Bacterial vaginosis and preterm birth: a prospective community-based cohort study.

BACKGROUND Preterm birth before 37 weeks' gestation is associated with 70% of perinatal morbidity and nearly half of long-term neurological morbidity. Hospital-based studies have shown that bacterial vaginosis is associated with preterm birth. AIM To estimate the relative risk of preterm birth in women with and without bacterial vaginosis, detected by self-administered vaginal swab at < 10 weeks' gestation. DESIGN Prospective cohort study. SETTING Thirty-two general practices and five family planning clinics in South London. PARTICIPANTS A total of 1216 women with bacterial vaginosis status established before 10 weeks' gestation, by analysis of Gram stained vaginal smears by two independent observers. METHOD All women who did not miscarry or have a termination of pregnancy before 16 weeks' gestation were sent a brief confidential questionnaire at 16 weeks and at term asking about pregnancy outcome. Data on non-responders were obtained by searches of hospital and general practice records and by telephone calls to patients. RESULTS Ascertainment was 87% (937/1072). The mean age of the women was 31 years. Thirteen per cent (122/925) had bacterial vaginosis and 5% (44/897) had a spontaneous preterm birth. The relative risk (RR) of preterm birth in women with bacterial vaginosis was 0.9 (95% confidence interval [CI] = 0.4 to 2.2). However, bacterial vaginosis was associated with late miscarriage at 13-23 weeks (R = 4.0, 95%CI = 1.3 to 12.1). Preterm birth was not associated with previous preterm birth, black ethnicity, age < 20 years, low social class, single marital status, or chlamydial infection. However, it was more common in women who reported smoking in pregnancy (RR = 2.9, 95% CI = 1.5 to 5.5). Of 867 responders, 552 (64%) said that providing a vaginal swab was at least as easy as providing a urine specimen. CONCLUSIONS In this low-risk community-based cohort, bacterial vaginosis was not a strong risk factor for preterm birth.