Share this article on:

Infection duration and inflammatory imbalance are associated with atherosclerotic risk in HIV-infected never-smokers independent of antiretroviral therapy

Desvarieux, Moïsea,b; Boccara, Franckc,d,g; Meynard, Jean-Lucd,e; Bastard, Jean-Philliped,f,g; Mallat, Ziadh,j; Charbit, Benyi; Demmer, Ryan T.a; Haddour, Nabilac,d; Fellahi, Sorayad,f,g; Tedgui, Alainh; Cohen, Arielc,d; Capeau, Jacquelined,f,g; Boyd, Andersk; Girard, Pierre-Maried,e,k

doi: 10.1097/QAD.0b013e3283634819
Epidemiology and Social

Objectives: To determine whether the reported increased atherosclerotic risk among HIV-infected individuals is related to antiretroviral therapy (ART) or HIV infection, whether this risk persists in never-smokers, and whether inflammatory profiles are associated with higher risk.

Design: Matched cross-sectional study.

Methods: A total of 100 HIV-infected patients (50 ART-treated >4 years, 50 ART-naive but HIV-infected >2 years) and 50 HIV-negative controls were recruited in age-matched never-smoking male triads (mean age 40.2 years). Carotid intima–media maximal thickness (c-IMT) was measured across 12 sites. Pro-inflammatory [highly sensitive C-reactive protein (hs-CRP), resistin, interleukin-6, interleukin-18, insulin, serum amyloid A, D-dimer) and anti-inflammatory (total and high molecular weight adiponectin, interleukin-27, interleukin-10) markers were dichotomized into high/low scores (based on median values). c-IMT was compared across HIV/treatment groups or inflammatory profiles using linear regression models adjusted for age, diabetes, hypertension, and, for HIV-infected patients, nadir CD4+ cell counts.

Results: Although adjusted c-IMT initially tended to be thicker in ART-exposed patients (P = 0.2), in post-hoc analyses stratifying by median HIV duration we observed significantly higher adjusted c-IMT in patients with longer (>7.9 years: 0.760 ± 0.008 mm) versus shorter prevalent duration of known HIV infection (<7.9 years: 0.731 ± 0.008 mm, P = 0.02), which remained significant after additionally adjusting for ART (P = 0.04). Individuals with low anti-inflammatory profile (<median versus >median score) had thicker c-IMT (0.754 ± 0.006 mm versus 0.722 ± 0.006 mm, P < 0.001), with anti-inflammatory markers declining as prevalent duration of HIV infection increased (P for linear trend <0.001).

Conclusion: Known HIV duration is related to thicker c-IMT, irrespective of ART, in these carefully selected age-matched never-smoking HIV-treated and ART-naive male individuals. Higher levels of anti-inflammatory markers appeared protective for atherosclerosis.

Supplemental Digital Content is available in the text

aDepartment of Epidemiology, Columbia University, Mailman School of Public Health, New York, USA

bInserm U-738 and École des Hautes Études en Santé Publique

cAP-HP Department of Cardiology, Hôpital Saint-Antoine, F-75012, Paris, France

dUPMC Univ Paris 06, Paris F75012

eAPHP Department of Infectious and Tropical Diseases, Hôpital Saint-Antoine

fAPHP Hôpital Tenon, Service de biochimie et hormonologie, Paris F75020

gInserm UMR_S938, Paris F-75012

hInserm U970, Paris-Cardiovascular Research Center, and Paris-Descartes University

iAP-HP Centre d’Investigation Clinique Paris-EST (9304), Hôpital Pitié-Salpêtrière, Paris, France

jDepartment of Medicine, University of Cambridge, Cambridge, UK

kInserm UMR-S707, Paris, France.

Correspondence to Moïse Desvarieux, MD, PhD, Department of Epidemiology, Mailman School of Public Health, Columbia University, 722 W168th Street, Suite 525, New York, NY 10032, USA. Tel: +1 212 305 5172; fax: +1 212 342 2756; e-mail:;

Received 13 January, 2013

Revised 3 May, 2013

Accepted 15 May, 2013

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

© 2013 Lippincott Williams & Wilkins, Inc.