Bacterial vaginosis is the most common vaginal infection. Although many cases remain undiagnosed, the prevalence is estimated to be between 10 and 15 percent of all women. The condition may be asymptomatic and have no apparent sequelae or may be associated with a range of serious conditions such as pelvic inflammatory disease, preterm labor, endometritis and chorioamnionitis. Although bacterial vaginosis may be sexually transmitted, there is disagreement concerning the benefit of treating male sexual partners of infected women. Potter conducted a comprehensive review of the medical literature to establish the evidence concerning such treatment.

The author searched medical databases for randomized controlled trials of treatment of sexual partners of women with bacterial vaginosis. This search identified five randomized controlled trials, and an additional trial was identified from the references of the other studies. Four of the trials had methodologic problems. All four of these trials concluded that there was no advantage in treating the male partner. One of the remaining two trials suggested that there was benefit in the treatment of male partners, but the study was flawed in several respects, principally the large numbers of patients lost to follow-up and the criteria used to define cases and outcomes. The remaining study was a large and well-designed trial of more than 200 monogamous women treated with tinidazole. This study concluded that there was no benefit in treating the male partners of women with bacterial vaginosis.

The author concludes that the available experimental evidence suggests that there is no benefit in treating the male partners of women with bacterial vaginosis, but this conclusion is largely based on studies that have methodologic problems or were conducted on specific populations.

editor's note: Why is such an apparently simple research question so difficult to answer? Bacterial vaginosis has widely accepted criteria for case definition (Amsel's criteria), so defining cases and outcomes should not be problematic. The treatment options are also non-controversial and are generally limited to metronidazole or clindamycin. The difficulties must concern the condition itself or the status of clinical investigation of common problems in primary care. The many significant issues that surround studying a disease that is possibly related to sexuality can be overcome, as shown by the progress this country has made against sexually transmitted diseases (STDs) and human immunodeficiency virus infection. Some authors do not regard bacterial vaginosis as an STD because it can occur in women who are not sexually active and has many of the characteristics of a change in vaginal environment rather than an infection. Most probably, the explanation for the paucity of evidence to support our management of bacterial vaginosis is that “uninteresting” problems of apparently low morbidity just don't get studied.—a.d.w.