Identifying twin gestations at low risk for preterm birth with a transvaginal ultrasonographic cervical measurement at 24 to 26 weeks' gestation

Abstract

Objective: Because twins are a high-risk group for preterm birth, many clinicians routinely use prophylactic interventions such as home bed rest, hospital bed rest, oral tocolytics, or home uterine activity monitoring to prevent preterm delivery. We sought to identify twin gestations at low risk for spontaneous preterm birth with transvaginal ultrasonography of the cervix to avoid the unnecessary use of prophylactic interventions in these pregnancies.

Study design: We measured cervical length at 24 to 26 weeks' gestation by transvaginal ultrasonography in women with twin gestations referred to our prematurity prevention clinic. Each delivery was classified as (1) spontaneous preterm birth < 34 weeks' gestation, (2) delivery at > or = 34 weeks' gestation with intervention, or (3) delivery at > or = 34 weeks' gestation without intervention. Intervention included strict bed rest at home or in the hospital, either parenteral or oral tocolysis, or both, or home uterine activity monitoring. Indicated preterm deliveries and patients with cerclage were excluded from this analysis. The ability of transvaginal cervical length to predict women who would deliver at > or = 34 weeks without intervention was evaluated. A cervical length of 35 mm was chosen by scatter diagram as the best cutoff to discriminate between the group delivered at term without intervention and the other two groups.

Results: Of 85 women with twin gestations who underwent ultrasonographic cervical length measurements at 24 to 26 weeks' gestation, 17 had spontaneous preterm birth at < 34 weeks, 23 were delivered at > or = 34 weeks but required intervention, and 45 were delivered at > or = 34 weeks without intervention. The mean cervical length for those delivered at > or = 34 weeks' gestation without intervention (36.4 +/- 5.8 mm) was significantly greater (p < 0.0001) than the mean for those delivered preterm (27.4 +/- 8.5) and those delivered at > or = 34 weeks' gestation who required intervention (27.7 +/- 10.5 mm). The sensitivity, specificity, and positive and negative predictive values of a cervical length > 35 mm for predicting delivery at > or = 34 weeks' gestation are 49%, 94%, 97%, and 31%, respectively.

Conclusion: A transvaginal ultrasonographic measurement of the cervix of > 35 mm at 24 to 26 weeks in twin gestations can identify patients who are at low risk for delivery before 34 weeks' gestation.