Objective:To evaluate the efficacy and safety of submucosal tunneling endoscopic resection (STER) and endoscopic submucosal excavation (ESE) for upper gastroenterology submucosal tumors (SMT) originating from the muscularis propria (MP) layer. Methods: Clinicopathological and endoscopic data of 42 cases with upper gastroenterology tumors originating from the MP layer who were treated with STER (n=28) or ESE (n=14) between April 2013 and December 2016 in Peking University Third Hospital were retrospectively analyzed. The treatment and complications of the two groups were compared. Results: In the study, 42 cases were all resected by therapeutic endoscopy successfully.There was no significant difference (STER vs. ESE) in gender, age, mean tumor size [1.5 (1.0-6.0) cm vs. 1.3 (0.5-2.0) cm,P=0.056]. STER was superior to ESE with reduced sutured time [3.5 (1.0-11.0) min vs. 8.0 (2.0-33.0) min, P=0.006], but more resection time [46.5 (11.0-163.0) min vs.19.5 (6.0-56.0) min, P=0.007]. There was statistical difference between the two groups in resection time or sutured time, but no significant difference (STER vs. ESE) in total operative time [52.0 (14.0-167.0) min vs. 31.5 (10.0-88.0) min, P=0.080]. En bloc resection rates (92.9% vs. 85.7%), hospital stay duration and complications (10.7 vs. 0.0) were similar in the STER and ESE groups. One case developed mediastinal emphysema and 2 pneumonia after operation in STER group, and all of them recovered uneventfully after conservative treatments; There were no complications in the ESE group. After operation, 28 cases of leiomyoma and 14 cases of stromal tumor were diagnosed by routine pathological and immunohistochemical staining. Among them, 6 cases of stromal tumors in group STER were all extremely low risk, 4 cases of stromal tumors in group ESE were extremely low risk, 4 cases of stromal tumors in group ESE were medium risk (the size of the lesion was about 1.0-2.0 cm, and mitotic figures counted (6-8)/50 high power field). The median follow-up time of all the patients was 46.5 (24-60) months, and the shortest follow-up time for medium risk stromal tumors was 32 months. No residual tumor, recurrence and implantation in the tunnel were observed. Conclusion: STER or ESE can be used as an effective and safe option for treatment of submucosal tumors originating from the muscularis propria of the upper digestive tract. Compared with STER, ESE had shorter resection time but longer wound closure time. There was no significant difference in total operation time.