Conventional versus hybrid knife endoscopic submucosal dissection in large colorectal laterally spreading tumors: A propensity score analysis

BACKGROUND: Colorectal endoscopic submucosal dissection (CR-ESD) has become a promising treatment for laterally spreading tumors (LSTs), but is accompanied by great challenges. This study aimed to evaluate the efficacy and safety of CR-ESD with a hybrid knife, versus the conventional technique for LSTs ≥30 mm in diameter, and analyze the risk factors for piecemeal resection and perforation. METHODS: Patients eligible for CR-ESD were divided into two groups according to the use of the hybrid knife (HK group) or the use of the conventional technique, with an interchange of injection and hook knife (C-group). We performed propensity score matching (PSM) to compare the HK group and the C-group. Risk predictors for perforation and piecemeal resection were identified. RESULTS: PSM identified 61 (132 patients) and 61 (129 patients) patients in the C-group and the HK group, respectively. Resection speed was significantly faster in the HK group than in the C-group (18.86 vs. 13.33 mm(2)/min, P < 0.001). The rate of knife exchange was significantly lower in the HK group than in the C-group (1.6% vs. 49.2%, P < 0.001). Multivariate analysis revealed that unfavorable locations, including the splenic flexure, hepatic flexure, or cecum, were predictive of piecemeal resection. The presence of severe fibrosis and a semilunar fold were independent risk factors for perforation. CONCLUSIONS: The use of a hybrid knife appears to increase CR-ESD resection speed. The indicators for piecemeal resection or perforation in CR-ESD identified herein might help to assess the technical difficulties of CR-ESD.