AI Versus MD: Evaluating the surgical decision-making accuracy of ChatGPT-4.

BACKGROUND: ChatGPT-4 is a large language model with possible applications to surgery education The aim of this study was to investigate the accuracy of ChatGPT-4's surgical decision-making compared with general surgery residents and attending surgeons. METHODS: Five clinical scenarios were created from actual patient data based on common general surgery diagnoses. Scripts were developed to sequentially provide clinical information and ask decision-making questions. Responses to the prompts were scored based on a standardized rubric for a total of 50 points. Each clinical scenario was run through Chat GPT-4 and sent electronically to all general surgery residents and attendings at a single institution. Scores were compared using Wilcoxon rank sum tests. RESULTS: On average, ChatGPT-4 scored 39.6 points (79.2%, standard deviation ± 0.89 points). A total of five junior residents, 12 senior residents, and five attendings completed the clinical scenarios (resident response rate = 15.9%; attending response rate = 13.8%). On average, the junior residents scored a total of 33.4 (66.8%, standard deviation ± 3.29), senior residents 38.0 (76.0%, standard deviation ± 4.75), and attendings 38.8 (77.6%, standard deviation ± 5.45). ChatGPT-4 scored significantly better than junior residents (P = .009) but was not significantly different from senior residents or attendings. ChatGPT-4 was significantly better than junior residents at identifying the correct operation to perform (P = .0182) and recommending additional workup for postoperative complications (P = .012). CONCLUSION: ChatGPT-4 performed superior to junior residents and equivalent to senior residents and attendings when faced with surgical patient scenarios. Large language models, such as ChatGPT, may have the potential to be an educational resource for junior residents to develop surgical decision-making skills.

A second chance for a new heart? The role of metabolic and bariatric surgery in patients with end-stage heart failure.

BACKGROUND: Obesity is an independent risk factor for heart failure (HF). Substantial weight loss has been shown to reverse obesity-related cardiomyopathy. This study aimed to report our institution's experience with laparoscopic sleeve gastrectomy (LSG) in patients with morbid obesity and end-stage HF. METHODS: Between 2018 and 2022, 26 patients with end-stage HF were referred for LSG. Of 26 patients, 16 underwent an operation, and 10 did not. After institutional review board approval, a retrospective electronic medical record review was performed to evaluate (i) age, (ii) preoperative weight, (iii) decrease in body mass index (BMI) score, (iv) whether the patient underwent heart transplantation, and (v) mortality. Data analysis was performed using Stata/SE (version 17.0; StataCorp). The Wilcoxon rank-sum test was used to compare continuous variables between the cohorts, and the Pearson chi-square test was used for binary variables with Bonferroni correction applied. RESULTS: The LSG and non-LSG cohorts had comparable ages (P = .088) and starting BMI score (P = .918), and a proportion of patients had a ventricular assist device (P = .191). Patients who underwent LSG lost significantly more weight than the patients who did not, with an average decrease in BMI score of 8.9 kg/m2 (SD, ±6.13) and 1.1 kg/m2 (SD, ±4.10), respectively (P = .040). Of note, 6 patients (37.5%) who underwent LSG eventually underwent transplantation, compared with 2 patients (20.0%) from the matched cohort (P = .884). Of the 26 patients, there were 6 deaths: 2 (12.5%) in the LSG cohort and 4 (40.0%) in the non-LSG cohort (P = .525). CONCLUSION: LSG may be safe and effective for weight loss in patients with HF. This operation may provide patients affected by obesity with end-stage HF the lifesaving opportunity to achieve transplant candidacy.

Hiatal hernia reporting: time to remove subjectivity?

INTRODUCTION: The size of a hiatal hernia (HH) is a key determinant of the approach for surgical repair. However, endoscopists will often utilize subjective terms, such as "small," "medium," and "large," without any standardized objective correlations. The aim of this study was to identify HHs described using objective axial length measurements versus subjective size allocations and compare them to their corresponding manometry and barium swallow studies. METHODS AND PROCEDURES: Retrospective chart reviews were conducted on 93 patients diagnosed endoscopically with HHs between 2017 and 2021 at Newton-Wellesley Hospital. Information was collected regarding their HH subjective size assessment, axial length measurement (cm), manometry results, and barium swallow readings. Linear regression models were used to analyze the correlation between the objective endoscopic axial length measurements and manometry measurements. Ordered logistic regression models were used to correlate the ordinal endoscopic and barium swallow subjective size allocations with the continuous axial length measurements and manometry measurements. RESULTS: Of the 93 endoscopy reports, 42 included a subjective size estimate, 38 had axial length measurement, and 12 gave both. Of the 34 barium swallow reads, only one gave an objective HH size measurement. Axial length measurements were significantly correlated with the manometry measurements (R2 = 0.0957, p = 0.049). The endoscopic subjective size estimates were also closely related to the manometry measurements (R2 = 0.0543, p = 0.0164). Conversely, the subjective size estimates from barium swallow reads were not significantly correlated with the endoscopic axial length measurements (R2 = 0.0143, p = 0.366), endoscopic subjective size estimates (R2 = 0.0481, p = 0.0986), or the manometry measurements (R2 = 0.0418, p = 0.0738). Mesh placement was significantly correlated to pre-operative endoscopic axial length measurement (p = 0.0001), endoscopic subjective size estimate (p = 0.0301), and barium swallow read (p = 0.0211). However, mesh placement was not significantly correlated with pre-operative manometry measurements (0.2227). CONCLUSIONS: Endoscopic subjective size allocations and objective axial length measurements are associated with pre-operative objective measurements and intra-operative decisions, suggesting both can be used to guide clinical decision making. However, including axial length measurements in endoscopy reports can improve outcomes reporting.

Does retrieval bag use during laparoscopic appendectomy reduce postoperative infection?

BACKGROUND: Appendectomy is the most commonly performed emergency operation in the United States, with approximately 370,000 patients undergoing the procedure every year. Although laparoscopic appendectomy is associated with decreased complications when compared with open appendectomy, the risk for infectious complications, including surgical site infection, intra-abdominal abscess, and sepsis, remains a significant source of postoperative morbidity and health care cost. The goal of this study is to determine whether the appendix retrieval technique during laparoscopic appendectomy affects risk of infectious complications. METHODS: The American College of Surgeons National Surgical Quality Improvement Program database and the Appendectomy Procedure Targeted database were used to conduct this retrospective study. Patients who underwent laparoscopic appendectomy in 2016 were identified. The primary outcomes were infectious complications of superficial site infection and intra-abdominal abscess. RESULTS: A total of 10,578 (92.2%) patients underwent laparoscopic appendectomy using an appendix retrieval bag and 897 (7.8%) patients underwent laparoscopic appendectomy without an appendix retrieval bag. There was no significant difference in preoperative sepsis, smoking status, wound class, complicated appendicitis, or American Society of Anesthesiologists class between patient groups (all P > .05). In the univariate analysis, there was no difference in the rate of superficial site infection (0.9% vs 0.6%, P = .28) or intra-abdominal infection (2.7% vs 3.8%, P = .06) between retrieval bag use and non-use. In the multivariable analysis, appendix retrieval bag use was an independent predictor of intra-abdominal infection and associated with a 40% decrease in intra-abdominal infection rates (odds ratio: 0.6, 95% confidence interval: 0.42-0.95, P = .03). CONCLUSION: Appendix retrieval bags are associated with a decreased risk of postoperative intra-abdominal abscess. The use of appendix retrieval bags should be the standard of care during laparoscopic appendectomy.