Current treatment guidelines recommend first-line HAART regimens containing a combination of two nucleos(t)ide reverse transcriptase inhibitors (NRTIs) . Mitochondrial toxicity (MtT) has been recognized as one of the most serious potential side effects of NRTI therapy, particularly with the use of the thymidine analogue NRTIs zidovudine
and stavudine (d4T) . Although the clinical manifestations of MtT vary between specific NRTIs, a serious and immediately life-threatening consequence of MtT is symptomatic hyperlactataemia (SHL) and lactic acidosis (LA) , often associated with exposure to d4T and didanosine (ddI) . Although infrequent with a reported incidence of three-to-four per 1000 patient-years (py) [6,7], LA is a serious condition associated with significant mortality . SHL without acidosis, although less serious, is more prevalent AG14699 five of 349 (1.4%) patients were affected in one study . Risk factors for SHL and LA have not been clearly defined, and vary depending on the study and the population exposed. Extremes of body mass index (BMI) [8,9], female gender  and lower CD4 T-cell count  have been reported to be associated
MLN0128 concentration with LA and SHL. The use of d4T and ddI in developed countries has markedly declined, with reductions in the prevalence of associated toxicities . However, these agents are still commonly used in resource-limited settings, where second the largest burden of HIV disease remains. Reports of clinical manifestations of MtT have been increasing in these regions, where there is also a difficulty in diagnosing MtT and a shortage of alternate antiretroviral agents [9,11]. As a result, there is a need for predictors for LA and SHL to identify
those who may be at higher risk. It is presumed that SHL and LA arise from an inability of liver and skeletal muscle to meet aerobic energy requirements secondary to mitochondrial dysfunction [12–14]. However, biopsy of these tissues is invasive and associated with significant risks. In contrast, peripheral blood mononuclear cells (PBMCs) are easier to sample and changes in markers of mitochondrial function such as mitochondrial DNA (mtDNA) and mitochondrial RNA (mtRNA) in PBMCs have been reported with both HIV infection itself and with exposure to HAART [15–17]. This offers the potential for changes in mtDNA and mtRNA in PBMCs to be used as markers of tissue mitochondrial dysfunction, although reports to date have been conflicting. In cross-sectional studies, HIV-infected individuals have been found to have lower PBMC mtDNA copy numbers than HIV-uninfected individuals, and those with SHL to have lower mtDNA copy numbers than untreated HIV-infected controls . However, other cross-sectional studies in HIV-infected subjects have failed to show an association between PBMC mtDNA and current or prior exposure to NRTIs  or mtDNA content in tissues such as muscle .