Published February, 2009

Heterosexual risk of HIV-1 infection per sexual-act: systematic review and meta-analysis of observational studies, Marie-Claude Boily et al., The Lancet (2009)

This study follows up on an earlier study by the same authors examining per-act heterosexual HIV transmission probabilities. It is a systematic review and analysis of all available study data related to the likelihood of heterosexual HIV transmission. The authors reviewed 43 published studies conducted in various countries that reported per-act heterosexual HIV-1 transmission probability estimates. The authors concluded that the average male to female risk of HIV transmission is .07 - .08% per vaginal sex act (which, in a large study, would mean approximately 7-8 cases of transmission for every 10,000 acts of unprotected vaginal sex) if there was no receptive anal intercourse, the HIV-positive person was asymptomatic, and there were no other cofactors present, such as other sexually transmitted infections.

The authors' three objectives were to provide summary estimates of HIV-1 transmission probabilities per heterosexual contact; do in-depth single variable and multivariable analysis to explore the reasons for different study results; and estimate the role of risk factors such as viral load and STIs on the likelihood of transmission.

The authors point out that putting a number on the actual likelihood of HIV transmission in a single sexual act is difficult to measure. The actual transmission to a partner, the number of unprotected sex acts, the length of the partner's exposure to HIV, and other potential co-factors among the people who participate in a study about their sex acts are rarely completely known and there are unreported factors, such as some participants actually having other STIs, which could affect the accuracy of studies. Of course, this is likely true of most, if not all, studies that attempt to base conclusions on what people report about their sex lives.

The authors found that, overall, female-to-male (.04% per act or, in theory, about 4 cases of HIV transmission per every 10,000 acts of vaginal sex with a woman who is HIV positive) and male-to-female (.08% per act or 8 cases of HIV transmission per every 10,000 acts of vaginal sex with a man who is HIV positive) transmission estimates in high-income countries show a low risk of infection even when the person with HIV is not on antiretroviral treatments.

Other findings showed that there were higher estimates of HIV transmission during receptive anal sex (1.7% per act or 17 cases of HIV transmission per every 1,000 acts of anal sex in which the "top" is HIV positive) as opposed to other sexual acts. There also were larger estimated risks of HIV transmission for sexual acts during the early (9.2 to times greater) and late phases (7.3 times greater) of a partner's HIV infection than for sexual acts during the asymptomatic phase of HIV disease. Finally, the authors state that commercial sex exposure and/or genital ulcers in either sexual party increased per-act risk of infection 5.3 times in comparison to the same acts in which sex partners did not have an STI.