BACKGROUND:Lifelong HIV and antiretroviral therapy may confer neurodevelopmental risk on the children of women with perinatally acquired HIV infection (PHIV). SETTING:We analyzed data from HIV-exposed uninfected (HEU) infants born to women with PHIV vs. non-perinatally acquired HIV (NPHIV) enrolled in the Surveillance Monitoring for Antiretroviral Therapy Toxicities (SMARTT) study. METHODS:Using the Bayley Scales of Infant and Toddler Development, third Ed. (Bayley-III), we compared neurodevelopmental outcomes at the age of 1 year in HEU infants born to women with PHIV vs. NPHIV. Those with valid Bayley-III data at the age of 1 year and a mother born after 1982 were included. Cognitive, language, and motor domains were assessed as continuous composite scores. Linear mixed effects models were fit to estimate the mean difference in Bayley-III scores between groups, adjusting for confounders. RESULTS:Five hundred fifty women with HIV gave birth to 678 HEU children (125 and 553 born to women with PHIV and NPHIV, respectively). Mean scores for each of the Bayley-III domains were not significantly different between infants born to women with PHIV vs. NPHIV in unadjusted models. After adjustment, infants of women with PHIV had lower language (91.9 vs. 94.8, P = 0.05) and motor (93.7 vs. 96.8, P = 0.03) composite scores, but no differences in cognitive composite scores. CONCLUSIONS:Cognitive domain outcomes of infants born to women with PHIV vs. NPHIV are reassuring. Differences in early language and motor functioning, while of modest clinical significance, highlight the importance of long-term monitoring of neurodevelopment in children of women with PHIV.