University of California, San Francisco Logo

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

Breastfeeding with maternal antiretroviral therapy or formula feeding to prevent HIV postnatal mother-to-child transmission in Rwanda
Global Health Sciences Literature Digest
Published May 10, 2010
Journal Article

Peltier CA, Ndayisaba GF, Lepage P. Breastfeeding with maternal antiretroviral therapy or formula feeding to prevent HIV postnatal mother-to-child transmission in Rwanda. AIDS. 2009 Aug 28.

In Context

Although perinatal transmission of HIV has been significantly reduced with prevention to mother-to-child transmission (PMTCT) programs in resource-constrained countries, (1, 2) postnatal transmission from breastfeeding reduces the efficacy of such programs.(3, 4) Formula feeding infants born to HIV-infected women, while reducing postnatal HIV transmission, is associated with increased mortality.(5, 6)


To assess the nine-month HIV-free survival of children using two strategies for PMTCT: breastfeeding with maternal antiretroviral therapy (ART) or formula feeding


Four public health centers in Rwanda: two urban, one semi-rural, and one rural

Study Design

Nonrandomized interventional cohort study


Between May 2005 and January 2007, all women entering PMTCT programs at study clinics and consenting at 28 weeks of gestation were included.


  1. incidence of mother-to-child HIV transmission;
  2. infant mortality;
  3. incidence of HIV-free survival: HIV infection or death, whichever came first.


Enrolled women received ART from 28 weeks of gestation and women with CD4 counts of <350 cells/µL or who met the World Health Organization (WHO) clinical criteria of stage 4 disease were eligible for lifelong treatment. Women electing formula feeding were counseled on proper preparation and were provided with free formula for six months. Women in the breastfeeding group were instructed to breastfeed exclusively for six months followed by rapid weaning. A supplemental food mixture was provided for one month while the infant was weaned. Breastfeeding women were given ART until seven months. All infants received nevirapine at birth and zidovudine for seven days. Cotrimoxazole was given from six weeks until nine months and was continued for HIV-infected infants.

Women and infants were followed at 15 days, 6 weeks, and 3, 6, 7, and 9 months. Adherence to ART and feeding method was assessed by interview and clinical examination. Breastfed infants who received any liquids or solids were classified as having received mixed feeding. Mixed feeding was an option for women who had difficulty breastfeeding their infants. All women had measurements of CD4 count and HIV-RNA viral load at delivery and six months post-partum.

Infants were tested for HIV DNA PCR at three, seven, and nine months. Infants positive for HIV DNA PCR within 48 hours of birth were considered to have been infected in utero. Infants who were negative at 48 hours and positive at six weeks were considered to have been infected peripartum. Infants who were negative at birth and positive after 15 days were considered to have become infected postnatally. All initially positive tests were confirmed.


A total of 562 HIV-positive women enrolled; 240 (42.7%) chose to breastfeed and take ART, and 322 (57.3%) elected to formula-feed their infants. After loss to follow-up, stillbirths, and deaths, 532 women-infant pairs were followed. Breastfeeding women were younger and had higher CD4 counts than the mothers who chose to formula feed. Formula-feeding mothers were more likely to have started lifelong ART. All women on ART remained on treatment throughout the study and adherence to exclusive breastfeeding was 94.2%.

There were seven children infected with HIV; six were infected in utero and one breastfed infant was infected between three and seven months. In the breastfeeding group, the cumulative probability of HIV-1 transmission at six weeks and nine months was 1.3% (95% confidence interval [CI]: 0.4-4.1%) and 1.8% (95% CI: 0.7-4.8%), respectively. In the formula-feeding group, these cumulative probabilities were similar at six weeks and nine months and estimated to be 1% (95% CI: 0.3-3.0%). Over the first nine months, the probability of HIV-1 transmission was not statistically different between both groups (log rank test, P=0.43). The one infant who acquired HIV infection in the breastfeeding group represented a cumulative risk of postnatal infection of 0.5% (95% CI: 0.1-3.4%; P=0.24) at nine months of life.

The mother of the infant who became infected had a number of factors that suggest poor absorption of ART and less than optimal adherence (severe gastritis, fasting, and weaning at five months). By nine months of age, seven (3.1%) children had died in the breastfeeding group and 17 (5.6%) in the formula-feeding group. For the 22 infants who died before nine months of age and were HIV-negative at birth, a negative polymerase chain reaction (PCR) result was available for all of them by three months prior to death. The nine-month cumulative probability of death for the breastfeeding group was 3.3% (95% CI: 1.6-6.9%) and 5.7% (95% CI: 3.6-9.2%) for the formula feeding group, with no statistically significant difference (log-rank test, P=0.20). Among the 532 exposed liveborn children, 29 were HIV-1 infected or dead at nine month of age. After adjustment for possible confounders, HIV survival was essentially the same in the breastfed and formula-fed groups (relative hazard 1.2, 95% CI: 0.2-2.9).


Providing ART to breastfeeding mothers offers an alternative to formula feeding among HIV-infected women in resource-constrained areas.

Quality Rating

This was a well-designed and well-conducted study. Although randomization would have strengthened the study, ethical issues prevented such a design. In addition, adherence to a feeding strategy not personally selected would likely have been compromised. The authors reported on loss to follow-up and used an intention to treat (feed) approach to the analysis, providing additional quality to the study.

Programmatic Implications

Feeding recommendations for HIV-infected mothers in sub-Saharan Africa have been difficult to develop. While avoidance of breastfeeding can eliminate the potential of postnatal HIV transmission, in such areas use of formula has been associated with increased morbidity and mortality. In Africa, ART is readily available and thus treating breastfeeding women for six months offers an alternative to formula feeding. It is worth noting that the mortality rates in the formula-fed infants was lower than reported in other studies, suggesting a benefit to carefully counseling women on proper formula preparation. Programs for PMTCT may wish to offer ART to breastfeeding HIV-infected mothers and to adopt better counseling and education for women electing to formula feed their infants.


  1. Coovadia HM, Rollins NC, Bland RM, et al. Mother-to-child transmission of HIV-1 infection during exclusive breastfeeding in the first 6 months of life: an intervention cohort study. Lancet 2007; 369:1107-16.
  2. Thior I, Lockman S, Smeaton LM, et al. Breastfeeding plus infant zidovudine prophylaxis for 6 months vs formula feeding plus infant zidovudine for 1 month to reduce mother-to-child HIV transmission in Botswana: a randomized trial: the Mashi Study. JAMA 2006; 296:794-805.
  3. Coutsoudis A, Dabis F, Fawzi W, et al. Late postnatal transmission of HIV-1 in breastfed children: an individual patient data meta-analysis. J Infect Dis 2004; 189:2154-66.
  4. Mbori-Ngacha D, Nduati R, John G, et al. Morbidity and mortality in breastfed and formula-fed infants of HIV-1-infected women: a randomized clinical trial. JAMA 2001; 286:2413-20.
  5. Kagaayi J, Gray RH, Brahmbhatt H, et al. Survival of infants born to HIVpositive mothers, by feeding modality, in Rakai, Uganda. PLoS ONE 2008; 3:e3877.
  6. Rollins NC, Becquet R, Bland RM, Coutsoudis A, Coovadia HM, Newell ML. Infant feeding, HIV transmission and mortality at 18 months: the need for appropriate choices by mothers and prioritization within programmes. AIDS 2008; 22:2349-57.