Arterial calcification at different sites and prediction of atherosclerotic cardiovascular disease among women and men

Background and aims: The sex-specific contributions of arterial calcification to atherosclerotic cardiovascular disease (ASCVD) risk prediction and stratification in the light of recent modifications by cardiovascular prevention guidelines remain unclear. We assessed the sex-specific value of calcification in different arteries, beyond the Pooled Cohort Equations (PCE) risk factors, for 10-year ASCVD risk prediction. Methods: From 2003 to 2006, participants from the population-based Rotterdam Study (n = 2167) underwent CT to quantify coronary artery calcification (CAC), aortic arch calcification (AAC), extracranial (ECAC) and intracranial carotid artery calcification (ICAC). Follow-up for ASCVD was complete on January 1, 2015. We refitted the PCE (base model), and categorized participants into low (<5%), borderline (5%-7.5%), intermediate (7.5%-20%), and high (>= 20%) ASCVD risk. We extended the models with calcifications and calculated c-statistics and net reclassification improvements for events (NRIe) and non-events (NRIne). Results: CAC predicted ASCVD in women [hazard-ratio (95%-CI) per 1-SD: 1.40 (1.14-1.73)] and men [1.62 (1.27-1.93)]. After addition of CAC to the base model, the c-statistic improved from 0.71 to 0.72 in women; from 0.65 to 0.68 in men. Addition of CAC led to NRIe of 14.3% in women, 4.8% in men and NRIne of 1.5% in women, 15.1% in men. Only in women, ICAC predicted ASCVD [hazard-ratio (95%-CI) per 1-SD: 1.62 (1.26-2.08)], and improved the model (c-statistic from 0.71 to 0.73, NRIe: 9.8% and NRIne: 5.9%). Conclusions: Assessment of CAC improves ASCVD risk prediction and stratification. In women, the added value of ICAC for ASCVD risk prediction is comparable to that of CAC.