S2253 Safety and Efficacy of Underwater Endoscopic Mucosal Resection for Superficial Non-Ampullary Duodenal Epithelial Tumors: An Updated Systematic Review and Meta-Analysis

Introduction: Superficial non-ampullary duodenal epithelial tumors (SNADET) are increasingly found during upper endoscopy. Underwater endoscopic mucosal resection (UEMR) allows resection without submucosal injection by leveraging the buoyancy of water to separate mucosal layers, potentially improving resection margins and minimizing complications. As further evidence emerges on the utility of UEMR in SNADETs, we aimed to conduct a systematic review and meta-analysis of its efficacy and safety for resection of SNADETs. Methods: A comprehensive search of MEDLINE, EMBASE, and Scopus was performed through April 2025 to identify studies reporting outcomes of UEMR for SNADETs. Both observational studies and randomized controlled trials were included. Outcomes assessed included en bloc resection, R0 resection, recurrence, delayed bleeding, perforation, and procedure time. Pooled effect estimates were calculated using a random-effects model and expressed as proportions with 95% confidence intervals (CI). Heterogeneity was assessed using the I² statistic. Meta-analyses were performed using the Review Manager (RevMan) software. Results: A total of 15 studies including 1608 lesions resected by UEMR were included in the analysis. Males comprised 69.9% of the population. Thirteen studies were retrospective, one was prospective, and one was a randomized controlled trial. Most of the lesions (72.3%) had elevated macroscopic morphology. The mean/median size of the lesions reported by individual studies ranged from 6 to 17 mm. The pooled en bloc resection rate was 85.4% (95% CI: 79.6–91.2; I² = 91.58%), and the R0 resection rate was 66.3% (95% CI: 60.6–72.1; I² = 74.62%). The delayed bleeding rate was 1.7% (95% CI: 0.1–2.3; I² = 0%), and the perforation rate was 0.6% (95% CI: 0.2–1.0; I² = 0%). The recurrence rate was 2.4% (95% CI: 1.1–3.8; I² = 0%), with minimal heterogeneity across studies. The pooled mean procedure time across studies was 6.3 minutes (95% CI: 5.1-7.5; I² = 95.3%). Conclusion: UEMR is a highly effective and safe modality for the management of SNADETs with high en bloc and R0 resection rates and low rates of recurrence and complications. These findings support UEMR as a minimally invasive alternative to surgery for these lesions. Further prospective studies with long-term follow-up are warranted to validate these outcomes and assess survival impact.