Unveiling the Potential of AI in Plastic Surgery Education: A Comparative Study of Leading AI Platforms' Performance on In-training Examinations.

Background: Within the last few years, artificial intelligence (AI) chatbots have sparked fascination for their potential as an educational tool. Although it has been documented that one such chatbot, ChatGPT, is capable of performing at a moderate level on plastic surgery examinations and has the capacity to become a beneficial educational tool, the potential of other chatbots remains unexplored. Methods: To investigate the efficacy of AI chatbots in plastic surgery education, performance on the 2019-2023 Plastic Surgery In-service Training Examination (PSITE) was compared among seven popular AI platforms: ChatGPT-3.5, ChatGPT-4.0, Google Bard, Google PaLM, Microsoft Bing AI, Claude, and My AI by Snapchat. Answers were evaluated for accuracy and incorrect responses were characterized by question category and error type. Results: ChatGPT-4.0 outperformed the other platforms, reaching accuracy rates up to 79%. On the 2023 PSITE, ChatGPT-4.0 ranked in the 95th percentile of first-year residents; however, relative performance worsened when compared with upper-level residents, with the platform ranking in the 12th percentile of sixth-year residents. The performance among other chatbots was comparable, with their average PSITE score (2019-2023) ranging from 48.6% to 57.0%. Conclusions: Results of our study indicate that ChatGPT-4.0 has potential as an educational tool in the field of plastic surgery; however, given their poor performance on the PSITE, the use of other chatbots should be cautioned against at this time. To our knowledge, this is the first article comparing the performance of multiple AI chatbots within the realm of plastic surgery education.

Prescription Opioids and Patient-Reported Outcomes and Satisfaction After Carpal Tunnel Release Surgery.

BACKGROUND: Amount of opioid use correlates poorly with procedure-related pain; however, prescription limits raise concerns about inadequate pain control and impacts on patient-reported quality indicators. There remain no consistent guidelines for postoperative pain management after carpal tunnel release (CTR). We sought to understand how postoperative opioid use impacts patient-reported outcomes after CTR. METHODS: This is a pragmatic cohort study using prospectively collected data from all adult patients undergoing uncomplicated primary CTR over 17 months at our center. Patients were categorized as having received or not received a postoperative opioid prescription, and then as remaining on a prescription opioid at 2-week follow-up or not. Questionnaires were completed before surgery and at 2-week follow-up. We collected brief Michigan Hand questionnaire (bMHQ) score, Patient-Reported Outcomes Measurement Information System Global Health score, satisfaction, and pain score. RESULTS: Of 505 included patients, 405 received a postoperative prescription and 67 continued use at 2-weeks. These 67 patients reported lower bMHQ, lower satisfaction, and higher postoperative pain compared to those that discontinued. Multivariable regressions showed that receiving postoperative prescriptions did not significantly influence outcomes or satisfaction. However, remaining on the prescription at 2 weeks was associated with significantly lower bMHQ scores, particularly in patients reporting less pain. CONCLUSIONS: Patients remaining on a prescription after CTR reported worse outcomes compared to those who discontinued. Unexpectedly, the widest bMHQ score gap was seen across patients reporting lowest pain scores. Further research into this high-risk subgroup is needed to guide policy around using pain and patient-reported outcomes as quality measures.Level of Evidence: Level III.

Risk Factors for Preventable Emergency Department Use After Outpatient Hand Surgery.

PURPOSE: Emergency department (ED) visits for postoperative concerns that could be safely addressed in outpatient clinics have an impact on cost, quality measures, and care workflows. Patient-reported data (PRD) may give unique insights into individual-level factors that predict overuse of health care resources, and guide opportunities for intervention and prevention. We investigated the relationship between preoperative PRD and preventable ED use after outpatient hand surgery to determine whether the preoperative PRD can be used to identify patients at higher odds of having preventable ED visits. METHODS: All adult patients undergoing outpatient surgery at our hand center between January 1, 2018, and December 31, 2019, were included. Questionnaires, including the Patient-Reported Outcomes Measurement Information System (PROMIS) upper extremity (UE) and pain interference (PI) scales, were completed before surgery. We used our regional health information exchange to identify ED visits within 90 days of surgery. RESULTS: Our cohort included 2,819 patients. Within 90 days after surgery, 106 (3.8%) had preventable ED visits. Race, insurance status, and transportation issues increased odds of a preventable ED visit. Multivariable models found that each 1-point increase in the preoperative PROMIS UE score was associated with 4% decreased odds of ED presentation (odds ratio, 0.96; 95% confidence interval, 0.94-0.99), and each 1-point increase in the preoperative PROMIS PI score was associated with 4% increased odds of ED presentation (odds ratio, 1.04; 95% confidence interval, 1.0-1.1). Any PROMIS UE or PI scores ≥1SDs worse than population norms increased the probability of a preventable ED visit, independent of other factors. CONCLUSIONS: Worse preoperative PROMIS UE and PI scores were associated with increased odds of preventable ED visits. Preoperative PRD may allow for identification of outliers at higher risk for preventable ED use, and facilitate preventative interventions. TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic IV.

Early Outcomes of Endoscopic Versus Open Carpal Tunnel Release.

PURPOSE: To compare the short-term outcomes of endoscopic carpal tunnel release (ECTR) and open carpal tunnel release (OCTR), including patient-reported outcomes, pain and satisfaction scores, return to work, and postoperative prescription pain medication use. METHODS: We included all patients over 18 years of age undergoing carpal tunnel release at a single hand center between January 2018 and December 2019. The carpal tunnel release method was driven by variations in surgeon practice. Data from patient-reported outcomes measurement information system (PROMIS) questionnaires and brief Michigan hand outcomes questionnaires and data on patient-reported pain levels, satisfaction with care, return to work, and postoperative prescription pain medication use were collected at preoperative visits and the first follow-up visit between postoperative days 7 and 14. RESULTS: We included 678 (586 ECTR and 92 OCTR) patients. The median age was 58 years, and 75% of the patients were women. At early follow up, patients who underwent OCTR reported significantly lower postoperative PROMIS upper-extremity scores than those who underwent ECTR (median, 32 vs 36 points, respectively) but similar postoperative PROMIS pain interference, global physical health, global mental health, and brief Michigan hand outcomes questionnaire scores. The postoperative pain and satisfaction scores were similar between the 2 groups. In multivariable models, patients who underwent OCTR had 62% lower odds of returning to work and 30% greater odds of remaining on a postoperative pain prescription at the first follow-up visit. CONCLUSIONS: This study found no evidence suggesting the definitive superiority of 1 surgical technique with regard to clinical outcomes in the early postoperative period. However, OCTR was associated with lower postoperative PROMIS upper-extremity scores of unclear clinical significance, higher odds of remaining on pain medication, and lower odds of returning to work by the first postoperative visit. Endoscopic carpal tunnel release may be preferred in patients who need to return to work within the first 2 weeks after the procedure. TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic IV.

Correlation of Patient-Reported Outcomes Measurement Information System Questionnaires With the Brief Michigan Hand Questionnaire in Patients With 5 Common Hand Conditions.

PURPOSE: We evaluated the concurrent validity of the Patient-Reported Outcomes Measurement Information System (PROMIS) Global Mental Health (GMH), Global Physical Health (GPH), Upper Extremity (UE), Pain Interference (PI), and Self-Efficacy for Managing Medications and Treatment (SE-MMT) by analyzing correlation to the brief Michigan Hand Questionnaire (bMHQ) in patients with 5 common hand conditions: carpal tunnel syndrome, Dupuytren contracture, trigger finger, thumb carpometacarpal osteoarthritis, and wrist ganglion cysts. METHODS: Our cohort included 11,782 unique visits representing 4,401 patients. Patient demographics, PROMIS Computer Adaptive Test questionnaires, and bMHQ were collected prospectively at all visits for all patients. Spearman rank correlation was used to evaluate the relationship between the PROMIS and bMHQ scores. Multivariable linear regression models were used to evaluate the relationship between questionnaires and patient demographics. RESULTS: The PROMIS UE and PI showed strong correlations to the bMHQ. The PROMIS GPH showed moderate correlation to the bMHQ. The PROMIS GMH and SE-MMT were weakly correlated with the bMHQ. These results for the overall group were consistent across subgroup analysis for each condition, and regression models confirmed these correlation findings when controlling for demographic variables. The bMHQ had the smallest ceiling and floor effects compared with the PROMIS questionnaires. The PROMIS UE, PI, and SE-MMT took significantly less time to complete than the bMHQ. CONCLUSION: Correlations between the PROMIS questionnaires and the bMHQ were similar regardless of condition. CLINICAL RELEVANCE: Given their moderate-to-strong correlations with the bMHQ and consistency across conditions, the PROMIS UE and PI may be adequate replacements for the bMHQ for evaluating these domains in both clinical and research applications in patients with these common upper extremity pathologies. The PROMIS GPH, GMH, and SE-MMT, in conjunction with the bMHQ, may provide more information regarding patient's physical and mental health and ability to manage medications and treatment without substantially increasing patient burden. Clinicians and researchers can use these findings to guide questionnaire selection based on the clinical or research question(s) of interest.

Impact of nicotine/smoking, alcohol, and illicit substance use on outcomes and complications of burn patients requiring hospital admission: systematic review and meta-analysis.

Burn patients admitted to the hospital with concurrent intoxication are believed to be at an increased risk of poor outcomes and the development of complications, however data varies within the literature and remains controversial. This systematic review and meta-analysis compared outcomes and complications between nicotine/smoking, alcohol, and/or substance use in 26,512 burn patients admitted to the hospital to 299,543 burn patients admitted without these characteristics. The PubMed, EMBASE, Cochrane Library, and Web of Science databases were systematically and independently searched. PRISMA and Cochrane guidelines were strictly followed. Clinical characteristics, nicotine/smoking use, alcohol use, substance use, outcomes and complications were recorded. Seventeen of the 27 studies included in the study, were eligible for meta-analysis, with results from 39 of the possible 84 outcomes and complications. In conclusion, this systematic review and meta-analysis found that compared to non-nicotine/smoking, non-alcohol, non-substance use burn patients, patients using nicotine/smoking, alcohol, and/or substances were associated with more burn related operations, higher rates of graft loss/failure, longer hospital LOS (length of stay), higher rates of intubation, longer ICU (intensive care unit) LOS, increased mortality, and increased wound/local skin infections. Patients using nicotine/smoking were associated with higher rates of intubation and wound/local skin infections. Patients consuming alcohol were associated with more days on a ventilator, had higher rates of intubation, higher rates of inhalation injury, longer ICU LOS, and increased mortality. Patients taking substances were associated with higher %TBSA (percent total body surface area) of burns, longer hospital LOS, higher rates of intubation, higher rates of inhalation injury, longer ICU LOS, and increased wound/local skin infections.