This study aimed to investigate the predictive value of real-time shear wave elastography (SWE) for spontaneous preterm birth (SPB). This study prospectively selected 175 women with singleton pregnancies at 16 to 36 weeks of gestation. Cervical length (CL) and uterocervical angle (UCA) were measured using transvaginal ultrasonography. Real-time shear wave elastography was used to measure Young's modulus values, including the average Young's modulus (Emean) and the maximum Young's modulus (Emax) at 4 points: point A on the inner lip of the cervical os, point B on the outer lip of the cervical os, point C on the inner lip of the external os, and point D on the outer lip of the external os. Receiver operating characteristic (ROC) curve analysis was performed to compare the accuracy of Young's modulus values at the 4 points, CL, and UCA in predicting SPB. Significant variables were used to construct a binary logistic regression model to predict the multifactorial predictive value of SPB, which was evaluated using an ROC curve. A total 176 valid cases, including 160 full-term pregnancies and 16 SPB, were included in this study. Receiver operating characteristic curve analysis revealed that Emean at point A, as well as Emean and Emax at point D, had a relatively high accuracy in diagnosing SPB, with area under the curve values of 0.704, 0.708, and 0.706, respectively followed by CL (0.670), SWE at point C (Emean 0.615, Emax 0.565), SWE at point B (Emean 0.577, Emax 0.584), and UCA (0.476). Binary logistic regression analysis showed that comorbidities during pregnancy (including diabetes mellitus, hypertension, cholestasis and thyroid dysfunction), CL, and Emean at point A were independent predictors of preterm birth. In addition, the AUC value of the logistic regression model's ROC curve was 0.892 (95% CI: 0.804-0.981), with a sensitivity of 0.867, specificity of 0.792, and Youden's index of 0.659, indicating that the regression model has good predictive ability for SPB. Real-time shear wave elastography showed a higher predictive value for SPB than CL and UCA. The SWE combined with CL and comorbidities during pregnancy model has a good predictive ability for SPB.
Read full abstract