Background:  Before the use of HAART to prevent perinatal HIV transmission, rates of congenital cytomegalovirus (CMV) were higher and postnatal rates lower among HIV-exposed infants compared to unexposed infants. This study examines perinatal CMV rates among HIV-infected and HIV-exposed but uninfected infants after HAART became standard of care for pregnant women in our clinic.

Methods:  Of 404 infants born from 1988 to 2002 to HIV-infected women cared for in our clinic, 257 (64%) were evaluated for congenital CMV infection (a positive oral or urine CMV test obtained in the first 3 weeks of life) and 385 (95%) for postnatal CMV (excluding those with congenital CMV, a positive test in the first 6 months of life). Prenatal maternal HAART was defined as triple ART. All infants with maternal HAART received at least 6 weeks of ZDV.

Results:  In our study population, 62% were Hispanic and 30% black. Mother’s median age at child’s birth was 28.1 years. Rates of congenital CMV (mean number of tests, 1.7) did not differ with HAART use (4.5% with HAART vs 3.7% without HAART) or by infant HIV infection status. However, rates of postnatal CMV transmission (mean number of tests, 3.5) were significantly (p <0.05) lower for those with HAART (2.6%) compared with those without HAART (8.9%). HAART remained significantly associated with reduced postnatal CMV (OR 0.252, 95%CI 0.073 to 0.865) controlling for infant’s gender, birth weight, and gestational age. Among the 31 HIV-infected infants, none received HAART and 23% had postnatal CMV. Among the 354 HIV-uninfected infants, postnatal CMV was lower for those with HAART (2.6%) than for those without HAART (7.1%), although statistical significance was not reached (p = 0.09). For HIV-uninfected infants without HAART, 11.8% of those with postnatal CMV had either symptoms lasting >2 months during the first year of life—failure to thrive, lymphadenopathy, splenomegaly, hepatomegaly, or neurologic impairment—compared with 1.8% of those without postnatal CMV (p <0.05). No HIV-uninfected infants with HAART had any of these symptoms.

Conclusions:  In our population, while rates of congenital CMV did not change, prenatal HAART has not only reduced the transmission of perinatal HIV, but was associated with reduced transmission of postnatal CMV. Importantly, the occurrence of any potential CMV-related medical symptoms occurring in the first year of life was also reduced in HIV-exposed but HIV-uninfected infants receiving HAART.