Neurodevelopmental Outcomes and In-Utero Antiretroviral Exposure in HIV-Exposed Uninfected Children

Micah Piske; Matthew A. Budd; Annie Q. Qiu; Evelyn J. Maan; Laura J. Sauvé; John C. Forbes; Ariane Alimenti; Patricia Janssen; Hélène C.F. Côté


AIDS. 2018;32(17):2583-2592. 

In This Article

Abstract and Introduction


Objectives:To assess and compare neurodevelopmental disorders in HIV-exposed uninfected (HEU) and HIV-unexposed uninfected (HUU) children in British Columbia, Canada. To determine associations between these outcomes and in-utero exposure to antiretroviral drugs.

Design: Retrospective controlled cohort study.

Methods: Data were collected on 446 HEU children and 1323 HUU children (matched ~1 : 3 for age, sex, and geocode) born between 1990 and 2012. Multivariable logistic regressions determined odds ratios of neurodevelopmental disorder diagnoses.

Results: HEUs had three times higher odds of being born preterm (P < 0.0001), and a more than two-fold increase in odds for autism, disturbance of emotions, hyperkinetic syndrome, and developmental delay compared with matched HUUs (P < 0.02) in unadjusted analysis. This association was reduced [adjusted neurodevelopmental disorder odds ratio (AOR) = 1.67; 95% confidence interval: 1.12–2.48; P = 0.011] after adjusting for maternal substance use and/or smoking (children born after April 2000). Regardless of antiretroviral exposure type (i.e. none, treatment with one or multiple drug classes), HEUs had higher odds of any neurodevelopmental disorders compared with matched HUUs; however, there was no evidence suggesting any specific classes of antiretroviral drugs or exposure durations increased their likelihood of neurodevelopmental disorders.

Conclusion: The results suggest no adverse associations between antiretroviral drugs and neurodevelopmental disorders within antiretroviral-exposed HEU children in our cohort. Prevalence of neurodevelopmental disorders is higher in HEUs; however, maternal substance use plays a role, as could other environmental factors not captured. These findings highlight a need for holistic support for pregnant women as well as careful developmental monitoring of HEUs past infancy, and access to early interventions, particularly among those born preterm and those exposed to addictive substances.


Globally, over half of all adults living with HIV (LWH) are women, giving birth annually to over 1.4 million children;[1] a majority of whom are HIV-exposed but uninfected (HEU).

Antiretroviral treatment for prevention of vertical transmission of HIV began in 1994, with zidovudine monotherapy during pregnancy[2] and has evolved from single drug class regimens, to combination antiretroviral therapy (cART). Over 97% of mothers LWH in Canada had received cART during pregnancy in 2014,[3] whereas guidelines for HIV-exposed infants recommended 6 weeks of prophylaxis after birth.[4] Recent guidelines for pregnant women LWH indicate all women receive lifelong cART.[4,5] With access to at least 4 weeks of cART prior to delivery, and without breastfeeding, less than 1% of infants born to women LWH acquire HIV in Canada.[4]

Many antiretroviral drugs affect human DNA and have been associated with various adverse effects,[6–8] whereas concerns remain that prenatal antiretroviral exposure could impair infant development. HEU studies, however, require careful consideration of sociodemographic and maternal risk factors which likely influence infant developmental health and are often prevalent among families affected by HIV.

In addition, preterm birth, low socioeconomic status, and prenatal exposure to opiates are known predictors of poorer cognitive outcomes in children.[9–12] A 2006 study of cART-exposed HEUs aged 18–36 months reported similar development and adaptive behavior scores to controls but only after adjusting for maternal substance use, confirming the influence of this exposure.[13] Few studies investigating neurodevelopmental outcomes among HEUs in high-income settings have accounted for such factors. The Pediatric HIV/AIDS Cohort Study reported lower cognitive and academic achievement scores among HEU youth in the United States compared with population norm means after adjusting for demographic covariates; however, the study lacked in-utero exposures data and demographically matched HIV-unexposed controls.[14] A Canadian study raised the possibility of negative effects on HEU neurodevelopment but did not observe an association with maternal substance use.[15] Our retrospective pilot study of HEU children in British Columbia showed concerning prevalences of autism, behavioral problems, and developmental delay.[16] However, studies in resource-limited settings, in which maternal substance use may be less prevalent, saw no difference in developmental outcomes between HEUs and HIV-unexposed uninfected (HUUs).[17,18] Taken together, these findings reflect environmental and maternal risk factors may influence and confound studies on HEU development.

Fetal genotoxicity, mitochondrial toxicity, lower infant birthweight and length, developmental delay, in addition to heart and neurological defects have been reported in children with prenatal antiretroviral exposure.[19–22] Several studies in both high-income and resource-limited settings have suggested a relationship between antiretroviral regimens and preterm birth,[21,23–26] an outcome that is associated with a spectrum of neurological disorders.[27–29] Maternal substance use involving alcohol, tobacco, and cocaine is also documented as a risk factor toward preterm birth.[30,31]

Long-term effects of perinatal exposure to antiretroviral drugs and a maternal HIV 'milieu' are largely unknown. Longitudinal studies in children exposed to HIV/antiretrovirals are confounded by factors disproportionately affecting families LWH, and often lack scope, statistical power, and/or length of follow-up past infancy.

The current study utilizes retrospective analysis from a controlled cohort of HEU children born in British Columbia, Canada, to capture long-term neurodevelopmental outcomes as diagnoses of neurodevelopmental disorders relative to HUU children. We hypothesize HEUs will have increased odds of neurodevelopmental disorders compared with HUU children. Possible associations between maternal substance use and antiretroviral exposures (type and duration) and these disorders were also investigated.