HIV treatment may prevent at least nine out of ten transmissions

This article is more than 12 years old. Click here for more recent articles on this topic

A study of HIV transmission between long-term, HIV-serodiscordant heterosexual couples in Africa has found that the chance of transmission is reduced by at least 90% if the HIV-positive partner is on antiretroviral therapy.

As a comparison, this is better than the efficacy of 100% attempted condom use, which is in the order of 85% (with a high margin of uncertainty).

There was one transmission from a partner who was taking HIV therapy, however, and presenter Deborah Donnell said that this indicated that the advice to serodiscordant couples that they should maintain safer sex should not change, even when the HIV-positive partner was on treatment.


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.


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.

herpes simplex virus (HSV)

A viral infection which may cause sores around the mouth or genitals.

inclusion criteria

The conditions which a person must meet to join a research study.


exclusion criteria

Defines who cannot take part in a research study. Eligibility criteria may include disease type and stage, other medical conditions, previous treatment history, age, and gender. For example, many trials exclude women who are pregnant, to avoid any possible danger to a baby, or people who are taking a drug that might interact with the treatment being studied.

The proportion of couples who had unprotected sex actually decreased when the HIV-positive partner started treatment, allaying fears about behaviour change, at least in this population and in the short term.

The other important finding from this study was that untreated partners with CD4 counts under 200 cells/mm3 were approximately five times more likely to transmit HIV than those with CD4 counts over 350 cells/mm3, strengthening the case for extending antiretroviral (ARV) provision to all people with low CD4 counts.

The Partners in Prevention study

This was a substudy in the Partners in Prevention study, a large randomised controlled study designed to see if treatment for the genital herpes virus HSV-2 could reduce HIV transmission.

The main study, as reported at last year’s IAS Conference in Cape Town, found that herpes treatment was ineffective as HIV prevention.

This substudy was purely observational – it did not randomise people to HIV therapy – so its results can’t be regarded as conclusive. Donnell remarked that for that we will have to await the results of the HTPN 052 study, which is currently underway.

In the study, 3381 serodiscordant couples from seven countries from south and east Africa were included. The average age of women in the study was 29 and men 37, and two-thirds of the HIV-positive partners were women. All the HIV-positive partners had HSV-2.

At baseline about 30% of partners reported having uprotected sex with their main partner in the previous month.

None were on HIV treatment at baseline, and one of the study inclusion criteria was that the positive partner had to have a CD4 count over 250 cells/mm3. The average baseline CD4 count was over 400 cells/mm3.

CD4 counts were taken every six months and HIV status assessed. ARV therapy was ascertained by self-report: there was no independent confirmation that people were indeed on HIV therapy. Women taking short-term therapy for the prevention of mother-to-child transmission (PMTCT) were not counted as being on ARVs, and about one-third of women in fact took ARVs for this purpose at some point.

During the study 349 people, about 10% of the total, initiated HIV treatment. Approximately half of people initiating treatment had CD4 counts under 200 cells/mm3 at initiation and one-third between 200 and 350 cells/mm3.

There were 151 new HIV infections in the study. One important aspect of the study was that HIV viruses in transmitting and infected partners were sequenced to show that the new infection had indeed come from the long-term partner, and 108 were thus linked: so 28.5% of infections came from someone who was not the primary partner. Five of these 108 transmissions were excluded because the partner’s ARV status was unknown, and one because the positive partner was a woman taking ARVs for PMTCT.

Only one of the transmissions came from a partner taking ARVs.

When HIV incidence was calculated in terms of person-years of follow-up, antiretroviral users and their partners had a transmission rate of 0.39 per 100 person-years (1 case ÷ 256 person-years) (95% confidence interval [CI], 0.09-2.18). Antiretroviral non-users and their partners had a transmission rate of 2.23 per 100 person-years (102 cases ÷ 4851 person-years) (95% CI, 1.84-2.70).

This meant the relative risk of transmission from a partner taking ARVs, when adjusted for time on study and CD4 count, was 0.08; a 92% reduction in HIV transmission.

Some significant differences were observed among subsets of study participants. A higher proportion of men (12%) than women (9%) initiated antiretroviral therapy (p = 0.01). Men initiated antiretroviral therapy at a median CD4 cell count of 192 cells/mm3, while the median for women was 204 cells/mm3 (p = 0.05).

The single case of transmission involved a man who initiated ARVs 18 days before his 12-month study visit. At this visit his partner tested positive for HIV, having been negative at month 9. His CD4 count was in the 200 to 350 cells/mm3 range.

Untreated partners were far more likely to transmit HIV if they had low CD4 counts. Annual HIV incidence among HIV-negative partners was 8.79% if their partner had a CD4 count under 200 cells/mm3, 2.79 for CD4 counts between 200 and 350 cells/mm3, 1.70 between 350 and 500 cells/mm3, and 1.82 for CD4s over 500 cells/mm3.

Unprotected sex declined when partners started ARVs. Before ARV treatment, 6.2% of partners reported unprotected sex in the previous month; 3.7% reported it after treatment initiation. There was no change in sexual frequency.

This study had a number of limitations: it was not randomised, ARV status relied on self-report, and transmission and behaviour data were only followed for a maximum of two years. Using a single transmission to calculate the risk of infection by a person on ARVs involves sophisticated statistical analysis and, as noted above, very wide confidence intervals.

Audience members also commented that the incidence of sexually transmitted infections was low (as, of course, were herpes symptoms) and that a similar study needed to be conducted in gay men.

Nonetheless, Donnell commented, ARVs appear to confer a significant prevention benefit across all CD4 ranges, and this study goes some way towards quantifying that more accurately.

Further information

You can view the abstract on the official conference website.


Donnell D et al. ART and risk of heterosexual HIV-1 transmission in HIV-1 serodiscordant African couples: a multinational prospective study. Seventeenth Conference on Retroviruses and Opportunistic Infections, San Francisco, abstract 136, 2010.