The introduction of highly active antiretroviral therapy for the treatment of human immunodeficiency virus (HIV) infection has led to substantial reduction in morbidity and near-complete suppression of HIV-1 replication. This progress has been tempered by a growing number of new adverse effects. Mitochondrial toxicity is one aspect of these long-term toxicities of antiretroviral drugs, with the role of nucleoside analogs being particularly underlined. Some cases of impaired mitochondrial function have been clearly identified, such as pancreatitis, neuropathy, miopathy and lactic acidosis. Beyond the inhibition of DNA polymerase-g using nucleoside analogs, it appears that several physiopathologic mechanisms interact to explain the observed toxicity. At present there is no reliable method to detect subclinical mitochondrial toxicity. There is no proven effective therapy for antiretroviral therapy-associated mitochondrial toxicity other than ceasing the implicated agent, and even with this strategy, resolution of symptoms may be incomplete. Therefore, investigation of mitochondrial toxicity of new compounds or new combinations is of growing interest for the clinical application of antiretroviral agents.