Impact of antiretroviral choice on hypercholesterolaemia events: the role of the nucleoside reverse transcriptase inhibitor backbone.

BACKGROUND: The use of highly active antiretroviral therapy (HAART) has profoundly altered the life expectancy of individuals infected with HIV. Metabolic abnormalities associated with antiretrovirals and cumulative exposure to combination antiretroviral therapy, including dyslipidaemia and insulin resistance, have been linked to an increased risk of myocardial infarction.

METHODS: Longitudinal data from a large prospectively collected clinical database were analysed. All patients who commenced first antiretroviral therapy (ART) [two nucleoside reverse transcriptase inhibitors (NRTIs)+one nonnucleoside reverse transcriptase inhibitor (NNRTI) or one active protease inhibitor (PI)] since 1996 were identified. Patients with elevated cholesterol levels [>5.5 mmol/L (215 mg/dL)] prior to therapy initiation were excluded. Quantitative data were categorized into quartiles and presented stratified by individuals developing abnormal levels of cholesterol during first-line HAART. Event time was defined as time from commencing first-line ART to either development of cholesterol level >6.5 mmol/L (254 mg/dL) or switch of first-line therapy. The Kaplan-Meier product limit survival method was used to estimate time to abnormal cholesterol level, and the chi2 test was used for comparisons between drug classes. Cox's proportional hazards regression analysis was used to identify factors predicting a likelihood of raised cholesterol level.

RESULTS: A total of 1664 patients were included in the study: 57.1% on two NRTIs+one NNRTI, 38.4% on two NRTIs+one PI, and 4.4% on two NRTIs+a boosted PI regimen. Regimens containing stavudine or PIs were associated with a significantly higher event risk and earlier time to event. No differences between efavirenz and nevirapine or between didanosine and lamivudine were observed. In 28 patients exposed to the combination of tenofovir+lamivudine+efavirenz, there were no episodes of elevated cholesterol level.

CONCLUSION: Dyslipidaemia has emerged as an important issue in HIV-infected individuals receiving antiretroviral therapy. This study demonstrates that age at start of therapy, baseline cholesterol level, stavudine use and PI use are all associated with increased risk of hypercholesterolemia on initial therapy. Both NRTI and NNRTI/PI choice influence risk of hypercholesterolaemia.