114. Birth Prevalence of Congenital Cytomegalovirus Infection and Language, Hearing, and Developmental Outcomes in a Cohort of HIV-Exposed, Uninfected Preschool Children

Abstract

BackgroundThe prevalence of congenital cytomegalovirus infection (cCMV) at birth is 0.5%–1% in the United States. Most cCMV newborns are asymptomatic at birth with 10%–15% subsequently developing sequelae, such as hearing loss. Higher cCMV prevalence (2.5%–11.4%) is reported in infants born to HIV-infected women, associated with maternal immune suppression and lack of antiretroviral therapy (ART), with few studies addressing neurodevelopmental (ND) outcomes in their offspring. We report birth prevalence of cCMV in a cohort of HIV-exposed, uninfected infants (HEU) born to women on combination ART with well-controlled HIV and describe ND outcomes through age 5 years.MethodsThe Surveillance Monitoring for ART Toxicities (SMARTT) study is an ongoing NICHD-funded observational multi-centered cohort study (United States and Puerto Rico) of growth and development of HEU children that commenced in 2007. As of August 1, 2017, participants with stored blood pellets collected ≤3 weeks after birth and at least 1 ND assessment ≥1 year of age had pellets tested by DNA PCR to establish cCMV. Comparisons of ND outcomes (defined in figure) at ages 1, 2, and 5 by cCMV status were made using Wilcoxon and Fisher’s Exact tests.ResultsOf 895 children meeting study criteria (55% black; 32% white; 40% Latino), 8 had cCMV, yielding a birth prevalence of 0.89% (95% CI 0.39–1.75%). All were asymptomatic and similar to CMV-uninfected infants in gestational age and anthropometric measurements at birth. The last HIV viral load prior to delivery was undetectable in 88% of women. The last available CD4% was <20% in 3/8 mothers of cCMV newborns compared with 112/873 in those without (38% vs. 13%, P < 0.07). The mean duration of follow-up (± standard deviation) of children with cCMV was 7.2 years (1.6) and those without 5.9 (2.3) years (P < 0.11). ND assessments for language development (CDI at 1, A&S at 2, TOLD-P:3 at 5), cognition (Bayleys-III at 1), intelligence (WPPSI-III at 5), and hearing (PTA at 5) did not differ by cCMV status (figure).ConclusionBirth prevalence of cCMV in HEU children born within the last decade approaches national US prevalence. Preschool HEU children with asymptomatic cCMV at birth did not show poorer language, hearing, and developmental outcomes compared with CMV-uninfected HEU children. Disclosures All authors: No reported disclosures.