Since the introduction of the HIV protease inhibitors in 1995, considerable progress has been made in the treatment of HIV-infected patients. However, treatment has not been without problems. Studies have demonstrated associations between protease inhibitor concentrations and efficacy and in some cases toxicity. As considerable inter-individual and intra-individual variations of protease inhibitor concentrations have been observed, it has been questioned to what extent this has clinical implications with regard to efficacy and toxicity. As a consequence the use of protease inhibitor concentration measurements to optimise HIV antiretroviral therapy (therapeutic drug monitoring--TDM) has been suggested. The objectives of this study, initiated in 2000, were: to establish a method for the simultaneous measurement of the available protease inhibitors; to explore the pharmacokinetics of the protease inhibitors in clinically relevant situations and in this context; to consider the applicability of TDM in protease inhibitor therapy. The presented studies and review demonstrate: 1) that it is feasible to measure protease inhibitor plasma concentrations in a clinical setting with precision and accuracy, and that protease inhibitor concentrations are very stable during different circumstances ex vivo; 2) that protease inhibitor plasma concentrations display limited long-term intra-individual variation but considerable inter-individual variation; 3) that protease inhibitor plasma concentrations display considerable intra-individual variations between morning and evening and in the case of drug-drug interactions; 4) that protease inhibitor drug-drug interactions can be unpredictable and adverse but also that protease inhibitor drug-drug interactions can be exploited to increase protease inhibitor concentrations or decrease protease inhibitor dose; 5) that increases in protease inhibitor plasma concentrations can enhance efficacy but also that decreases can reduce toxicity; 6) that the concentration-efficacy associations which have been established by others can be confirmed in clinical trials but that concentration-toxicity associations are more difficult to establish and confirm. The experiences with protease inhibitor therapy and the understanding of protease inhibitor pharmacokinetics have resulted in new treatment principles and the development of new and better protease inhibitors. Most patients achieve concentrations several folds higher than the minimum effective concentration with the regimens that are used currently (2007). As a result TDM in protease inhibitor therapy has become less relevant in HIV-infected patients receiving their first protease inhibitor. In protease inhibitor experienced patients, harbouring HIV with varying degrees of resistance/reduced susceptibility to protease inhibitors, the combination of TDM and genotypic resistance testing, seems to be a promising tool, but prospective studies are needed. In some patients with certain conditions or in certain circumstances known to be associated with considerable inter-individual or intra-individual variation of protease inhibitor concentrations (drug-drug interactions, gastrointestinal disease, pregnancy or children) TDM might also be of benefit. However, no studies have investigated these patients specifically in randomised TDM trials.