Prophylactic clipping after endoscopic mucosal resection of large nonpedunculated colorectal lesions: A meta-analysis

Abstract

Background and aim: It is not clear whether prophylactic clipping after endoscopic mucosal resection (EMR) of large nonpedunculated colorectal lesions (LNPCLs) prevents delayed bleeding (DB). We aimed to conduct a meta-analysis to clarify the efficacy of prophylactic clipping in prevention of DB following EMR of LNPCLs.

Methods: We searched PubMed, EMBASE, Web of Science, ScienceDirect, Cochrane Library databases, and ClinicalTrials.gov for studies that compared clipping versus (vs) nonclipping in prevention of DB following EMR of LNPCLs. Pooled odds ratio (OR) was determined using a random effects model. The pooled ORs of DB, perforation, and post-polypectomy syndrome in the clipping group compared with the nonclipping group comprised the outcomes. Subgroup analyses based on study design, polyp location, and completeness of wound closure were performed.

Results: Five studies with a total of 3112 LNPCLs were extracted. Prophylactic clipping reduced the risk of DB compared with nonclipping (3.3% vs 6.2%, OR: 0.494, P = 0.002) following EMR of LNPCLs. In subgroup analysis, prophylactic clipping reduced DB of LNPCLs at proximal location (3.8% vs 9.8%, P = 0.029), but not of them at distal location (P = 0.830). Complete wound closure showed superior efficacy to prevent DB compared with partial closure (2.0% vs 5.4%, P = 0.004). No benefit of clipping for preventing perforation or post-polypectomy syndrome was observed (P = 0.301 and 0.988, respectively).

Conclusions: Prophylactic clipping can reduce DB following EMR of LNPCLs at proximal location. Besides, complete wound closure showed superior efficacy to prevent DB compared with partial closure. Further cost analyses should be conducted to implement the most cost-effective strategies.

Keywords: clip; colonoscopy; endoscopic mucosal resection; hemorrhage; meta-analysis.