Most people know that an HIV-positive mother can infect her baby through breastfeeding. If the HIV-positive mother is not taking anti-retroviral drugs, the rate is about 0.5% per month.
But there are other risks with breastfeeding. Specifically, a child with HIV can infect anyone who breastfeeds him or her, whether it’s the mother or another woman. And any woman with HIV can infect any child she breastfeeds “in person” or through expressed breastmilk. HIV can be found in breastmilk or in blood or pus from cracked nipples or mouth sores.
PREVENTION RECOMMENDATION: In any community with infected women and babies, sharing babies for breastfeeding is a risk for both women and babies. At least until the HIV epidemic is quelled, sharing babies around for breastfeeding is something to avoid!
EXCEPTION: Breastmilk and wet-nurses can save babies’ lives. Breastmilk may be heat-treated to kill HIV. If a wet-nurse is required, both the wet-nurse and the baby can be tested for HIV (a baby that tests positive for antibodies to HIV should be tested for the virus, because antibodies may be from the mother, and the baby may not be infected).
HIV transmission from child to breastfeeding woman
Good evidence from investigated outbreaks in Russia and Libya, where health care infected hundreds of children with HIV, suggests HIV goes faster from babies to breastfeeding mothers than from mothers to breastfeeding babies. In Russia, 22 breastfeeding mothers got HIV from their babies. In Libya, at least 15 breastfeeding mothers got HIV from their babies..
Experts at the US Centers for Disease Control estimate that 40%-60% of Russian and Libyan children who breastfed after getting HIV from health care infected their mothers. Evidence from Russia suggests that child-to-mother transmission was faster when children had mouth sores or mothers had cracked nipples.
HIV transmission from babies to breastfeeding mothers (or to a surrogate breastfeeder) no doubt occurs in Africa, but because health care experts have generally ignored risks for babies to get HIV from health care, they similarly ignore the risk babies might infect mothers. Without outbreak investigations, there has been no chance to see or document child-to-mother transmission.
HIV transmission from breastfeeding women to others’ babies
In August 2012, the Lancet reported a baby girl in South Africa infected by breastfeeding with her HIV-positive aunt. The girl’s mother was HIV-negative. The child, who was 10 weeks old when found to have HIV, had breastfed intermittently with her aunt from age 6 weeks. HIV from the aunt and child were similar and therefore linked; one of them had infected the other. Laboratory records showed the aunt was infected more than a year before she began to breastfeed her niece, making it clear the direction of infection was from aunt to child.
The frequency of surrogate or shared breastfeeding varies from one community to another. In a 2005 national survey in South Africa, 3.5% of women reported breastfeeding another’s baby. On the other hand, 40% of women surveyed in Gabon reported sometimes breastfeeding others’ infants, and 40% of infants breastfed at times with other women.
Several studies of HIV in children with HIV-negative mothers have asked about breastfeeding with other women. For example, among 11 such children in Kenya, only one had breastfed with another woman, but the woman was not traced and tested, so the source of the child’s infection was not known.
Puzzling results a 2005 study in Free State, South Africa
A 2005 study in Free State, South Africa, reported puzzling and unconfirmed evidence that breastfeeding with a woman other than the mother is an extremely high risk for babies to get HIV. The study tested 2,457 children with HIV-negative mothers, of which 7 children were infected. According to the study, 15 of 2,457 children had breastfed with another woman. Four of these 15 children were HIV-positive; 2 had breastfed twice with another woman, and two had done so regularly. According to these data, breastfeeding with another women increased the child’s risk to be HIV-positive by 437 times (crude odds ratio = 437; see table 9 in the HSRC link below).
This result is curious. Among the 15 women who breastfed others’ babies, it is likely not more than 5 had HIV (cf: 27.9%-30.3% of antenatal women were HIV-positive in Free State during 2000-2005). If these 5 women infected 4 babies through breastfeeding, the transmission rate would be 80% through even occasional breastfeeding. This is far greater than the 35% rate of HIV transmission from mother-to-child through birth and 2 years of breastfeeding.
Is it possible that a lactating woman could infect other babies so much faster than she infects her own baby (about 0.5% per month)? Whether breastmilk comes from an HIV-positive mother or from another HIV-positive woman, it contains antibodies, and only some contains HIV.
Unfortunately, the Free State study did not trace and test the handful of women who breastfed HIV-positive children with HIV-negative mothers. Thus, the study leaves an unsubstantiated supposition – that surrogate breastfeeders are high speed transmitters.
Similarly, the study reported that children with HIV-negative mothers who took milk from a milkroom were much more likely to be HIV-positive than children who did not. But what happened? All such children had been inpatients and likely had many skin-piercing blood risks as well. Unfortunately, the study did not trace and test other inpatient children and their mothers to find others who were infected from the hospital and thereby to identify the specific pathways for HIV transmission – infusions, injections, milk, or other?
 Little KM, Kilmarx PH, Taylor AW, et al. A Review of Evidence for Transmission of HIV From Children to Breastfeeding Women and Implications for Prevention. Pediatr Infect Dis J 2012; 31: 938-942. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22668802 (accessed 27 August 2012).
 Goedhals D, Rossouw I, Hallbauer U, et al. The tainted milk of human kindness. Lancet 2012; 380: 702.
 Okinyi M, Brewer DD, Potterat JJ. Horizontally-acquired HIV infection in Kenyan and Swazi children. Int J STD AIDS 2009; 20: 852-857.