BACKGROUND: The pregnancy-related adverse effects of antiretroviral therapy (ART) have yielded discordant results, which could be explained in part by the heterogeneity of ART protocols. The objective of our study was to explore whether lopinavir/ritonavir (LPV/r) exposure during pregnancy is associated with adverse outcomes. METHODS: Data on 100 consecutive HIV type-1 (HIV-1)-infected women receiving LPV/r during pregnancy and who delivered after 15 weeks gestational age (GA) between January 2003 and June 2007 in a single centre were analysed. For each HIV-1-infected woman, two uninfected women matched by age, parity and geographical origin were selected among patients delivering during the same period. Preterm delivery (PTD), vasculoplacental complications, gestational glucose intolerance and post-partum complication rates were compared between cases and controls. Factors associated with PTD and post-partum complications were assessed in HIV-1-infected women by a logistic regression model. RESULTS: Rates of vasculoplacental complication and gestational glucose intolerance were not higher among HIV-1-infected women than in controls. PTD was higher in HIV-1-infected women (21%) than in controls (10%; P<0.01). In HIV-1-infected women, PTD was associated with HIV-1 RNA level > or =50 copies/ml at delivery (adjusted odds ratio 6.15, 95% confidence interval 1.83-20.63; P=0.003). No association was found between occurrence of PTD and LPV/r exposure before 14 weeks GA. CONCLUSIONS: In this population of HIV-1-infected pregnant women receiving LPV/r, the risk of PTD was higher than in HIV-1-uninfected controls. As PTD risk was not associated with early exposure to LPV/r, these data support current guidelines to initiate ART earlier in pregnancy.