An Observational Study of Vaginal Antibacterial Activity and Outcomes after Antibiotic Prescription for Bacterial Vaginosis.

BACKGROUND: Bacterial vaginosis (BV) is the most common cause of vaginal discharge syndrome and usually treated with metronidazole. BV frequently recurs. Published studies have not linked measured vaginal antibiotic activity (VAA) or antibiotic concentrations to clinical or microbiological BV outcomes. The objective is to determine if VAA in vaginal fluid after diagnosis of BV is associated with persistent/recurrent BV or changes in concentrations of vaginal BV-associated bacteria (BVAB). METHODS: VAA was assessed using a bioassay to measure inhibition of growth of BVAB in culture. Persistent/recurrent BV was measured by Amsel clinical criteria at days 6-35 (early) and 36-100 (late), concentrations of indicator BVAB in the vagina by quantitative PCR, and condom use by questionnaire. RESULTS: 87 participants who experienced 130 episodes of BV were prescribed metronidazole.No VAA was detected in 40/130 (31%) episodes of BV in the 10 days post-diagnosis. VAA was linked to significantly greater declines in vaginal concentrations of indicator BVAB (p<.001). Persistent/recurrent BV was present in 38/129 (29%) episodes at days 6-35 and 45/63 (71%) episodes at days 36-100. In a multivariable model, VAA for majority of prescribed days was associated with a reduced risk for recurrent BV in days 6-35, in the absence of condomless sex (RR 0.31, 95% CI 0.11-0.89, p=.03), but not when condomless sex was reported (p=0.49). No associations between VAA and BV recurrence in days 36-100 were found. CONCLUSION: The combination of metronidazole detection on the majority of prescribed days and condom use is associated with reduced risk of BV recurrence.