Antiretroviral Therapy May Lower HIV Transmission Risk Within Couples

H. Ball

First published online:

Antiretroviral therapy may greatly reduce the risk of HIV-1 transmission, according to a study of serodiscordant heterosexual couples in seven African countries.1 Antiretroviral therapy was associated with a 92% reduction in the transmission rate—from 2.2 cases per 100 person-years among nonusers to 0.4 per 100 person-years among participants who had initiated treatment (incidence rate ratio, 0.08). The transmission rate was highest (8.8 cases per 100 person-years) among participants not on antiretroviral therapy who had CD4 cell counts lower than 200 per microliter (i.e., those with the weakest immune systems); it was also elevated among individuals with the highest viral loads.

The data were collected in Botswana, Kenya, Rwanda, South Africa, Tanzania, Uganda and Zambia as part of a randomized, placebo-controlled trial of aciclovir, a drug used to treat herpes simplex virus type 2 (HSV-2). The trial was designed to examine whether treatment with aciclovir reduces the incidence of HIV transmission in couples where one partner is infected with both viruses. Although aciclovir did not prevent HIV transmission, the researchers conducted a secondary analysis to see whether antiretroviral therapy (which some infected individuals began using) was associated with reduced transmission rates.

Couples in which both partners were 18 or older were eligible to participate if one partner was HIV-1 seronegative and the other had tested positive for HIV-1 and HSV-2, had a CD4 count of 250 per microliter or higher, did not have AIDS and was not receiving anti-retroviral therapy. Couples were also required to have had vaginal intercourse at least three times in the three months prior to screening. A total of 3,381 couples participated in the study for up to 24 months between November 2004 and October 2008.

Each month, HIV-1–positive participants received study drugs, a clinical examination and a behavioral risk assessment; their HIV-1 viral load was measured at baseline; at months 3, 6 and 12; and at their last study visit. In addition, their CD4 counts were assessed every six months, and those whose counts indicated a need for antiretroviral therapy (according to national guidelines) were referred to local clinics for treatment. Every three months, HIV-1–positive participants reported whether they had taken any antiretroviral drugs in the previous three months; treatment was assumed to be ongoing after initiation, whether or not participants reported it as such at subsequent visits. Uninfected partners were tested for HIV-1 every three months. All participants received condoms, STI treatment, and HIV and risk-reduction counseling throughout the study. Some 98% of HIV-1–positive participants completed at least one follow-up assessment of HIV status; 89% of uninfected partners participated for at least 12 months, and 84% participated for the full 24 months. The researchers used logistic regression to assess associations between antiretroviral therapy use and HIV-1 transmission, and genetic testing to establish whether new infections were the result of transmissions within couples.

Sixty-eight percent of HIV-positive participants were women. The average age at baseline was 32 for participants with HIV-1 and 33 for uninfected partners. CD4 cell counts were lower and viral loads higher among infected men than among infected women. Ten percent of HIV-1–infected participants began antiretroviral therapy during the trial, after a median of 13 months' participation; 13% of antiretroviral therapy users reported nonuse at a subsequent follow-up visit.

Genetically confirmed HIV-1 transmissions occurred among 102 of the 3,032 couples in which the HIV-1–positive partner did not initiate antiretroviral therapy, corresponding to a rate of 2.2 per 100 person-years. Only one case occurred among the 349 couples in which the infected partner was using antiretroviral therapy (0.4 per 100 person-years). After adjustment for time since enrollment and CD4 count, use of antiretrovirals was associated with a 92% reduction in the incidence of transmission (incidence rate ratio, 0.08). No transmissions occurred among antiretroviral therapy users with CD4 counts below 200 per microliter during 132 person-years of follow-up.

For couples who had not initiated antiretroviral therapy, HIV-1 transmission was highest (8.8 cases per 100 person-years) if the infected partner had a CD4 count below 200 per microliter; otherwise, transmission rates were 1.7–2.8 per 100 person-years. In addition, transmission rates among nonusers were positively associated with plasma HIV-1 concentrations; for example, among those with CD4 counts of 200–349 cells per microliter, transmission rates were 4.7 cases per 100 person-years for participants with plasma HIV-1 concentrations of at least 50,000 copies per milliliter, compared with 0.3–1.9 per 100 person-years among those with lower concentrations.

The proportion of HIV-1–infected participants who reported having had unprotected sex in the previous month dropped from 29% at baseline to an average of 7% at all follow-up visits. The proportion decreased even further among participants who received antiretroviral therapy: They reported unprotected sex at 6% of follow-up visits before initiation of treatment and only 4% of visits afterward (odds ratio, 0.6)

Because antiretroviral therapy was associated with reduced transmission in the sickest study participants, the researchers argue that providing antiretroviral therapy to patients with CD4 cell counts lower than 200 should be a priority "for both treatment and prevention of HIV-1." However, they recommend that countries prioritize treatment as well for those with high viral loads. In addition, they note that "counselling is needed to reinforce understanding that potential for HIV-1 transmission to partners remains after [antiretroviral therapy] initiation."—H. Ball


1. Donnell D et al., Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: a prospective cohort analysis, Lancet, 2010, 375(9731):2092–2098.