The effectiveness and safety of endoscopic submucosal dissection compared with endoscopic mucosal resection for early gastric cancer: a systematic review and metaanalysis

Endoscopic submucosal dissection (ESD) allows en bloc resection of the entire lesion, permitting a higher curative resection rate and increased quality of life by minimizing the resection size compared with that of endoscopic mucosal resection (EMR). Although ESD has been implemented at most university hospitals in Korea, potential complications of ESD such as bleeding and perforation raise doubts in the therapeutic decision on use of the ESD procedure for early gastric cancer patients and in reimbursement decision making. This systematic review aimed to address both the effectiveness and safety outcomes of ESD versus EMR for early gastric cancer. MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL), and Koreamed were searched using the primary keywords stomach neoplasm AND endoscopic submucosal dissection AND endoscopic mucosal resection. To assess the quality of selected studies, the methodologic approach of the Scottish Intercollegiate Guidelines Network was used. Five effectiveness-relevant and three safety-relevant outcome measures were extracted. Bibliography management and metaanalysis for each outcome were conducted using Review Manager 5.0. Three nonconcurrent cohort studies and nine retrospective cohort studies were identified. Metaanalyses showed ESD to be significantly more effective than EMR for en bloc resection (odds ratio [OR], 8.43; 95% confidence interval [CI], 5.20-13.67), complete resection (OR, 14.11; 95% CI, 10.85-18.35), curative resection (OR, 3.28; 95% CI, 1.95-5.54), and local recurrence (risk ratio [RR], 0.13; 95% CI, 0.04-0.41). Whereas intraoperative bleeding (RR, 2.16; 95% CI, 1.14-4.09), perforation risk (RR, 3.58; 95% CI, 1.95-6.55), and operation time (standard mean difference [SMD], 1.55; 95% CI, 0.74-2.37) were significantly greater for ESD, overall bleeding risk (RR, 1.22; 95% CI, 0.76-1.98) and all-cause mortality (RR, 0.65; 95% CI, 0.08-5.38) did not differ significantly between ESD and EMR. Considering that bleeding risk did not differ significantly between ESD and EMR and that perforation risk usually does not lead to life-threatening disease, the effectiveness benefit of ESD can outweigh the overall harm compared with EMR on the condition that ESD is performed by experienced practitioners.