Predictors of technical difficulty for complete closure of mucosal defects after duodenal endoscopic resection

Background and Aims: It has been reported that the prophylactic closure of mucosal defects after duodenal endoscopic resection (ER) can reduce delayed adverse events; however, under certain circumstances, this can be technically challenging. Therefore, the aim of this study was to determine the predictors of difficulty during the complete closure of mucosal defects after duodenal ER. Methods: This was a retrospective study of duodenal lesions that underwent ER between July 2010 and May 2020. We reviewed the endoscopic images and analyzed the relationships between the degree of closure or closure time and clinical features of the lesions using univariate and multivariate analyses. Results: We analyzed 698 lesions. The multivariate analysis revealed that lesion location in the medial or anterior wall (odds ratio, 2.8; 95% confidence interval, 1.36-5.85; P < .01) and a large lesion size (odds ratio, 1.4; 95% confidence interval, 1.07-1.89; P = .03) were independent predictors of an increased risk of incomplete closure. Moreover, a large lesion size (β coefficient, .304; P < .01), an occupied circumference over 50% (β coefficient, .178; P < .01), intraoperative perforation (β coefficient, .175; P < .01), treatment period (β coefficient, .143; P < .01), and treatment with endoscopic submucosal dissection (β coefficient, .125; P < .01) were independently and positively correlated with a prolonged closure time in the multiple regression analysis. Conclusions: This study revealed that lesion location in the medial or anterior wall and lesion size affected the incomplete closure of mucosal defects after duodenal ER, and lesion size, occupied circumference, intraoperative perforation, treatment period, and treatment method affected closure time.