Background and AimEndoscopic resection (ER) is the preferred approach to treat early gastric cancer (EGC) in patients without suspected lymph node involvement and that meet the criteria for ER. Surgery is a more aggressive treatment, but it may be associated with less recurrence and the need for reintervention. Previous meta-analyses comparing ER with surgery for EGC did not incorporate the most recent studies, making accurate conclusions not possible.MethodsThis systematic review and meta-analysis aimed to examine complete resection, length of hospital stay (LOHS), adverse events (AEs), serious AEs, recurrence, 5-year overall survival (OS), and 5-year cancer-specific survival (CSS) in patients with EGC.ResultsA total of 29 cohorts studies involving 20559 patients were included. The ER (n = 7709) group was associated with a lower incidence of AEs (RD = -0.07, 95%CI = -0.1, -0.04, p < 0.0001) and shorter LOHS (95% CI -5.89, -5.32; p < 0,00001) compared to surgery (n = 12850). However, ER was associated with lower complete resection rates (RD = -0.1, 95%CI = -0.15, -0.06; p < 0.00001) and higher rates of recurrence (RD = 0.07, 95%CI = 0.06; p < 0.00001). There were no significant differences between surgery and ER in 5-year OS (RD = -0.01, 95%CI = -0.04, 0.02; p = 0.38), 5-year CSS (RD = 0.01, 95%CI = 0.00, 0.02; p < 0.17), and incidence of serious AEs (RD = -0.03, 95%CI = -0.08, 0.01; p = 0.13).ConclusionsER and surgery are safe and effective treatments for EGC. ER provides lower rates of AEs and shorter LOHS compared to surgery. Although ER is associated with lower complete resection rates and a higher risk of recurrence, the OS and CSS were similar between both approaches.Systematic Review Registrationhttps://www.crd.york.ac.uk/PROSPERO/, identifier CRD42021255328.