OBJECTIVE To explore whether steady-state plasma pharmacokinetics of ritonavir and saquinavir change during long-term treatment in HIV-1-infected patients on antiretroviral treatment including ritonavir and saquinavir. METHODS The pharmacokinetics of ritonavir and saquinavir were assessed during an 8-h period on two occasions in six HIV-1 infected patients on stable twice daily treatment with ritonavir 400 mg, saquinavir 400 mg and stavudine 40 mg with or without lamivudine 150 mg twice daily. RESULTS The first study day was 4-12 months (median 7 months) after the start of the current regimen. The second study day was 9-15 months (median 10 months) later. No significant differences were observed for the ritonavir pharmacokinetics between the first and second study day. However, median change in plasma trough level of saquinavir between the two study days was -30% (range -79 to +11%; P = 0.06). Median change in maximum plasma concentration was -40% (range -62 to +34%; P = 0.09). The median change in area under the plasma concentration versus time curve over 0-8 h was -33% (range -53 to +21%; P = 0.06). CONCLUSIONS The exposure to saquinavir decreased over time in HIV-infected patients on stable antiretroviral therapy. These data suggest that regular monitoring of plasma drug concentrations should become part of routine patient care even in apparently compliant patients.