Abstract
Introduction: Endoscopic mucosal resection (EMR) is considered an effective endoscopic treatment of sessile polyps and non-polypoid colorectal neoplasia. A limitation of this technique is the risk of incomplete endoscopic resection, which can lead to local residual neoplasia development.
Aim: Identification of the risk factors associated with local residual neoplasia (LRN) onset. Methods: Retrospective analysis was performed on colorectal neoplasia EMRs in one high-volume tertiary-referral endoscopic center in 2013-2015.
Individuals with at least one follow-up colonoscopy after the initial EMR were included. LRN was defined as the histopathological presence of neoplastic tissue at the post-EMR site.
Univariate and multivariate analysis of factors associated with LRN were performed. Results: 280 EMRs of sessile polyps and non-polypoid colorectal neoplasia (size >= 10 mm) including laterally spreading tumors (LST) were analyzed and surveillance endoscopy was carried out on 186 lesions (66.4% of all EMRs) in 163 patients (66.3% male; mean age 67 years).
The mean follow-up interval was 7.8 months. LRN was verified in 33 lesions (17.7%) resected by EMR.
Single variate analysis showed evidence of an increased risk of residual neoplasia for lesions >= 20 mm (p = 0.006), LST with granular type (p = 0.002), villous component of adenomas with low grade dysplasia (p = 20 mm, villous component of adenomas, piece meal EMR technique, and LST lesions of the granular type. In these cases, earlier endoscopic post-EMR surveillance or alternative endoscopic or surgical techniques should be considered.