Fackelmayer, Oliver J.

Abstract Context Literature suggests patients with thyroid cancer have unmet informational needs in many aspects of care. Patients often turn to online resources for their health-related information, and generative artificial intelligence programs such as ChatGPT are an emerging and attractive resource for patients. Objective To assess the quality of ChatGPT's responses to thyroid cancer-related questions. Methods Four endocrinologists and 4 endocrine surgeons, all with expertise in thyroid cancer, evaluated the responses to 20 thyroid cancer-related questions. Responses were scored on a 7-point Likert scale in areas of accuracy, completeness, and overall satisfaction. Comments from the evaluators were aggregated and a qualitative analysis was performed. Results Overall, only 57%, 56%, and 52% of the responses “agreed” or “strongly agreed” that ChatGPT's answers were accurate, complete, and satisfactory, respectively. One hundred ninety-eight free-text comments were included in the qualitative analysis. The majority of comments were critical in nature. Several themes emerged, which included overemphasis of diet and iodine intake and its role in thyroid cancer, and incomplete or inaccurate information on risks of both thyroid surgery and radioactive iodine therapy. Conclusion Our study suggests that ChatGPT is not accurate or reliable enough at this time for unsupervised use as a patient information tool for thyroid cancer.

Background The American Association for the Surgery of Trauma (AAST) has developed a grading system for emergency general surgery (EGS) conditions. We sought to validate the AAST EGS grades for patients undergoing urgent/emergent colorectal resection. Methods Patients enrolled in the “Eastern Association for the Surgery of Trauma Multicenter Colorectal Resection in EGS—to anastomose or not to anastomose” study undergoing urgent/emergent surgery for obstruction, ischemia, or diverticulitis were included. Baseline demographics, comorbidity severity as defined by Charlson comorbidity index (CCI), procedure type, and AAST grade were prospectively collected. Outcomes included length of stay (LOS) in-hospital mortality, and surgical complications (superficial/deep/organ-space surgical site infection, anastomotic leak, stoma complication, fascial dehiscence, and need for further intervention). Multivariable logistic regression models were used to describe outcomes and risk factors for surgical complication or mortality. Results There were 367 patients, with a mean (± SD) age of 62 ± 15 years. 39% were women. The median interquartile range (IQR) CCI was 4 (2-6). Overall, the pathologies encompassed the following AAST EGS grades: I (17, 5%), II (54, 15%), III (115, 31%), IV (95, 26%), and V (86, 23%). Management included laparoscopic (24, 7%), open (319, 87%), and laparoscopy converted to laparotomy (24, 6%). Higher AAST grade was associated with laparotomy ( P = .01). The median LOS was 13 days (8-22). At least 1 surgical complication occurred in 33% of patients and the mortality rate was 14%. Development of at least 1 surgical complication, need for unplanned intervention, mortality, and increased LOS were associated with increasing AAST severity grade. On multivariable analysis, factors predictive of in-hospital mortality included AAST organ grade, CCI, and preoperative vasopressor use (odds ratio (OR) 1.9, 1.6, 3.1, respectively). The American Association for the Surgery of Trauma emergency general surgery grade was also associated with the development of at least 1 surgical complication (OR 2.5), while CCI, preoperative vasopressor use, respiratory failure, and pneumoperitoneum were not. Conclusion The American Association for the Surgery of Trauma emergency general surgery grading systems display construct validity for mortality and surgical complications after urgent/emergent colorectal resection. These results support incorporation of AAST EGS grades for quality benchmarking and surgical outcomes research.