Abstract
BACKGROUND: Coronary artery calcium scoring (CACS) may improve risk stratification of diabetics, in whom coronary artery disease (CAD) is often more diffuse and progressive. We examined the long-term value of CACS in outcome prediction of asymptomatic patients with type 2 diabetes. METHODS: A population-based cohort recruited between 2006 and 2008 of type 2 diabetics aged 55-74 years, with at least one additional risk factor and no history or symptoms of CAD. The association of overall (n = 735) and individual vessel (n = 596) CACS with the occurrence of myocardial infarction (MI), ischemic stroke, or all-cause death (MACE) was assessed over a median follow-up of 17.3 years. RESULTS: Increasing levels of CACS independently predicted MACE and its components, particularly MI, with no events of MI observed in those with zero CACS, and an adjusted hazard ratio (95 % confidence interval) of 2.31 (1.30-4.11) for MI in those with CACS > 300 compared to 1-100 Agatston units. The addition of CACS to MESA 10-year coronary heart disease risk score increased significantly the discriminatory capacity for MI [combined AUC 0.75 (0.72-0.78)]. For each additional coronary artery with any CACS, the adjusted HR for MI was 1.68 (1.34-2.10), and for MACE 1.18 (1.08-1.30), p < 0.001. The hazard for MI or MACE did not increase in proportion to the percentage of CACS localized to the left-main coronary artery. CONCLUSIONS: The predictive value of CACS in asymptomatic diabetics is retained over 17 years of follow-up, particularly for MI. The number of coronary arteries with calcified plaques may provide additional prognostic information.