Infants start ART too late but improvement over time in Southern Africa

Infants start antiretroviral treatment (ART) too late – with advanced HIV and at older ages than recommended – according to data from the International epidemiologic Databases to Evaluate AIDS (IeDEA) presented at the Southern African Clinicians Society Conference 2014. But, there has been improvement since the 2010 World Health Organisation (WHO) guidelines recommended universal treatment for this age group.

Mireille Porter from the University of Cape Town presented findings from an examination of baseline characteristics of infants starting first line ART in routine care sites within Southern Africa. The study was conducted across 11 IeDEA sites in South Africa, Malawi, Zambia and Zimbabwe.

In IeDEA, routine data is collected prospectively. Sites with infant ART initiation before and after 1 January 2010 (2004-2012) were included. Infant inclusion criteria for the study were: HIV infected (with PCR diagnosis), ART naive (except for PMTCT exposure), first-line ART and recorded date before 1st birthday.

The investigators evaluated mortality, loss to follow up (no visit for >9 months), transfer out and virological suppression in the participants.

Overall the median age of 4945 infants starting ART during the study period was 5.9 months (3.7-8.7), 76.5% were WHO stage 3 or 4, their median CD4 percentage was 18.5% and 87.2% had severe immunosuppression.

There was a modest improvement in baseline characteristics over time. Infant age, proportion WHO 3 or 4, CD4 percentage and proportion with severe immune suppression during 2004-2009 vs 2010-2012 were respectively: 6.1 vs 5.4 months, 81.2% vs 63.4%, 18% vs 20,7% and 89.2 vs 81.3%.

The majority (69.7%) of infants started with d4T-contaning regimens, 15.2% started with AZT and 14.9% abacavir. Most (68.1%) received a protease inhibitor and just over half (57.9%) were exposed to antiretrovirals through PMTCT.

At three years, 39.4% of children were alive and in care. Three-year mortality probability was 13.1%; loss to follow up was 23.2% and transfer out 24.5%. In multivariate analysis, severe immune suppression, HR 2.19 (95% CI1.44-3.33), p<0.001; WHO stage 3 or 4, HR 1.36 (95% CI 1.04-1.78), p=0.023; and lower weight for age z-score less than -3, HR 2.23 (95% CI 1.78-2.8), p<0.001, were associated with higher mortality. Starting treatment from 2010 was associated with lower mortality, HR 0.75 (95% CI 0.59-0.94), p=0.015.

In a subset of 1364 infants with baseline and one other viral load result the probability of viral suppression was 21% at 6 months and 41% at 12 months. Starting treatment from 2010 was the only predictor of viral suppression.

Dr Porter noted a “very different picture to the CHER trial”. Infants continue to start treatment late and at older ages with high mortality and suboptimal outcomes.

But there was an improvement from the start of 2010, which suggests that WHO 2010 guidelines did lead to earlier initiation of ART with better outcomes.

Infants start ART too late but improvement over time in Southern Africa

Comment