Factors Associated with Complete Endoscopic Resection of Large Non-ampullary Duodenal Adenomas

Abstract

Introduction: Endoscopic mucosal resection for large (>20 mm) duodenal polyps (DEMR) has gained favor as an alternate treatment to surgery. Several studies have evaluated the safety and efficacy of endoscopic resection but factors predicting the success of endoscopic resection in large polyps remain unclear. We hypothesized that degree of luminal involvement and location within the duodenum determines complete resection success. Methods: Using an electronic endoscopic database, patients with large (> 20 mm) non-ampullary duodenal adenomas referred for endoscopic therapy from 2006 to 2016 were identified. The primary outcome was complete endoscopic resection at the time of the index endoscopy, as defined visually by the endoscopist. Univariate and multivariate analyses of clinical and endoscopic factors determining complete resection were performed. Results: 55 patients with 55 polyps were included. The mean adenoma size was 31.8 ± 10.6 mm. 8 (14.5%) had undergone partial resection attempts prior to their referral. Complete endoscopic resection was reported in 38 of 55 (67%) patients at the time of their index procedure; of which 81.6% achieved biopsy proven successful resection during a mean follow-up of 460 days (range 84-3710). Of the remaining 17 patients with incomplete endoscopic resection, only 59% achieved biopsy-proven successful resection after a mean of 3 (range 2 - 11) resection attempts. On univariate analysis, there was no difference in gender, history of FAP, or complications between complete and incomplete resections. Age, polyp size, location in D3, 2/3 lumen involvement, prior resection attempt, and reported difficulty lifting showed statistical significance (see Table 1). On multivariate regression, size in quartiles (OR 0.28 95% CI 0.116, 0.692) and lack of prior resection attempt (OR 0.065 95% CI 0.008, 0.493) were associated with successful resection. Based on our regression model, the probability of successful resection ranges from 52% to 98% as the polyp size reduces from >40mm toTable 1: Baseline characteristics of non-ampullary duodenal adenoma patients achieving complete versus incomplete endoscopic resection after the first attempt at endoscopic mucosal resection (EMR).Figure 1Conclusion: The success of DEMR for large non-ampullary adenomas is significantly impacted by the size of the lesion and history of prior partial resection. Complete resection at index endoscopy should be prioritized to maximize the probability of biopsy-proven eradication.