Diagnostic yields of endoscopic ultrasound-guided fine-needle tissue acquisition according to the gastric location.

The diagnostic accuracy of endoscopic ultrasound-guided fine-needle tissue acquisition (EUS-FNTA) according to the gastric location of subepithelial tumors (SETs) has not been well established. We aimed to evaluate the efficacy of EUS-FNTA for the diagnosis of gastric SETs according to tumor location.Thirty-three patients diagnosed with gastric SETs via EUS-FNTA from January 2016 to May 2018 were analyzed retrospectively. Patient demographics, diagnostic yields, and complications were evaluated.Nineteen patients (57.6%) were female, with a mean age of 57.7 years. Endoscopic ultrasound revealed a mean longitudinal diameter of 25.6 mm. The most common location of SETs was in the gastric body (n = 18, 54.5%), followed by cardia and fundus (n = 10, 30.3%), and antrum (n = 5, 15.2%). A 20-gauge biopsy needle was most frequently used (90.9%). The diagnostic yield was obtained in 23 patients (69.7%). The most common diagnosis was gastrointestinal stromal tumor (73.9%), followed by leiomyoma (17.4%). The diagnostic yield of SETs in gastric antrum (0/5, 0%) was significantly lower than that in the gastric body and cardia (23/28, 82.1%, P = .001). A case of immediate bleeding after EUS-FNTA occurred in 1 patient (3.0%) who recovered uneventfully. According to related literature, the overall diagnostic yield of SETs in gastric antrum was significantly lower than that in the gastric body, fundus, and cardia (29.7% vs 71.4%, P < .001, n = 191).EUS-FNTA is ineffective in the diagnosis of SETs in the gastric antrum. Although EUS-FNTA is an advanced diagnostic tool for gastric SETs, it is essential to develop more effective methods for the diagnosis of antral SETs.