Background: We investigated associations between Atopobium vaginae and bacterial vaginosis (BV) and the role that A. vaginae plays in recurrent BV after oral metronidazole therapy.
Methods: Women with abnormal vaginal discharge or odor were enrolled in a cross-sectional study (n=358); the proportion of those infected with Gardnerella vaginalis and A. vaginae was determined by polymerase chain reaction. Women with BV (Nugent score [NS] 7-10 or 4-6 with > or =3 Amsel criteria; n=139) were treated with oral metronidazole (400 mg twice a day for 7 days) and examined at 1, 3, 6, and 12 months or until they reached an NS of 7-10 and recurrence of A. vaginae and G. vaginalis infection was established.
Results: A. vaginae and G. vaginalis were highly sensitive for BV--96% (95% confidence interval [CI], 91%-98%) and 99% (95% CI, 97%-100%), respectively. However, A. vaginalis was more specific for BV (77% [95% CI, 71%-82%]) than was G. vaginalis (35% [95% CI, 29%-42%]). G. vaginalis was detected in 100% and A. vaginae in 75% of women with recurrent BV; higher organism loads were present in women with recurrent BV. A. vaginae was rarely detected without G. vaginalis, and women in whom both organisms were detected had higher rates of recurrent BV (83%) than women infected with G. vaginalis only (38%) (P<.001).
Conclusions: Infection with A. vaginae is more specific for BV than infection with G. vaginalis. The higher recurrence rates in women in whom both A. vaginae and G. vaginalis were detected suggest that A. vaginae makes a significant contribution to BV. However, its etiological role remains unclear.