Clinicopathological factors associated with clinical outcomes of endoscopic submucosal dissection for colorectal epithelial neoplasms

Abstract

Background and study aims: Endoscopic submucosal dissection (ESD) permits removal of colorectal epithelial neoplasms en bloc, but a substantial risk of procedure-related perforation has been reported. We sought to unravel the clinicopathological factors associated with the clinical outcomes of ESD for colorectal epithelial neoplasms in a large series. Patients and methods: ESD was done in 278 patients with 292 colorectal tumors that fulfilled the inclusion criteria. The criteria for ESD were: lesion greater than 20 mm in size, lesion with fibrotic scarring, locally residual colorectal lesion, or invasive carcinoma with slight submucosal penetration. Resection was assessed as en bloc or piecemeal, complete (en bloc with tumor-free lateral and basal margins) or incomplete. Complications including perforation and bleeding were assessed, and factors related to each were analyzed using logistic regression. Patients underwent multiple follow-up endoscopic examinations (mean 4.6; median 4; range 2 – 9; total number 1010). Results: En bloc resection was achieved in 90.1 % of lesions (263/292) and resection was deemed to be complete in 233 (79.8 %). Right-side colonic location and the finding of fibrosis were the significant contributors to incomplete resection. Perforation was seen in 24 cases (8.2 %), and was associated with large tumor size and the presence of fibrosis. When the contributive factors for each were combined, the risks of incomplete resection and perforation were substantially increased. Conclusion: The present study provides useful information for predicting risks for incomplete resection and complication in colorectal ESD.