Bacterial vaginosis: efficacy and safety of intravaginal metronidazole treatment

Abstract

Objective: We evaluated the effectiveness and safety of intravaginal metronidazole in the treatment of bacterial vaginosis. Demographic data, Gram stain score, and individual clinical diagnostic criteria for bacterial vaginosis were evaluated for association with therapeutic outcome.

Methods: According to accepted diagnostic criteria and blinded Gram stain scoring for quality control, a multicenter, prospective, double-blinded trial of patients randomized to either 0.75% metronidazole gel 5 gm twice daily for 5 days or placebo was undertaken. Therapeutic response after completion of treatment was evaluated by examination at 4 to 16 days and again at 28 to 32 days for patients initially cured.

Results: After metronidazole treatment, 38 (78%) of 49 patients were cured initially, compared with 11 (27%) of 41 placebo recipients (p < 0.001). One month after treatment 31 (91%) of 34 in the group initially cured by metronidazole remained cured. Side effects were uncommon and mild. Significant intercenter differences were noted for vaginal discharge assessment but not for other diagnostic criteria. Among patients cured initially with metronidazole, mean Gram stain score was 0.58 among those in whom cure was maintained versus 2.33 for those in whom recurrent bacterial vaginosis developed 1 month after treatment (p = 0.03). Curative metronidazole treatment was found initially in 4 (100%) of 4 patients with sperm present on gram stain compared with 34 (76%) of 45 with sperm absent (p = 0.56) and in 6 (100%) of 6 and 25 (89%) of 28, respectively, at the final visit (p = 1.0).

Conclusions: Intravaginal metronidazole is effective, safe, well-tolerated, durable therapy for bacterial vaginosis. Characterization of vaginal discharge appears to be the most variable among the diagnostic criteria for bacterial vaginosis. Gram stain score may be able to identify patients cured early after therapy who are at risk for later recurrence. Exposure to semen does not appear to affect efficacy of intravaginal metronidazole.