Abstract 16795: Improving the Coronary Artery Calcium Score - Addition of Regional Measures of Calcium Distribution: Multi-Ethnic Study of Atherosclerosis (MESA)

Abstract

Background: Conventional CAC scoring ignores information on the regional distribution of calcified plaque by coronary vessel and location. We investigated whether inclusion of simple measures of calcified plaque distribution might improve the predictive power of the traditional Agatston CAC score.

Methods: Of 6,544 participants in MESA who underwent a baseline CAC scan scored on a per-vessel basis, we studied the 3,262 (50%) with CAC >0. Multi-territory CAC was defined by the number of vessels with CAC (scored 1 to 4, including the left main). We first tested for heterogeneity between CAC score group and the number of calcified vessels. We then performed Cox regression and AUC analyses to determine whether inclusion of number of calcified vessels added to the CAC score for prediction of CHD events. We further tested whether score in individual vessels, particular the left main, added information beyond the total CAC score.

Results: Mean age of the population was 66 ± 10 years, with 42% women. Median follow-up was 7.6 years. Considerable heterogeneity existed between CAC score group and number of calcified vessels (CAC 1-100: 57% 1-vessel, 30% 2-vessel, 12% 3-vessel, 1% 4-vessel; CAC 101-400: 9% 1-vessel, 31% 2-vessel, 47% 3-vessel, 14% 4-vessel). Addition of number of vessels with calcium significantly improved the predictive power of the traditional CAC score (considered as a categorical, continuous, or log-transformed variable) in both survival and AUC analyses (Table). In multivariable models, inclusion of Framingham risk variables attenuated, but did not abolish, the improvement. CAC score in individual coronary arteries, including the left main, did not improve predictive value of the overall total CAC score (all p=NS).

Conclusion: The number of coronary arteries with calcified plaque, indicating multi-vessel increasingly “diffuse” subclinical atherosclerosis, adds significantly to the standard Agatston CAC score for the prediction of CHD events.