Background: The association between combination antiretroviral (cARV) therapy use by human immunodeficiency virus (HIV)-infected women during pregnancy and risk of prematurity is still controversial. We explored this question, focusing on the initiation of ritonavir-boosted protease inhibitors (PIs) during pregnancy, which is now standard care.
Methods: Trends in prematurity (<37 gestational weeks) were studied among all singleton pregnancies in the Agence Nationale de Recherche sur le SIDA (ANRS) French Perinatal Cohort from 1990 through 2009 (n = 13 271). In-depth analysis was conducted in a more detailed substudy of the cohort, among women starting PI-based ARV therapy during pregnancy (n = 1253). Multivariable analysis adjusted for immunovirological status and known risk factors for prematurity.
Results: Prematurity increased from 9.2% during 1990-1993 (no therapy) and 9.6% during 1994-1996 (mostly zidovudine monotherapy) to 12.4% during 1997-1999 (dual-nucleoside analog therapy) and 14.3% during 2005-2009 (routine cARV therapy; P < .01). Prematurity was associated with cARV therapy, compared with zidovudine monotherapy, with an adjusted odds ratio of 1.69 (95% confidence interval [CI], 1.38-2.07; P < .01) when accounting for maternal age, intravenous drug use, geographic origin, and CD4 cell count. During 2005-2009, the prematurity rate was higher with boosted than with nonboosted PI therapy started during pregnancy (14.4% vs 9.1% [P = .05]; adjusted hazard ratio, 2.03 [95% CI, 1.06-3.89; P = .03] in multivariate analysis). The difference concerned mainly induced preterm delivery for maternal or fetal indications (5.6% vs 1.6%; P = .02),
Conclusions: The prematurity rate among HIV-infected pregnant women was twice that in the general population in France; this was not entirely explained by sociodemographic characteristics. Prematurity was independently associated with cARV therapy and, particularly, with the initiation of ritonavir-boosted PI therapy during pregnancy.