Of infants born to HIV-infected women participating in a study in Kenya, those assigned to be formula-fed were no more likely than those assigned to be breastfed to die before their second birthday, even when infection with HIV was taken into account. HIV-free survival at two years, however, was significantly more frequent among infants in the formula-fed group than among those in the breastfeeding group.1 Diarrhea, pneumonia and malnutrition were the most common causes of death, but none occurred more often in one feeding group than in the other.

To compare the effects of breastfeeding and formula-feeding on mortality, morbidity and nutrition among infants of HIV-infected women in developing countries, researchers recruited a sample of HIV-infected women from antenatal clinics in Nairobi between 1992 and 1998. Half were randomly assigned to breastfeed their newborns, while the other half were instructed to use formula; participants were followed for two years after delivery. During clinic visits, doctors performed physical examinations of the children and obtained information on their health and development; infants were tested for HIV once a month during their first year and once every three months during their second year. Infants who were sick, dehydrated or malnourished received care from study physicians or from Kenyatta National Hospital.

Overall, 425 HIV-infected women participated in the study: 212 were assigned to breastfeed and 213 to formula-feed. All women had access to drinkable water, a reliable supply of formula, training on how to properly prepare formula and access to medical care for their infants. The women in the breastfeeding group delivered 197 liveborn singletons and firstborn twins, 185 of whom were followed up for two years; the women in the formula-feeding group delivered 204 infants, of whom 186 completed the full two years. Eighty-four infants died during the course of the study, 45 in the breastfeeding group and 39 in the formula-feeding group; nearly twice as many breastfed infants as formula-fed infants became infected with HIV (61 vs. 31).

The overall mortality rates of the breastfeeding group at 12 and 24 months (15% and 20%) did not differ significantly from those of infants in the formula-feeding group (17% and 24%). Among infants who did not acquire HIV during the study, the two-year mortality rate was 10% in the formula-feeding group and 8% in the breastfeeding group; among HIV-infected infants, however, these rates were 40% and 46%, respectively. In analyses that controlled for HIV status, the mortality rates of infants in the formula and breastfeeding groups were not significant different. The proportion of infants who died or were infected with HIV within the first two years of life was significantly higher among infants in the breastfeeding group than among those in the formula-feeding group (42% vs. 30%, respectively).

Pneumonia, diarrhea and malnutrition were the most common contributing causes of death, although none of the three was significantly associated with either feeding method. The proportions of infants in the formula-feeding group who died of sepsis or of neonatal noninfectious causes (20% and 11%, respectively) were significantly higher than the proportions among infants in the breastfeeding group (2% and 0%). Within the first six months of life, infants assigned to be breastfed were significantly better nourished overall than were infants assigned to be formula-fed; after adjusting for HIV status, breastfed infants were also significantly better nourished during the first two years.

The researchers note that the study had several limitations. Not all of the mothers complied with their designated feeding method: Thirty percent of infants in the formula-feeding group had some exposure to breastmilk. In addition, the rates of morbidity may have been underestimated because of the reliance on maternal reporting of illnesses between clinic visits.

The researchers comment that "two years of follow-up was sufficient to capture any potential adverse consequences of formula feeding but not all of the adverse consequences of breastfeeding with respect to HIV-1 related mortality." They add that "Of children infected after two months of age, only 9% had died after two years but most of the remaining children would be expected to die sometime during childhood." The researchers conclude that "the use of formula to prevent HIV-1 transmission can be a safe and viable option even in resource poor settings, if maternal education, clean water, a supply of formula and access to health care are available."--J.Rosenberg


1. Mbori-Ngacha D et al., Morbidity and mortality in breastfed and formula-fed infants of HIV-1-infected women, Journal of the American Medical Association, 2001, 286(19):2413-2420.