Skip to main content
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: JACC Cardiovasc Imaging. 2020 Oct 28;14(1):219–229. doi: 10.1016/j.jcmg.2020.06.047

Table 4.

AUC analysis for prediction of long-term CAC=0 among individuals with metabolic syndrome or type 2 diabetes.

C-Statistic (95% CI) Model P-value Change in C-Statistic* AUC Contrast P-value*
Demographics
Age, Sex, Race 0.625 (0.577, 0.673) <0.001 - -
Traditional Risk Factors/Risk Scoring
Individual Risk Factors 0.671 (0.625, 0.718) <0.0001 +0.046 <0.01
Pooled Cohort Equations ASCVD Risk 0.599 (0.551, 0.648) <0.01 −0.026 0.17
Novel Risk Factors/Risk Scoring
Individual Risk Factors + Metabolic Syndrome Severity Score 0.673 (0.627, 0.719) <0.001 +0.048 <0.01
Individual Risk Factors + hs-cTnT, NT-proBNP, Lp(a), hs-CRP, eGFR, Absence of Urine Protein 0.686 (0.625, 0.747) 0.05 +0.061 0.12
Imaging
Individual Risk Factors + Absence of Extra-Coronary Atherosclerosis 0.698 (0.650, 0.740) <0.0001 +0.073 <0.01
Demographics, Traditional Risk Factors, Novel Risk Factors, and Imaging § 0.709 (0.650, 0.768) <0.01 +0.084 0.03

All models, except pooled cohort equations model, include age, sex, and race as covariates.

*

=ROC curves were compared to the reference model including age, sex, and race.

= systolic blood pressure, diastolic blood pressure, total cholesterol, HDL cholesterol, fasting blood glucose, waist circumference, triglycerides, antihypertensive medication, glucose-lowering medication,

lipid-lowering medication.

= carotid plaque and thoracic calcification.

§

= all individual risk factors listed above, except the pooled cohort equations ASCVD risk and the metabolic syndrome severity score.

ASCVD=atherosclerotic cardiovascular disease; AUC=area-under-the-curve; eGFR=estimated glomerular filtration rate; hs-CRP=high sensitivity C-reactive protein; hs-cTnT=high-sensitivity cardiac troponin T; Lp(a)=lipoprotein a;

NT-proBNP=n-terminal prohormone brain natriuretic peptide.