Therapeutic drug monitoring (TDM) has been proposed as a means to optimise response to highly active antiretroviral therapy (HAART) in HIV infection. Protease inhibitors (PIs) and the non-nucleoside reverse transcriptase inhibitors (NNRTIs) efavirenz and nevirapine satisfy many criteria for TDM. Nucleoside reverse transcriptase inhibitors (NRTIs) are not suitable candidates for TDM, since no clear plasma concentration-effect relationships have been established for these drugs. Several important limitations to the application of TDM for antiretroviral drugs should be recognised, including uncertainty about the best pharmacokinetic predictor of response and insufficient validation of target concentrations for individual PIs and NNRTIs. Data from two clinical trials support the use of TDM in treatment-naive HIV-infected patients who start with an indinavir- or nelfinavir-based regimen. TDM either prevented virological failures (presumably by preventing the development of resistance) or treatment discontinuations due to concentration-related toxicity. Application of routine TDM in other patient groups (treatment-experienced patients) or for drugs other than indinavir or nelfinavir (NNRTIs, other PIs, combination of PIs) is speculative at this moment. However, TDM can be used in selected patient groups (children, pregnant women, patients with renal or hepatic dysfunction) to confirm adequate drug concentrations, and for management of drug-drug interactions.TDM in treatment-experienced patients may be optimally used in conjunction with resistance testing. The integration of pharmacological and virological measures in the inhibitory quotient (IQ) needs to be standardised and elaborated further. TDM should be accompanied by careful assessment of adherence and can itself help identify non-adherence, although a drug concentration only reflects the last few drug doses taken by a patient. Additional clinical trials are needed before routine TDM can be adopted as standard of care in the treatment of HIV infection.