A Randomized Controlled Trial of Metronidazole for the Prevention of Preterm Birth in Women Positive for Cervicovaginal Fetal Fibronectin: The PREMET Study

This double blind, randomized, placebo-controlled study was undertaken at 14 hospitals in the United Kingdom in asymptomatic women who were at high risk of preterm birth because of risk factors such as previous midtrimester pregnancy loss or preterm delivery, uterine abnormality, cervical surgery, or cerclage. The criterion for entry into the study was the presence of fetal fibronectin (fFN) in vaginal secretions in the second trimester of pregnancy. This factor appears to be more predictive of preterm birth than bacterial vaginosis or other putative biochemical markers. A total of 900 pregnancies were screened at 24 and 27 weeks gestation. Women with positive results were assigned to receive either oral metronidazole (400 mg 3 times a day) or a placebo for 1 week. The primary outcome was delivery before 30 weeks gestation, and secondary outcomes included delivery before 37 weeks. The study was stopped at an early stage when 21% of 53 women given metronidazole and 11% of 46 placebo recipients delivered before 30 weeks gestation. The risk ratio (RR) was 1.9, with an 85% confidence interval (CI) of 0.72-5.09. Preterm delivery (earlier than 37 weeks gestation) took place in 62% of women given metronidazole and in 39% of the placebo group (RR, 1.6; 95% CI, 1.05-2.4). The positive and negative predictive values of fFN for early preterm birth, estimated at 24 weeks gestation for the risk of delivery before 30 weeks, were 26% and 99%, respectively. The fFN reverted to negative in approximately half of the actively treated and placebo groups. Birth weights averaged 366 g lower for infants whose mothers received metronidazole, and 56% more infants in this group were smaller than 2.5 kg at birth compared with those who mothers received placebo. The proportions of low 1- and 5-minute Apgar scores were comparable in the metronidazole and placebo groups. This is the first prospective trial showing that metronidazole is not an effective means of preventing preterm delivery in high-risk women and that it may have adverse effects on the newborn infant. The investigators do not recommend the continued use of this drug with the intent of preventing preterm delivery.