Comparison of endoscopic resection and gastrectomy for the treatment of early gastric cancer: a meta-analysis

Background Endoscopic resection methods, including endoscopic mucosal resection and endoscopic submucosal dissection, have become standard treatment modalities for patients with early gastric cancer (EGC) and absolute indications, with en bloc resection being more frequent with the latter. Endoscopic resection, however, has been associated with higher recurrence and metachronous cancer rates than gastrectomy. This meta-analysis compared the efficacy and safety of endoscopic resection and gastrectomy for EGC. Methods PubMed, EMBASE and Web of Science were electronically searched for relevant studies comparing endoscopic resection and gastrectomy for EGC from 1976 through March 2015. The primary endpoints were en bloc resection and histologically complete resection rates. The secondary endpoints were duration of hospital stay and rates of complications, recurrence, metachronous cancer and overall survival. Results This meta-analysis enrolled 10 studies with 2070 patients: 993 patients who underwent endoscopic resection and 1077 who underwent gastrectomy. Endoscopic resection was associated with shorter hospital stay (standardized mean difference −2.02; 95 % confidence interval [CI] −2.64 to −1.39) and lower complication rate (relative risk [RR] 0.41; 95 % CI 0.22–0.76) than gastrectomy. However, endoscopic resection was associated with lower rates of en bloc resection (odds ratio [OR] 0.05; 95 % CI 0.02–0.16) and histologically complete resection (OR 0.04; 95 % CI 0.01–0.11) and higher rates of recurrence (RR 5.23; 95 % CI 2.43–11.27) and metachronous cancer (RR 5.22; 95 % CI 2.40–11.34) than gastrectomy. Overall survival rate (OR 1.18; 95 % CI 0.76–1.82) was similar. Conclusions Endoscopic resection is minimally invasive and as effective as surgery, suggesting that the former be considered standard treatment for EGC. It should be recommended as standard treatment for EGC with indications. Additional randomized controlled trials from more countries are required.