Utility of coronary artery calcium scoring in low-risk patients: The Multi-Ethnic Study of Atherosclerosis (MESA)

Background: Guidelines recommend consideration of coronary artery calcium (CAC) scoring in intermediate atherosclerotic cardiovascular disease (ASCVD) risk patients, but its utility in lower-risk individuals is less clear. Methods: Data from 6712 participants from MESA was used with 10-year ASCVD risk defined by the pooled cohort equations (PCE) and AHA PREVENT equations. The association between CAC, CHD and ASCVD risk was evaluated using Cox proportional hazard models. Risk prediction improvement was evaluated using Harrell's C-index and net reclassification improvement (NRI). Results: Amongst all participants (mean age 62.2 ± 10.2 years, 52.8 % women), the ASCVD event rate per 1000 person years was 14.3 vs. 4.1 with and without CAC over a median of 16.7 years. CAC score was most strongly associated with increased ASCVD risk in low and borderline-risk individuals (HR 1.35, 95 % CI 1.22-1.50 and 1.30, 1.16-1.46). Among these individuals, addition of the Agatston score to the PCE improved the C-index (SE) for ASCVD from 0.593(0.029) to 0.640(0.031) and 0.558(0.037) to 0.663(0.036), respectively. Category-free NRI was also significant in low (0.3268, 95 % CI 0.0960-0.5408) and borderline (0.4283, 0.2319-0.7332) risk individuals with similar results using the AHA PREVENT equations. Using a statin eligibility threshold of 7.5 %, the addition of CAC correctly reclassified a net of 10.1 % of low/borderline risk individuals vs the PCE and 16.7 % vs. PREVENT. Conclusions: CAC is associated with increased ASCVD risk in lower-risk individuals. The addition of CAC scoring to the PCE and AHA PREVENT equations improved risk prediction, suggesting potential utility in this population.