OBJECTIVE Pharmacokinetic differences, contributing to drug-related side effects, between men and women have been reported for HIV protease inhibitors. As only limited and inconclusive data on ritonavir-boosted atazanavir are available, we evaluated the respective steady-state pharmacokinetics in 48 male and 26 female HIV-1-infected adults receiving atazanavir/ritonavir 300/100 mg once-daily as part of their antiretroviral therapy. METHODS Pharmacokinetic profiles (24 h) of atazanavir/ritonavir were assessed and measured by HPLC/tandem mass spectrometry. Geometric mean (GM; ANOVA) of minimum and maximum plasma drug concentrations (C(min) and C(max)), area under the concentration-time curve (AUC) and total clearance (CL(total)) were compared between the sexes and correlated to demographic (age, gender and ethnicity), physiological (weight and body mass index) and clinical (CD4+ cell count, HIV-RNA, co-medication and hepatitis serology) co-factors. RESULTS The GM of the atazanavir AUC, C(max) and C(min) of men versus women were 32 643 versus 36 232 ng.h/mL [GM ratio (GMR) = 1.11, P = 0.435], 2802 versus 3211 ng/mL (GMR = 1.15, P = 0.305) and 398 versus 470 ng/mL (GMR = 1.18, P = 0.406), respectively. Although weight (80.6 versus 63.9 kg, P = 0.001) and body weight-adjusted atazanavir dose (3.84 versus 4.60 mg/kg, P = 0.013) were different between the sexes, no significant correlation to atazanavir pharmacokinetics was observed. A linear regression analysis detected significant correlations of atazanavir C(min) with ritonavir AUC (P < 0.001) and the co-administration of methadone oral solution (P = 0.032), and inverse correlations with the time since the first HIV infection diagnosis (P = 0.003) and the number of previous antiretroviral treatments (P = 0.022). CONCLUSIONS Atazanavir/ritonavir steady-state pharmacokinetics was comparable in men and women, despite gender-related significant differences in atazanavir dose/body weight. The administration of atazanavir/ritonavir is pharmacokinetically safe; 95% of all trough samples were above the recommended plasma concentration of 150 ng/mL.