Neonatal Predictors of Infection Status and Early Death Among 332 Infants at Risk of HIV-1 Infection Monitored Prospectively From Birth

Abstract

Background and Methods. Differences in newborn outcome measures for human immunodeficiency virus (HIV)-1-infected and HIV-1-exposed but uninfected infants have been found in several studies, but not in others. Eighty-four infected and 248 uninfected children born to HIV-1-seropositive mothers followed prospectively in a multicenter, perinatal HIV-1 transmission cohort study were compared for differences in maternal demographics, health status, and newborn outcome measures, including delivery complications, physical examination findings, neonatal complications, and laboratory results. Results. Mothers of HIV-1-infected infants were more likely than those of uninfected infants to have acquired immunodeficiency syndrome (AIDS) diagnosed through 2 weeks postpartum (21% vs 11%, P = .04); the transmission rate for the 38 women with AIDS was 37% compared with 22% for the 245 women without AIDS. Two of 27 (7%) women receiving zidovudine during pregnancy had infected infants compared with 73 (27%) of 275 women who did not receive zidovudine (P = .033). Mean gestational age was significantly lower among HIV-1-infected (37 weeks) than among uninfected infants (38 weeks; P < .001). Infected infants had significantly higher rates of prematurity (gestational age less than 37 weeks) (33% vs 19%, P = .01) and extreme prematurity (gestational age less than 34 weeks) (18% vs 6%, P = .001) than uninfected infants. Infection was associated with lower birth weight (2533 g vs 2862 g, P < .001) and smaller head circumference (32.0 cm vs 33.1 cm, P = .001). HIV-1-infected infants were significantly more likely to be small for gestational age (26% vs 16%, P = .04) and low birth weight (less than 2500 g) (45% vs 29%, P = .006) than infants who were uninfected. Twenty-two (26%) HIV-1-infected children died during a median follow-up of 27.6 months (range 1.9 to 98.3 months). Prematurity was predictive of survival: by Kaplan-Meier, an estimated 55% (95% confidence interval, 31% to 72%) of preterm infected children survived to 24 months compared with 84% (95% confidence interval, 70% to 92%) of full-term infected children (P = .005). Conclusion. Infants born to women with AIDS are at higher risk for HIV-1 infection than are infants born to HIV-1-infected women with AIDS not yet diagnosed. Women receiving zidovudine appear less likely to transmit HIV-1 to their infants. Significantly higher rates of prematurity and intrauterine growth retardation were found among HIV-1-infected infants than among those in the uninfected, HIV-1-exposed control group. Prematurity was associated with shortened survival in HIV-1-infected infants. Measures of intrauterine growth and gestation appear to be important predictors of HIV-1 infection status for seropositive infants and of prognosis for the infected infant.