Diarrhoea in uninfected infants of HIV-positive mothers who stop breastfeeding at 6 Months
Polly Clayden, HIV i-Base
Athena Kourtis and coworkers present findings from a study, evaluating diarrhoea among uninfected infants of HIV-positive mothers in the Breastfeeding Antiretrovirals Nutrition (BAN) Study currently ongoing in Lilongwe, Malawi. 
BAN is a randomised, controlled clinical trial, of an ART and a nutritional intervention among breastfeeding HIV-positive mothers with CD4 counts over 200 cells/mm3 and their infants. Mothers exclusively breastfeed followed by rapid weaning by 28 weeks. Mothers and infants are randomised to a maternal or an infant ART (maternal: NVP or PI containing regimen; infant: NVP only) or to standard of care during breastfeeding. In addition, mothers are randomised to receive or not a nutritional supplement during breastfeeding. Approximately 170 women in each arm were included in this interim analysis, final numbers will be 371 in each arm.
In this preliminary analysis, the investigators looked at rates of infant diarrhoea and of hospitalisation and death due to diarrhoea, in HIV-uninfected infants enrolled in the BAN trial to-date. These data were compared with national Malawi data in infants who follow local feeding practices according to which infant diarrhoea prevalence is 9.2% in the first 6 months, rising to 41.2% between 6 to 11 months (with prolonged breastfeeding).
The investigators reported that between April 2004 and June 2006, 771 HIV-uninfected infants had been enrolled in BAN, of whom 225 uninfected infants had reached 28 weeks of age. They noted an abrupt rise around the time of weaning that continued through the end of the first year of life. Over one-third had diarrhoea in the 8th and 9th months of life, and nearly two-thirds in the 10th month. Hospitalisations due to diarrhoea/dehydration also peaked around the 8th month. There was a higher probability of infant diarrhoea in the rainy, compared to the non-rainy season (p <0.001).
The overall HIV-uninfected infant mortality (8/1000 at 1 month, 20/1000 at 6 months and 43/1000 at 12 months) was much lower than that reported in the Malawi Demographic and Health Survey (76/1000 live births). Additionally, mortality due to diarrhoea was low in this cohort (2/1000 and 1 month, 2/1000 at 6 months and 14/1000 at 12 months) and accounted for 10% of deaths at 6 months in HIV-negative infants and 32% of all deaths at 12 months.
The investigators wrote: We observed an increase in diarrhoea during and following weaning among exclusively breastfed infants reportedly weaned at 6 months. This is consistent with the pattern seen in populations who practice prolonged breastfeeding, as this time coincides with the introduction of complementary infant foods. Greater emphasis should be placed on hygienic preparation of weaning foods and water purification in order to decrease infant diarrhoeal morbidity in resource-limited settings.
They noted that the lack of a control group with prolonged breastfeeding precludes the drawing of conclusions about the relative safety or lack of safety of an early weaning approach.
- Kourtis A, Fitzgerald D, Hyde L et al. Diarrhoea in uninfected infants of HIV-infected mothers who stop breastfeeding at 6 months: the BAN study experience. 14th CROI, 25-28 February 2007, Los Angeles. Abstract 772.