Safety of short-term breastfeeding among children born to HIV-infected mothers in Cote d’Ivoire: lessons for urban Africa?

Short-term breastfeeding lasting no more than four months is no more likely to lead to serious illness or death for children of HIV-positive mothers than replacement feeding using formula feed, and neither method of feeding leads to a higher risk of death within 18 months of birth when compared to the extended period of breastfeeding that is the norm in sub-Saharan Africa. The findings, from the Ditrame Plus study, are published today in the January edition of PLoS Medicine.

The current UNAIDS guidelines recommend the avoidance of breast-feeding by HIV-infected mothers when replacement feeding is acceptable, feasible, affordable, sustainable and safe. When breast-feeding is unavoidable, short-term exclusive breast-feeding is recommended during the first months. However, studies about the safety of these feeding practices have yielded rather conflicting results.

The findings of the Ditrame Plus study were first presented at the Second International AIDS Society conference in Paris in 2003, and it has taken three and a half years for them to appear in a peer reviewed journal. During that period a clinical trial in Botswana has suggested that formula feeding is associated with a higher risk of death in the first six months of life when compared to exclusive breastfeeding, but that difference disappeared after 18 months of follow-up

In the Ditrame Plus study, a team of Cote d’Ivoirean and French investigators assessed morbidity and mortality among short-term breast-fed and formula-fed children born to HIV-infected mothers in an urban setting with access to clean water and breast milk substitutes.

The Ditrame Plus study took place in health facilities in two districts of Abidjan, Cote d’Ivoire. Between March 2001 and March 2003, any HIV-infected pregnant woman aged 18 years or above was recruited and presented at delivery with antiretroviral prophylaxis and after delivery with a choice of nutritional interventions to prevent MTCT. The latter consisted of either complete avoidance of breast feeding or exclusive breast-feeding with early cessation from the fourth month. Mothers made a choice of feeding method based on their own personal circumstances and preference.

After birth, nutritional, psychosocial, and biological follow up of mother-infant pairs of 262 breast-fed and 295 formula-fed children were carried out up to the second birthday. At each visit, the medical staff recorded the clinical events which occurred since the last visit as well as infant feeding practices. Anthropometric measurements including height and weight were taken. Infant feeding counselling was provided whenever needed. Life-threatening illnesses were treated at a university hospital in one of the study regions.

Blood samples were collected at specified intervals until 18 months of age or until two months after complete cessation of breast-feeding. HIV diagnosis by serology was carried out in all children at 18 months. Pediatric HIV infection was defined as a positive RNA PCR at any age or positive HIV serology if aged 18 months and above. Cotrimoxazole prophylaxis was provided to HIV-infected children upon diagnosis.

Formula-feeding mothers were more educated, were less likely to have co-spouses or live in shared housing, and had access to tap water at home than breast-feeding mothers. Other baseline characteristics, compliance with the feeding practice agreed upon, and the rate of stopping follow up were comparable between the two groups.

The occurrence of adverse health outcomes, severe events, and validated morbidity were comparable between formula-feeding and short-term breast-feeding mothers even after adjusting for possible confounders. However, the incidence of diarrhoea and acute respiratory infection were significantly higher among formula-fed children while the incidence of malnutrition was significantly higher in breast-fed children.

The investigators also compared mortality rates in infants recruited from an earlier study - in which breast-feeding lasted longer (a mean duration of 8 months) - with mortality rates in the short-term breast-feeding intervention. The 18-month probability of survival was similar in formula-fed and short-term and long-term breast-fed HIV-uninfected children. The occurrence of death was not associated with the infant’s feeding option.

In summary, there was no difference in adverse events, severe illness, hospitalisation or death over two years of follow-up between short-term breast-fed and formula-fed children. The fact that the observed differences in the socio-demographic backgrounds of the mothers did not result in any significant difference in morbidity and mortality between the two groups is surprising. The authors conclude that given appropriate nutritional counselling and care, access to clean water, and an adequate supply of breast milk substitutes, early weaning and formula-feeding were safe for HIV-exposed children.

An accompanying commentary by Grace John-Stewart of the University of Washington, Seattle, published in the same journal highlights the conclusions about the safety of replacement feeding, which has been all but abandoned in many African settings. The commentary finds that the study conclusions are probably valid given the meticulous and robust analyses with large cohorts, a long follow-up, and the rigour of morbidity assessment, but she was less sanguine about the generalisability of the research cohort findings to other African settings in normal maternal-child health clinic settings.

For example, the reduced morbidity and mortality in formula-fed children might have been due to the close monitoring and surveillance during follow up of the research cohort. By comparison, some prevention of mother to child HIV transmission (PMTCT) programs have reported increased hospitalisation rates in children on replacement feeding.

Translation of the findings from a research setting to clinic setting might not be practicable since easy access to clean water and free availability of breast milk might not be available to all and sundry in Africa. The barriers to safe replacement feeding must be overcome if the safety benefits reported by Becquet et al. are to trickle down to mother-infant pairs in dirt poor African rural communities.