Role of bacterial vaginosis in spreading HIV requires more study

This article is more than 16 years old. Click here for more recent articles on this topic

In many regions hardest hit by the epidemic, 15% of HIV infections may be attributable to bacterial vaginosis, estimates a report in the July 31st edition of AIDS. There is an urgent need, say investigators, to undertake prospective studies to better define the role of this common gynaecological condition in HIV transmission.

Bacterial vaginosis is the most common vaginal inflammation in women of childbearing age, and it has been associated with increased risk of several conditions including upper genital tract infections and pelvic inflammatory disease. Studies of its impact on HIV infections have yielded mixed results, with most showing a varying degree of increased risk of infection. To better understand the association, investigators undertook a meta-analysis of published reports.

The meta-analysis included 23 studies of a total of 30,739 women and resulting in 29 different estimates on 25 populations. The studies took place predominantly in Africa, but also in Thailand and the United States. Four studies were prospective and compared rates of vaginosis between women who became infected with HIV and those who remained HIV-negative during the study. The remaining studies assessed at a single time point the association of HIV and vaginosis.


bacterial vaginosis

A condition caused by the overgrowth of certain species of the bacteria that are normally present in the vagina.

prospective study

A type of longitudinal study in which people join the study and information is then collected on them for several weeks, months or years. 

odds ratio (OR)

Comparing one group with another, expresses differences in the odds of something happening. An odds ratio above 1 means something is more likely to happen in the group of interest; an odds ratio below 1 means it is less likely to happen. Similar to ‘relative risk’. 

relative risk

Comparing one group with another, expresses differences in the risk of something happening. For example, in comparison with group A, people in group B have a relative risk of 3 of being ill (they are three times as likely to get ill). A relative risk above 1 means the risk is higher in the group of interest; a relative risk below 1 means the risk is lower. 


When the statistical data from all studies which relate to a particular research question and conform to a pre-determined selection criteria are pooled and analysed together.

For all 23 studies, the pooled prevalence of vaginosis was 33%. Results varied widely, from 11% in a study of young women in the US to 70% among South African women with symptomatic sexually transmitted infections (STIs). Vaginosis was recorded by clinical criteria, laboratory diagnosis or both.

When investigators analysed the four incidence studies, they found that having vaginosis increased the relative risk of being infected with HIV by 60% (RR = 1.61; 95% CI: 1.21 - 2.13). Using a model that incorporated this result and the finding of an average prevalence of vaginosis of 30% in a population, investigators estimate that 15% of all HIV infections would be due to bacterial vaginosis. This number, they say, could rival the impact of other sexually transmitted infections, which have been reported as having a higher relative risk (in the order of 200% to 500%) but also a lower prevalence.

In more concrete terms, the investigators calculate that in a population where the baseline risk of seroconversion is 2%, one additional case of HIV infection would occur for every 80 to 90 cases of vaginosis. “These data,” they say, “suggest that greater attention needs to be given to bacterial vaginosis in the global fight against HIV infection.”

Analysing the prevalence study data proved more difficult. The studies tended to show a higher prevalence of HIV among women with vaginosis, but results were heterogeneous, with most showing an increased prevalence odds ratio from 1.1 to 3.7. However, a 1999 American study showed a decreased risk (prevalence odds ratio of 0.77), and the investigators could find no obvious factor in the study that would have led to such a result.

The investigators did note that prevalence studies including women at low risk for HIV showed a slightly stronger association between vaginosis and HIV, but this may simply be due to the fact that women in high-risk groups are more likely to become infected regardless of the presence of vaginosis.

More prospective studies about the risk of HIV infection attributable to vaginosis are needed, the investigators write in conclusion. “This information could be helpful in identifying specific sub-populations, with a stronger association between bacterial vaginosis and HIV, in whom to target bacterial vaginosis control measures.” In addition, treating vaginosis should also be explored. “Randomized clinical trials (RCT) to determine the effect of bacterial vaginosis control measures on HIV acquisition may be worth considering,” they add.


Atashili J et al. Bacterial vaginosis and HIV acquisition: a meta-analysis of published studies. AIDS 22:1493 – 1501, 2008.