Abstract Objective To evaluate the efficacy and safety of cold snare polypectomy (CSP) vs cold snare endoscopic mucosal resection (CS-EMR) for colorectal polyps measuring 6–15 mm Paris IIa/Is/Isp+Ip(P) by using full-thickness serial sectioning. Methods This single-blind, randomized-controlled trial enrolled 214 consecutive patients with solitary colorectal polyps undergoing polypectomy at the Digestive Endoscopy Center of Xiamen Humanity Hospital between 3 September and 1 December 2024. Patients were randomized to undergo either CSP or CS-EMR. We assessed en-bloc resection and endoscopic complete resection to exclude cases converted to hot snare polypectomy (HSP) or conventional EMR. The primary outcome was the pathologic complete resection rate. The secondary outcomes included the muscularis mucosae coverage rate, submucosal coverage rate, residual lesions coverage rate at the basal margin, intraprocedural bleeding rate, delayed bleeding rate, and perforation rate. Results All 214 cases achieved en-bloc resection and endoscopic complete resection without complications; 8 required HSP conversion and 7 required EMR conversion. Per-protocol analysis showed comparable pathologic complete resection rates between the CSP and CS-EMR groups (91% vs 89%, P = 0.654). No significant differences were found between the CSP and CS-EMR groups in the pathologic complete resection rate (100% vs 100%), submucosal coverage rate (0% vs 0%), and residual lesions coverage rate (0% vs 0%) (all P 0.05). CSP demonstrated a significantly lower intraprocedural bleeding rate (25% vs 44%, P = 0.006). Intention-to-treat analysis, which included converted HSP and EMR cases, yielded consistent results. Conclusions For Paris IIa/Is/Isp+Ip(P) colorectal polyps measuring 6–15 mm, CSP and CS-EMR showed comparable efficacy and safety as cold resection techniques. However, CSP had lower intraprocedural bleeding rates. CS-EMR did not demonstrate superiority over CSP. Considering cost-effectiveness and operational efficiency, CSP is recommended as the preferred method.
Chen et al. (Tue,) studied this question.