Role of cervical length measurement for preterm delivery prediction in women with threatened preterm labor and cervical dilatation

Objectives - To compare the accuracy and cutoff points for cervical length for predicting preterm delivery in women with threatened preterm labor between those with a closed cervix and cervical dilatation. Methods - We conducted a retrospective cohort study of women with singleton pregnancies with threatened preterm labor before 34 weeks. The accuracy of cervical length for predicting preterm delivery was compared between women with cervical dilatation (0.5-3 cm) and those with a closed cervix. The predictive accuracy of cervical length for spontaneous preterm delivery was analyzed with several outcome-specific thresholds. Results - Overall, 1068 women with threatened preterm labor met the inclusion criteria; of them, 276 (25.8%) had cervical dilatation, and 792 (74.2%) had a closed cervix. The risk of preterm delivery before 37 weeks was significantly higher in the cervical dilatation group than the closed cervix group, as well as a shorter assessment-to-delivery interval of within 14 days (P = .001 and .004, respectively). On a multivariable analysis, cervical length was independently associated with the risk of preterm delivery in both groups. There was no significant difference between women with cervical dilatation and those with a closed cervix regarding the area under the receiver operating characteristic curves of cervical length for prediction of preterm delivery before 37 (0.674 versus 0.618; P = .18) and 34 (0.628 versus 0.640; P = .88) weeks and an assessment-to-delivery interval of 14 days (0.686 versus 0.660; P= .72). The negative predictive value of cervical length ranged from 77.4% to 95.7% depending on the different thresholds used. Conclusions - Cervical length was significantly associated with the risk of preterm delivery in women presenting with threatened preterm labor and cervical dilatation of less than 3 cm. However, the predictive accuracy of cervical length as a single measure was relatively limited.