Abstract 14960: Coronary Artery Calcium Distribution Patterns Are Independently Predictive of Adverse Cardiovascular Events: Observations From the Dallas Heart Study

Introduction: The coronary artery calcium (CAC) diffusivity index (DI) may improve risk stratification for those with Agatston CAC > 0 in older individuals. We aimed to assess whether the distribution of CAC would improve risk prediction of cardiovascular (CV) events beyond the traditional Agatston CAC score in a younger population.

Methods: Among Dallas Heart Study 2 (DHS2) participants with Agatston CAC > 0 and no history of CVD (35% of entire cohort), we studied 1013 participants with CAC DI at DHS2. The raw CAC DI was calculated as 1 - (CAC in most affected vessel/total CAC) and categorized into concentrated (<25th%), standard (25-75th%), and diffuse (>75th%) patterns. Multivariable Cox proportional hazards regression and C statistics were calculated for coronary events (CHD) and all CVD, and all-cause mortality. Models were adjusted for total Agatston score, race, gender, BSA, smoking history, diabetes, and statin use.

Results: Mean age of the DHS2 study cohort was 55 ± 9.4 years (48% women, 45% Black). The median follow up was 9 ± 2.5 years (55 CHD, 97 CVD, and 120 all-cause mortality events) (Figure, A) . CAC DI was modestly correlated with Agatston CAC (Spearman rho = 0.60, p<0.01). In models adjusted for total Agatston score, risk factors, and demographics, compared with a concentrated pattern, a diffuse CAC pattern was associated with incident CHD (HR [95%CI]: 5.32 [1.68-16.86]); CVD (HR [95%CI]: 3.21 [1.66-6.21]), and all-cause mortality (HR [95%CI]: 2.02 [1.18-3.47] (Figure, B) . Findings were similar for continuous CAC DI. Addition of continuous CAC DI to adjusted models including Agatston CAC improved the C-statistic for CHD (0.79 to 0.81, p<0.01) and CVD (0.71 to 0.73, p<0.01).

Conclusions: Beyond the validated Agatston score, higher CAC DI at baseline improves risk prediction of incident CHD and CVD events in a younger population-based cohort. Further studies are warranted to evaluate the clinical utility of assessing distribution patterns of CAC.