Mid-trimester sonographic cervical consistency index to predict spontaneous preterm birth in a low-risk population.

Objectives: To investigate the effectiveness of mid-trimester sonographic cervical consistency index (CCI) for the prediction of spontaneous preterm birth (sPTB) in low-risk pregnancies and to compare its performance with that of mid-trimester sonographic cervical-length (CL) measurement.

Methods: This was a prospective cohort study of women with a singleton pregnancy examined by ultrasound at 19 + 0 to 24 + 6 weeks' gestation. All women underwent transvaginal ultrasound examination of the cervix, but CCI and CL were measured, offline, only in women without a risk factor for sPTB. Staff and participants were blinded to CL and CCI results. CCI was obtained by calculating the ratio between the anteroposterior diameter of the uterine cervix at maximum compression and at rest. The primary outcome was prediction of sPTB before 37 + 0 weeks. Receiver-operating characteristics (ROC) curves were produced and sensitivity and specificity were calculated for the optimal cut-off based on the ROC curve and for the 1 , 5 and 10 centiles of CCI and CL. Intraclass correlation coefficients (ICC) and Bland-Altman plots were used to estimate intra- and interobserver agreement and reliability for measurement of CCI and CL.

Results: Of the 749 women who underwent ultrasound examination of the cervix, 532 were included for analysis. The rates of sPTB before 37 + 0 and before 34 + 0 weeks were 4.1% (22/532) and 1.3% (7/532), respectively. The rates of short cervix < 25 mm and ≤ 20 mm were 0.9% (5/532) and 0.4% (2/532), respectively. The area under the ROC curve (AUC) with regard to predicting sPTB before 37 + 0 weeks was 0.84 (95% CI, 0.75-0.93) for CCI compared with 0.68 (95% CI, 0.56-0.81) for CL (P = 0.03). The optimal cut-off based on the ROC curve was 64.6% for CCI (sensitivity, 77.3%; specificity, 82.7%) and that for CL was 37.9 mm (sensitivity, 72.7%; specificity, 61.2%). The AUC with regard to predicting sPTB before 34 + 0 weeks was 0.87 (95% CI, 0.71-1.0) for CCI compared with 0.71 (95% CI, 0.47-0.94) for CL (P = 0.25). The optimal cut-off based on the ROC curve was 63.6% for CCI (sensitivity, 85.7%; specificity, 84.0%) and that for CL was 37.9 mm (sensitivity, 85.7%; specificity, 61.3%). Intraobserver ICC was > 0.90 both for CCI and CL, while interobserver ICC was 0.89 for CCI and 0.90 for CL.

Conclusions: Second-trimester CCI is a better predictor of sPTB < 37 weeks in low-risk pregnancies than is CL. External validation is needed as well as studies assessing the value of CCI as a screening tool in unselected and high-risk populations. Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.