You are here
CAROTID ARTERY ATHEROSCLEROSIS IS ASSOCIATED WITH MORTALITY IN HIV+ WOMEN AND MEN
David B. Hanna1, Jee-Young Moon1, Kathryn Anastos1, Sabina Haberlen2, Audrey French3, Frank J. Palella4, Stephen J. Gange2, Mallory Witt5, Seble Kassaye6, Jason Lazar7, Phyllis Tien8, Lawrence Kingsley9, Wendy Post10, Robert C. Kaplan1, Howard Hodis11
1Albert Einstein College of Medicine, Bronx, NY, USA,2The Johns Hopkins University, Baltimore, MD, USA,3Cook County Health & Hospitals System, Chicago, IL, USA,4Northwestern University, Chicago, IL, USA,5Harbor–UCLA Medical Center, Torrance, CA, USA,6Georgetown University, Washington, DC, USA,7SUNY Downstate Medical Center, Brooklyn, NY, USA,8University of California San Francisco, San Francisco, CA, USA,9University of Pittsburgh, Pittsburgh, PA, USA,10Johns Hopkins Hospital, Baltimore, MD, USA,11University of Southern California, Los Angeles, CA, USA
Using carotid artery intima-media thickness (cIMT) measured by ultrasound as a surrogate marker for cardiovascular disease is standard, yet long-term studies of carotid artery ultrasound parameters predicting major health events in persons with HIV are lacking. We evaluated associations of carotid artery measurements with all-cause mortality in the Women's Interagency HIV Study (WIHS) and the Multicenter AIDS Cohort Study (MACS).
Participants without self-reported coronary heart disease underwent B-mode carotid artery ultrasound in 2004-2006, with measurement of 1) cIMT at the common carotid artery; 2) plaque (focal cIMT >1.5 mm) at the common or internal carotid arteries or carotid bifurcation; and 3) Young's modulus of elasticity, a measure of arterial stiffness. Participants were followed for a median 9 years (total 22,432 person-years), and death was ascertained by active surveillance and the National Death Index. Cox models estimated the association of each measure at baseline with time to death, controlling for HIV status and demographic, behavioral, cardiometabolic, and HIV-related factors. We tested interactions by cohort and HIV status.
Among 1,722 women (median age 40 years, 88% black or Hispanic, 71% HIV+, 62% on ART at baseline) and 880 men (median age 49, 35% black or Hispanic, 66% HIV+, 72% on ART), 10% (206 women, 83 men) died during follow-up. In adjusted analyses, cIMT was not associated with mortality. Presence of carotid artery plaque was associated with 56% greater mortality risk (95% CI 1.13-2.15) and varied by cohort (HR 1.25 among women, 95% CI 0.83-1.89; HR 2.48 among men, 95% CI 1.35-4.38; p for interaction 0.045). The highest quartile of Young's modulus, indicating greatest stiffness, was associated with 58% greater mortality risk compared with the lowest quartile (95% CI 1.05-2.38, p for interaction by cohort 0.29). While the association of plaque with mortality was more pronounced in HIV- vs. HIV+ participants (p for interaction=0.01), potentially owing to AIDS deaths in the HIV+ group, the relationship was statistically significant in each group (Table). The association of Young's modulus showed a similar pattern, but the interaction by HIV status was marginally significant (p=0.08).
Carotid artery measurements were independently associated with all-cause mortality in both HIV+ and HIV- persons. To our knowledge our study is the first to show that carotid artery plaque is predictive of major health events in HIV+ adults.