Three regimens of highly active antiretroviral therapy (HAART) were equally effective in suppressing HIV-1 during pregnancy and breast-feeding, according to a randomized trial in Botswana.1 The regimens, which were administered beginning no later than the 34th week of gestation and continued through up to six months of breast-feeding, suppressed the virus at similarly high rates (92–96%) and together achieved a 1.1% rate of mother-to-child transmission at six months—the lowest recorded in a breast-feeding population.

Although HAART is one of the most successful methods for reducing HIV-1 transmission during pregnancy and childbirth, its effectiveness in preventing transmission during breast-feeding has not been established. In the current trial, researchers examined outcomes for three HAART regimens in a sample of 730 HIV-1–positive pregnant women.

The 560 women with CD4 counts of 200 or more cells per ml were randomly assigned to one of two groups, each of which received a different HAART regimen beginning at 26 to 34 weeks' gestation and continuing through weaning or six months postpartum, whichever occurred first. One group (285 women) received a nucleoside reverse-transcriptase inhibitor (NRTI) regimen consisting of abac- avir, zidovudine and lamivudine; the other group (275 women) received a protease-inhibitor regimen consisting of lopinavir, ritonavir, zidovudine and lamivudine. A third group of 170 women who had CD4 counts below 200 cells per ml or had an illness indicative of AIDS were not randomized but received the standard regimen for AIDS patients in Botswana: a twice-daily cocktail of nevirapine, zidovudine and lamivudine. Treatment for this "observational group" began at 18 to 34 weeks' gestation and continued indefinitely. All infants in the study received a single dose of nevirapine at birth and a twice-daily dose of zidovudine from birth through four weeks.

Investigators evaluated women's CD4 counts at enrollment, delivery, and three and six months postpartum. They also monitored viral suppression, pregnancy outcomes and side effects in mothers and newborns. Their main goals were to compare rates of viral suppression (defined as having plasma HIV-1 RNA levels of less than 400 copies per ml) in the NRTI and protease-inhibitor groups and to determine rates of mother-to-child transmission in all three groups. They used logistic regression to identify characteristics associated with lack of suppression.

Overall, study participants had 709 live births and 24 stillbirths. Stillbirth rates were 7% in the observational group and 2–3% in the other two groups. The proportion of births that were premature was higher in the protease-inhibitor group (23%) than in the NRTI (15%) or observational group (10%). However, the NRTI, protease-inhibitor and observational groups had similar proportions of low-birth-weight infants (13%, 17% and 15%, respectively).

The median duration of HAART before delivery was 11 weeks in the two randomized groups and 13 weeks in the observational group. Nearly all of the women began breast-feeding while receiving HAART, and 71% continued for at least five months. The NRTI and protease-inhibitor groups had similarly high rates of HIV-1 suppression at delivery (96% and 93%, respectively) and throughout breast-feeding (92% and 93%, respectively). Suppression rates were also high among women in the observational group—94% at delivery and 95% throughout six months of breast-feeding. In both randomized groups, women who began the trial late in the third trimester or had a higher baseline level of HIV-1 had an elevated risk of not achieving suppression.

In an additional analysis, the researchers assessed the proportion of women who had attained a higher degree of viral suppression—i.e., to fewer than 50 copies of HIV-1 RNA per ml. This level of suppression was more common in the NRTI group (81%) than in the protease-inhibitor group (69%) at delivery, but the two groups did not differ during breast-feeding (83% and 77%, respectively). In the observational group, the rate of suppression to fewer than 50 copies per ml was 77% at delivery and 84% throughout six months of breast-feeding.

Mother-to-child transmission rates were low: Among the three treatment groups, only eight newborns (1.1%) were infected by age six months; six were infected during pregnancy and two during breast-feeding. No other study of breast-feeding women has recorded such a low rate.

The trial's limitations, the researchers note, include its lack of sufficient statistical power to detect differences among study groups in rates of mother-to-child transmission. Additionally, because only women with CD4 cell counts of 200 or more were randomized, the results may not be applicable to women with lower counts. Moreover, the study did not examine whether HAART protects against mother-to-infant transmission if women breast-feed for longer than six months.

Nonetheless, given that the rates of HIV-1 suppression in the trial matched or exceeded those in studies of nonpregnant women, the findings suggest that "neither pregnancy nor breast-feeding adversely affects achievable rates of virologic suppression," the authors note. They attribute the high suppression rates to good compliance and to the initiation of therapy before 30 weeks' gestation, a factor that also may have reduced the risk of transmission in utero. They caution, however, that the birth outcomes data from this and previous studies suggest that providers should anticipate a mild increase in prematurity among women using protease inhibitors and that pregnant women with a history of low birth-weight could experience problems while taking HAART.

A commentary accompanying the study emphasizes that HAART's success ultimately depends on the presence of services that identify, care for and treat HIV-1-positive women and their infants.2 "We now have the tools to make a considerable difference in controlling the pediatric HIV-1 epidemic," the author notes.—A. Kott


1. Shapiro RL et al., Antiretroviral regimens in pregnancy and breast-feeding in Botswana, New England Journal of Medicine, 2010, 362(24):2282–2294.

2. Mofenson LM, Protecting the next generation—eliminating perinatal HIV-1 infection, New England Journal of Medicine, 2010, 362(24):2316–2318.