Clinical impact of cardiovascular calcifications on stroke incidence in primary prevention: analysis in NADESICO study.

Abstract

The utility of assessment of cardiovascular calcifications for predicting stroke incidence remains unclear. This study assessed the relationship between cardiovascular calcifications including coronary artery calcification (CAC), aortic valve (AVC), and aortic root (ARC) assessed by coronary computed tomography (CT) and stroke incidence in patients with suspected CAD. In this multicenter prospective cohort study, 1187 patients suspected of CAD who underwent coronary CT were enrolled. Cardiovascular events including stroke were documented. Hazard ratio (HR) and confidence interval (CI) were assessed by Cox proportional hazard model adjusted for the Framingham risk score. C statistics for stroke incidence were also examined by models including cardiovascular calcifications. A total of 980 patients (mean age, 65 ± 7years; females, 45.8%) were assessed by the CAC, AVC, and ARC Agatston scores. During a median follow-up of 4.0years, 19 patients developed stroke. Cox proportional hazard model showed severe CAC (Agatston score ≥ 90th percentile [580.0 value]) and presence of AVC and ARC were associated with stroke incidence (HR; 10.33 [95% CI; 2.08-51.26], 3.08 [1.19-7.98], and 2.75 [1.03-7.30], respectively). C statistic in the model with CAC and AVC severity for predicting stroke incidence was 0.841 (95% CI; 0.761-0.920), which was superior to the model with CAC alone (0.762 [95% CI; 0.665-0.859], P < 0.01). CAC, AVC, and ARC were associated with stroke incidence in patients suspected of CAD. Assessment of both CAC and AVC may be useful for prediction of stroke incidence.