Is a change in the vaginal flora associated with an increased risk of preterm birth?

Objective The purpose of this study was to determine if a change in the vaginal flora was associated with an increased risk of preterm birth, and to determine if metronidazole therapy before 32 weeks increased the risk of preterm birth. Study design We compared cultures taken at 23 to 26 weeks of gestation with cultures taken at delivery from women enrolled in the Vaginal Infections and Preterm Birth study to analyze the association of changes in the vaginal flora with preterm birth. Results Metronidazole therapy before 32 weeks was associated with an increased risk of preterm birth (OR 1.5, 95%CI 1.05-2.1) in an unadjusted model. A change to heavy growth of Escherichia coli or Klebsiella pneumoniae at delivery was found to be associated with preterm birth (OR 2.4, 95%CI 1.6-3.8). After controlling for race, parity, prepregnancy weight <100 pounds, smoking or drinking during pregnancy, Trichomonas vaginalis, bacterial vaginosis, chlamydia, mycoplasmas, group B streptococcus, metronidazole therapy before 32 weeks, vaginal pH >5.0, and an increase in E coli or K pneumoniae, only prepregnancy weight <100 pounds (adjusted odds ratio [AOR] 2.07, 95%CI 1.01-4.21) and increased E coli or K pneumoniae in the vagina at delivery (AOR 2.99, 95%CI 1.37-6.53) were found to be significantly associated with preterm birth. Conclusion An increase in E coli or K pneumoniae in the vagina is an independent risk factor for preterm birth. Changes in the vaginal flora may explain the increased risk of preterm birth seen with vaginal clindamycin or oral metronidazole therapy.