Long-Term Prognosis After Coronary Artery Calcification Testing in Asymptomatic Patients

Study Questions:

What is the relationship between the extent of coronary artery calcification (CAC) and long-term outcome in persons without symptoms of coronary artery disease (CAD)?


CAC scoring and binary risk factor data were collected in a single-center outpatient cardiology site (Nashville, TN) in 9,715 asymptomatic patients physician referred from 1996-1999. Each patient paid $69 out of pocket for the procedure at the time of service. The primary endpoint was time to all-cause mortality (median follow-up, 14.6 years). Univariate and multivariable Cox proportional hazards models were used to compare survival distributions. The net reclassification improvement statistic was calculated.


A total of 86% were white, 8% African American, 4% Hispanic, and 2% were Asian. The median annual per capita income was $33,000. In Cox models adjusted for risk factors for CAD, the CAC score was highly predictive of all-cause mortality (p < 0.001). Overall 15-year mortality rates ranged from 3% to 28% for CAC scores from 0 to 1,000 or greater (p < 0.001). The relative hazard for all-cause mortality ranged from 1.68 for a CAC score of 1-10 (p < 0.001) to 6.26 for a score of 1,000 or greater (p < 0.001). The categorical net reclassification improvement using cut points of <7.5% to ≥22.5% was 0.21 (95% confidence interval, 0.16-0.32).


The extent of CAC accurately predicts 15-year mortality in a large cohort of asymptomatic patients. Long-term estimates of mortality provide a unique opportunity to examine the value of novel biomarkers, such as CAC, in estimating important patient outcomes.


While the study is limited by the availability of cardiovascular risk factors only at baseline, the lack of data on treatment and cause of deaths, and physician referral bias, as with short-term (5-year) clinical reports that demonstrated the relationship between CAC and cardiovascular events, the relationship between CAC and long-term (15-year) total mortality could have been negatively influenced by treatment of those with high-risk and neglect of those with low-risk CAC scores. This observational study demonstrates the ‘warranty’ of the zero calcium score (3% 15-year all-cause mortality), and the utility of adding the CAC to risk factors for net reclassification improvement for all-cause mortality.

Clinical Topics: Noninvasive Imaging, Prevention, Vascular Medicine, Atherosclerotic Disease (CAD/PAD)

Keywords: Arteriosclerosis, Biological Markers, Cardiovascular Diseases, Cause of Death, Coronary Artery Disease, Diagnostic Imaging, Primary Prevention, Prognosis, Proportional Hazards Models, Risk, Risk Factors, Vascular Calcification

< Back to Listings