(1200) Gestational Age at Cervical Length Measurement and the Risk of Preterm Birth in Twin Pregnancies

Sonographic cervical length (CL) is a well-established predictor of spontaneous preterm birth (sPTB) in twin pregnancies. However, its predictive accuracy may vary by the gestational age (GA) at the time of CL measurement. Earlier screening may yield higher positive predictive value (PPV), as early cervical shortening is associated with higher sPTB risk, but lower sensitivity, since some patients only experience shortening later in gestation. Data on the week-specific predictive accuracy of CL in twin pregnancies are limited. Therefore, this study aimed to assess the effect of GA at CL measurement on the predictive accuracy of CL for sPTB in twin pregnancies and to develop week-specific risk estimates to facilitate patient counseling.

Study Design:

This was a retrospective study of twin pregnancies followed at a single center (2012-2024). CL was routinely measured every 2-3 weeks from 16-18 weeks to 28-30 weeks. For each 2-week interval between 16-29 weeks, we evaluated the discriminative ability of CL for predicting sPTB < 32 weeks using the area under the ROC curve (AUC), and described its predictive accuracy at the fixed threshold of ≤25mm using sensitivity, specificity, and positive and negative predictive values.

Results:

A total of 1,760 patients who underwent 7,392 CL measurements met the study criteria. The AUC of CL for sPTB < 32 weeks increased with GA at measurement, from 0.636 (0.565-0.706) at 16-17 weeks to 0.830 (0.741-0.919) at 28-29 weeks. Sensitivity increased from 10.3% to 69.0% while PPV declined from 41.2% to 12.9% as GA at measurement increased from 16-17 weeks to 28-29 weeks (Figure 1). Week-specific sPTB risk estimates by CL were developed (Figure 2).  

Conclusion:

In twin pregnancies, the predictive performance of CL for sPTB varies by GA at measurement. These findings can inform patient counseling based on CL and GA at CL measurement.