Treatment of HIV-Positive Pregnant Women Lowers Infection Risk of Infants

M. Klitsch

First published online:

When HIV-positive women are treated in late pregnancy and during labor with a short course of zidovudine, their child's likelihood of later becoming infected with HIV is significantly reduced, even when that child has been breastfed.1 Overall, about two years after being born, children of HIV-infected mothers who had been treated with zidovudine around the time of delivery were about 25% less likely to be HIV-positive than children of women who had received a placebo. The protective effect of zidovudine treatment, however, could be detected only among the children of women who had had higher levels of circulating CD4 cells at study enrollment; the relative reduction in HIV risk among these children was about 60%. In contrast, mother-to-child transmission was unaffected when the mothers had had lower circulating levels of CD4 cells, indicating more advanced immunosuppression.

Postnatal maternal transmission of HIV to a newborn child is a continuing problem in the developing world, particularly in areas such as Africa where breastfeeding is common. Research has shown that maternal use of the drug zidovudine around the time of delivery substantially reduces mother-to-child transmission of HIV when the mother is not breastfeeding her newborn. The long-term benefits of zidovudine treatment are not as clear, however, for breastfed infants.

To assess zidovudine's effectiveness at preventing maternal transmission of HIV in a breastfeeding population, researchers pooled data collected in two studies conducted by France's National AIDS Research Agency (ANRS), the Côte d'Ivoire Ministry of Health and the U.S. Centers for Disease Control and Prevention (CDC). Both were double-blinded, placebo-controlled randomized trials: One, sponsored by the ANRS, was conducted in Abidjan, Côte d'Ivoire, and in Bobo-Dioulasso, Burkina Faso; the other, sponsored by the CDC, took place only in Abidjan.

Pregnant HIV-positive women were recruited to participate in the studies between September 1995 and February 1998. Beginning at 36-38 weeks of gestation, all were instructed to take a tablet (either 250 or 300 mg of zidovudine or a placebo) twice per day until labor began. At this point, the studies diverged slightly: In the ANRS study, women in the treatment group took one dose of 500-600 mg of zidovudine during labor and then took the same amount each day for seven days after delivery; in the CDC study, women in labor were given 300 mg of zidovudine every three hours until they delivered their baby, but received none thereafter. The newborns were not treated in either study.

Researchers performed lymphocyte counts on blood samples taken from the women at enrollment. Blood was taken from the infants one week after delivery, at age four or six weeks, at three months of age and every three months thereafter. Blood samples were tested for the presence of HIV DNA; infants were deemed HIV-negative if they had no evidence of infection 60 days or more after breastfeeding had ended.

Over the period of the study, the researchers enrolled 701 pregnant HIV-positive women (421 in the ANSR study and 280 in the CDC study). Of these, 678 delivered and 23 were lost to follow-up. When stillbirths and second births among twin deliveries were excluded, there were 662 live-born infants. HIV status could not be determined for 21 of these children. Among the remaining 641 infants, 13 had been bottle-fed from birth or their age at weaning was unknown, leaving 628 breastfed children. Nearly half of the mothers in the study had had CD4 cell counts at enrollment of fewer than 500 per ml, a level indicative of advanced immune system disruption.

By two years after delivery, 68 children born to 319 women who had received zidovudine were HIV-infected, compared with 94 children born to 322 women who had been given a placebo. The cumulative proportions of children who were HIV-infected were 23% in the zidovudine group and 30% in the placebo group. Thus, use of zidovudine appears to have reduced the overall risk of HIV transmission by 26%.

When the researchers examined the results according to the mothers' CD4 counts, however, they found a substantial difference. Among women who had had lower levels of CD4 cells (fewer than 500 per ml) when they entered the study, the 24-month HIV infection levels among their children did not differ by treatment status: 40% in the treatment group and 41% in the placebo group. In contrast, infection rates at 24 months among children whose mothers had had higher CD4 levels (at least 500 per ml) were significantly lower in the treatment group (9%) than in the placebo group (22%). Thus, in this group, the efficacy of zidovudine treatment in reducing the overall risk of mother-to-child HIV transmission appears to have been 59%. There was no difference between the two trials in the impact of treatment.

Overall mortality was substantially lower at 24 months among the children whose mothers had been treated with zidovudine than among those whose mothers had received a placebo. Thirty-nine children in the treatment group had died by 24 months after birth (12%), compared with 68 children in the placebo group (21%).

The researchers observe that the efficacy of zidovudine in preventing HIV-1 infection or death was similar to its efficacy in preventing infection alone. In addition, they point out that the risk of postnatal transmission from untreated mothers with lower prepartum CD4 counts at study enrollment was substantial, while the risk of transmission from untreated mothers with higher CD4 counts was relatively low.

"It is indeed encouraging," the investigators comment, "to confirm a substantial reduction of mother-to-child transmission in breastfeeding women with high CD4 cell counts." They also note, however, that "continued antiretroviral therapy for mothers after giving birth, and perhaps also for their infants, should be considered, especially for mothers with lower CD4 cell counts." They warn that the efficacy of a short-course zidovudine regimen for breastfeeding African women with somewhat advanced immunodeficiency may be limited, and that the overall public health impact of such treatment may depend on the level of immunodeficiency in the population. Nevertheless, while further study of different regimens is necessary, they conclude that "wide-scale implementation of programs to prevent mother-to-child transmission of HIV-1 in Africa should not be delayed, as there are already many obstacles to surmount" if recently promulgated United Nations goals for reduction of mother-to-child transmission are to be achieved.--M. Klitsch


1. Leroy V et al., Twenty-four month efficacy of a maternal short-course zidovudine regimen to prevent mother-to-child transmission of HIV-1 in West Africa, AIDS, 2002, 16(4):631-641.