Clinically silent upper genital tract microbial infections have been strongly associated with spontaneous preterm delivery, especially prior to 32 weeks gestational age. This condition represents a potentially preventable or reversible cause of preterm birth using safe, readily available, and inexpensive antibiotics. Fetal fibronectin detected in cervical and vaginal secretions between 22 to 24 weeks' gestation is a powerful predictor of subsequent risk for spontaneous preterm delivery. Evidence indicates that the detection of fetal fibronectin in cervical/vaginal secretions could represent a marker for clinically silent upper genital tract microbial infection. Thus, the National Institute of Child Health and Human Development Maternal-Fetal Medicine Units (NICHD MFMU) Network performed a randomized clinical trial to determine if antibiotic treatment (metronidazole plus erythromycin) of asymptomatic women with a positive cervical or vaginal fetal fibronectin test in early pregnancy will reduce the risk of spontaneous preterm delivery. No significant difference was observed between the antibiotic- and placebo-treated groups for spontaneous preterm delivery before 37 week's gestation (relative risk 1.17, 95% confidence interval 0.80-1.70). Therefore, use of the fetal fibronectin test to identify women for antibiotic treatment in an effort to prevent preterm birth should be discouraged, and current clinical use of the fetal fibronectin test should be limited to its negative predictive value. © 2003 Elsevier Inc. All rights reserved.