LAUSR

Background Coronary artery calcium (CAC) has been demonstrated as a noninvasive, low-cost means of assessing atherosclerotic burden and risk of major cardiovascular events. While it has been previously shown that CAC progression predicts all-cause mortality, we sought to quantify this association by examining a large cohort over a follow-up period of 1–22 years. Methods We studied 3260 persons aged 30–89 years referred by their primary physician for CAC measurement, with a follow-up scan at least 12 months from the initial scan. Receiver operator characteristic (ROC) curves assessed a level of annualized CAC progression that predicted all-cause mortality. Multivariate analyses using Cox proportional hazards models were used to compute hazard ratios and 95% confidence intervals (CIs) for the association between annualized CAC progression and death after adjusting for relevant cardiovascular risk factors. Results The average time between scans was 4.7 ± 3.2 years with an additional average of 9.1 ± 4.0 years of follow-up time. The average age of the cohort was 58.1 ± 10.5 years, 70% being male, and 164 deaths occurred. Annualized CAC progression of 20 units optimized sensitivity (58%) and specificity (82%) in ROC curve analysis. Annualized CAC progression of 20 units was significantly associated with mortality while adjusting for age, sex, race, diabetes, hypertension, hyperlipidemia, smoking, baseline CAC level, family history, and time between scans, hazard ratio 1.84 (95% CI, 1.28–2.64) P = 0.001. Conclusion Annualized CAC progression of greater than 20 units per year significantly predicts all-cause mortality. This may add clinical value in encouraging close surveillance and aggressive treatment of individuals within this range.