Prenatal exposure to zidovudine and risk for ventricular septal defects and congenital heart defects: data from the Antiretroviral Pregnancy Registry.

OBJECTIVE

To assess the risk for ventricular septal defects and congenital heart defects following zidovudine exposure during pregnancy using data from the Antiretroviral Pregnancy Registry.

STUDY

DESIGN

Data on 16,304 prospectively reported pregnancies were analyzed to estimate the frequency and risk of ventricular septal defects and congenital heart defects, comparing exposure between zidovudine-containing regimens and non-zidovudine antiretroviral regimens. The numerator includes defect cases in outcomes at ≥ 20 weeks of gestational age. The denominator includes live birth outcomes. Infants with chromosomal anomalies were excluded.

RESULTS

There were 15,451 live birth outcomes; 13,073 were prenatally exposed to zidovudine-containing regimens and 2378 to non-zidovudine containing regimens. There were 36 ventricular septal defect cases: 31 exposed to prenatal zidovudine and 5 unexposed. Nine of the zidovudine-exposed cases had earliest exposure in the first trimester; 22 had second/third trimester exposure. Of the 5 ventricular septal defect cases not exposed to zidovudine, 2 had earliest exposure to non-zidovudine antiretroviral regimens in the first trimester, and 3 had exposure in the second/third trimester. The prevalence of ventricular septal defect was 0.24% (95% confidence interval: 0.16, 0.34) for infants exposed to zidovudine-containing regimens and 0.21% (95% confidence interval: 0.07, 0.49) for non-zidovudine regimens. The relative risk comparing the 2 was 1.13 (95% confidence interval: 0.44, 2.90). There were a total of 90 congenital heart defect cases; 78 were exposed prenatally to zidovudine-containing regimens, and 12 were unexposed. Twenty-six of the zidovudine-exposed cases had earliest exposure in the first trimester and 52 had second/third trimester exposure. Six congenital heart defect cases with non-zidovudine antiretroviral regimens had earliest exposure in the first trimester and 6 had exposure in the second/third trimester. The prevalence of congenital heart defects was 0.60% (95% confidence interval: 0.47, 0.74) for infants exposed to zidovudine-containing regimens and 0.50% (95% confidence interval: 0.26, 0.88) for non-zidovudine regimens. The relative risk comparing the 2 was 1.18 (95% confidence interval: 0.64, 2.17).

CONCLUSIONS

The prevalence and risk of ventricular septal defects and congenital heart defects among infants exposed to zidovudine-containing regimens is not significantly different from the prevalence and risk in infants exposed to non-zidovudine containing regimens.

CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT01137981.