Abstract

Objective: To explore the application value of CT imaging in differentiating gastric stromal tumors (GST) from gastric leiomyomas (GLMs). Methods: CT images of patients with GST (n=65) or GLMs (n=13, maximum diameter of tumor ≤5 cm) proved by surgery and pathology were retrospectively analyzed. The tumor size, location, contour, growth pattern, degree and pattern of enhancement, calcification, necrosis, surface ulceration, lymph nodes, and patient clinical data were evaluated by two independent reviewers. Receiver operating characteristic (ROC) curves were employed to assess the measurement and calculation parameters in the differentiation of GST and GLMs. Results: Between the GST and GLMs groups, there was no statistically significant difference in the contour, growth pattern, calcification, surface ulceration, and patient's sex (P>0.05). CT values of in plain scans, degree of enhancement in arterial phase (DE1), size, location and pattern of enhancement were found to be different between GST and GLMs (P<0.05). When the cutoff value of the maximum tumor diameter was 3.2 cm, the area under ROC curve, sensitivity and specificity were 0.707, 92.3%(12/13) and 60.6%(40/66), respectively. When the cutoff value of age was 59 years, the area under ROC curve, sensitivity and specificity were 0.773, 92.3% (12/13) and 46.2% (30/65), respectively. Taking the cutoff value of 10.9 HU as the degree of enhancement in arterial phase (DE1), the area under ROC curve, sensitivity and specificity were 0.774, 84.6% (11/13) and 77.3% (51/66), respectively. Using a cutoff value of 30.3 HU, the sensitivity, specificity, and the area under ROC curve were 84.6% (11/13), 65.2% (43/66), and 0.731, respectively. Conclusions: CT examination in addition to clinical data can be very helpful for the differential diagnosis of GLMs from GSTs in maximum diameter ≤5 cm.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call