Sonographic Measurement of Cervical Length ad Fetal Fibronectin Testing in Threatened Preterm Labor

Abstract

When measured in women with preterm contractions, both the presence or absence of fetal fibronectin in cervicovaginal secretions and sonographically measured cervical length have been reported to distinguish true from false labor, essentially between women who will and will not deliver within the next 7 days. The present study, which prospectively enrolled 195 women with singleton pregnancies who were seen at 24 to 36 weeks gestation, attempted to determine whether combining these 2 tests is more predictive than either test by itself. Participants presented with regular painful uterine contractions but were less than 3 cm dilated and had intact membranes. Fetal fibronectin concentrations were measured by immunoassay in cervicovaginal secretions taken from the posterior fornix or endocervix followed by digital assessment of cervical dilation. Transvaginal sonography was then performed to measure cervical length. The clinician caring for each patient was blinded to the test results. The women presented at a median gestational age of 31 weeks; 19 of them (9.7%) delivered within 1 week. The rate of preterm delivery was 51.4% in women whose cervical length was less than 15 mm and 0.6% in those having a longer cervix. Similarly, the preterm delivery rate for women with positive fetal fibronectin was 21.2% compared with 0.9% for fibronectin-negative women. The positive and negative likelihood ratios for delivery within 1 week were 9.8 and 0.05, respectively, for a cervical length less than 15 mm, and 2.6 and 0.08 for a positive fibronectin test. The rate of delivery within 7 days increased exponentially as cervical length decreased. Cervical length correlated inversely with positive fetal fibronectin; the shorter the cervix, the more likely that fibronectin was positive. On multiple logistic regression analysis, the only factor that independently and significantly predicted preterm delivery was cervical length. Excluding cervical length, positive fibronectin contributed significantly to predicting delivery within 1 week. Other predictive factors included Afro-Caribbean ethnicity and gestational age at the time of presentation. Analysis of receiver-operating characteristic curves indicated that predictions based on cervical length were better than those provided by the fibronectin test. In women presenting with threatened preterm labor, a short cervix on sonography more accurately identifies those at high risk than does the presence of fetal fibronectin in cervicovaginal secretions. Measurements of cervical length can provide a basis for hospitalizing women and administering a tocolytic agent or steroid. If sonography is not available, risk status may be inferred from ethnic origin, gestational age, and the results of a fetal fibronectin test.