Second analysis concludes that we can't eliminate the long-term possibility of HIV transmission from someone on treatment

This article is more than 10 years old. Click here for more recent articles on this topic

A study that estimates the risk that someone living with HIV and taking antiretroviral therapy could transmit the virus reports that, on the basis of the few transmissions from heterosexual partners on treatment that have been reported, it is not possible to dismiss the risk of infection as zero.

The analysis by French researchers in Clinical Infectious Diseases estimates that the highest-likely risk of HIV being transmitted is between 8.7 and 13 transmissions per 100,000 sex acts; in other words, from one in about 11,500 to one in about 7700 acts. However, the researchers stressed to that this is the highest-likely risk: the actual risk may be lower than this and could indeed be zero.

This implies that the accumulated highest-likely risk of HIV transmission would rise to 1% after between 195 and 389 occasions of sex: a couple who have vaginal sex around six times a month would take two and a half years to have sex 195 times, or five and a half years to have sex 389 times.



How well something works (in a research study). See also ‘effectiveness’.


A serodiscordant couple is one in which one partner has HIV and the other has not. Many people dislike this word as it implies disagreement or conflict. Alternative terms include mixed status, magnetic or serodifferent.

voluntary male medical circumcision (VMMC)

The surgical removal of the foreskin of the penis (the retractable fold of tissue that covers the head of the penis) to reduce the risk of HIV infection in men.


Receptive anal intercourse refers to the act of being penetrated during anal intercourse. The receptive partner is the ‘bottom’.

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.

This is the second recent study to find that the long-term risk from a partner on antiretroviral therapy (ART), while very much lower than from a partner not on treatment, may not be negligible in the long term.

The other study, by the Centers for Disease Control and Prevention (CDC) in the USA, used a mathematical model to calculate the one- and ten-year risks of HIV infection in heterosexual and gay couples. It then added in the mitigating effects of ART, condom use, circumcision and pre-exposure prophylaxis (PrEP). It used estimates of the likelihood of transmission, and the efficacy of the different prevention methods, from various studies.

The French researchers tackled the question by searching out the few actual reported cases of HIV transmission within a heterosexual couple where the partner living with HIV was on ART, and where the virus was unequivocally shown to have come from them. They then calculated the highest-likely probability of transmission from someone on ART based on these cases.

The researchers found six studies that were set up sufficiently well to document such cases. They identified four cases of viral transmission from a partner on ART during 2773 person-years in 1672 heterosexual, serodiscordant couples. (An additional 182 transmissions occurred when people were not taking ART.) Four of the studies took place in Africa and one each in Spain and Brazil. Between 70 and 100% of study participants had an undetectable viral load at various time points. At the start of the studies, sexual frequency in participants varied from three to twelve times a month; the American model assumed an unvarying frequency of six times a month.

In three of these transmissions, which were proven to come from the HIV-positive partner by genetic analysis, that partner had been taking ART for less than six months. In the fourth transmission, the person had been taking ART for less than a year. As the Swiss statement says that people who have had an undetectable viral load and no sexually transmitted infections for more than six months may be regarded as non-infectious, the researchers did two calculations for the likelihood of transmission risk, based on whether the transmission in that study had taken place less or more than six months after the start of therapy. This explains the two figures cited for the highest-likely risk of transmission of 8.7 or 13 transmissions per 100,000.

The researchers’ calculation that the chance of transmission from a partner on treatment in a heterosexual couple could rise up to 1% after 195 to 389 occasions of sex allows a comparison with the American model. The CDC estimated the ten-year risk of HIV transmission from a partner on ART to be 2%. According to the French researchers, the highest-likely risk after 720 sex acts (equivalent to ten years in the US model) was either 1.85% or 3.7% (depending on the timing of that one transmssion). This is compatible with the American estimates, though the CDC study computes an average risk of transmission from rather conservative assumptions about the efficacy of different prevention methods, while the French study computes a range of risk, from zero (the lowest-likely risk) to the uppermost-likely risk quoted.

The researchers argue that we may never be able to get a more precise answer for the long-term risk of transmission than this. Because transmission from someone on treatment is so rare, if the highest-likely ‘true’ risk is, say, one in 100,000, it would have taken the HPTN 052 study, which provided an answer of "at least 96%" for the reduction in infections conferred by ART, 27 years to establish such a fact.

This French study tells us nothing about the risk of transmission within a gay couple. The American model suggests that the long-term risks could be very much higher for gay men simply because the risk of transmission via anal sex (where the HIV-negative partner is the receptive one) is 18 times higher than in vaginal sex. But we do not know if a partner on ART is 18 times more likely to transmit HIV – because no completely undisputed and verified transmission from a partner on ART in a gay couple has been documented.

It is important to reiterate that the true likelihood of a person on fully suppressive ART transmitting the virus may be much closer to zero than these two studies suggest. Nonetheless, the ten-year risk may not be negligible, and research into even more effective prevention methods is still needed.


Supervie V et al. Heterosexual risk of HIV transmission per sexual act under combined antiretroviral therapy: systematic review and Bayesian modelling. Clinical Infectious Diseases, early online edition. See abstract here.