PURPOSE: Evidences on diagnostic ability of esophagogastroduodenoscopy (EGD), CT, and endoscopic ultrasonography (EUS) for gastric submucosal tumor (SMT) are insufficient. For investigating a suitable diagnostic strategy, we retrospectively investigated diagnostic accuracy of EGD and CT for SMT, in comparison to EUS findings. METHODS: Ninety-three consecutive lesions in which gastric SMT was suspected by EGD were investigated by EUS from March 2009 to June 2011, including 55 lesions evaluated also by CT 4 months before or after EUS. Frequency of possibly malignant SMTs as a detection rate of EGD, and sensitivity and specificity of CT in detection of SMT were retrospectively analyzed. Factors influencing the ability for CT to detect SMT were also investigated. RESULTS: EUS revealed possibly malignant SMT in 60 cases (64.5%). In 32 out of 55 cases, possibly malignant SMT was suspected by CT. Sensitivity and specificity of CT was 80.6% and 84.2%, respectively. The mean size of possibly malignant SMT detected and not detected by CT was 27.4 and 11.0 mm, respectively, with a statistically significant difference (p = 0.0001). CONCLUSION: This study elucidated that approximately one-third of suspected SMTs by EGD might not need clinical care. EUS for suspected SMTs is indispensable for extracting possibly malignant SMTs from others, and CT may be insufficient as an alternative to EUS in detecting especially small ones.
PURPOSE: Evidences on diagnostic ability of esophagogastroduodenoscopy (EGD), CT, and endoscopic ultrasonography (EUS) for gastric submucosal tumor (SMT) are insufficient. For investigating a suitable diagnostic strategy, we retrospectively investigated diagnostic accuracy of EGD and CT for SMT, in comparison to EUS findings. METHODS: Ninety-three consecutive lesions in which gastric SMT was suspected by EGD were investigated by EUS from March 2009 to June 2011, including 55 lesions evaluated also by CT 4 months before or after EUS. Frequency of possibly malignant SMTs as a detection rate of EGD, and sensitivity and specificity of CT in detection of SMT were retrospectively analyzed. Factors influencing the ability for CT to detect SMT were also investigated. RESULTS: EUS revealed possibly malignant SMT in 60 cases (64.5%). In 32 out of 55 cases, possibly malignant SMT was suspected by CT. Sensitivity and specificity of CT was 80.6% and 84.2%, respectively. The mean size of possibly malignant SMT detected and not detected by CT was 27.4 and 11.0 mm, respectively, with a statistically significant difference (p = 0.0001). CONCLUSION: This study elucidated that approximately one-third of suspected SMTs by EGD might not need clinical care. EUS for suspected SMTs is indispensable for extracting possibly malignant SMTs from others, and CT may be insufficient as an alternative to EUS in detecting especially small ones.