Up to half of infants infected with HIV in nine major studies of mother to child HIV transmission contracted the virus through breastfeeding, and the risk of infection did not fall throughout the breastfeeding period, according to a meta-analysis published in the June 15th edition of theJournal of Infectious Diseases.
In an accompanying editorial experts from the US Centers for Disease Control emphasise that antiretroviral treatment should be targeted at breastfeeding mothers with advanced HIV infection as a key element in the package of measures designed to reduce mother to child HIV transmission.
Male infants and children of mothers with low CD4 cell counts (3) were at greater risk of infection through breastfeeding, suggesting that substantial efforts will be needed to identify mothers with advanced HIV disease who would benefit from antiretroviral treatment. The reasons for the greater vulnerability of male children is still unclear.
The cumulative probability of late postnatal transmission (more than 28 days after birth) through breastfeeding at 18 months was 9.3% in a study of 4085 HIV-1 exposed, breastfed infants from nine clinical trials conducted in high prevelance resource-poor settings. Of the 24% of children infected with HIV, 42% had late postnatal transmission, with the overall risk of late postnatal transmission at 8.9 transmissions/100 child years of breastfeeding.
Making breastfeeding safer
Breastfeeding protects against malnutrition and infant morbidity and mortality in both the developed and developing world. This is particularly true in areas where the water supply is unsafe and infant mortality remains high. The cost of supplying the extra nutrients needed by a breastfeeding woman – around 500 calories a day - is far less expensive than using animal milk or formula to feed the child. It also helps to delay the return to fertility, thereby helping with child spacing. Breastfeeding is therefore good for the health of both mother and child. However, breastfeeding also carries the risk of transmission of HIV.
Mother-to-child transmission (MTCT) of HIV-1 can occur in the womb, during delivery and postnatally through breastfeeding. In many parts of the world complete avoidance of breastfeeding is not feasible, so an international collaboration set out to identify the factors that might increase the risk of transmission during breastfeeding, and also to determine whether the risk of transmission is higher during the early months of breastfeeding or not.
The meta analysis
In this study the authors conducted an individual patient data meta-analysis of transmission of HIV-1 through breastfeeding to estimate the contribution of later postnatal transmission of HIV-1 to the overall risk of MTCT and to characterise the timing and determinants of late postnatal transmission. Data from nine clinical trials of short course antiretroviral therapy for prevention of mother to child transmission were pooled in the hope that greater statisical power would shed more light on the some of the contradictory findings of previous studies.
Detailed analyses of late postnatal transmission included breastfed children with negative HIV-1 viral load tests at 4 weeks of age, some of whom eventually acquired HIV-1 infection through late postnatal transmission.
Of the 4085 children enrolled in this study, 993 (24%) were definitely infected, with similar proportions in the placebo group or the short course antiretroviral therapy group. Early acquisition of infection occurred in 314 children (32%), with 122 being positive from birth (12%). 225 (23%) had late postnatal transmission. The timing of transmission was unknown for 454 children (46%), as 343 were without a prior negative HIV-1 viral load test result whilst 111 had a negative HIV-1 test result before 28 days of age. The authors then assumed that the timing among these 454 children was similar to that for children with known timing of infection. They therefore concluded that 58% of the HIV-1 infected children acquired infection early and 42% acquired infection through late postnatal transmission. Late postnatal transmission was defined as seroconversion after day 28 in those children with a negative HIV-1 result at 4 weeks of age who were breastfed to at least 28 days of age. Therefore 223 of the HIV-1 infected children had late postnatal transmission, and 2082 remained uninfected.
A number of factors that could affect the risk of late postnatal transmission through breastfeeding were analysed including maternal age, parity and CD4 count and infant birth weight and sex. Other potential risk factors such as maternal breast abnormalities, children’s oral candidiasis and type of breastfeeding (exclusive vs. mixed feeding) could not be examined as they were not collected systematically in all participating trials.
Increased risk for male infants still unexplained
Maternal age, parity and infant birth weight were not found to be associated with transmission, but maternal CD4 count and male sex were associated with an increased risk of late postnatal transmission. Male infants may be at increased risk due to increased duration and frequency of breastfeeding when compared to female infants (although no difference in duration was found in this study).
Feeding type may also vary between the sexes, with exclusive breastfeeding more common for female infants. Exclusive breastfeeding has been shown to carry less risk of HIV transmission than mixed feeding (in which solid food and other liquids are given), probably due to early exposure to allergens in food that cause inflammation of the gut wall. Inflammation in the gut would increase the risk of contracting HIV from breast milk. Unfortunately this study was not able to analyse differences in feeding methods, frequency of feeding or volume of milk ingested.
However, the apparently higher risk of late postnatal transmission in male infants may simply reflect the fact that female infants have a higher vulnerability to HIV infection in the womb. In this meta analysis however, there was no difference in the proportion of male and female infants identified as HIV-positive by day 28.
A third possibility is that male infants are more vulnerable to HIV infection through breastfeeding because of differences in immune defences, a hypothesis that could not be tested in this analysis due to lack of data.
True risk may be underestimated
The authors point out that the risk of MTCT through breastfeeding estimated in this analysis may be an underestimation of the true risk, with the duration of breastfeeding in the mostly urban-based clinical trial populations being generally shorter than most areas of sub Saharan Africa. However, the results of the meta analysis are consistent with those of the randomised clinical trial of breastfeeding vs. formula feeding conducted in Kenya, in which 44% of MTCT in breastfeeding arm was attributable to breastfeeding (both early and late postnatal transmission).
The findings strongly support strategies being pursued by programmes such as MTCT-Plus which seek to provide short course antiretroviral treatment around the time of delivery and also treatment for mothers with advanced HIV disease, both to reduce the risk of mother to child transmission and to improve the chances that the baby will be raised by its birth mother.
References
The Breastfeeding and HIV International Transmission Study Group. The late postnatal transmission of HIV-1 in breast-fed children: An individual patient data meta-analysis. JID 189: 2154-2166, 2004.
Bulterys M et al. Prevention of mother-to-child transmission of HIV-1through breast-feeding: Past, present and future. JID 189: 2149-2153, 2004.