We review below the evidence from the literature about the relationships between bacterial vaginosis, its treatment and pregnancy outcome. The literature indicates that there is a well-defined association between the presence of bacterial vaginosis during pregnancy and the risk of premature membrane rupture and preterm birth. Less definite is the role of the treatment of such pathology in reducing the frequency of preterm birth and/or premature rupture of membranes. The results of the controlled clinical trials are not entirely consistent. Of the most studied therapies, clindamycin seems to have shown favourable results most consistently but the published data are limited to trials including just a few hundred subjects. Only the availability of further data from controlled clinical trials will clarify the role of such treatment for bacterial vaginosis in pregnancy.