Absence of association between individual thymidine analogues or nonnucleoside analogues and lipid abnormalities in HIV-1-infected persons on initial therapy.

Changes in levels of triglycerides and cholesterol during antiretroviral therapy raise concerns regarding an increased future risk of atherogenic disease and may precede the appearance of fat redistribution. Hypotheses regarding the impact of nucleoside analogues on adipocytes provide a possible explanation for metabolic and clinical fat disturbances. It is unclear whether the choice of nucleoside analogue combination or coadministration of nonnucleoside agents influences change in lipids. We performed a cross-sectional analysis of 135 persons receiving their first nucleoside analogue plus nonnucleoside-based combination antiretroviral regimen for at least 1 month and for whom cholesterol and triglyceride values were available on therapy. Univariate and multivariate regression models were used to explore the relation between cholesterol and triglycerides, as continuous variables with other variables. Both significant and nonsignificant variables from univariate analyses were evaluated in multivariate models to limit possible confounders. No association with drug choice was observed, either when comparing thymidine analogues (stavudine or zidovudine), all nucleoside analogue combinations or choice of either efavirenz or nevirapine as nonnucleoside. Age and triglyceride levels were found in a multivariate analysis to be associated with higher cholesterol. Only higher cholesterol was associated with higher triglyceride levels. In conclusion, no differences were observed between choice of drug or combination on cholesterol or triglyceride values during therapy. Older individuals may be more likely to have elevated cholesterol values.