Background:  Use of antiretrovirals (ARV) during pregnancy for treatment of the mother’s own HIV infection or for prevention of mother-to-child transmission may be associated with adverse events in HIV-exposed but uninfected children. We assessed temporal trends in the use of ARV during pregnancy in the Surveillance Monitoring for ART Toxicities (SMARTT) study cohort.

Methods:  SMARTT is a cohort study at 22 sites in the US and Puerto Rico that enrolls HIV-uninfected children born to HIV-infected mothers. ARV regimens were classified as follows:  none; zidovudine (ZDV) alone; 2 nucleoside reverse transcriptase inhibitors (NRTI); combination ARV regimens representing HAART; and other. A multivariate logistic regression model was used to assess associations of maternal and other factors with use of HAART during pregnancy.

Results:  Prenatal ARV exposure data were available for 1621 HIV-exposed but uninfected children born between 1995 and 2009. Prenatal HAART exposure increased from 21% in 1997 to 91% in 2009, while use of ZDV alone decreased from 58% in 1997 to <1% from 2003 to 2009. In 1997, 8.3% of children had no prenatal ARV exposure, but this decreased to only 2.0% from 2007 to 2009. Of children born in 2009, 99% had prenatal exposure to NRTI (most commonly ZDV 70%, lamivudine 69%, tenofovir 42%, emtricitabine 40%, and abacavir 22%). Exposure to protease inhibitors increased from 19% in 1997 to 87% in 2009 (most commonly ritonavir 81%, lopinavir 57%, and atazanavir 22%). Exposure to non-NRTI declined from 33% in 2003 to 11% in 2009. Children with the first maternal plasma HIV RNA concentration (viral load) during pregnancy >10,000 copies/mL had increased odds of prenatal HAART exposure (OR 1.7, 95%CL 1.2 to 2.5, P <0.001) compared to those with viral load <1000 copies/mL. Children born after 2002 had greater odds of HAART exposure than those born between 1998 and 2002 (OR 1.9, 95%CI 1.4 to 2.6, P <0.001), as did children whose mothers had a first trimester CD4 or viral load measurement compared to those with first measurement during the third trimester (OR 1.8, 95%CL 1.1 to 3.0, P = 0.03). Neither race/ethnicity nor substance use during pregnancy was significantly associated with HAART exposure.

Conclusions:  ARV use during pregnancy continues to evolve, and children with prenatal exposure to ARV should be monitored for possible long-term effects. Prenatal HAART exposure has increased, but still is not universal. Higher maternal viral loads are associated with HAART exposure.