Associations between Nugent-bacterial vaginosis and preterm birth and other adverse pregnancy outcomes in rural northwestern Bangladesh.

OBJECTIVE

Bacterial vaginosis, defined by Nugent score (Nugent-BV), has been associated with preterm birth and other adverse pregnancy outcomes. However, few studies have longitudinally described the associations between Nugent-BV and adverse pregnancy outcomes at varying times in pregnancy.

METHODS

We assessed associations between Nugent-BV at two separate time points in pregnancy and preterm birth (<37 weeks of gestation), low birth weight (LBW, <2500 g), and small-for-gestational-age (SGA, <10th percentile) using data from a community-based randomized trial in rural northwestern Bangladesh from 2001 to 2007. Pregnant women provided self-collected vaginal swabs in early (first or second trimester) and late (third trimester) pregnancy for Nugent scoring. Nugent-BV was categorized as 7-10 (ref: 0-6) or 4-10 (ref: 0-3). We used multivariable regression techniques to model relationships between Nugent-BV and incidence of preterm birth, LBW, and SGA, adjusted for the trial intervention and demographic, socioeconomic, and pregnancy history factors.

RESULTS

A total of 1243 women provided ≥1 vaginal swab with a Nugent score and had a live birth. Preterm birth incidence was 23.5% (n = 271/1151). Nugent-BV 7-10 in early (adjusted relative risk (aRR): 0.84, 95% confidence interval [CI]: 0.42, 1.66) and late (aRR: 1.04, 95% CI: 0.44, 2.41) pregnancy were not associated with preterm birth. Nugent-BV 4-10 at either time point was not associated with preterm birth in the final adjusted models. From early to late pregnancy, a transition from Nugent score category 4-10 to 0-3, relative to 0-3 at both timepoints, was associated with decreased risk of preterm birth (aRR 0.20, 95% CI: 0.06, 0.63). In early and late pregnancy, neither Nugent-BV categorization was associated with LBW or SGA in the final models.

CONCLUSION

Resolution of Nugent-BV between early and late pregnancy might be associated with a decreased risk of preterm birth. Longitudinal studies with frequent sampling in pregnancy and using molecular techniques are needed to better describe changes in the vaginal microbiota and associated risk for adverse pregnancy outcomes.