Abstract 11332: Abdominal Aortic Calcium, Coronary Artery Calcium, and Cardiovascular Morbidity and Mortality in the Multi-Ethnic Study of Atherosclerosis

Abstract

We compared abdominal aortic calcium (AAC) with coronary artery calcium (CAC) in predicting cardiovascular disease (CVD) events and mortality in 1974 men and women randomly selected from a multi-ethnic cohort with CT scans of both the abdominal aorta and the chest. AAC and CAC were scored by the Agatston method, and were defined categorically as < 50^th %tile, 51-75 %tile, and 76-100 %tile. Outcome event categories were hard coronary heart disease (CHD), hard CVD, CVD mortality, and total mortality. During a mean of 4.7 years of follow-up there were 46 CHD events, 72 total CVD events, 19 of which were fatal, and 82 total deaths. Categorical unadjusted analyses showed sharp CHD and CVD gradients for both AAC and CAC of a similar magnitude. In contrast, for CVD mortality and total mortality, the gradients across categories were markedly stronger for AAC than CAC. In Cox models adjusted for the General Framingham Risk Score (GFRS), and with AAC and CAC categories both in the model, AAC and CAC were each significantly and independently predictive of CHD and CVD, with hazard ratios (HR) of 2.7 to 3.4 for the 4^th quartile compared to <50^th percentile. For CVD mortality the HR was striking for AAC, 12.7, while CAC showed no association, HR=1.1. For total mortality, AAC was significant, HR=3.0, while CAC was not, HR=1.5. Both area under the ROC curve and Net Reclassification Index analyses showed improvement over the GFRS for both AAC and CAC for CHD and CVD, but only AAC showed improvement for CVD and total mortality. The results of analyses using log transformed continuous measures of AAC and CAC mirrored these results. In this study, AAC and CAC both predicted CHD and CVD events, but only AAC was independently related to CVD mortality and total mortality. Previous studies have not adjusted AAC and CAC for each other, and these results suggest AAC has a much stronger association with fatal CVD events than CAC.