Untreated HIV infection and large and small artery elasticity

Abstract

Background: Untreated HIV infection may increase risk for cardiovascular disease, and arterial elasticity is a marker of cardiovascular risk and early disease.

Methods: HIV-infected participants not taking antiretroviral therapy (n=32) were compared with HIV-negative controls (n=30). Large and small artery elasticity (LAE and SAE) were estimated via analysis of radial pulse waveforms. Differences in LAE and SAE by HIV status were compared using analysis of covariance, with and without adjustment for Framingham risk (model 1); covariates that differed between groups [smoking, injection drug use, hepatitis C, and high-density lipoprotein cholesterol (HDLc); model 2]; or age, sex, race/ethnicity, smoking, injection frug use, hepatitis C, HDLc, and non-HDLc (model 3).

Results: HIV infection was associated with impaired LAE (-2.55 mL/mm Hg x 10; P=0.02) and SAE (-1.50 mL/mm Hg x 100; P=0.02). Associations with traditional risk factors were often stronger for SAE than LAE, including with Framingham score (per 1% higher; SAE -0.18, P=0.01; LAE -0.19, P=0.13). Fasting lipid levels were not significantly associated with LAE and SAE. After adjustment, differences between HIV-infected and HIV-uninfected participants were similar in model 1 (-2.36 for LAE, P=0.04; -1.31 for SAE, P=0.04), model 2 (-2.67 for LAE, P=0.02; -1.13 for SAE, P=0.07) and model 3 (-2.91 for LAE, P=0.02; -1.34 for SAE, P=0.03). CD4 count and HIV RNA level were not associated with LAE and SAE among HIV-infected participants.

Conclusions: Untreated HIV infection is associated with impaired arterial elasticity, of both the large and small vasculature, after controlling for additional risk factors. Pulse waveform analysis is a noninvasive technique to assess cardiovascular disease risk that should be evaluated in larger studies of HIV-infected persons.

FIGURE 1

Representative radial artery pulse waveforms are shown for an HIV-negative control (A) and an HIV-infected participant (B). Waveforms are plotted with BP along the y axis over time along x axis. Differences in the pulse contour, including the diastolic decay, can be appreciated. Resting BP along with the corresponding large and small arterial elasticity (LAE and SAE) measurements is reported, and the beginning of systole and diastole is estimated.

FIGURE 2

The distribution of LAE (A) and SAE (B) estimates are presented, stratified by HIV status. Box plots represent the IQR with median values designated by a horizontal line. Error bars constitute minimum and maximum values. A, One high outlier (value >1.5 χ IQR higher than the third quartile) is designated for LAE within the HIV-infected group. P values represent a significantly lower mean LAE and SAE for HIV-infected participants compared with HIV-negative controls.

FIGURE 2

The distribution of LAE (A) and SAE (B) estimates are presented, stratified by HIV status. Box plots represent the IQR with median values designated by a horizontal line. Error bars constitute minimum and maximum values. A, One high outlier (value >1.5 χ IQR higher than the third quartile) is designated for LAE within the HIV-infected group. P values represent a significantly lower mean LAE and SAE for HIV-infected participants compared with HIV-negative controls.