To evaluate the combination of transvaginal conventional ultrasonography and shear wave elastography (SWE) on the diagnosis of endometrial cancer, and establish a predictive Logistic regression model.

Clinical information collection, transvaginal conventional ultrasonography, and SWE check were performed in 112 patients who were post-menopausal vaginal bleeding with ≥5 mm thick endometrium. The Emax and Emean of Young′s modulus for the endometrium were obtained. Pathology was used as the gold standard, ROC curve was plotted, which could be used to evaluate the Young′s modulus on the diagnostic effectiveness on endometrial cancer. Single factor analysis and bring logistic regression methods were applied to assess the values of the clinical variables, transuaginal conventional ultrasonography variables, and Young′s modulus in the identification of endometrial cancer.

In 112 cases of endometrial lesions diagnosed by pathology, there were 84 cases of benign lesions (benign group) and 28 cases of cancer(malignant group). Both Emax and Emean in malignant group were larger than benign group[(53.00±16.07)kPa vs (31.99±13.89)kPa, (27.25±9.28)kPa vs (19.94±10.37)kPa, all P<0.001]. In the logistical regression analysis, body mass index, endometrial thickness, blood flow grading and Young′s modulus were identified as independent risk factors for endometrial cancer. The accuracy, sensitivity and specificity of the logistic regression model in the prediction of endometrial cancer were 89.29%, 82.14%, and 91.67%, respectively. The area under the ROC curve was 0.928.

Transvaginal conventional ultrasonography combined SWE technique has an important value in the diagnosis of endometrial cancer.

Ultrasonography; Shear wave elastography; Endometrial cancer; Young′s modulus; Logistic regression