Perinatal antiretroviral treatment and hematopoiesis in HIV-uninfected infants.

Background: The perinatal prophylactic administration of zidovudine is associated with rapidly reversible macrocytic anemia in infants. However, a recent study suggests that there may be more persistent inhibition of hematopoietic stem cells. Objective: To study hematopoiesis in uninfected infants, born to HIV-1 seropositive mothers, including those exposed and those not exposed to perinatal zidovudine alone or in combination. Methods: Longitudinal study, from 0 to 18 months, of hemoglobin, platelets, poly- nuclear neutrophils, total lymphocytes, and CD4+ and CD8+ lymphocytes in more than 4000 infants of the French Perinatal Study. Modeling of repeated measures and non-linear evolution with age, with models combining natural cubic B-splines and random effects. Results: The hemoglobin level was transiently reduced in newborns exposed to zidovudine. Multivariate analysis taking into account age, prematurity, geographical origin, maternal drug use and maternal CD4 cell count, indicated that levels of the three other lineages were slightly lower until age 18 months in exposed than not exposed infants (P , 0.0001 for each lineage).There was a negative relationship between the duration of exposure and each hematological variable. Combinations of antiretroviral treatments were associated with larger decreases than monotherapy up to 15 months of age. Similar, but less pronounced, patterns were found for the CD4+ and CD8+ subpopulations of lymphocytes. Conclusions: Zidovudine administered during the perinatal period may result in a small but significant and durable effect on hematopoiesis up to the age of 18 months.