This study aimed to quantify the inhibition of cytochrome P450 (CYP3A), CYP2D6, and P-glycoprotein in human immunodeficiency virus (HIV)-infected patients receiving an antiretroviral therapy (ART) containing ritonavir boosted lopinavir, and to identify factors influencing ritonavir and lopinavir pharmacokinetics. We measured activities of CYP3A, CYP2D6, and P-glycoprotein in 28 patients before and during ART using a cocktail phenotyping approach. Activities, demographics, and genetic polymorphisms in CYP3A, CYP2D6, and P-glycoprotein were tested as covariates. Oral midazolam clearance (overall CYP3A activity) decreased to 0.19-fold (90% confidence interval (CI), 0.15-0.23), hepatic midazolam clearance and intestinal midazolam availability changed to 0.24-fold (0.20-0.29) and 1.12-fold (1.00-1.26), respectively. In CYP2D6 extensive metabolizers, the plasma ratio AUC(dextromethorphan)/AUC(dextrorphan) increased to 2.92-fold (2.31-3.69). Digoxin area under the curve (AUC)(0-12) (P-glycoprotein activity) increased to 1.81-fold (1.56-2.09). Covariates had no major influence on lopinavir and ritonavir pharmacokinetics. In conclusion, CYP3A, CYP2D6, and P-glycoprotein are profoundly inhibited in patients receiving ritonavir boosted lopinavir. The covariates investigated are not useful for a priori dose selection.