Cervical length, multifetal pregnancy reduction, and prediction of preterm birth

PURPOSE: To evaluate the application of transvaginal sonography assessment of cervical length before fetal reduction for predicting spontaneous preterm birth in triplet gestations reduced to twins. METHODS: This retrospective study was conducted at the ultrasound unit of a university-affiliated municipal hospital. The study cohort consisted of 25 women with triplet gestations following ovulation induction or assisted-reproduction techniques who underwent fetal reduction to twins. Cervical length was assessed via transvaginal sonography before fetal reduction, and data on pregnancy outcome were retrieved from maternal records and/or maternal interviews. RESULTS: Cervical length (mean +/- SD) at reduction was 4.0 +/- 0.85 (range: 1.2-5.5). Five women were excluded from statistical evaluation because pregnancy complications precluded spontaneous delivery. Two of 3 (67%) women with a cervical length of <3.5 cm delivered prior to 33 weeks' gestation compared with 1/17 (6%) women with a cervical length > or = 3.5 cm. This difference was statistically significant (P < 0.05). The sensitivity, specificity, positive predictive value, and negative predictive value of cervical lengths <3.5 cm to predict delivery prior to 33 gestational weeks was 67%, 94%, 67%, and 94%, respectively. CONCLUSIONS: Measurement of cervical length in triplet pregnancies before fetal reduction provides useful predictive information on the risk for preterm delivery.