Endoscopic versus surgical resection of GI stromal tumors in the upper GI tract

Background and Aims: Endoscopic resection has been performed for treatment of GI stromal tumors (GISTs) in the upper GI tract. However, the therapeutic roles of the endoscopic procedure remain debatable. We aimed in this retrospective study to evaluate the feasibility and long-term follow-up results of endoscopic resection of GISTs in the upper GI tract, compared with surgery. Methods: Between March 2005 and August 2014, 130 cases of GIST in the upper GI tract were resected. We compared baseline characteristics and clinical outcomes including R0 resection rate and recurrence rate between the endoscopy group (n = 90) and surgery group (n = 40). Results: The most common location of GIST was the stomach body in the endoscopy group, whereas it was the duodenum in the surgery group (P = .001). Tumor size was significantly smaller (2.3 vs 5.1 cm; P < .001), and procedure time (51.8 +/- 36.2 vs 124.6 +/- 74.7 minutes; P < .001) and hospital stay (3.3 +/- 2.4 vs 8.3 +/- 5.4 days; P < .001) were significantly shorter in the endoscopy group than in the surgery group. The R0 resection rate was 25.6% in the endoscopy group, whereas it was 85.0% in the surgery group (P = .001), and 50.0% of resected tumors belonged to a very low-risk group in the endoscopy group, whereas 35.0% and 30.0% belonged to low-risk and high-risk in the surgery group (P = .001). However, during 45.5 months of follow-up, the recurrence rate was not significantly different between the 2 groups (2.2% vs 5.0%; P = .586). Conclusions: Endoscopic resection might be an alternative therapeutic modality for GISTs in the upper GI tract in selective cases.