HIV viral load in breast milk: no easy predictors in South African study

A simple formula for predicting which mothers may be at highest risk of transmitting HIV through breastfeeding may not be available, according to a South African study reported in the February 14^th edition of AIDS.

The study set out to explore factors related to HIV viral load in breast milk and the risk of transmission from mothers to babies. It also tried to explain why babies that are exclusively breast-fed appear less likely to become HIV positive than babies fed with breastmilk and other food and drink (including water).

The study was carried out in Durban between 1997 and 1999, and recruited 145 breast-feeding HIV positive women. At several time-points after birth (1, 6 and 14 weeks) women expressed milk from each of their breasts which was then tested for HIV and indicators of low-level inflammation (mastitis) which is suspected of boosting HIV viral load in milk.

Information on the HIV status of the babies was available for 131 whose mothers took part in another research study, on the effects of vitamin A supplementation on HIV transmission. Three women left the study early, when their babies died, and 15 others were lost to contact.

While the study confirmed that there is a highly significant correlation between mastitis, as shown by the sodium/potassium ratio in the milk exceeding 1.0, it also found that this explained only a small part of the variation in viral load which was found in the different breastmilk samples.

The study also confirmed that viral load in milk varies over time and between breasts, while being strongly correlated with the mother’s CD4 count – so that women with more advanced HIV disease have higher viral load in their breasts. More advanced disease may also increase the risk of mastitis, and cause vitamin deficiencies, making cause-and-effect hard to sort out,

There were too few babies with substantially mixed feeding to be clear about the impact of this practice on viral load, but there was some evidence that poor breastfeeding practice – especially when this meant leaving milk in the breast – did boost viral load through promoting mastitis.

Another explanation for the link between mixed feeding and transmission may be that mothers who are unwell are less able to breast-feed and therefore more likely to give their babies additional water and/or milk.

Lower viral load in breast milk was signficantly associated with greater weight-gain in babies. This was also put forward as an explanation for why viral load was higher when the baby was a girl than when it was a boy, since boys fed more vigorously (and may well have been fed more by some mothers).

This study was not large enough to show the relationship between viral load in breast milk and HIV transmission, with six babies clearly infected from breastmilk during the study compared to 88 who remained uninfected. (Other babies were infected at birth or very shortly afterwards.) There was a trend, however, for babies that became infected to have been exposed to milk with a higher viral load at each of the time points when samples were taken.

According to this research, treatments (to deal promptly with infections) and other measures (such as nutritional supplements and education on how best to breast-feed babies) to prevent low-level mastitis might be expected to prevent as few as 5% of cases of HIV transmission to babies. Nonetheless, because these measures are of low cost and can benefit mothers and infants regardless of HIV status, the case for implementing them is strong.

The need for further research, including studies to look at the impact of ARV treatment of breastfeeding women on the levels of HIV in their milk, is clear.