Hydrogen peroxide (H₂O₂) produced by vaginal lactobacilli is generally believed to protect against bacteria associated with bacterial vaginosis (BV), and strains of lactobacilli that can produce H₂O₂ are being developed as vaginal probiotics. However, evidence that led to this belief was based in part on non-physiological conditions, antioxidant-free aerobic conditions selected to maximize both production and microbicidal activity of H₂O₂. Here we used conditions more like those in vivo to compare the effects of physiologically plausible concentrations of H₂O₂ and lactic acid on a broad range of BV-associated bacteria and vaginal lactobacilli.

Anaerobic cultures of seventeen species of BV-associated bacteria and four species of vaginal lactobacilli were exposed to H₂O₂, lactic acid, or acetic acid at pH 7.0 and pH 4.5. After two hours, the remaining viable bacteria were enumerated by growth on agar media plates. The effect of vaginal fluid (VF) on the microbicidal activities of H₂O₂ and lactic acid was also measured.

Physiological concentrations of H₂O₂ (< 100 μM) failed to inactivate any of the BV-associated bacteria tested, even in the presence of human myeloperoxidase (MPO) that increases the microbicidal activity of H₂O₂. At 10 mM, H₂O₂ inactivated all four species of vaginal lactobacilli but only one of seventeen species of BV-associated bacteria. Moreover, the addition of just 1% vaginal fluid (VF) blocked the microbicidal activity of 1 M H₂O₂. In contrast, lactic acid at physiological concentrations (55-111 mM) and pH (4.5) inactivated all the BV-associated bacteria tested, and had no detectable effect on the vaginal lactobacilli. Also, the addition of 10% VF did not block the microbicidal activity of lactic acid.

Under optimal, anaerobic growth conditions, physiological concentrations of lactic acid inactivated BV-associated bacteria without affecting vaginal lactobacilli, whereas physiological concentrations of H₂O₂ produced no detectable inactivation of either BV-associated bacteria or vaginal lactobacilli. Moreover, at very high concentrations, H₂O₂ was more toxic to vaginal lactobacilli than to BV-associated bacteria. On the basis of these in vitro observations, we conclude that lactic acid, not H₂O₂, is likely to suppress BV-associated bacteria in vivo.