Abstract
AIM: The SCORE2 algorithm is recommended to estimate risk of cardiovascular disease (CVD). Coronary artery calcification (CAC) score is expensive but improves the risk prediction. This study aims to determine and compare the additive value of CAC-score and 19 biomarkers in risk prediction. METHODS: Traditional cardiovascular (CV) risk factors, CAC-score, and a wide range of biomarkers (including lipids, calcium-phosphate metabolism, troponin, inflammation, kidney function and ankle brachial index (ABI)) were collected from 1211 randomly selected middle-aged men and women in this multicenter prospective cohort in 2009-2010. 10-year follow-up data on CV-events were obtained via the Danish Health Registries. CV-event was defined as stroke, myocardial infarction, hospitalization for heart failure, coronary artery revascularization or death from CVD. The association between SCORE2, CAC-score, biomarkers, and CV-events was assessed using cox proportional hazard rates (HR) and compared using AUC-calculation of ROC-curves. Finally, net reclassification improvement (NRI) was calculated. RESULTS: 92 participants had CV-events. Adjusted for risk factors, CAC-score was significantly associated with events (adjusted HR 1.9 (95%CI:1.1; 3.3), 3.6 (95%CI:1.9; 6.8), and 5. (95%CI:2.6; 10.3) for CAC-score 1-99, CAC-score 100-399 and CAC-score ≥400, respectively. HR for the highest quartile of CRP was 2.3 (95%CI:1.2; 4.5), while none of the remaining biomarkers improved HR. Adjusted for SCORE2, the CAC-score improved AUC (AUC(CAC): 0.72, AUC(SCORE2): 0.67, p<0.01). A combination of selected biomarkers (total cholesterol, low-density lipoprotein, phosphate, troponin, CRP, and creatinine) borderline improved AUC (AUC(Biomarkers + SCORE2): 0.71, AUC(SCORE2): 0.67, p=0.06). NRI for CAC score was 63 % (p<0.0001). CONCLUSION: CAC-score improved prediction of CV-events, however the selected biomarkers did not.