Archives
The reduction in mother to child transmission MTCT of HIV
The reduction in mother-to-child transmission (MTCT) of HIV is one of the greatest success stories of the antiretroviral therapy (ART) era. However, evidence is accumulating that, despite avoiding HIV, HIV-exposed uninfected (HEU) infants have more infectious morbidity and mortality than HIV-unexposed infants. Because antenatal HIV prevalence is not decreasing as rapidly as reductions in MTCT, the number of HEU infants is increasing. Evidence-based interventions for this population are therefore important to improve child health outcomes, particularly in sub-Saharan Africa. WHO guidelines recommend co-trimoxazole prophylaxis for all infants of HIV-infected mothers, regardless of infant HIV status, from age 4–6 weeks until the end of HIV exposure (usually the end of breastfeeding). This approach was first recommended in 2000, with the aim of preventing mortality from pneumonia in infants who become infected. In HIV-infected children, co-trimoxazole also reduces invasive bacterial infections; given the increased burden of morbidity and mortality in HEU infants, particularly from pneumonia, co-trimoxazole might have similar benefits for this population. However, the evidence thrombin inhibitors for co-trimoxazole prophylaxis in HEU infants is weak, and current guidelines have therefore been debated. Co-trimoxazole reduces the incidence of malaria in HEU infants in malaria-endemic regions, but the benefits for HEU infants in non-malarial regions remain unclear. In areas with high antenatal HIV prevalence, such as KwaZulu-Natal, South Africa, as many as 40% of infants are exposed to HIV and therefore receive long-term co-trimoxazole. Such widespread use of prophylaxis drives antimicrobial resistance at a population level, increases costs and the complexity of supply-chain logistics for health systems, and increases individual risks such as potential toxicity and adverse effects on the microbiota during a developmentally sensitive period. Given the uncertain risk–benefit ratio, further evidence is needed to guide recommendations for prophylaxis in HEU infants. The clinical trial by Shahin Lockman and colleagues in is therefore an important and timely addition to the literature. This randomised, placebo-controlled, double-blind trial in 2848 HIV-exposed infants in a non-malarial region of southern Botswana found no survival benefit of co-trimoxazole prophylaxis up to age 18 months, and recruitment was stopped early on the advice of the data safety monitoring board. This was a well conducted trial, with high retention; however, infant mortality was much lower than anticipated in the placebo group and the study was therefore only powered to detect a 50% reduction in mortality between groups. Nevertheless, results were consistent across a range of outcomes (eg, hospital admission, pneumonia, diarrhoea), and the reduction in the prespecified composite secondary endpoint of death, hospital admission, or grade 3–4 clinical diagnoses was not significant. Of note, only 20% of infants in this trial were breastfed because free formula is provided to HIV-infected mothers in Botswana; although no evidence of a difference in the efficacy of co-trimoxazole by feeding method was found, the trial was not powered to assess this interaction. A non-inferiority trial is currently underway in South Africa (also a non-malarial region) among breastfed HEU infants, who are being randomly assigned to receive co-trimoxazole or not; however, an even smaller benefit of co-trimoxazole in breastfed HEU infants might be expected than in Lockman and colleagues\' trial, given the significant reduction in all-cause mortality associated with breastfeeding. What do these results therefore mean for policy makers? On the basis of this trial, Lockman and colleagues suggest that, in non-malarial settings with low HIV transmission and relatively low child mortality, co-trimoxazole prophylaxis might not be needed. This recommendation seems logical in view of the trial findings. Even in other settings, the only consistently reported benefit of co-trimoxazole for HEU infants is reduced malaria. The anti-malarial properties of co-trimoxazole would probably confer benefits even to HIV-unexposed infants, and are insufficient justification for long-term prophylaxis. However, in many regions, co-trimoxazole might continue to have important benefits for the 16% of HIV-exposed infants who still acquire HIV during pregnancy, delivery, or breastfeeding. Although the obvious solution is to improve MTCT prevention and early infant diagnosis, there is still some way to go: in 2013, only 44% of pregnant women in low-income and middle-income countries received an HIV test, and only 42% of HIV-exposed infants underwent HIV testing within 2 months of birth. Furthermore, definitive HIV testing at the end of the breastfeeding period is frequently missed. Until coverage of interventions aimed at paediatric HIV elimination is improved, co-trimoxazole prophylaxis might continue to be important in many countries.