Background: Current international guidelines strongly advise endoscopic submucosal dissection (ESD) instead of endoscopic mucosal resection (EMR) for the endoscopic resection of sessile colorectal tumours >20 mm.
Aims: To compare the safety and efficacy of EMR and ESD for treating large non-invasive colorectal lesions.
Material and methods: We performed a systematic review using electronic databases (MEDLINE/PubMed, EMBASE, and Cochrane Library) on February 21st, 2021 and a meta-analysis to assess en-bloc and R0 rates, and related adverse events.
Results: Twenty-four studies were included, comparing 3,424 ESD and 5,122 EMR procedures. The en-bloc resection rate was 90.8% in the ESD and 33.0% in the EMR group (p < .001). The R0 resection rate was 85.0% in the ESD and 64.6% in the EMR group (p = .005). The rate of perforation was 5.1% in the ESD and 1.67% in the EMR group (p < .001). The bleeding rate was 4.3% in the ESD and 3.6% in the EMR group (p = .008). The overall need for surgery, including oncologic reasons and complications, was 5.9% in the ESD and 3.1% in the EMR group (p < .001).
Conclusions: ESD for large non-pedunculated colorectal lesions allows a higher rate of R0 resections than EMR, at the cost of a higher perforation rate and the need for additional surgery.
Keywords: C-EMR; Colorectal cancer; ESD; U-EMR; adenoma detection.