Background: We decide whether to perform endoscopic submucosal dissection (ESD) or gastrectomy for early gastric cancer (EGC) by endoscopic examination. We cannot estimate the correct tumor size with naked eye during upper endoscopic examination. And endoscopic size of gastric tumor is different from histologic size of it. The goal of this prospective study is to compare macroscopic and pathologic diameters of gastric tumors resected by ESD. Methods: We prospectively enrolled gastric adenomas and adenocarcinomas and performed ESD for the tumors from July 2013 to November 2013. The tumor diameter was measured macroscopically by the same investigator with a hand ruler just after removal. Thereafter, it was measured pathologically by the same pathologist on mapping charts after processing specimen. Results: We included 50 gastric tumors (gastric adenomas were 39 and gastric adenocarcinomas were 11). In all tumors, mean of the macroscopic diameter (20.44 ± 8.28 mm) was signifi cantly larger than mean of the pathologic diameter (14.92 ± 8.33 mm, p < 0.001). But, in adenocarcinoma group, there was not signifi cant difference between mean of two diameters (19.80 ± 6.41 mm, 16.80 ± 5.73 mm, p > 0.05). The lesions of adenocarcinoma in adenoma were signifi cant difference (n=5, 19.80 ± 6.41 mm, 16.80 ± 5.73 mm, p < 0.05). Conclusions: In adenocarcinoma, tumor size was not decreased significantly after formalin fixation and slicing. But, the lesions of adenocarcinoma in adenoma was decreased signifi cantly. This study showed that ESD may be performed for adenocarcinoma in adenoma that crosses absolute indication of ESD for tumor size.