Endoscopy vs surgery in the treatment of early gastric cancer: Systematic review.

Aim: To report a systematic review, establishing the available data to an unpublished 2a strength of evidence, better handling clinical practice.
Methods: A systematic review was performed using MEDLINE, EMBASE, Cochrane, LILACS, Scopus and CINAHL databases. Information of the selected studies was extracted on characteristics of trial participants, inclusion and exclusion criteria, interventions (mainly, mucosal resection and submucosal dissection vs surgical approach) and outcomes (adverse events, different survival rates, mortality, recurrence and complete resection rates). To ascertain the validity of eligible studies, the risk of bias was measured using the Newcastle-Ottawa Quality Assessment Scale. The analysis of the absolute risk of the outcomes was performed using the software RevMan, by computing risk differences (RD) of dichotomous variables. Data on RD and 95%CIs for each outcome were calculated using the Mantel-Haenszel test and inconsistency was qualified and reported in χ(2) and the Higgins method (I (2)). Sensitivity analysis was performed when heterogeneity was higher than 50%, a subsequent assay was done and other findings were compiled.
Results: Eleven retrospective cohort studies were selected. The included records involved 2654 patients with early gastric cancer that filled the absolute or expanded indications for endoscopic resection. Three-year survival data were available for six studies (n = 1197). There were no risk differences (RD) after endoscopic and surgical treatment (RD = 0.01, 95%CI: -0.02-0.05, P = 0.51). Five-year survival data (n = 2310) showed no difference between the two groups (RD = 0.01, 95%CI: -0.01-0.03, P = 0.46). Recurrence data were analized in five studies (1331 patients) and there was no difference between the approaches (RD = 0.01, 95%CI: -0.00-0.02, P = 0.09). Adverse event data were identified in eight studies (n = 2439). A significant difference was detected (RD = -0.08, 95%CI: -0.10--0.05, P < 0.05), demonstrating better results with endoscopy. Mortality data were obtained in four studies (n = 1107). There was no difference between the groups (RD = -0.01, 95%CI: -0.02-0.00, P = 0.22).
Conclusion: Three-, 5-year survival, recurrence and mortality are similar for both groups. Considering complication, endoscopy is better and, analyzing complete resection data, it is worse than surgery.