Oral clindamycin for asymptomatic bacterial vaginosis in early pregnancy reduces premature births

Question Does treatment with oral clindamycin for bacterial vaginosis before 22 weeks' gestation lead to decreased premature births and late miscarriages?

Synopsis Previous studies of screening for bacterial vaginosis in pregnancy performed the screening late in the second trimester and treated bacterial vaginosis with metronidazole. In this randomised controlled double blinded trial, a general population of 6120 asymptomatic pregnant women in the United Kingdom was screened in antenatal outpatients for bacterial vaginosis by using Gram stain and Nugent scoring (0-10), and 494 (8.1%) women positive for bacterial vaginosis with a Nugent score higher than 3 who were randomised (allocation concealed) to oral clindamycin 300 mg twice a day for five days (n = 249) or placebo (n = 245). By chance the treated group had a history of fewer miscarriages (26% v 34%) but not fewer preterm deliveries (10% v 9%). Spontaneous preterm deliveries (24-37 weeks' gestation) were 11/244 (5%) in the treated group versus 28/241 (12%) in the placebo group, and late miscarriages (13-24 weeks) were 2/244 (1%) compared with 10/241 (4%) (P = 0.001 for the combined end point; number needed to treat = 10). The number needed to screen to prevent one preterm birth or late miscarriage was approximately 120. There were no differences in mean birth weight, low birth weight, stillbirths, or mean gestational age at delivery. There was a trend to increased gastrointestinal side effects in the treated group (7% v 3%; P = 0.10). Admissions to neonatal intensive care were 8% in the clindamycin group compared with 10% in the placebo group (P = 0.41), which would be clinically significant if the study had been large enough to show a statistically significant difference.

Bottom line Women with asymptomatic bacterial vaginosis who were treated with oral clindamycin before 22 weeks' gestation had fewer second trimester miscarriages and preterm births, with a number needed to treat of 10. The study was not large enough to show a difference in neonatal admissions to intensive care units. The Gram stain screening method and generic antibiotic treatment are simple and inexpensive. The potential here for getting a big bang for a buck in our use of healthcare dollars is very attractive. Previous studies using metronidazole and treatment at later gestational age have not found screening to be beneficial. This study needs to be replicated in a larger trial before introducing widespread screening for bacterial vaginosis in pregnancy.

Level of evidence 1b (see www.infopoems.com/resources/levels.html; individual randomised controlled trials (with narrow confidence interval).

Ugwumadu A, Manyonda I, Reid F, Hay P. Effect of early oral clindamycin on late miscarriage and preterm delivery in asymptomatic women with abnormal vaginal flora and bacterial vaginosis: a randomised trial.Lancet 2003;361: 983-8

©infoPOEMs 1992-2003 www.infoPOEMs.com/informationmastery.cfm