General infections

One of the most important benefits of breast milk is its ability to protect against common childhood infections such as diarrhoea, pneumonia, neonatal sepsis and acute otitis media (Golding, 1998; Duncan et al.,1993; Goldman, 1993; Ashraf et al., 1991; Huffman et al., 1990; Lucas A., 1990; Habicht et al., 1986 & 1988; Victora et al., 1987; Hanson et al., 1985). It has been assumed, but not proven, that the breast milk of HIV-infected women also protects infants against these infections.

In a study in Kinshasa of 19 infected children, development of clinical AIDS was not associated with two particular types of infant feeding practice (Ryder et al., 1991). However, morbidity was significantly higher in 237 non-HIV-infected children (of both infected and uninfected mothers) who were not exclusively breastfed, compared with 81 uninfected infants who were exclusively breastfed during the first six months of life (Ryder et al., 1991). In Durban, South Africa, exclusively breastfed infected children had a slower rate of progression to AIDS than those on mixed feeds (Bobat et al., 1997).

Two recent studies from South Africa compared partially breastfed and exclusively formula-fed HIV-infected infants (Bobat et al., 1997; Gray et al., 1996). In these studies, both groups had similar frequencies of failure to thrive, diarrhoea, and pneumonia. Uninfected infants of HIV-positive mothers also had a comparable frequency of these conditions, whether they were partially breastfed or exclusively formula-fed. However, these results should be interpreted with great caution since the failure to detect a difference in health outcomes between breastfed and formula-fed infants may reflect factors specific to these studies. These include: short duration of exclusive breastfeeding and the inclusion of infants that had stopped breastfeeding in the breastfeeding group; a relatively safe environment (water, electricity, sanitation etc.) that minimized the risks of formula feeding; and a relatively literate, urban study population with access to continual health care, as part of a research study design. It is unlikely that these findings would be replicated in studies from other settings in sub-Saharan Africa without additional support being given to women who choose not to breastfeed.

HIV infection

Breast milk contains maternal antibodies. All basic forms of immunoglobulins IgG, IgM, IgA, IgD, and IgE are present in breast milk. The most abundant is usually secretory IgA (Lawrence, 1994). The role of HIV-specific antibodies in breast milk in inhibiting HIV transmission through breastfeeding has been investigated. Breast milk in women with established HIV infection has been found to have HIV-specific IgG, with its wide spectrum of activity against HIV proteins, comparable to HIV-specific IgG in serum. The spectrum of activity of serum IgA against HIV has been found to be similar to that of serum IgG, but the spectrum of activity of HIV-specific secretory IgA (sIgA) in breast milk is directed against only a limited number of viral proteins (env protein, gp 160, core proteins).

In a study of breast-milk samples from 215 HIV-infected women in Rwanda (Van de Perre et al., 1993), the most frequently identified HIV-specific antibody in breast milk was IgG (in >95% of samples), the next was IgM (in 41-78% of samples) and the least frequent was IgA (in 23-41% of samples). Lack of persistence of HIV-specific IgM in breast milk collected at 18 months was associated with a high risk of transmission of HIV. Of 20 children receiving breast milk with detectable HIV DNA in samples collected at day 15, but without detectable IgM in later samples, 47% were infected with HIV. In those with detectable DNA in breast milk samples at day 15, and with IgM in later samples only 30% became infected. This suggests that IgM may protect against breast-milk transmission of HIV. The rate of transmission was 18% in infants of mothers whose breast-milk sample at day 15 had undetectable HIV DNA, regardless of IgM levels (Van de Perre et al., 1993).

Other components of breast milk are protective against viral infections. Human lactoferrin has been shown in vitro to have an inhibitory activity against HIV (Harmsen et al., 1995), and lipid-dependent antiviral activity directed at HIV and other enveloped viruses and bacteria has also been described (Orloff et al., 1993; Isaacs and Thormar, 1990). An additional factor that has also been identified in breast milk, possibly a sulphated protein, glycoprotein mucin or glycosaminoglycan, appears to inhibit the binding of HIV to CD4 receptors (Newburg et al., 1992).