Incident pregnancy and pregnancy outcomes among HIV-infected women in Uganda and Zimbabwe.

OBJECTIVE To describe pregnancy outcomes among HIV-infected women and examine factors associated with live birth among those receiving and not receiving combination antiretroviral therapy (cART). METHODS The present analysis included women with HIV from Uganda and Zimbabwe who participated in a prospective cohort study during 2001-2009. Incident pregnancies and pregnancy outcomes were recorded quarterly. The Kaplan-Meier method was used to estimate incident pregnancy probabilities; factors associated with live birth were evaluated by Poisson regression with generalized estimating equations. RESULTS Among 306 HIV-infected women, there were 160 incident pregnancies (10.1 per 100 women-years). The pregnancy rate was higher among cART-naïve women than among those receiving cART (10.7 vs 5.5 per 100 women-years; P=0.047), and it was higher in Uganda than in Zimbabwe (14.4 vs 7.7 per 100 women-years; P<0.001). Significant associations were noted between a live birth and prenatal care (relative risk 3.9; 95% confidence interval 2.2-6.9) and illness during pregnancy (relative risk 0.8; 95% confidence interval 0.7-1.0). CONCLUSION Women not receiving cART have higher pregnancy rates than do those receiving cART, but cART use might not affect the risk of adverse pregnancy outcomes. Timely prenatal care and monitoring of illnesses during pregnancy should be incorporated into treatment services for HIV-infected women.