Faculty Bibliography

BACKGROUND:Comparative trials evaluating surgical outcomes are critical in guiding treatment for hallux valgus. However, the statistical stability of these outcomes is not well documented. Purpose The purpose of this study was to evaluate the statistical fragility of comparative studies analyzing minimally invasive surgery (MIS) versus open techniques for hallux valgus correction. STUDY/METHODS:Design A systematic review identified comparative studies assessing MIS versus open hallux valgus correction. METHODS:Outcome data were extracted with Fragility Index (FI) and Continuous Fragility Index (CFI) calculated for significant outcomes, and reverse FI (rFI) and reverse CFI (rCFI) for nonsignificant outcomes. Fragility Quotient (FQ) was calculated for each and compared to the number of patients lost to follow-up (LTFU). RESULTS:Of 628 studies screened, 18 met inclusion criteria, totaling 1,369 patients. Among 88 dichotomous outcomes, the median FI was 2, FQ was 0.021, rFI was 4, and rFQ was 0.072. For 236 continuous outcomes, the median CFI was 8, CFQ was 0.116, rCFI was 19, and rCFQ was 0.280. CONCLUSION/CONCLUSIONS:This is the first study to evaluate fragility in comparative trials on MIS versus open hallux valgus correction and among the first to assess reverse fragility in continuous outcomes. Significant results were more fragile than nonsignificant data and dichotomous outcomes were more fragile than continuous ones with nearly a quarter having an FI lower than the number of patients LTFU. Both outcome types demonstrated considerable statistical fragility supporting the cautious interpretation of MIS vs open hallux valgus findings and the reporting of statistical fragility data alongside P-values to better contextualize the robustness of clinical research.

BACKGROUND:Epigenetic regulators represent a novel strategy to modulate the tumor immune microenvironment in pancreatic ductal adenocarcinoma (PDAC). In preclinical models, DNA hypomethylating agents enhance cytotoxic T-cell infiltration, synergize with PD-1 blockade, and improve survival when combined with immune checkpoint blockade. This single-institution, phase II study evaluated the safety, efficacy, and biomarkers of azacitidine plus pembrolizumab in patients with previously treated PDAC. METHODS:Patients with locally advanced or metastatic PDAC after one prior regimen received 50 mg/m2 subcutaneous azacitidine on days 1-5 of a 28-day cycle, starting week 1, and pembrolizumab 200 mg intravenously every 3 weeks starting week 3. Baseline and on-treatment blood and tumor was collected for exploratory biomarker analysis. RESULTS:Thirty-six patients enrolled between October 2017 and September 2021 (median age: 62.5 years); 34 were evaluable for safety; 31 for efficacy. Treatment was generally well-tolerated, with Grade 1-2 fatigue and diarrhea most common AEs. Three patients (9.7%) had a partial response, and the disease control rate was 35.5%. Median progression-free and overall survival was 1.51 and 4.83 months, respectively. Exploratory analysis suggested higher baseline CD8+ T cells and lower tumor Ki-67 was associated with response, whereas low baseline CD8+ T cell and Granzyme B infiltration correlated with higher exponential tumor growth rate. PD-L1 and CD68 expression were not predictive of benefit. CONCLUSION/CONCLUSIONS:Azacitidine plus pembrolizumab demonstrated limited clinical activity in second line, locally advanced or metastatic PDAC. Biomarker analysis suggests higher baseline CD8+ T-cell infiltration and lower proliferative index may identify patients more likely to benefit. (NCT03264404).

BACKGROUND:The number of graduates from accelerated 3-year MD (A3YP) programs has increased over the past decade. Previous work showed that A3YP graduates perform comparably to non-accelerated (4-year) graduates from the same medical schools on mid-year and end-year Accreditation Council of Graduate Medical Education (ACGME) harmonized milestones. In shifting to the residency program perspective, it remains unclear how the performance of A3YP graduates compares to non-accelerated graduates including traditional 4-year, international, and osteopathic medical school graduates. OBJECTIVE:To compare the intern performance of A3YP graduates compared with non-accelerated graduates using mid-year and end-year ACGME milestones in Internal Medicine (IM) residency programs. DESIGN/METHODS:The study employed a retrospective cohort design, hypothesizing that graduates from A3YPs were comparable to non-accelerated graduates in the same program. PARTICIPANTS/METHODS:108 interns who graduated from A3YP were compared to 3,542 interns from non-accelerated programs at the same 34 IM residency programs. MAIN MEASURES/METHODS:Descriptive statistics were provided for ACGME milestone performance. Cross-classified random-effects regression was used to account for residency program effects and estimate group differences. KEY RESULTS/RESULTS:After controlling for residency programs, the milestone ratings of A3YP graduates were higher in all competency domains at mid-year except practice-based learning and improvement (PBLI) at .04 (P = .089) (coefficients ranged from 0.08 for medical knowledge (MK) (P < .001) to 0.23 in professionalism (PROF) (P < .001)). These differences persisted at the end-year period (coefficients ranged from 0.05 in PBLI (P = .039) to 0.17 in PROF (P < .001)) except MK at .02 (P = .656). Patient care differences were 0.15 (P < .001) at mid- and 0.14 (P = .005) at end-year. CONCLUSIONS:This study contributes to the literature demonstrating that interns graduating from A3YP are at least equivalent in terms of milestone assessment and possibly better in the competencies of PC and PROF than their non-accelerated counterparts.

OBJECTIVE:Exposure to antidepressants, particularly agents that work through serotonin-reuptake inhibition, may increase potential for bleeding, especially among patients with other bleeding risk factors. There is limited guidance for clinicians in the use of serotonin reuptake inhibitors (SRIs) and other antidepressants in the setting of increased bleeding risk. METHODS:A PubMed literature search was conducted for English-language articles (1992-2024) examining the bleeding risk associated with antidepressants. Physicians from the Association of Medicine and Psychiatry then convened to develop consensus recommendations. RESULTS:Consensus recommendations were established for managing antidepressant use in patients with medical and psychiatric comorbidities. Additionally, a clinical decision algorithm was created to assist clinicians in assessing the appropriateness of antidepressant prescribing in patients at risk for bleeding. CONCLUSIONS:The proposed algorithm can aid clinicians in determining whether antidepressant (including SRI) initiation, discontinuation, or dose adjustment should be considered for patients susceptible to bleeding. These guidelines preserve a role for clinical judgment in selection of treatments that balance the risks and benefits, which may be particularly relevant for patients with complex medical and psychiatric comorbidities. Additional studies are needed to better guide clinical decision making.

INTRODUCTION/BACKGROUND:Biliary strictures (BS) are a significant challenge, with malignant strictures frequently diagnosed at advanced stages, limiting curative options. Digital single-operator cholangioscopy (D-SOC) enables high-resolution, direct visualization of the bile duct, yet with suboptimal accuracy. Artificial intelligence (AI) has shown promise for detection and differentiation of BS in frame-level analysis and small clinical series. This study aimed to validate a deep learning model for AI-assisted D-SOC image analysis. METHODS:This multicenter study included 135 D-SOC exams from 129 patients (61 with malignant BS) across 14 centers in the United States, Brazil, Spain, Colombia, Australia, and Saudi Arabia. For each exam, up to 25 clinically relevant frames were selected and uploaded to a web-based platform for AI analysis. The model performed both detection and differentiation of BS: detection was assessed by comparing AI-generated bounding boxes with expert-defined annotations using intersection-over-union (IoU), while differentiation was benchmarked against histopathology. Performance metrics included accuracy, sensitivity, specificity, and positive and negative predictive values (PPV and NPV). RESULTS:At the patient level, malignant BS were identified with 86.0% accuracy, 84.1% sensitivity and 85.7% specificity, with an AUC of 0.904. The model demonstrated robust detection performance, achieving a mean IoU of 70.3%. Performance was maintained across demographic variables and centers. DISCUSSION/CONCLUSIONS:This first multicentric validation study demonstrates real-world performance of AI-assisted D-SOC analysis across multiples continents and devices, with robust accuracy for BS detection and differentiation. These findings support AI as an adjunctive tool in D-SOC, enhancing a more accurate evaluation of patients with indeterminate BS.

BACKGROUND:2-octyl cyanoacrylate (2-OCA) topical skin adhesives are widely used for surgical wound closure but are increasingly associated with allergic contact dermatitis (ACD). We conducted a systematic review and meta-analysis to define the incidence, clinical features, and risk factors for 2-OCA-associated ACD. STUDY DESIGN/METHODS:A PRISMA systematic review of PubMed, Embase, and Web of Science (2008-2025) identified studies reporting cutaneous hypersensitivity to 2-OCA in human wound closure. Randomized, observational, and case-based reports were included. Risk of bias was assessed using ROBINS-I and RoB 2. Incidence from analytic cohorts was pooled using a random-effects model with prespecified subgroup analyses by surgical specialty. RESULTS:Seventy-four studies comprising 26,330 exposed patients were included; 20 analytic cohorts (25,442 patients) contributed to meta-analysis. The pooled ACD incidence was 4% (95% CI 3-5%) with substantial heterogeneity (I²=94.5%; prediction interval 0-12%). Incidence was 4% in orthopedic cohorts and 8% in plastic surgery cohorts, with lower rates in dermatology and obstetrics/gynecology (p=0.015 for subgroup differences). Re-exposure markedly increased risk, with reaction rates rising from 1-3% after initial exposure to >20% in staged or repeat procedures in several cohorts. Prior adhesive/contact allergy and cosmetic acrylate exposure were also strong risk factors. Diagnosis was primarily clinical, with selective patch testing. Management typically involved adhesive removal and topical corticosteroids; systemic therapy was reserved for severe cases. CONCLUSIONS:ACD to 2-OCA is a clinically meaningful and likely under-recognized complication of surgical wound closure. Re-exposure is strongly associated with increased postoperative reaction rates, supporting preoperative risk assessment and caution in repeat adhesive use.

OBJECTIVE:We discuss implications of potential ascertainment biases for studies examining diabetes risk following SARS-CoV-2 infection using electronic health records (EHRs). We quantitatively explore sensitivity of results to misclassification of COVID-19 status using data from the U.S.-based Diabetes in Children, Adolescents and Young Adults (DiCAYA) Network on children (≤17 years) and young adults (18-44 years). MATERIALS AND METHODS/METHODS:In our retrospective case study from the DiCAYA Network, SARS-CoV-2 was identified using labs and diagnoses from June 1, 2020 to December 31, 2021. Patients were followed through December 31, 2022 for new diabetes diagnoses. Sites examined incident diabetes by COVID-19 status using Cox proportional hazards models. Results were pooled in meta-analyses. A bias analysis examined potential impact of COVID-19 misclassification scenarios on results, guided by hypotheses that sensitivity would be <50% and would be higher among those who developed diabetes. RESULTS:Prevalence of documented COVID-19 was low overall and variable across sites (children: 4.4%-7.7%, young adults: 6.2%-22.7%). Individuals with documented COVID-19 were at higher risk of incident diabetes compared to those with no documented infection, but results were heterogeneous across sites. Findings were highly sensitive to COVID-19 misclassification assumptions. Observed results could be biased away from the null under several differential misclassification scenarios. DISCUSSION/CONCLUSIONS:Although EHR-based documentation of COVID-19 was associated with incident diabetes, COVID-19 phenotypes likely had low sensitivity, with considerable variation across sites. Misclassification assumptions strongly impacted interpretation of results. CONCLUSION/CONCLUSIONS:Given the potential for low phenotype sensitivity and misclassification, caution is warranted when interpreting analyses of COVID-19 and incident diabetes using clinical or administrative databases.