Effect of pre-existing malnutrition on growth parameters in HIV-infected children commencing antiretroviral therapy

Background: Growth in HIV-infected children generally improves with antiretroviral therapy (ART). Little is known about the effect of pre-existing malnutrition on response to treatment. Aim: To evaluate prospectively the effect of pre-existing malnutrition on growth of HIV-infected children commenced on ART compared with those without pre-existing malnutrition. Methods: Inclusion criteria were children (2 months to 8.5 years) who were commenced on ART. Exclusion criteria were pre-treatment with ART, virological non-responders and co-existing tuberculous infection. Weight-for-age (WAZ) and height-for-age Z scores (HAZ) <= -2 at the initiation of treatment were the criteria for malnutrition. Monthly height and weight measurements were made for 18 months (2007-2008) after initiation of ART. Z scores were used to express changes in standard deviation (SD) units for each of the children at 0 and 18 months of the study using WHO height and weight reference curves for age and gender. The changes in Z score were compared within the groups by paired t-test and in both groups by the Mann-Whitney U test. Results: The subjects were grouped as malnourished (G1) and not malnourished (G2) before initiation of treatment. There was a significant increase in HAZ score (mean - 0.15, p = 0.006) and WAZ score (mean - 0.09, p = 0.034) in G1 and also HAZ score (mean - 0.29, p < 0.001) and WAZ score (mean - 0.30, p = 0.001) in G2. There was a trend toward a significantly greater Z-score change in G2 than in G1 for height (p = 0.027) and weight (p = 0.046). Conclusion: In HIV-infected children, pre-existing malnutrition may impair nutritional response to ART.