GPT-4 Artificial Intelligence Model Outperforms ChatGPT, Medical Students, and Neurosurgery Residents on Neurosurgery Written Board-Like Questions.

BACKGROUND: Artificial intelligence (AI) and machine learning have transformed health care with applications in various specialized fields. Neurosurgery can benefit from artificial intelligence in surgical planning, predicting patient outcomes, and analyzing neuroimaging data. GPT-4, an updated language model with additional training parameters, has exhibited exceptional performance on standardized exams. This study examines GPT-4's competence on neurosurgical board-style questions, comparing its performance with medical students and residents, to explore its potential in medical education and clinical decision-making. METHODS: GPT-4's performance was examined on 643 Congress of Neurological Surgeons Self-Assessment Neurosurgery Exam (SANS) board-style questions from various neurosurgery subspecialties. Of these, 477 were text-based and 166 contained images. GPT-4 refused to answer 52 questions that contained no text. The remaining 591 questions were inputted into GPT-4, and its performance was evaluated based on first-time responses. Raw scores were analyzed across subspecialties and question types, and then compared to previous findings on Chat Generative pre-trained transformer performance against SANS users, medical students, and neurosurgery residents. RESULTS: GPT-4 attempted 91.9% of Congress of Neurological Surgeons SANS questions and achieved 76.6% accuracy. The model's accuracy increased to 79.0% for text-only questions. GPT-4 outperformed Chat Generative pre-trained transformer (P < 0.001) and scored highest in pain/peripheral nerve (84%) and lowest in spine (73%) categories. It exceeded the performance of medical students (26.3%), neurosurgery residents (61.5%), and the national average of SANS users (69.3%) across all categories. CONCLUSIONS: GPT-4 significantly outperformed medical students, neurosurgery residents, and the national average of SANS users. The mode's accuracy suggests potential applications in educational settings and clinical decision-making, enhancing provider efficiency, and improving patient care.

Perioperative mortality for radical cystectomy in the modern Era: experience from a tertiary referral center

ABSTRACT Purpose To evaluate the perioperative mortality and contributing variables among patients who underwent radical cystectomy (RC) for bladder cancer in recent decades, with comparison between modern (after 2010) and premodern (before 2010) eras. Materials and Methods Using our institutional review board-approved database, we reviewed the records of patients who underwent RC for primary urothelial bladder carcinoma with curative intent from January 2003 to December 2019. The primary and secondary outcomes were 90- and 30-day mortality. Univariate and multivariable logistic regression models were applied to assess the impact of perioperative variables on 90-day mortality. Results A total of 2047 patients with a mean±SD age of 69.6±10.6 years were included. The 30- and 90-day mortality rates were 1.3% and 4.9%, respectively, and consistent during the past two decades. Among 100 deaths within 90 days, 18 occurred during index hospitalization. Infectious, pulmonary, and cardiac complications were the leading mortality causes. Multivariable analysis showed that age (Odds Ratio: OR 1.05), Charlson comorbidity index ≥ 2 (OR 1.82), blood transfusion (OR 1.95), and pathological node disease (OR 2.85) were independently associated with 90-day mortality. Nevertheless, the surgical approach and enhanced recovery protocols had no significant effect on 90-day mortality. Conclusion The 90-day mortality for RC is approaching five percent, with infectious, pulmonary, and cardiac complications as the leading mortality causes. Older age, higher comorbidity, blood transfusion, and pathological lymph node involvement are independently associated with 90-day mortality.

Perioperative mortality for radical cystectomy in the modern Era: experience from a tertiary referral center.

PURPOSE: To evaluate the perioperative mortality and contributing variables among patients who underwent radical cystectomy (RC) for bladder cancer in recent decades, with comparison between modern (after 2010) and premodern (before 2010) eras. MATERIALS AND METHODS: Using our institutional review board-approved database, we reviewed the records of patients who underwent RC for primary urothelial bladder carcinoma with curative intent from January 2003 to December 2019. The primary and secondary outcomes were 90- and 30-day mortality. Univariate and multivariable logistic regression models were applied to assess the impact of perioperative variables on 90-day mortality. RESULTS: A total of 2047 patients with a mean±SD age of 69.6±10.6 years were included. The 30- and 90-day mortality rates were 1.3% and 4.9%, respectively, and consistent during the past two decades. Among 100 deaths within 90 days, 18 occurred during index hospitalization. Infectious, pulmonary, and cardiac complications were the leading mortality causes. Multivariable analysis showed that age (Odds Ratio: OR 1.05), Charlson comorbidity index ≥ 2 (OR 1.82), blood transfusion (OR 1.95), and pathological node disease (OR 2.85) were independently associated with 90-day mortality. Nevertheless, the surgical approach and enhanced recovery protocols had no significant effect on 90-day mortality. CONCLUSION: The 90-day mortality for RC is approaching five percent, with infectious, pulmonary, and cardiac complications as the leading mortality causes. Older age, higher comorbidity, blood transfusion, and pathological lymph node involvement are independently associated with 90-day mortality.

Association between early postradical cystectomy kidney injury and perioperative outcome in enhanced recovery era.

OBJECTIVE: To evaluate the incidence and predictors of early postoperative acute kidney injury (EP-AKI) during index hospitalization following radical cystectomy and its association with postoperative outcomes. METHODS: All patients with bladder cancer who underwent radical cystectomy with intent-to-cure at our center between 2012 and 2020 were reviewed. EP-AKI during index hospitalization was evaluated using the Acute Kidney Injury Network criteria. The association between EP-AKI and demographics, clinicopathologic features, and perioperative outcomes, including length of hospital stay, complication rate, and readmission rate, were examined. A logistic regression analysis was performed to evaluate the predictors of EP-AKI. RESULTS: Overall, 435 patients met eligibility, of whom 112 (26%) experienced EP-AKI during index hospitalization (90 [21%] stage 1, 17 [4%] stage 2, and 5 [1%] stage 3). EP-AKI was associated with a longer mean operative time (6.8 vs. 6.1 hours; P < 0.001), higher mean length of hospital stay (6.3 vs. 5.6; P = 0.02), 30-day complication rate (71% vs. 51%; P < 0.001), 90-day complication rate (81% vs. 69%; P = 0.01) and 90-day readmission rate (37% vs. 33%; P = 0.04). The rate of complications increased at higher stages of AKI. On multivariable analysis, perioperative blood transfusion (OR: 1.84, P = 0.02) and continent diversion (OR: 3.29, P < 0.001) were independent predictors of EP-AKI. CONCLUSION: A quarter of cystectomy patients experience acute kidney injury during index hospitalization, which is associated with higher length of stay, postoperative complication, and readmission rates. Perioperative blood transfusion and continent diversion are independent predictors of such injury.

Distal ureter and bladder cuff excision using the "Keyhole Technique" during Robotic Radical Nephroureterectomy

ABSTRACT Introduction: Upper tract urothelial carcinoma (UTUC) accounts for 5-10% of all urothelial tumors (1). Radical nephroureterectomy (RNU) remains the standard treatment for high, and low-grade UTUC (2). Although the open approach has been considered the gold standard, robotic techniques have shown comparable oncological outcomes with potential advantages in terms of peri-operative morbidity (3). Materials and Methods: We present a novel "Keyhole" technique for management of distal ureter and bladder cuff during robotic RNU. This technique allows the surgeon to directly visualize the ureteric orifices, delineate resection borders, and maintain oncologic principles of en-bloc excision without necessitating secondary cystotomy incision or concomitant endoscopic procedure. Descriptive demographic characteristics, surgical, pathological, and oncological outcomes were analyzed. Complications were reported using the Clavien-Dindo classification system. Results: Between 2015 and 2020, ten patients underwent robotic RNU with bladder cuff excision using the Keyhole technique (single-dock, single-position). Median age was 75 years. Eight patients underwent surgery for right-sided tumors. Median operative time, estimated blood loss, and length of hospital stay were 287 min, 100 mL, and 3 days, respectively. No intraoperative complications occurred, and one grade II complication occurred during the 90-day postoperative period. All patients had high-grade UTUC, being 90% pure urothelial. Bladder recurrences occurred in 30% of patients with an overall median follow-up of 11.2 months. Conclusions: Keyhole technique for the management of distal ureter and bladder cuff during RNU represents a feasible approach with minimal 90-day complications and low bladder recurrence rate at centers of experience.

Distal ureter and bladder cuff excision using the "Keyhole Technique" during Robotic Radical Nephroureterectomy.

INTRODUCTION: Upper tract urothelial carcinoma (UTUC) accounts for 5-10% of all urothelial tumors (1). Radical nephroureterectomy (RNU) remains the standard treatment for high, and low-grade UTUC (2). Although the open approach has been considered the gold standard, robotic techniques have shown comparable oncological outcomes with potential advantages in terms of peri-operative morbidity (3). MATERIALS AND METHODS: We present a novel "Keyhole" technique for management of distal ureter and bladder cuff during robotic RNU. This technique allows the surgeon to directly visualize the ureteric orifices, delineate resection borders, and maintain oncologic principles of en-bloc excision without necessitating secondary cystotomy incision or concomitant endoscopic procedure. Descriptive demographic characteristics, surgical, pathological, and oncological outcomes were analyzed. Complications were reported using the Clavien-Dindo classification system. RESULTS: Between 2015 and 2020, ten patients underwent robotic RNU with bladder cuff excision using the Keyhole technique (single-dock, single-position). Median age was 75 years. Eight patients underwent surgery for right-sided tumors. Median operative time, estimated blood loss, and length of hospital stay were 287 min, 100 mL, and 3 days, respectively. No intraoperative complications occurred, and one grade II complication occurred during the 90-day postoperative period. All patients had high-grade UTUC, being 90% pure urothelial. Bladder recurrences occurred in 30% of patients with an overall median follow-up of 11.2 months. CONCLUSIONS: Keyhole technique for the management of distal ureter and bladder cuff during RNU represents a feasible approach with minimal 90-day complications and low bladder recurrence rate at centers of experience.