S2173 Underwater EMR vs Conventional EMR for Superficial Nonampullary Duodenal Epithelial Tumors (SNADETs): A Systematic Review and Meta-Analysis

Introduction: Superficial nonampullary duodenal epithelial tumors (SNADETs), most frequently encountered in the second portion of the duodenum just distal to the major papilla, are increasingly identified with HD endoscopy are technically challenging to resect. Limited scope maneuverability, thin duodenal wall, retroperitoneal location, occasional non-lifting lesions due to submucosal fibrosis make conventional endoscopic mucosal resection (CEMR) suboptimal in many cases. Underwater endoscopic mucosal resection (UEMR) eliminates the need for submucosal injection. By filling the lumen with water, the mucosa and lesion float away from the muscularis propria, facilitating safer and more effective en bloc snaring. Methods: A systematic search of PubMed, EMBASE, and the Cochrane Library was performed for studies published through May 2025. Both comparative and single-arm studies were included. Primary outcomes were en bloc resection, R0 resection, and recurrence. Secondary outcomes included immediate perforation, delayed bleeding, and procedure time. A random-effects model was used to calculate pooled proportions, risk ratios (RR), and mean differences (MD), with 95% confidence intervals (CI). Heterogeneity was assessed using the I² statistic. Funnel plots were examined to assess publication bias. Results: Eighteen studies - 7 comparative and 11 single-arm observational studies, with a total of 1,357 lesions resected using UEMR and 539 lesions using CEMR were included. The pooled en bloc resection rate was 83% (95% CI: 69-93%, I² = 93%), with no significant difference compared to CEMR (RR: 1.00; 95% CI: 0.89-1.13, I² = 75%). The R0 resection rate was 65% (95% CI: 58-71%, I² = 73%), similar to CEMR (RR: 0.97; 95% CI: 0.81-1.15, I² = 66%). Recurrence after UEMR was 3.3% (95% CI: 0.7-7.4%, I² = 69%), with no significant difference from CEMR (RR: 0.88; 95% CI: 0.52-1.50, I² = 0%). Immediate perforation (0%) and delayed bleeding (1.3%; 95% CI: 0.6-2.2%) rates were low and comparable between groups. Procedure time was significantly shorter with UEMR (MD: -2.99 minutes; 95% CI: -5.71 to -0.28; P = 0.03). Funnel plots showed no substantial asymmetry, suggesting low publication bias. Conclusion: UEMR safe and effective for the resection of SNADETs. It offers comparable resection and recurrence outcomes with very low complication rates and a significant reduction in procedure time. However, the limited number of comparative studies and heterogeneity in study designs warrant cautious interpretation of these findings.