Nucleoside analogs can induce mitochondrial toxicity by inhibiting the human DNA polymerase gamma. This can lead to a wide range of clinical toxicities, from asymptomatic hyperlactatemia to death. Despite their technical and physiological variability, we propose that random venous lactate measurements can be useful to monitor the development of nucleoside-related mitochondrial toxicity. Recently, we have developed an assay that can measure changes in mitochondrial DNA levels in peripheral blood cells. Using this assay we have characterized changes in mitochondrial DNA (mtDNA) relative to nuclear DNA (nDNA) in peripheral blood cells of patients with symptomatic nucleoside-induced hyperlactatemia. Our results demonstrate that symptomatic hyperlactatemia was associated with markedly low mtDNA/nDNA ratios, which were on average 69% lower than HIV-uninfected controls and 45% lower than HIV-infected asymptomatic/antiretroviral naive controls. A statistically significant (p = .016) increase in mtDNA/nDNA ratio was observed following discontinuation of antiretroviral therapy. The mtDNA/nDNA ratio remained stable among selected patients who reintroduced antiretroviral therapy with stavudine (d4T)-sparing regimens. Of note, the decline in mtDNA preceded the increase in venous lactate levels. More recently we have evaluated changes in the mtDNA/nDNA ratio in relation to selected antiretroviral drug regimens in a cross-sectional study on a non-random sample of participants within the British Columbia Centre for Excellence in HIV/AIDS Drug Treatment Program. Eligible patients had continuously received saquinavir plus ritonavir with either nevirapine (n = 20), lamivudine (n = 15), d4T (n = 53) or lamivudine + d4T (n = 69), for 4 to 30 months. d4T-sparing regimens were associated with a higher median mtDNA/nDNA ratio than d4T-containing regimens (p = .016), despite the fact that study patients had received d4T-containing regimens for a shorter median time than patients taking d4T-sparing regimens (13 versus 25 months, p = .002). In summary, mtDNA levels are significantly decreased among patients who develop symptomatic, nucleoside-related hyperlactatemia, an effect reversed upon therapy discontinuation. Furthermore, mtDNA/nDNA ratios were statistically significantly lower in patients taking d4T-containing regimens than in those taking selected d4T-sparing regimens in a population setting. These results suggest that measurement of this parameter should be investigated as a potential clinical management tool.