Background: Artificial intelligence, particularly chatbot systems, is becoming an instrumental tool in health care, aiding clinical decision-making and patient engagement. Objective: This study aims to analyze the performance of ChatGPT-3.5 and ChatGPT-4 in addressing complex clinical and ethical dilemmas, and to illustrate their potential role in health care decision-making while comparing seniors’ and residents’ ratings, and specific question types. Methods: A total of 4 specialized physicians formulated 176 real-world clinical questions. A total of 8 senior physicians and residents assessed responses from GPT-3.5 and GPT-4 on a 1-5 scale across 5 categories: accuracy, relevance, clarity, utility, and comprehensiveness. Evaluations were conducted within internal medicine, emergency medicine, and ethics. Comparisons were made globally, between seniors and residents, and across classifications. Results: Both GPT models received high mean scores (4.4, SD 0.8 for GPT-4 and 4.1, SD 1.0 for GPT-3.5). GPT-4 outperformed GPT-3.5 across all rating dimensions, with seniors consistently rating responses higher than residents for both models. Specifically, seniors rated GPT-4 as more beneficial and complete (mean 4.6 vs 4.0 and 4.6 vs 4.1, respectively; P<.001), and GPT-3.5 similarly (mean 4.1 vs 3.7 and 3.9 vs 3.5, respectively; P<.001). Ethical queries received the highest ratings for both models, with mean scores reflecting consistency across accuracy and completeness criteria. Distinctions among question types were significant, particularly for the GPT-4 mean scores in completeness across emergency, internal, and ethical questions (4.2, SD 1.0; 4.3, SD 0.8; and 4.5, SD 0.7, respectively; P<.001), and for GPT-3.5’s accuracy, beneficial, and completeness dimensions. Conclusions: ChatGPT’s potential to assist physicians with medical issues is promising, with prospects to enhance diagnostics, treatments, and ethics. While integration into clinical workflows may be valuable, it must complement, not replace, human expertise. Continued research is essential to ensure safe and effective implementation in clinical environments.

This study demonstrates that anal relaxation on anorectal manometry significantly correlates with rectal evacuation on defecography, supporting its physiological relevance. Balloon expulsion in the left lateral position shows high specificity but low sensitivity for defecatory dysfunction. Notably, balloon expulsion test (BET) failure was not associated with anatomical abnormalities. An integrated diagnostic approach using anorectal manometry (ARM), BET, and defecography enhances accuracy in distinguishing functional from structural causes of pelvic floor dysfunction.

Background: Acute pain after video-assisted thoracoscopic surgery (VATS) promotes respiratory splinting, impaired cough, and pulmonary complications, and predicts persistent opioid use. Surgeon-administered intercostal nerve block (ICNB) is a simple regional technique, but its independent effect on early pain and opioid requirement in a contemporary uniportal VATS (UVATS) pathway is incompletely defined. Methods: We performed a retrospective cohort study of 456 consecutive patients undergoing UVATS at a single Israeli center between 2017 and 30 May 2025. Patients receiving an intercostal block were compared with those who did not. Baseline covariates were balanced by 1:1 nearest-neighbor propensity-score matching (caliper 0.2 SD of the logit propensity score). The primary endpoints were pain on postoperative day (POD) 1 (visual analog scale, VAS) and postoperative opioid use; secondary endpoints included later pain, analgesic regimen, postoperative pneumonia, and mortality. Results: Matching yielded 159 patients per group (n = 318) with all clinically relevant covariates balanced (standardized mean difference [SMD] < 0.13). Median POD1 VAS was lower with the block (4 [IQR 3–4] vs. 5 [5–7]; p < 0.001), and 76.1% of block patients were opioid-free versus 10.7% who were not (p < 0.001). The effect was concentrated early and attenuated by POD3. In multivariable analysis the block was independently associated with lower POD1 VAS (adjusted β = −1.64, 95% CI −2.00 to −1.29; p < 0.001). Postoperative pneumonia was less frequent in the block group (5.7% vs. 20.1%; p < 0.001). Thirty-day and one-year mortality did not differ significantly. Conclusions: In UVATS, a surgeon-placed intercostal nerve block was associated with lower early postoperative pain that persisted after adjustment for operating surgeon and surgical era, concordant with pooled meta-analytic estimates; associated reductions in opioid use and pneumonia were confounded with surgeon and secular trend and are hypothesis-generating. These single-center, retrospective findings require prospective, protocol-randomized confirmation.

Adult Bochdalek hernias represent a rare clinical entity that often presents diagnostic challenges due to their non-specific symptomatology. We report the case of a 42-year-old male who presented with acute abdominal pain and was found to have a left-sided Bochdalek hernia with strangulated small bowel. The patient underwent a successful laparoscopic repair using right lateral decubitus positioning, which facilitated optimal access to the diaphragmatic defect. The herniated bowel was reduced and found to be viable, and the defect was closed primarily with mesh reinforcement. This case highlights the importance of maintaining a high index of suspicion for diaphragmatic hernias in adults with vague abdominal and respiratory symptoms, the value of prompt imaging in establishing the diagnosis, and the efficacy of minimally invasive surgical techniques with innovative patient positioning for definitive management. The patient recovered well, with complete resolution of symptoms and no recurrence at the 4-year follow-up.

BACKGROUND Artificial intelligence, particularly chatbot systems, is becoming an instrumental tool in health care, aiding clinical decision-making and patient engagement. OBJECTIVE This study aims to analyze the performance of ChatGPT-3.5 and ChatGPT-4 in addressing complex clinical and ethical dilemmas, and to illustrate their potential role in health care decision-making while comparing seniors’ and residents’ ratings, and specific question types. METHODS A total of 4 specialized physicians formulated 176 real-world clinical questions. A total of 8 senior physicians and residents assessed responses from GPT-3.5 and GPT-4 on a 1-5 scale across 5 categories: accuracy, relevance, clarity, utility, and comprehensiveness. Evaluations were conducted within internal medicine, emergency medicine, and ethics. Comparisons were made globally, between seniors and residents, and across classifications. RESULTS Both GPT models received high mean scores (4.4, SD 0.8 for GPT-4 and 4.1, SD 1.0 for GPT-3.5). GPT-4 outperformed GPT-3.5 across all rating dimensions, with seniors consistently rating responses higher than residents for both models. Specifically, seniors rated GPT-4 as more beneficial and complete (mean 4.6 vs 4.0 and 4.6 vs 4.1, respectively; P<.001), and GPT-3.5 similarly (mean 4.1 vs 3.7 and 3.9 vs 3.5, respectively; P<.001). Ethical queries received the highest ratings for both models, with mean scores reflecting consistency across accuracy and completeness criteria. Distinctions among question types were significant, particularly for the GPT-4 mean scores in completeness across emergency, internal, and ethical questions (4.2, SD 1.0; 4.3, SD 0.8; and 4.5, SD 0.7, respectively; P<.001), and for GPT-3.5’s accuracy, beneficial, and completeness dimensions. CONCLUSIONS ChatGPT’s potential to assist physicians with medical issues is promising, with prospects to enhance diagnostics, treatments, and ethics. While integration into clinical workflows may be valuable, it must complement, not replace, human expertise. Continued research is essential to ensure safe and effective implementation in clinical environments.

Background: Bile leaks complicate 5–25% of liver transplants and are among the most common early postoperative biliary events. Their management within the first 14 postoperative days has traditionally been framed around the timing of intervention. Objective: To evaluate whether leak characteristics and patient stability provide more consistent guidance than intervention timing for early post-transplant bile leaks. Methods: A structured narrative review using a dual-source search (AI-assisted semantic search [Elicit Plus] and PubMed/MEDLINE) was conducted. A PICO-informed (Population, Intervention, Comparator, Outcome) framework guided inclusion; two independent reviewers preformed screening, with senior adjudication, yielding 37 sources. Because “early” was defined heterogeneously—by leak onset in most studies (definitions exceeding 14 days) and by intervention timing in others—and no study compared timing within a 14-day window, quantitative pooling was not undertaken. Results: No included study directly compared intervention timing within the first 14 postoperative days. In the largest available analysis of ERCP timing, resolution and adverse event rates did not differ by whether endoscopy was performed within one day, on days two to three, or after three days; endoscopic success was consistently high (80–98%) across temporal windows. Leak type was a consistent correlate of management intensity and resolution: anastomotic leaks were associated with higher surgical-intervention rates (40% vs. 8.3% for T-tube exit-site leaks in one cohort) and lower resolution rates (58–69%) than non-anastomotic leaks (90–100%). Within endoscopic therapy, bridging stent placement—not leak location—was the strongest independent predictor of resolution, and any bile leak, irrespective of subtype, independently predicted subsequent biliary stricture (pooled adjusted OR ≈ 4). Ultra-early leaks (≤72 h), frequently technical failures, formed the one subset favoring early surgical re-exploration. Conclusions: The absence of a demonstrated timing effect reflects a lack of direct comparative data rather than proof that timing is unimportant. Leak type, patient stability, and management modality offer more consistent and actionable guidance than temporal urgency, forming the basis for a leak-stratified framework. Prospective multicenter registries with standardized timing strata and leak type stratification are needed to test whether timing independently influences outcomes.