Our objective was to identify possible differences in protease inhibitor plasma concentrations between and within three protease inhibitor regimens (indinavir, saquinavir and lopinavir all in combination with low-dose ritonavir) and to relate these differences to safety and efficacy. Data originated from pre-defined pharmacokinetic substudies within two randomized 48-week trials evaluating the safety and efficacy of three protease inhibitor regimens. At weeks 4 and 48, plasma was collected and minimum drug plasma concentrations, C(min), were obtained. Out of 656 randomized patients, 283 patients had available C(min) at week 4. Indinavir, saquinavir and lopinavir C(min) were high when combined with low-dose ritonavir. No significant difference in the proportion of patients experiencing treatment failure could be found according to the C(min) within any treatment arm. A saquinavir C(min) > 2000 ng/ml was associated with an increased risk of gastrointestinal grade 3 or 4 adverse events and higher total cholesterol. Overall, there were no changes in C(min) from week 4 to week 48 in patients who remained on therapy. No association between treatment failure and the C(min) could be demonstrated. Associations between high C(min) and toxicity were identified in the saquinavir arm; therefore, dose reductions may be appropriate in certain patients with C(min) several times above the minimum effective concentration.