Quantitative fetal fibronectin for prediction of preterm birth in asymptomatic twin pregnancy.

Introduction: To evaluate cervicovaginal fluid quantitative fetal fibronectin, measured by a bedside analyzer, to predict spontaneous preterm birth in twin pregnancy before 30 weeks of gestation.

Material And Methods: In a prospective cohort study, we studied the accuracy of quantitative fetal fibronectin measured between 18 and 27 weeks of gestation in high-risk asymptomatic women with twin pregnancies, to predict spontaneous preterm birth before 30 weeks of gestation. Predefined fetal fibronectin thresholds were ≥10, ≥50 and ≥200 ng/mL. Predictive statistics were also calculated to evaluate accuracy of "early" tests, performed between 18 and 21 weeks and "standard" tests performed between 22 and 27 weeks of gestation in the same cohort. Subgroup analysis was performed according to cervical length measurement. In addition, we compared accuracy of prediction with quantitative fetal fibronectin measured during the standard test period in asymptomatic twin pregnancy with no additional risk factors, to twin pregnancies with one or more additional risk factors for spontaneous preterm birth.

Results: Of 130 eligible women identified with quantitative fetal fibronectin tests undertaken during the standard testing period, 9% delivered before 30 weeks of gestation. Quantitative fetal fibronectin was significantly related to outcome before 30/40 (ROC curves of 0.8 [95% CI 0.7-1]). Early tests were not significantly predictive; ROC area 0.53 (95% CI 0.29-0.81). There was a trend towards better predictive accuracy when one or more additional risk factors for spontaneous preterm birth or cervical length were considered.

Conclusions: Quantitative fetal fibronectin measured from 22 to 27 weeks of gestation accurately predicts spontaneous preterm birth at <30 weeks of gestation. Tests undertaken earlier are of limited value. Consideration of cervical length or prior history in addition to quantitative fetal fibronectin strengthens prediction.