University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Heterosexual HIV-1
Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: a prospective cohort analysis
Global Health Sciences Literature Digest
Published August 6, 2010
Journal Article

Donnell D, Baeten JM, Kiarie J, et al. Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: a prospective cohort analysis. Lancet. 2010 Jun 12;375(9731):2092-8.

In Context

Plasma viral load is a major determinant of HIV transmission. Although it is known that antiretroviral therapy (ART) reduces the concentration of HIV in plasma and genital fluids,(1, 2, 3, 4, 5) little is known about the effect of viral load suppression on sexual transmission of HIV. In addition, studies of transmission risk in the presence of ART were conducted in which ART was initiated at CD4 counts of ≤350 cells/mm3. The benefits of ART for all HIV-infected persons, regardless of CD4 count, are not yet known. If this approach results in a significant decrease in the risk of sexual transmission of HIV, it may be more widely recommended.


To measure the effect of ART on sexual transmission of HIV in heterosexual, serodiscordant couples


Fourteen clinical sites in Botswana, Kenya, Rwanda, South Africa, Tanzania, Uganda, and Zambia

Study Design

Secondary analysis of data collected in a randomized controlled trial


People who were HIV and herpes simplex virus type 2 (HSV-2) seropositive and their HIV-seronegative heterosexual partners who were enrolled in a study to measure the effect of suppressive HSV-2 therapy on HIV transmission. Participants were aged 18 years and older and reported sex in the past three months. Seropositive participants had CD4 counts of ≥250 cells/mm3 at enrollment and, consistent with national treatment guidelines at that time, were not receiving ART.


HIV transmission from infected persons to their uninfected partner


HIV-infected participants were seen once a month, at which time they received study drugs (acyclovir or placebo) and assessment of clinical status and behavioral risk. CD4 counts were measured every six months. Uninfected participants were tested for HIV every three months. Participants who had CD4 counts of <200-250 cells/mm3 or World Health Organization (WHO) stage 4 disease were referred for HIV medical care and followed in the study for 24 months. Duration of ART use was collected for all participants who initiated treatment during the study.

Partners of seropositive participants were tested for HIV and, for those who had seroconverted, genetic sequencing of the env and gag genes was performed to assess transmission between partners. Laboratory technicians were not aware of the treatment and ART-use assignment of the participants.

Analysis: Use of ART was analyzed as a time-dependent variable. Any use of ART during the three-month assessment period was defined as ART use for that entire time period and as an ART-exposed period for the uninfected partner. It was assumed that participants who initiated ART continued to receive treatment for the duration of the study. Women who received a short course of ART for the prevention of mother-to-child transmission were excluded. Uninfected participants whose seropositive partners were lost to follow-up were excluded. Poisson regression analysis was used to calculate the incidence rate ratio and confidence intervals for HIV transmission, with adjustment for duration of time enrolled and CD4 counts (<200 cells/mm3 or ≥200 cells/mm3). Transmission risk was calculated overall and by CD4 cell count strata. Sexual behavior was compared before and after ART initiation. Conditional logistic regression was used to assess changes in unprotected sexual behaviors. Negative binomial regression with generalized estimating equations and robust standard error estimation was used to model the number of episodes of sex. Paired t-tests were used to compare plasma viral load at the most recent visit before starting ART and at the final visit. If viral load measurements were undetectable, a value of 120 copies/mL was assigned.


A total of 3381 heterosexual, HIV-serodiscordant couples were eligible for this analysis. Baseline CD4 cell counts were lower and viral load levels were higher among HIV-infected men compared with women. Ninety-eight percent of participants had at least one follow-up visit, and there was a total of 5017 person-years of follow-up. Among HIV-uninfected participants, retention was 89% at 12 months and 84% at 24 months. Retention was 96% among the HIV-infected participants, excepting participants excluded for receiving ART for the prevention of mother-to-child transmission.

Ten percent (349) of participants initiated ART. There were 103 genetically linked HIV transmission events for which ART status was known and which resulted in an incidence rate of 2.13 per 100 person-years. An additional 39 unlinked transmission events occurred from persons not enrolled in the study. Of the 103 genetically linked transmissions, only one was from a participant who was receiving ART. After adjustment for the time on ART, treatment was associated with a 92% reduction in the risk of transmission.

The rate of transmission from untreated persons was highest for those with CD4 cell counts of <200 cells/mm3 and was similar among three other categories of CD4 counts (200-349 cells/mm3, 350-499 cells/mm3, ≥500 cells/mm3).

The HIV concentration fell from 4.88 log10 copies/mL (3.97-5.4 log10 copies/mL) before ART initiation to <2.38 log10 copies/mL (<2.38-3.53 log10 copies/mL) at the end of the study (P<0.0001) among the 344 participants with measurements before and after treatment initiation. The median time from starting treatment until the final study visit was 7.3 months.

Sexual risk behaviors decreased after enrollment; at follow-up visits, only 7% of HIV-infected participants reported having had unprotected sex. After initiating ART, the proportion of visits during which unprotected sex was reported decreased from 6.2% to 3.7% (adjusted odds ratio 0.63, 95% confidence interval [CI]: 0.41-0.96). The mean number of episodes of sex did not change.

Among participants with CD4 counts >200 cells/mm3, the risk of transmission was greatest where viral loads were more than 50,000 copies/mL; this did not vary with higher CD4 cell counts.


Rate of HIV transmission is lower among persons receiving ART and is reduced with high CD4 counts and low viral loads.

Quality Rating

This was a high-quality cohort study. Although there is no information regarding the representativeness of the cohort, participants were included from seven countries and the sample size was large, suggesting a reasonable degree of representativeness. Ascertainment of laboratory values was unbiased, and evidence of infection and lack of infection at baseline is likely to be accurate. Reports of sexual behavior may be biased because of social desirability effects. Follow-up was sufficient to measure the outcome, but longer follow-up would be beneficial, particularly to determine changes in adherence and risk behaviors and how these then affect transmission. A major strength of the study was the genetic linking of transmission events.

Programmatic Implications

The 92% reduction in transmission risk is substantial and this result is supported by a prior meta-analysis.(6) Although the greatest absolute reduction in transmission risk occurred among persons who initiated ART when CD4 cell count was <200/mm3, reduction in transmission occurred among persons who initiated ART at higher counts as well. Given that <50% of HIV-infected persons with CD4 counts of <200/mm3 have received ART, expanded efforts to begin treatment of this group would likely provide substantial clinical and prevention benefits. In addition, the data from this study support the proposed approach of treating all HIV-infected persons, regardless of CD4 levels, which would be a prevention tool in addition to the improved health outcomes among those already infected. However, in settings with insufficient resources to treat all HIV-infected persons, ART should be prioritized for those with CD4 cell counts of <200/mm3. The increased risk of transmission among persons with CD4 cell counts of ≥200 cells/mm3 and viral loads >50,000 copies/mL show the need for viral load testing in clinical settings. As point-of-care testing options increase, better clinical and prevention outcomes can be achieved.


  1. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000;342:921-9.
  2. Chaisson RE, Keruly JC, Moore RD. Association of initial CD4 cell count and viral load with response to highly active antiretroviral therapy. JAMA 2000;284:3128-9.
  3. Graham SM, Holte SE, Peshu NM, et al. Initiation of antiretroviral therapy leads to a rapid decline in cervical and vaginal HIV-1 shedding. AIDS 2007;21:501-7.
  4. Gupta P, Mellors J, Kingsley L, et al. High viral load in semen of human immunodeficiency virus type 1-infected men at all stages of disease and its reduction by therapy with protease and nonnucleoside reverse transcriptase inhibitors. J Virol 1997;71: 6271-75.
  5. Marcelin AG, Tubiana R, Lambert-Niclot S, et al. Detection of HIV-1 RNA in seminal plasma samples from treated patients with undetectable HIV-1 RNA in blood plasma. AIDS 2008;22:1677-9.
  6. Attia S, Egger M, Muller M, Zwahlen M, Low N. Sexual transmission of HIV according to viral load and antiretroviral therapy: systematic review and meta-analysis. AIDS 2009;23:1397-404.