Higher total bilirubin predicted incident cardiovascular disease (CVD) in an analysis of 98,405 U.S. veterans with or without HIV infection. The association remained significant in veterans without liver disease.
Hyperbilirubinemia may protect against (CVD) by lowering lipids, decreasing oxidative stress and inhibiting platelet activation. The HIV protease inhibitor atazanavir can cause unconjugated hyperbilirubinemia. ACTG studies 5257 and 5260s linked atazanavir/ritonavir (Reyataz/Norvir) to lower levels of the cardiovascular markers hsCRP and D-dimer and to slower progression of carotid artery intima-media thickening. Veterans Aging Cohort Study (VACS) investigators conducted a new analysis to determine whether higher total bilirubin is associated with lower risk of (1) heart failure, (2) acute myocardial infarction (MI), (3) ischemic stroke or (4) all three outcomes (CVD).
The analysis involved veterans without CVD at baseline, which was the first VACS visit after April 2003. Follow-up continued until the first CVD event (identified by ICD-9 code), death, the last study visit, or September 2012. The investigators collected total bilirubin levels from lab data recorded as close as possible to baseline; they divided participants into four quartiles from lowest to highest total bilirubin. The VACS team used Cox regression to estimate CVD risk after adjusting for baseline confounders including demographics and classic cardiovascular risk factors.
Related: Classic Risk Factors Drive Heart Attack Rates More Than HIV Variables
The study group included 31,418 HIV-positive and 66,987 HIV-negative veterans, 97% of them men and 48% African American. Age averaged 48 years. Through an average follow-up of six years, there were 3,843 heart failures; 1,931 acute MIs; 2,112 strokes; and 6,603 total CVD events. CVD incidence measured 10.21 per 1,000 person-years in HIV-negative veterans, 11.95 per 1,000 in HIV-positive veterans not taking antiretroviral therapy (ART), 13.02 per 1,000 in HIV-positive veterans taking ART with atazanavir and 13.33 per 1,000 in veterans taking ART without atazanavir. The analysis had limited power to assess the impact of atazanavir alone on CVD incidence.
The four total bilirubin quartiles were ≤0.4 mg/dL, 0.5 to 0.6 mg/dL, 0.7 to 0.8 mg/dL and ≥0.9 mg/dL. In the entire study group, compared with the lowest quartile, total bilirubin in each of the three higher quartiles was independently associated with incident CVD (for the highest quartile, adjusted hazard ratio [aHR] 0.76, 95% confidence interval [CI] 0.70 to 082, P < .001). In the entire study group, each of the three higher total bilirubin quartiles was also independently associated with lower risk of heart failure, acute MI or stroke (all P < .01), except for the 0.7-0.8 quartile and acute MI risk (aHR 0.89, 95% CI 0.78 to 1.03).
With the analysis limited to HIV-positive veterans, each of the three higher total bilirubin quartiles independently predicted CVD (for the highest quartile, aHR 0.75, 95% CI 0.66 to 0.86, P < .001). Compared with the lowest quartile, the highest quartile independently predicted heart failure (aHR 0.75, 95% CI 0.63 to 0.88, P < .01) and stroke (aHR 0.68, 95% CI 0.54 to 0.86, P = .009) but not acute MI (aHR 0.84, 95% CI 0.67 to 1.06, P = .16).
Among all veterans without liver disease (excluding those with hepatitis C, alcohol abuse and FIB4 ≥1.45), each of the three higher total bilirubin quartiles independently predicted CVD (for the highest quartile, aHR 0.68, 95% CI 0.59 to 0.77, P < .001).
The VACS investigators note that their analysis did not include many women, measures of direct and indirect bilirubin, classification of MI or heart failure by type or time-updated analyses. With those limitations in mind, they conclude that higher bilirubin is associated with lower CVD risk in HIV-positive and negative veterans and in those without liver disease.