Effect of multiple micronutrient supplementation on survival of HIV‐infected children in Uganda: a randomized, controlled trial

Abstract

Background Micronutrient deficiencies compromise the survival of HIV‐infected children in low‐income countries. We assessed the effect of multiple micronutrient supplementation on the mortality of HIV‐infected children in Uganda. Methods In a randomized, controlled trial, 847 children aged one to five years and attending HIV clinics in Uganda were stratified by antiretroviral therapy (ART, n = 85 versus no ART, n = 762). The children were randomized to six months of either: twice the recommended dietary allowance of 14 micronutrients as the intervention arm (vitamins A, B1, B2, niacin, B6, B12, C, D and E, folate, zinc, copper, iodine and selenium); or the standard recommended dietary allowance of six multivitamins (vitamins A, D2, B1, B2, C and niacin) as a comparative “standard‐of‐care” arm. Mortality was analyzed at 12 months of follow up using Kaplan Meier curves and the log rank test. Results Mortality at 12 months was 25 out of 426 (5.9%) children in the intervention arm and 28 out of 421 (6.7%) in the comparative arms: risk ratio 0.9 (95% CI 0.5 ‐ 1.5). Two out of 85 (2.4%) children in the ART stratum died compared with 51 out of 762 (6.7%) in the non‐ART stratum. Of those who died in the non‐ART stratum, 25 of 383 (6.5%) were in the intervention arm and 26 of 379 (6.9%) in the comparative arm; risk ratio 1.0 (95% CI 0.6 ‐ 1.6). There was no significant difference in survival at 12 months (p = 0.64, log rank test). In addition, there was no significant difference in mean weight‐for‐height at 12 months; 0.70 ± 1.43 (95% CI 0.52 ‐ 0.88) for the intervention versus 0.59 ± 1.15 (95% CI 0.45 ‐ 0.75) in the comparative arm. The mean CD4 cell count; 1024 ± 592 (95% CI 942 ‐ 1107) versus 1060 ± 553 (95% CI 985 ‐ 1136) was also similar between the two groups. Conclusions Twice the recommended dietary allowance of 14 micronutrients compared with a standard recommended dietary allowance of six multivitamins for six months was well tolerated, but it did not significantly alter mortality, growth or CD4 counts. Future intervention studies should carefully consider: (1) the composition and dosing of the supplements; and (2) the power needed to detect a difference between arms. Trial Registration ClinicalTrials.gov Identifier: NCT00122941