Difficulties in implementing infant feeding guidelines

Ineffective implementation of current WHO/UNICEF guidelines for infant feeding for HIV-positive mothers might be associated with a higher risk of mother-to-child transmission (MTCT) of HIV, according to the findings of a cohort published in the August edition of AIDS. The study reported that only 13% of women were able to adhere to exclusive breastfeeding and only 29% to exclusive formula feeding.

Breastfeeding is practiced by the majority of mothers in sub-Saharan Africa and yet it contributes to 50% of all MTCT of HIV. Under ideal conditions with running water and a high standard of hygiene, exclusive replacement feeding can reduce the risk of postnatal transmission of HIV. However, in resource-poor countries with an escalating HIV problem, conditions are far from ideal.

According to published studies, the risk of MTCT due to replacement feeding depends on individual and environmental circumstances. While a Kenyan study reported that formula feeding reduced the risk of MTCT by 40%, a Botswana study reported that formula feeding increased mortality.

However, the results from a Cote D’Ivoirian study were more sanguine. Formula feeding in an urban setting with access to clean water, free healthcare, free transport to health facilities, and free formula feeding supplies, was not associated with any increased risk of deaths or illness at 24 months.

The WHO/UNICEF guidelines recommend that all breastfeeding must be avoided only if replacement feeding is acceptable, affordable, sustainable, and safe. The problem is that there are not clear guidelines on the implementation of these criteria in the field. With the scaling up of prevention of MTCT programs (PMTCT) on the continent, more and more HIV-positive African women must make appropriate infant-feeding choices based on sound infant-feeding counselling.

Individual mothers must make appropriate choices guided by effective and validated guidelines as well as high quality infant-feeding counselling. In order to identify individual and environmental criteria that might guide appropriate infant-feeding choices in resource-poor settings, a team of Swedish, US, and South African investigators followed HIV-positive South African women and their infants in order to assess vertical MTCT.

The study was a prospective cohort study which followed up mother-infant pairs in three national PMTCT sites in South Africa. The antenatal HIV prevalence at these sites ranged from 15-39%. The study participants were pregnant women attending an antenatal clinic in one of the three sites between October 2002 and September 2003. The inclusion criteria included being 16 years of age, being pregnant at least 28 weeks, agreeing to HIV counselling, and having a positive HIV test.

The PMTCT consisted of antenatal counselling and HIV testing, antenatal infant-feeding counselling, free infant formula to women who chose not to breast-feed, and single dose nevirapine. Infant feeding counselling was according to WHO/UNICEF guidelines.

Overall, 635 HIV-positive mother-infant pairs were recruited of which 19% were lost to follow up yielding a final sample size of 514. Home visits to collect data were made by trained field workers before and after birth every fortnight until twelve weeks then monthly until 36 weeks. Infant feeding intentions were recorded prior to discharge from the delivery ward. At each postnatal visit, infant-feeding practices were recorded. A verbal autopsy questionnaire was used to record infant mortality and maternal HIV status as well as maternal viral load was determined using standard procedures.

During the first postnatal visit at three weeks, the appropriateness of formula use was assessed by gathering socioeconomic data related to the WHO/UNICEF guidelines. This information included having piped water in the house or yard, disclosure of HIV status by three weeks after birth, having someone in the home being employed, and having access to a fridge for storage of the prepared formula.

When infant-feeding intentions and actual feeding practices were compared, only 13% of women who had chosen to breastfeed were exclusively breastfeeding at twelve weeks; the majority were either mixed feeding, predominantly breastfeeding, or exclusively formula feeding. Among women who chose to formula feed, 29% gave milk at some point between birth and nine months. The free formula supplies from the clinics were not reliable; instead a high proportion of mothers gave their infants breast milk, purchased formula, and sugar water.

Overall, 95 of 311 women whose socioeconomic data met the WHO/UNICEF criteria (30.5%) chose to breastfeed and 195 of 289 women who did not meet the criteria (67.4%) chose to formula feed. Infants of women who chose to formula feed without fulfilling the criteria had the highest risk of HIV transmission and death (hazard ratio, 3.63; 95% confidence interval, 1.48-8.89).

The findings of this study highlight a potential calamitous situation which occurs when the WHO/UNICEF guidelines for infant-feeding are not implemented effectively under operational conditions in the field. Counselling of mothers should include an assessment of personal and family circumstances to guide appropriate infant feeding choices, the authors say.