Background: Endoscopic resection (ER) has been accepted as standard treatment of EGC. However, data about long-term outcomes especially related with recurrence have not yet been enough and appropriate surveillance strategy is not established. We evaluated the patterns and risk factors of recurrence, through evaluating the long-term outcomes of ER for treatment of EGCs comparing conventional and expanded indication criteria. Methods: We performed a retrospective review of prospectively and consecutively collected database of 1,855 EGCs in 1,803 patients [1,409 lesions in the conventional indication (CI) and 446 lesions in expanded indication (EI)] underwent ER for EGCs from April 2000 to June 2011. Results: Local recurrence was occurred from five tumors (0.35%) in the CI group and one tumor (0.22%) in the EI group. The median local recurrence interval was 13.5 months (range 2-34 months). 48 (3.1%) patients developed metachronous EGCs during follow up period. The median duration until metachronous recurrence was occurred, was 27.3 months (range, 12.25-115.29 months). There was not significantly different between metachronous group and non-metachronouos group except sex and histologic differentiation. Extragastric recurrence of tumor occurred from two patients in each of the CI and EI group. The extragastric recurrence interval was 62 months in the CI group and 48 months in the EI group. At that time, endoscopic examination showed no evidence of local recurrence around ESD scar area. Conclusions: The expanded indication of ER for EGCs showed acceptable long-term outcomes. More than 5 years of close follow up is needed and CT should be performed together with endoscopy for surveillance of recurrence after ER for EGC in the conventional indication group as well as the expanded indication group.