February 13–16, 2017

Conference Dates and Location: 
February 13–16, 2017 | Seattle, Washington
Abstract Number: 



Vincent C. Marconi1, Kaku So-Armah2, Janet Tate3, Joseph Lim4, Vincent Lo Re5, Adeel A. Butt6, Matthew B. Goetz7, Maria Rodriguez-Barradas8, Amy Justice3, Matthew Freiberg9

1Emory Univ, Atlanta, GA, USA,2Boston Univ, Boston, MA, USA,3VA Connecticut Hlthcare System, West Haven, CT, USA,4Yale Univ, New Haven, CT, USA,5Univ of Pennsylvania, Philadelphia, PA, USA,6Weill Cornell Med Coll, New York, NY, USA,7VA Greater Los Angeles Hlth Care System, Los Angeles, CA, USA,8Baylor Coll of Med, Houston, TX, USA,9Vanderbilt Univ, Nashville, TN, USA

Abstract Body: 

Hyperbilirubinemia may protect against cardiovascular disease (CVD) by reducing oxidative stress and via its anti-atherogenic properties. Whether elevated bilirubin reduces risk of CVD events including among HIV positive (HIV+) people independently of atazanavir (ATV) use is unclear. ATV can cause unconjugated hyperbilirubinemia by competitive inhibition of the uridine diphosphate-glucuronosyl transferase (UGT) 1A1 enzyme. We assessed whether elevated bilirubin independently reduced the risk of heart failure (HF) and acute myocardial infarction (AMI) among participants in the Veterans Aging Cohort Study (VACS).

VACS participants, without CVD at baseline (first clinical visit after 4/1/2003) were included. The total bilirubin from clinical labs closest to baseline was categorized into quartiles. HF and AMI were assessed using VA, VA Fee For Service and Medicare ICD-9 codes. Participants were followed until first HF or AMI, death, last VA encounter or 12/31/2011. Cox regression was used to estimate HF/AMI risk adjusting for baseline confounders including HIV and hepatitis C, demographics, Framingham CVD risk factors, and substance abuse/dependence. Analyses restricted to HIV+ participants were further adjusted for HIV viremia and ATV use.

There were 96,373 participants; mean age was 49 years, 48% were African American, and 97% were men. There were 3,844 incident HF events (median follow up 6.9 years) and 1,932 incident AMI events over (median follow up 7.4 years). Among the HIV+ participants with bilirubin >75th percentile (≥0.9 mg/dL), 9% were on ATV compared to 3%, with bilirubin <25th percentile (<0.04 mg/dL). In unadjusted models, bilirubin was inversely associated with HF and AMI risk. After adjusting for CVD risk factors, higher bilirubin was associated with a lower hazard ratio (HR) for HF and AMI (Table). When the same model included only HIV+ participants (N=30,425), results persisted for HF and similar but non-significant associations were observed for AMI. Death rates were highest in the lowest and highest bilirubin quartiles.

VACS participants with elevated bilirubin had lower risk of incident HF and AMI events after adjusting for known risk factors. This association persisted for HF among HIV+ people but was attenuated for AMI. Future studies, should investigate how this apparently protective effect of elevated bilirubin may be harnessed to reduce CVD risk or improve CVD risk estimation among HIV+ people.

Session Number: 
Session Title: 
Presenting Author: 
Vincent Marconi
Presenter Institution: 
School of Medicine, Emory University