430: Cervical length shortening between pregnancies and preterm delivery

Abstract

To test whether a decrease in midtrimester cervical length between pregnancies is associated with preterm delivery in a subsequent pregnancy. This is a cohort study of women who had two consecutive singleton births of at least 20 weeks’ gestation at one institution who had a midtrimester cervical length measured in each pregnancy. Cervical length change between pregnancies was measured as the difference in centimeters (cm) between cervical lengths using the measurement taken closest to 20 weeks 0 days’ gestation in each pregnancy. Cervical length shortening was defined as present if the cervical length decreased by one standard deviation or more between index and subsequent pregnancies. Bivariable analyses and multivariable logistic regression were used to determine whether cervical length shortening between pregnancies was associated with preterm delivery. Breslow-Day tests for homogeneity were used to assess for differences in the odds of preterm birth after stratification by preterm birth in the index pregnancy or a short cervix (< 2.5 cm) in the subsequent pregnancy. Among the 1,693 women who met inclusion criteria, mean change in midtrimester cervical length between pregnancies was -0.1 ± 0.9 cm; 130 (7.7%) experienced a preterm delivery in the subsequent pregnancy. Compared to women whose subsequent pregnancy cervical length remained stable or increased, women whose cervical length shortened were more likely to experience a preterm delivery (11.5% vs 6.8%, p=0.006). In the multivariable model, cervical length shortening remained significantly associated with preterm delivery even when accounting for interdelivery interval, cervical length in the subsequent pregnancy, or a preterm delivery or short cervix in the index pregnancy (Table). The magnitude of the increased risk in preterm birth did not significantly differ between women with a most recent preterm vs term birth (p-value for Breslow-Day = 0.07) or between women with a normal vs short cervical length in the subsequent pregnancy (p-value for Breslow-Day = 0.64). Midtrimester cervical length shortening between pregnancies is independently associated with an increased risk of preterm delivery. The increased risk for preterm birth associated with cervical length shortening was similar in women with a most recent term birth and in women with a normal cervical length in the subsequent pregnancy.