Endoscopic submucosal dissection for gastric gastrointestinal stromal tumors: a retrospective cohort study

Background and aim Endoscopic submucosal dissection (ESD) has been used to remove submucosal tumors. However, with regard to the potential malignant behavior of gastrointestinal stromal tumors (GISTs), whether ESD can be recommended for treatment is still controversial. Therefore, we evaluated the efficacy and safety of ESD for removal of GISTs in the muscularis propria (MP) layer and to assess the possible risk factors for a gastric-wall defect (GWD). Methods For 168 GISTs located in the MP layer from 168 consecutive patients, the baseline information, complications, and therapeutic outcomes were recorded. Subsequently, risk factors for a GWD were analyzed. Results Most GISTs (106/168) were located in the fundus of the stomach. Tumor shapes in 154 patients (91.7%) were regular, and the median size of the tumor was 1.5 (range 0.5–6.0) cm. The en bloc resection rate was 100% and the median procedure time was 46.5 (33–181) min. A GWD was observed in 71 patients (42.3%) and delayed bleeding occurred in 2 patients (1.2%), and they were treated by clips. A total of 117 patients with a GIST were at very low risk, 37 patients were at low risk, and 14 patients were at mild risk. No local recurrences or distant metastases were observed during a median follow-up of 25 (6–67) months. Univariate and multivariate logistic regression analyses identified the tumor type to be an independent risk factor for a GWD during ESD (odds ratio 29.82, 95% confidence interval 10.87–81.80, P < 0.001). Conclusion ESD is a safe and feasible method for gastric GISTs, especially for tumor types A and B. Endoscopic ultrasound can aid evaluation of the tumor type before ESD, which is an independent risk factor correlated with a GWD upon ESD.