatment of GERD. ChatGPT-3.5, ChatGPT-4, Copilot, Google Bard, and Perplexity AI were queried on November 16th, 2023, for recommendations regarding the surgical management of GERD. Accurate chatbot performance was defined as the number of responses aligning with SAGES guideline recommendations. Outcomes were reported with counts and percentages. RESULTS: Surgeons were given accurate recommendations for the surgical management of GERD in an adult patient for 5/7 (71.4%) KQs by ChatGPT-4, 3/7 (42.9%) KQs by Copilot, 6/7 (85.7%) KQs by Google Bard, and 3/7 (42.9%) KQs by Perplexity according to the SAGES guidelines. Patients were given accurate recommendations for 3/5 (60.0%) KQs by ChatGPT-4, 2/5 (40.0%) KQs by Copilot, 4/5 (80.0%) KQs by Google Bard, and 1/5 (20.0%) KQs by Perplexity, respectively. In a pediatric patient, surgeons were given accurate recommendations for 2/3 (66.7%) KQs by ChatGPT-4, 3/3 (100.0%) KQs by Copilot, 3/3 (100.0%) KQs by Google Bard, and 2/3 (66.7%) KQs by Perplexity. Patients were given appropriate guidance for 2/2 (100.0%) KQs by ChatGPT-4, 2/2 (100.0%) KQs by Copilot, 1/2 (50.0%) KQs by Google Bard, and 1/2 (50.0%) KQs by Perplexity. CONCLUSIONS: Gastrointestinal surgeons, gastroenterologists, and patients should recognize both the promise and pitfalls of LLM's when utilized for advice on surgical management of GERD. Additional training of LLM's using evidence-based health information is needed.

The Elephant in the Room: Racial Microaggressions and Implicit Bias in Surgical Training.

OBJECTIVE: The aim of this study was to examine the association between race, experience of microaggressions, and implicit bias in surgical training. BACKGROUND: There is persistent underrepresentation of specific racial and ethnic groups in the field of surgery. Prior research has demonstrated significant sex differences among those who experience microaggressions during training. However, little research has been conducted on the association between race and experiences of microaggressions and implicit bias among surgical trainees. METHODS: A 46-item survey was distributed to general surgery residents and residents of surgical subspecialties through the Association of Program Directors in Surgery listserv and social media platforms. The questions included general information/demographic data and information about experiencing, witnessing, and responding to microaggressions during surgical training. The primary outcome was the prevalence of microaggressions during surgical training by self-disclosed race. Secondary outcomes were predictors of and adverse effects of microaggressions. RESULTS: A total of 1624 resident responses were obtained. General surgery residents comprised 825 (50.8%) responses. The female-to-male ratio was nearly equal (815:809). The majority of respondents identified as non-Hispanic White (63.4%), of which 5.3% of residents identified as non-Hispanic Black, and 9.5% identified as Hispanic. Notably, 91.9% of non-Hispanic Black residents (n=79) experienced microaggressions. After adjustment for other demographics, non-Hispanic Black residents were more likely than non-Hispanic White residents to experience microaggressions [odds ratio (OR): 8.81, P <0.001]. Similar findings were observed among Asian/Pacific Islanders (OR: 5.77, P <0.001) and Hispanic residents (OR: 3.35, P <0.001). CONCLUSIONS: Race plays an important role in experiencing microaggressions and implicit bias. As the future of our specialty relies on the well-being of the pipeline, it is crucial that training programs and institutions are proactive in developing formal methods to address the bias experienced by residents.

Artificial Intelligence-enabled Decision Support in Surgery: State-of-the-art and Future Directions.

OBJECTIVE: To summarize state-of-the-art artificial intelligence-enabled decision support in surgery and to quantify deficiencies in scientific rigor and reporting. BACKGROUND: To positively affect surgical care, decision-support models must exceed current reporting guideline requirements by performing external and real-time validation, enrolling adequate sample sizes, reporting model precision, assessing performance across vulnerable populations, and achieving clinical implementation; the degree to which published models meet these criteria is unknown. METHODS: Embase, PubMed, and MEDLINE databases were searched from their inception to September 21, 2022 for articles describing artificial intelligence-enabled decision support in surgery that uses preoperative or intraoperative data elements to predict complications within 90 days of surgery. Scientific rigor and reporting criteria were assessed and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines. RESULTS: Sample size ranged from 163-2,882,526, with 8/36 articles (22.2%) featuring sample sizes of less than 2000; 7 of these 8 articles (87.5%) had below-average (<0.83) area under the receiver operating characteristic or accuracy. Overall, 29 articles (80.6%) performed internal validation only, 5 (13.8%) performed external validation, and 2 (5.6%) performed real-time validation. Twenty-three articles (63.9%) reported precision. No articles reported performance across sociodemographic categories. Thirteen articles (36.1%) presented a framework that could be used for clinical implementation; none assessed clinical implementation efficacy. CONCLUSIONS: Artificial intelligence-enabled decision support in surgery is limited by reliance on internal validation, small sample sizes that risk overfitting and sacrifice predictive performance, and failure to report confidence intervals, precision, equity analyses, and clinical implementation. Researchers should strive to improve scientific quality.

Stop the Bleed®: Medical Students as Instructors of the Bleeding Control Basics Course.

BACKGROUND: The Bleeding Control Basics (B-Con) Course was developed to teach lifesaving hemorrhage control techniques to the public. Currently, medical students (MS) without prior clinical experience (CE) may not act as autonomous instructors, limiting the instructor pool. PURPOSE: To assess the bleeding control knowledge of MS (phase I) and compare the knowledge of students taught by a certified instructor vs a medical student (phase II). METHODS: Phase I: 20 MS, 6 with prior CE and 14 without clinical experience (NCE) completed a pre-course and post-course knowledge assessment. Results were assessed by independent sample t-tests. Phase II: 91 first-year MS were taught the B-Con Course by either a third-year MS (n = 45) or certified instructor (n = 46). An analysis of covariance (ANCOVA) was performed to compare scores by instructor type (certified vs MS) using prior CE and pretest scores as confounding variables. RESULTS: In Phase I, the CE group scored higher on the pretest assessment compared to the NCE group (P = .003). All students improved in posttest scoring, and there was no difference in posttest scores between the groups (P = .597). In Phase II, despite no difference in pretest scores between groups, the MS taught learners scored significantly higher on the posttest compared to the certified instructor group (P < .01). Prior CE did not correlate to posttest scores (P = .719). DISCUSSION: Medical students are as effective as certified instructors at conveying the B-Con learning objectives. Based on near-perfect assimilation of content by students, MS should be permitted to teach B-Con Courses.

Preoperative Nutritional Optimization of the Oncology Patient: A Scoping Review.

BACKGROUND: Malnutrition is common among patients with cancer and is a known risk factor for poor postoperative outcomes; however, preoperative nutritional optimization guidelines are lacking in this high-risk population. The objective of this study was to review the evidence regarding preoperative nutritional optimization of patients undergoing general surgical operations for the treatment of cancer. METHODS: A literature search was performed across the Ovid (MEDLINE), Cochrane Library (Wiley), Embase (Elsevier), CINAHL (EBSCOhost), and Web of Science (Clarivate) databases. Eligible studies included randomized clinical trials, observational studies, reviews, and meta-analyses published between 2010 and 2020. Included studies evaluated clinical outcomes after preoperative nutritional interventions among adult patients undergoing surgery for gastrointestinal cancer. Data extraction was performed using a template developed and tested by the study team. RESULTS: A total of 5,505 publications were identified, of which 69 studies were included for data synthesis after screening and full text review. These studies evaluated preoperative nutritional counseling, protein-calorie supplementation, immunonutrition supplementation, and probiotic or symbiotic supplementation. CONCLUSIONS: Preoperative nutritional counseling and immunonutrition supplementation should be considered for patients undergoing surgical treatment of gastrointestinal malignancy. For malnourished patients, protein-calorie supplementation should be considered, and for patients undergoing colorectal cancer surgery, probiotics or symbiotic supplementation should be considered.

Implementation of a standardized robotic assistant surgical training curriculum.

Since 2000, robotic-assisted surgery has rapidly expanded into almost every surgical sub-specialty. Despite the popularity of robotic surgery across the United States, a national consensus for standardized training and education of robotic surgeons or surgical teams remains absent. In this quality improvement initiative, a novel, stepwise iterative Robotic Assistant Surgical Training (RAST) curriculum was developed to broaden and standardize robotic bedside assistant training. Thirteen voluntary participants, capable of fulfilling the bedside assistant role, were evaluated to determine if RAST enhanced the learner's self-perceived level of confidence and comfort in their role as bedside assistant. A pre- and post-RAST training survey and a between-stages repeated-measures survey were conducted. All learner participants reported statistically significant increases in confidence and comfort after RAST training, (p = < 0.001), and between each stage, F (2, 24 = 60.47, p < .001; [Formula: see text] = 0.834). Participant feedback regarding curriculum improvement was obtained, suggesting the desire for more training and practice, in smaller groups of 2-3 participants. One hundred percent of participants felt RAST was beneficial and that it should be implemented as standardized training during onboarding for all robotic bedside assistants. Thus, a standardized, stepwise iterative robotic bedside assistant curriculum increases learner preparedness, comfort, and confidence, safely away from the patient bedside.

Commentary on drivers of distress and well-being amongst pediatric surgeons.

This is a commentary on the manuscript titled "Drivers of Distress and Well-Being Amongst Pediatric Surgeons" by Rialon, Mueller, Ottosen, et al.

Surgeon Experience with Parental Leave Policies Varies Based on Practice Setting.

BACKGROUND: The increase in female surgeons has resulted in scrutiny of widely variable parental leave policies. We hypothesized that academic and private practice surgeons have different experiences based on difference in workplace expectations. METHODS: A 25-question survey was disseminated via social media and through the Association of Women Surgeons social media platforms from June 1 to September 15, 2017. An analysis of attending surgeons working in the USA in an academic or private practice setting was performed. RESULTS: Of 1115 total respondents, 477 were attending surgeons practicing in the USA. Practice distribution was 34% private and 47% academic. There was no difference in marital status, work status, or the number who report having been pregnant between the groups. Compared to academic surgeons, private practice surgeons were statistically less likely to have paid leave (p < 0.001) and were more likely to continue to pay benefits while on leave (p < 0.001). Private practitioners were more likely to return to work sooner than desired due to financial (p = 0.022) and supervisor (p = 0.004) pressures and were more likely to leave a job (p = 0.01). Academic surgeons were more likely to experience a delay in job advancement (p = 0.031). On multivariate analysis, more than two pregnancies were associated with an increased risk of perception of a bias and discrimination against pregnancy in the workplace. CONCLUSIONS: Parental leave policies and attitudes vary between academic and private practice, creating unique challenges for female surgeons and different issues for family planning depending on employment model.

Rural General Surgery Experience as a Valuable Adjunct to an Academic Based General Surgery Residency.

OBJECTIVE: Few general surgery residencies offer rural rotations. We aim to evaluate the contribution of our institution's rural rotation to meeting the Accreditation Council for Graduate Medical Education (ACGME) minimum case requirements for graduation, and residents' perceptions of the educational value of this rotation. DESIGN: ACGME case log data were obtained from categorical general surgery residents who had completed at least 1 month-long rural surgery rotation and 1 month-long general surgery rotation at our academic medical center within the same clinical year. Cases were classified per ACGME defined categories. For each category, the number of cases per month per resident was calculated, and the means for each educational setting were compared using the paired t-test. Residents also completed a 10-question Likert scale survey regarding their perceptions of the rotation. SETTING: Residents rotated at Vidant Medical Center, a tertiary AMC1 affiliated with East Carolina University in Greenville, NC, and at Vidant Chowan, a critical access hospital within the Vidant Health hospital system located in Edenton, NC. PARTICIPANTS: Categorical general surgery residents eligible to rotate through the rural surgery rotation and the general surgery rotation at the AMC. RESULTS: Eleven total residents completed 23 months of rural surgery (mean 2.1 months per resident) and 39 months at the AMC (mean 3.5 months per resident). Significantly more endoscopic cases, hernia repairs, breast cases, and vascular cases were performed on the rural surgery rotation. More abdominal and alimentary tract cases in addition to endocrine, thoracic, and head/neck cases were performed at the AMC. Frequencies of biliary and soft tissue cases were not significantly different. Survey responses regarding the rural rotation were universally positive including more hands-on experience, increased satisfaction with patient care and continuity, and operative confidence and competence. CONCLUSIONS: At our institution, residents benefit from an enriching rural surgery rotation that provides case numbers different from the comparative AMC general surgery rotation.

Crafting an Evidence-Based, Accreditation Council of Graduate Medical Education-Compliant Lactation Policy for Residents and Fellows.

Background: New Accreditation Council of Graduate Medical Education (ACGME) requirements mandate lactation accommodations for resident physicians and fellows. However, to date, few training programs have developed and reported robust lactation support programs or policies. Objective: The authors aimed to develop an evidence-based, ACGME-compliant policy to optimize lactation support for residents and fellows at their institution. Methods: Six Sigma process improvement methodology was utilized to structure this 2018-2019 project. Qualitative methods included stakeholder analysis, feedback sessions, formal needs assessments, and a thorough review of breastfeeding law, societal guidelines, and best practices. Quantitative methods included use of a standardized grading tool for lactation facilities. Quality assurance efforts are ongoing to ensure successful implementation of the developed policy. Results: The authors present a framework for improving lactation support for residents and fellows and share an institutional policy suitable for implementation by other graduate medical education departments. Conclusions: To ensure compliance with ACGME requirements and address breastfeeding challenges faced by medical trainees, it is crucial that U.S. residencies and fellowships implement lactation policies to support trainees. The authors welcome the modification and utilization of the evidence-based, ACGME-compliant policy reported herein.

Policies and practice regarding pregnancy and maternity leave: An international survey.

BACKGROUND: As women become a larger part of the surgical workforce, policies surrounding maternity and parental leave play a role in professional practice. Little is known about leave policies worldwide. METHODS: A de novo survey distributed internationally to women surgeons assessed leave polices for surgeons, inclusive of the regulatory body or source of applicable policies, changes in surgical practice due to pregnancy, and duration of leave for both parents. RESULTS: The 1111 survey respondents in 53 different countries describe diverse policies ranging from loss of operating room privileges early in pregnancy to maintenance of full surgical schedules until term delivery. Policy creators include national governments (42.38%), employers/hospitals (60.46%), supervisors (18.06%). Self-determined (9.12%), and unknown (8.7%). Paid parental leave was available to 64.44% of women surgeons and 38.68% of partners. CONCLUSION: Maternity and parental leave policies vary markedly across the global surgical workforce with implications for professional practice.

The Teachers of Quality Academy: Evaluation of the Effectiveness and Impact of a Health Systems Science Training Program.

This project aimed to evaluate the effectiveness of a faculty development program in health systems science (HSS)-the Teachers of Quality Academy (TQA). Participants in TQA and a comparison group were evaluated before, during, and 1 year after the program using self-perception questionnaires, tests of HSS knowledge, and tracking of academic productivity and career advancement. Among program completers (n = 27), the mean self-assessed ratings of knowledge and skills of HSS topics immediately after the program, as compared to baseline, increased significantly compared to controls (n = 30). Participants demonstrated progressive improvement of self-perceived skills and attitudes, and retention of HSS knowledge, from baseline to completion of the program. Participants also demonstrated substantially higher HSS scholarly productivity, leadership, and career advancement compared to the comparison group. The TQA effectively created a faculty cadre able to role model, teach, and create a curriculum in HSS competencies for medical students, resident physicians, and other health professionals.

The use of veno-venous extracorporeal membrane oxygenation for perinatal support of an infant with d-transposition of the great arteries, intact atrial and ventricular septa, and flow-restricted ductus arteriosus.

Prenatal assessment of a fetus with D-transposition of the great arteries demonstrated an absence of mixing between systemic and pulmonary circulations, and predicted lethal postnatal hypoxemia. A multidisciplinary meeting evaluated therapeutic options. After cesarean delivery, veno-venous extracorporeal membrane oxygenation was instituted in preparation for open atrial septectomy. The infant subsequently underwent an arterial switch procedure. Prenatal delineation of pulmonary and systemic circulations in the fetus with D-transposition of the great arteries influences postnatal management. Multidisciplinary planning enhanced the perinatal outcome.

State of the practice for pediatric surgery--career satisfaction and concerns. A report from the American Pediatric Surgical Association Task Force on Family Issues.

BACKGROUND: There has been increasing interest and concern raised in the surgical literature regarding changes in the culture of surgical training and practice, and the impact these changes may have on surgeon stress and the appeal of a career in surgery. We surveyed pediatric surgeons and their partners to collect information on career satisfaction and work-family balance. METHODS: The American Pediatric Surgical Association Task Force on Family Issues developed separate survey instruments for both pediatric surgeons and their partners that requested demographic data and information regarding the impact of surgical training and practice on the surgeon's opportunity to be involved with his/her family. RESULTS: We found that 96% of pediatric surgeons were satisfied with their career choice. Of concern was the lack of balance, with little time available for family, noted by both pediatric surgeons and their partners. CONCLUSION: The issues of work-family balance and its impact on surgeon stress and burnout should be addressed in both pediatric surgery training and practice. The American Pediatric Surgical Association is positioned to play a leading role in this effort.

Myelomeningocele: prenatal diagnosis, pathophysiology and management.

Myelomeningocele (MMC) is a common birth defect that is associated with significant lifelong morbidity. Little progress has been made in the postnatal surgical management of the child with spina bifida. Postnatal surgery is aimed at covering the exposed spinal cord, preventing infection, and treating hydrocephalus with a ventricular shunt. In utero repair of open spina bifida is now performed in selected patients and presents an additional therapeutic alternative for expectant mothers carrying a fetus with MMC. Early fetal intervention may improve neurologic outcome and reduce the hindbrain herniation associated with the Arnold-Chiari II malformation. These changes may improve long-term neurologic function and limit requirements for shunt placements and other surgical interventions. Further research is needed to better understand the pathophysiology of MMC, the ideal timing and technique of repair, and the long-term impact of in utero intervention. A prospective, randomized clinical trial is planned comparing prenatal MMC repair with postnatal repair.

Correction of hindbrain herniation and anatomy of the vermis after in utero repair of myelomeningocele in sheep.

BACKGROUND/PURPOSE: In utero repair of myelomeningocele (MMC) in humans spares distal neurologic function, reverses the hindbrain herniation component of the Arnold-Chiari II malformation (ACM), and reduces the rate of postnatal shunt placement. The authors hypothesized that extravasation of cerebrospinal fluid (CSF) from the lumbar spinal cord results in herniation. This hypothesis was tested by assessing the impact of a spinal cord myelotomy on hindbrain anatomy in fetal sheep. METHODS: A MMC lesion was created surgically in 34 fetal sheep at 75 days' gestation by excision of the L1-L5 lamina, the exposed dura, and surrounding tissues. A lumbar level myelotomy was performed in 28 of the 34 fetuses to open the central canal of the spinal cord to enhance egress of CSF through the MMC defect and potentially induce hindbrain herniation. At 102 days' gestation, a repair of the MMC lesion was performed in 14 fetuses with a myelotomy. Fetuses underwent autopsy at 102, 114, 120, or 140 days' gestation. Control animals underwent 2 unrelated fetal surgical procedures at approximately 70 and 110 days' gestation. The incidence of hindbrain herniation, ventricular size, biparietal diameter, brain weight, and brain anatomy were compared between the different animal groups. RESULTS: After MMC creation, significant cerebellar tonsillar herniation was observed in 85% of fetuses that underwent creation of a myelotomy; none of the lambs without a myelotomy (n = 6) had hindbrain herniation. At autopsy, cerebellar tonsillar herniation was present at the time of MMC repair (102 days' gestation), 2 weeks after MMC repair, but was reversed 3 weeks post-MMC repair. At birth, tonsillar herniation was absent, and hindbrain anatomy was restored in 88% of the fetuses with a myelotomy that underwent fetal MMC repair. No significant differences in brain weight and ventricular size was observed between animals with and without MMC repair. CONCLUSIONS: Adding a myelotomy to the sheep model of MMC leads to hindbrain herniation that is similar to that observed in the human ACM. These experiments support the hypothesis that leakage of CSF through the exposed central canal alters the normal CSF hydrodynamics, resulting in cerebellar tonsillar herniation. Fetal MMC repair reverses hindbrain herniation and restores gross anatomy of the vermis.