Objective: Half of all preterm births occur in women without clinical risk factors. Our goal was to assess fetal fibronectin assay, Bishop score, and cervical ultrasonography as screening tests to predict which low-risk pregnancies will end in preterm birth.
Study design: We performed a secondary analysis of data collected at 22 to 24 weeks' gestation from low-risk subjects enrolled in the Preterm Prediction Study, an observational study of risk factors for preterm birth conducted by the National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. Analysis was limited to primigravid women and to women who did not have a history of preterm birth or spontaneous pregnancy loss at <20 weeks' gestation. Bishop score (> or =4), fetal fibronectin level (> or =50 ng/mL), and cervical length (< or =25 mm) at 24 weeks' gestation were evaluated alone and in sequence as tests to predict spontaneous delivery before 35 weeks' gestation.
Results: Of the 2929 subjects enrolled in the original study, 2197 (1207 primigravid women and 900 low-risk multiparous women) met criteria for this analysis. There were 64 spontaneous births before 35 weeks' gestation (3.04%). All three tests were significantly related to birth before 35 weeks' gestation (high Bishop score: relative risk, 3.6; 95% confidence interval, 2.1-6.3; fetal fibronectin detection: relative risk, 8.2; 95% confidence interval, 4.8-13.9; short cervical length: relative risk, 6.9; 95% confidence interval, 4.3-11.1). However, the sensitivities of the tests alone were low (23.4% for high Bishop score, 23.4% for fetal fibronectin detection, and 39.1% for short cervix), as were the sensitivities for Bishop score followed by cervical ultrasonography (14.1%) and fetal fibronectin assay followed by cervical scan (15.6%).
Conclusion: In the setting of low-risk pregnancy, fetal fibronectin assay and cervical ultrasonography have low sensitivity for preterm birth before 35 weeks' gestation. Sequential screening with Bishop score or fetal fibronectin assay followed by cervical ultrasonography further decreased sensitivity to only 15% among low-risk women.