HIV, HAART, and lipoprotein particle concentrations in the Women's Interagency HIV Study

Abstract

Background: Changes in lipoprotein particle concentrations, especially greater small low-density lipoprotein particle (LDL-p) and lower small high-density lipoprotein particle (HDL-p) may provide information regarding cardiovascular disease (CVD) risk above and beyond that which is provided by standard lipids. We quantified the association HIV and HAART use had with LDL-p and HDL-p.

Methods: Cross-sectional study of 1077 individuals classified by HIV/HAART status (361 HIV-uninfected, 128 HIV-infected/HAART naive, 588 HIV-infected/on HAART) enrolled in the Women's Interagency HIV Study. Nuclear magnetic resonance spectroscopy estimated total and subclass lipoprotein particle concentrations. Quantile regression models were used to estimate differences in the 10th, 25th, 50th, 75th, and 90th percentiles of the distributions of lipoprotein particle concentrations between the two HIV-infected exposure groups and HIV-uninfected controls after adjustment for demographic, nonlipid metabolic factors, and standard lipids.

Results: Compared with HIV-uninfected women, those on HAART had greater 75th and 90th percentiles of small LDL-p; there was little difference between HIV-infected/HAART naive and HIV-uninfected women. After adjustment for triglycerides, the association of HAART with greater 75th and 90th percentiles of small LDL-p was attenuated and no longer significant. In contrast, after adjustment for triglycerides, HIV infection irrespective of HAART status remained associated with significantly lower 10th, 25th, 50th, 75th, and 90th percentiles of small HDL-p.

Conclusion: Our findings suggest that testing for LDL-p confers little additional information beyond triglycerides levels when assessing CVD risk. Further investigation is needed to examine the role of HDL-p in the link between HIV and CVD in women.

Figures

Figure 1

Distribution of total LDL particle concentration (panel A), small LDL particle concentration (panel B), total HDL particle concentration (panel C), and small HDL particle concentration (panel D) by HIV and HAART status. Three-hundred and sixty-one HIV-uninfected women, one-hundred and twenty-eight HIV-infected/HAART naïve, and five-hundred and eighty-eight HIV-infected women on HAART were included for each outcome. The numerical values of the 25th, 50th, and 75th percentiles are given. The 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles of the distribution are denoted. Values <2.5th percentile or >97.5th percentile are denoted by closed circles.

Figure 2

Differences in the distributions of total LDL particle concentration (panel A), small LDL particle concentration (panel B), total HDL particle concentration (panel C), and small HDL particle concentration (panel D) between HIV-infected, HAART naïve women (open diamond) and HIV-infected women on HAART (closed diamond) compared with HIV-uninfected controls. The estimated 10th, 25th, 50th, 75th, and 90th percentiles for HIV-uninfected women who were 40 years of age, African-American, not menopausal, reported smoking for 15 years, HCV RNA negative, did not report using lipid-lowering medications, had a waist size of 90 centimeters, a HOMA value of 3.5 μU/mL, and did not have diabetes are shown along the top of each panel as a reference. Point estimates and 95% confidence intervals for differences in the 10th, 25th, 50th, 75th, and 90th percentiles are plotted. All estimates were adjusted for age, race/ethnicity, menopause status, smoking history, hepatitis C infection status, lipid lowering therapy, waist size, HOMA, and diabetes status.