Abstract P227: Cardiovascular Disease (CVD) Risk Factors Have Opposite Associations with Abdominal Aortic Calcium (AAC) Density and Volume: The Multi-Ethnic Study of Atherosclerosis (MESA)
Background: AAC predicts future CVD events independently of CVD risk factors. The standard AAC score, the Agatston method, up-weights for greater calcium density, thus models higher calcium density as associated with increased CVD risk. We evaluated this model by investigating associations of CVD risk factors with AAC density and volume separately.
Methods: Multivariable linear regression was used to investigate the independent cross-sectional associations of CVD risk factors, markers of inflammation, and potential promoters of calcification with AAC density and ln(AAC volume). AAC density was calculated as:
Density = Agatston / (volume, mm3 / CT scan slice thickness, mm). Agatston = area, mm2 X density.
Results: In 1413 MESA participants with prevalent AAC, mean age was 65 ± 9 years, mean AAC density was 3.0 ± 0.6, 52% were men, 44% were European-, 24% were Hispanic-, 18% were African-, and 14% were Chinese Americans (EA, HA, AA, and CA respectively). In fully adjusted models, older age was non-significantly associated with lower AAC density, but significantly associated with higher ln(AAC volume) (Figure). Compared to EA, we observed AAC density was significantly higher in CA, non-significantly higher HA and AA, but ln(AAC volume) was significantly lower in CA, HA, and AA. Also, smoking, alcohol use, family history of myocardial infarction, higher systolic blood pressure, elevated total cholesterol, and interluken-6 were significantly associated with higher ln(AAC volume), but not AAC density. Lastly, lower body mass index, higher serum HDL and calcium were significantly associated with higher AAC density, but associations with ln(AAC volume) were in the opposite direction.
Conclusion: A greater burden of CVD risk factors was associated with higher AAC volume, but not AAC density. Thus, the Agatston method of up-weighting AAC scores for greater density in CVD risk prediction may be inappropriate. Investigators should evaluate separately associations of AAC density and volume with CVD events.
Author Disclosures: N.I. Forbang: None. J.H. Ix: None. R.A. Remigo-Baker: None. M.A. Allison: None. R.L. McClelland: None. V. Sandfort: None. M.H. Criqui: None.
- © 2016 by American Heart Association, Inc.