Transvaginal sonographic cervical length in first and second trimesters in a low-risk population: a prospective study.

OBJECTIVES: To assess cervical length (CL) longitudinally between the first and second trimesters and to determine the proportion of women with short CL. The study also aimed to assess if women with short CL at 19-24 weeks' gestation could be identified at the time of combined first-trimester screening (cFTS) at 11-14 weeks' gestation, in order to determine the potential value of implementation of CL screening for prediction of preterm delivery in a Danish population.

METHODS: This was a prospective longitudinal study of women with singleton pregnancy attending three University Hospitals in Denmark for combined first-trimester screening from 1 November 2013 to 1 December 2014. Exclusion criteria were multiple pregnancy, uterine anomaly, cerclage or progesterone treatment at inclusion. CL was measured on transvaginal sonography at 11-14 weeks (Cx1), 19-21 weeks (Cx2) and 23-24 weeks (Cx3), by trained operators as a straight line from external to internal os. Women with CL ≤ 25 mm were referred to a maternal-fetal medicine specialist for treatment according to a standardized management protocol.

RESULTS: Of the 4904 eligible women, 3477 (71%) participated and had Cx1 recorded. Of those, 3232 (93.0%) had CL measured on all three scans. Median Cx1 was 37 mm, and median Cx2 and Cx3 were 40 mm. The proportion of women with CL ≤ 25 mm increased with gestational age, from 0.41% (95% CI, 0.19-0.62%) at Cx1 to 1.79% (95% CI, 1.34-2.24%) at Cx3. In total, the proportion of women with second-trimester CL (Cx2 or Cx3) ≤ 25 mm was 2.0% (n = 67), of which 38.8% (n = 26) were detected at 19-21 weeks. The probability of short CL between 19 and 24 weeks was greater for those with shorter first-trimester CL. It was nearly nine-fold higher for women with Cx1 ≤ 25 mm compared with Cx1 ≥ 35 mm (17% vs 2%). The performance of Cx1 for prediction of short second-trimester CL was 50% at a 10% false-positive rate. It was found that more than 1500 women would need to be screened for short CL at 19-21 weeks to prevent one case of spontaneous preterm delivery before 34 weeks in a population such as the one in this study.

CONCLUSIONS: There is an association between first-trimester CL and risk of short cervix in the second trimester. Once short CL was observed, risk of preterm delivery was greatly increased. However, whether universal CL screening should be implemented in this low-risk population depends on cost-benefit analysis taking into account the low proportions of women with short CL and at risk for preterm delivery. Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.