Prophylactic clip closure after endoscopic mucosal resection of large non-pedunculated polyps in the right colon — a systematic review and meta-analysis

Clinically significant post-EMR bleeding (CSPEB) is the most common serious adverse event after endoscopic mucosal resection (EMR) of large (≥ 20 mm) non-pedunculated colorectal polyps (LNPCPs) and is concentrated in the right colon. Randomized trials and an individual-patient-data (IPD) meta-analysis performed largely at tertiary referral centres reported that prophylactic clip closure reduces CSPEB, but the 2025 pragmatic, multicentre CLIPPER trial found no benefit in routine practice. We synthesized randomized evidence restricted to right-colon LNPCPs to determine whether a protective effect persists once pragmatic data are incorporated. We searched MEDLINE, Embase, Cochrane CENTRAL, ClinicalTrials.gov, and the WHO ICTRP from inception to 15 March 2026, without language restriction, for randomized controlled trials comparing prophylactic clip closure with no closure after EMR of right-colon LNPCPs ≥ 20 mm and reporting CSPEB. All trials were analysed as randomized (intention-to-treat). A random-effects model (Mantel–Haenszel pooling, REML τ² estimation, Hartung–Knapp–Sidik–Jonkman adjustment) was the pre-specified primary analysis, with risk ratio (RR) as the primary effect measure. Reporting followed PRISMA 2020; risk of bias was assessed with RoB 2 and certainty of evidence with GRADE. Four RCTs were included (1,334 patients in the primary right-colon analysis; 1,741 in the whole-cohort safety analyses). Under the pre-specified random-effects model, prophylactic clipping did not significantly reduce right-colon CSPEB (RR 0.53, 95% CI 0.17–1.66; prediction interval 0.06–5.07). A common-effect model applied to the same data yielded RR 0.54 (95% CI 0.36–0.80). Heterogeneity was substantial (I²=68.0%, τ²=0.3631, p = 0.025) and was attributable predominantly to CLIPPER (Baujat plot). Clipping had no significant effect on perforation (RR 0.61, 95% CI 0.22–1.68; I²=0%) or on post-polypectomy syndrome/abdominal pain (RR 1.52, 95% CI 0.40–5.76; I²=19.4%). Certainty of evidence for the primary outcome was low. Once the pragmatic CLIPPER trial is included, the randomized evidence for prophylactic clipping in the right colon is uncertain rather than clearly protective, with no offsetting safety signal in either direction. Because all trials were analysed as randomized, these estimates describe a strategy of prophylactic clipping and not the efficacy of successful complete defect closure, which may still be protective. Whether selective, risk-targeted clipping is preferable to universal clipping is a hypothesis requiring direct evaluation, ideally through an IPD meta-analysis incorporating CLIPPER. PROSPERO CRD420261432990. Reported in accordance with the PRISMA 2020 statement.