The Value of Double Contrast-Enhanced Ultrasonography in the Differential Diagnosis of Gastric Submucosal Tumors and the Assessment of the Malignant Risk of Gastrointestinal Stromal Tumors.

To explore the clinical application value of double contrast-enhanced ultrasound (DCEUS), which combines oral contrast-enhanced ultrasound (OCEUS) with intravenous contrast-enhanced ultrasound (ICEUS), for the differential diagnosis of gastric submucosal tumors (SMTs) and assessment of their malignant risk, with a focus on distinguishing gastrointestinal stromal tumors (GISTs) from benign SMTs. We retrospectively analysed 276 patients with gastric SMTs admitted to our hospital from January 2019 to January 2024, all of whom completed gastric DCEUS before surgery or endoscopic biopsy. Pathological findings served as the gold standard, with all malignant/potentially malignant lesions confirmed as GISTs. Patients were divided into two groups: benign SMTs and GISTs. We analyzed clinical and imaging features, and performed binary logistic regression to identify independent risk factors and construct a prediction model. The model's performance was evaluated using receiver operating characteristic (ROC) curves, and internal validation was performed. Time-intensity curves (TIC) were generated using VueBox. Of 276 patients, 183 (66.3%) had GISTs, and 93 (33.7%) had benign SMTs. Multivariate logistic analysis showed that age, gender, location, layer of origin, enhancement degree, and homogeneity were independent predictors for distinguishing GISTs from benign SMTs. The prediction model had an area under the curve (AUC) of 0.84, accuracy of 80%, sensitivity of 73%, specificity of 84%, positive predictive value (PPV) of 70%, and negative predictive value (NPV) of 86%. Internal validation confirmed stability. TIC analysis showed higher PE and shorter RT in GISTs than benign lesions. DCEUS enhances the ability to distinguish GISTs from benign SMTs by integrating tumor morphology, hemodynamics, and clinical features. Heterogeneous high enhancement may serve as a potential imaging marker for intermediate-high-risk GIST stratification. The Nomogram model provides a quantitative tool for individualized risk assessment.