Lack of definitive severe mitochondrial signs and symptoms among deceased HIV-uninfected and HIV-indeterminate children < or = 5 years of age, Pediatric Spectrum of HIV Disease project (PSD), USA

Abstract

Background: In response to recent reports of mitochondrial dysfunction in HIV-uninfected infants exposed to antiretroviral (ARV) prophylaxis, the Perinatal Safety Review Working Group reviewed deaths in five large HIV-exposed perinatal cohorts in the United States to determine if similar cases of severe mitochondrial toxicity could be detected. We describe the results of this review for the PSD cohort.

Methods: Hospitalization, clinic and death records for deceased HIV-uninfected and HIV-indeterminate children who were less than 5 years of age were reviewed. Standard definitions were used to classify HIV infection status and the likelihood that signs and symptoms were related to mitochondrial dysfunction. Children were classified as having signs and symptoms that were considered (1) unrelated, (2) unlikely, (3) consistent with, or (4) likely related to mitochondrial disease. SIDS deaths were put into a separate category.

Results: 8,465 of 13,125 HIV-exposed children were either HIV-uninfected or HIV-indeterminate. Among the 84 deaths in the subgroup of 8,465 children, 9 were considered in Class 2 (unlikely), 4 were considered in Class 3 (consistent with), and none were considered in Class 4 (likely). 97% of those children who received ARV prophylaxis received zidovudine alone. None of the HIV-uninfected deaths were classified in 2, 3, or 4; and only one of these was exposed to ARV prophylaxis. Among the 3 HIV-indeterminate children who were classified in 3 (consistent with), 2 had no or unknown ARV exposure before 1994 when use of ZDV prophylaxis became the standard of care. Both HIV-uninfected and HIV-indeterminate children with ARV exposure or unknown exposure had lower mortality rates than children without ARV exposure.

Conclusion: Monoprophylaxis with ZDV was not associated with higher death rates in the cohort of 8,465 children or with any findings likely consistent with mitochondrial dysfunction among the 85 deaths. Ongoing monitoring of drug safety in large multi-site prospective cohort studies of HIV-exposed children is essential in the era of highly active antiretroviral therapy.