MORBIDITY AND MORTALITY OF HIV EXPOSED UNINFECTED CHILDREN IN SUB-SAHARAN AFRICA

Background: Several sub-Saharan Africa countries are on a fast-track to elimination of mother-to-child transmission of HIV through rapid scale-up of antiretroviral therapy (ART) programs. Peripartum-ART exposures have been associated with increased childhood morbidity. Objectives: To compare anthropometric, and clinical outcomes among exposed (maternal-HIV and ART), versus unexposed children. Methods: Prospective cohort of exposed children enrolled from the PROMISE randomized-clinical-trial (combination-ART (cART) versus non-cART), and age-and-gender-matched controls separately enrolled from child-well clinics, in Malawi, and Uganda. WHO growth-standards (2006) were used to derive weight-for-age (WAZ); length-for-age (LAZ); weight-for-length (WLZ); and head-circumference-for-age (HCAZ) z-scores; and the DAIDS toxicity tables (version 1.0, 2004/2009), to classify hematological parameters. Wilcoxan Rank-Sum/Fischer’s exact tests were used to compare variables, and Generalized-Estimating-Equations, and Cox proportional hazards models to measure associations. Results: Overall, 471(50.5%) exposed and 462(49.5%) control-children were enrolled. Ugandan exposed verses controls had lower mean-Z-scores: LAZ (p0.05). Adjusted relative-risk (RR), 95% confidence interval (CI) of stunting was higher among exposed versus control-children: 2.11 (1.14, 3.90), p=0.017, at 12-months, and 1.83 (1.03, 3.24), p=0.039, at 24-months-of-age, in Uganda; and 1.57 (1.18, 2.10), p=0.002, at 24-months-of-age, in Malawi. Relative-risk of HCAZ below WHO median was higher among exposed versus controls at 24-months-of-age, RR (95 CI) = 1.78 (1.10, 2.90), p=0.019, in Malawi; and 1.28 (0.82, 2.01), p=0.279, in Uganda. Hematological parameters, and hospitalization risks, were similar (p>0.05) across exposure groups, or more favorable among exposed versus controls. Grade 2 or higher anemia risk was lower among exposed versus control-children: adjusted RR (95% CI) = 0.33 (0.17, 0.64), p=0.001 in Uganda, and RR (95% CI) = 0.56 (0.26, 1.16), p=0.119, in Malawi. Similar trends were observed with grade 3 or higher risk. Risk-estimates were homogeneous across cART and non-cART exposure-groups (p>0.05); and in-utero versus cumulative (in-utero and postpartum)) models. Conclusions: In-utero, but not postpartum, exposures to maternal-HIV and ART, are associated with lower LAZ (including stunting), WAZ and HCAZ at 24 months-of-age. Hematological patterns and hospitalization risk were homogeneous across exposure groups