Objective: Our purpose was to determine whether treatment of bacterial vaginosis with metronidazole in patients with preterm delivery in the penultimate pregnancy from preterm labor or premature rupture of membranes reduces the risk of subsequent preterm birth.
Study design: From January 1989 to June 1992 patients with a singleton gestation between 13 and 20 weeks and a history of preterm birth in the preceding pregnancy from either idiopathic preterm labor or premature rupture of membranes were screened for bacterial vaginosis. Those with a positive screen were randomized to receive 250 mg of metronidazole three times a day for 7 days or placebo in a double-blind design. Data were analyzed with Student t and chi 2 tests, and differences considered significant at p < 0.05.
Results: Of 94 eligible patients, 80 were enrolled and completed the study, of which 44 received metronidazole. Both groups were comparable in number of entry variables. Compared with the placebo group, patients in the metronidazole group had significantly fewer hospital admissions for preterm labor, 12 (27%) versus 28 (78%); preterm births, eight (18%) versus 16 (39%); births of infants weighing < 2500 gm, six (14%) versus 12 (33%); and premature rupture of membranes, two (5%) versus 12 (33%).
Conclusion: Treatment of bacterial vaginosis with metronidazole was effective in reducing preterm births in patients with a history of prematurity in the preceding pregnancy.