FETAL FIBRONECTIN AS A PREDICTOR OF PRETERM BIRTH IN PATIENTS WITH SYMPTOMS - A MULTICENTER TRIAL

OBJECTIVE: Our aim was to determine whether the presence of fetal fibr onectin in vaginal secretions of patients with symptoms suggestive of preterm labor predicts preterm delivery. STUDY DESIGN: Patients who we re examined-at the hospital between 24 weeks' and 34 weeks 6 days' ges tation with intact membranes, no prior tocolysis, symptoms suggestive of preterm labor, and cervical dilation <3 cm were recruited at 10 sit es. Swabs of the posterior fornix were assayed for the presence of fet al fibronectin by monoclonal antibody assay, with a positive result de fined as greater than or equal to 50 ng/ml. Results were not available to the managing physicians. Tocolysis was used when clinically indica ted after specimen collection. RESULTS: A total of 763 patients had fe tal fibronectin results and pregnancy outcome data available for analy sis. Fetal fibronectin was detected in specimens from 150 (20%) patien ts. Compared with patients who had negative results, patients who had positive results for fetal fibronectin were more likely to be delivere d within 7 days (relative risk 25.9 [95% confidence interval 7.8 to 86 ]), within 14 days (relative risk 20.4 [95% confidence interval 8.0 to 53]), and before 37 completed weeks (relative risk 2.9 [95% confidenc e interval 2.2 to 3.7]). The negative predictive values for delivery w ithin 7 days, within 14 days, and at <37 weeks were 99.5%, 99.2%, and 84.5%, respectively. When we used multiple logistic regression analysi s to control for potential confounding variables among singleton pregn ancies, only the presence of fetal fibronectin (odds ratio 48.8, 95% c onfidence interval 7.4 to 320), prior preterm birth (odds ratio 8.3, 9 5% confidence interval 1.5 to 46.6), and tocolysis (odds ratio 4.1, 95 % confidence interval 1.0 to 16.0) were associated with birth within 7 days; fetal fibronectin (odds ratio 3.6, 95% confidence interval 2.2 to 5.9), prior preterm birth (odds ratio 2.5, 95% confidence interval 1.4 to 4.4), cervical dilatation >1 cm (odds ratio 2.9, 95% confidence interval 1.6 to 5.2), and tocolysis (odds ratio 4.5, 95% confidence i nterval 2.8 to 7.2) were all independently associated with delivery be fore 37 weeks. CONCLUSION: In a population of patients with symptoms, the presence of fetal fibronectin in vaginal secretions best defines a subgroup at increased risk for delivery within 7 days; the high negat ive predictive value of fetal fibronectin sampling supports less inter vention for patients with this result.