Abstract MP054: Coronary Calcification, C-reactive Protein and Prediction of Cardiovascular and All-Cause Mortality in Diabetes: Diabetes Heart Study

Abstract

Objective: Does coronary artery calcium (CAC) and C-reactive protein (CRP) improve discrimination and risk stratification of cardiovascular (CVD) and all-cause mortality in diabetes. Background: Inflammation is associated in the pathogenesis and progression of atherosclerosis. Since diabetes is characterized by both atherosclerosis and high CRP levels, we hypothesized that CAC and CRP will predict CVD and all-cause mortality in diabetes beyond Framingham risk factors (FRS). Method: A total of 961 participants, ages 39-86, with complete data on diabetes, vascular imaging, and inflammation in the Diabetes Heart Study were followed for an average of 7.4 years. CAC was assessed by CT at baseline. CRP was measured using enzyme-linked immunosorbent assay. Risk estimates for CVD and all-cause mortality were obtained using logistic regression models. Model 1 included FRS; age, gender, smoking, systolic blood pressure, antihypertensive medication use, total and HDL cholesterol, and ethnicity. Model 2 included FRS plus CRP or CAC. Model 3 included FRS plus CRP and CAC. Models for CVD and all-cause mortality were compared using c-statistics. Results: During follow-up there were 183 (19%) deaths of which 80 (8%) were CVD deaths. In FRS adjusted models both CRP and CAC were independently associated with all-cause mortality, however only CAC was associated with CVD mortality. The addition of CAC to a model with FRS and CRP improves discrimination for CVD and all-cause mortality (both p=0.001). However, the addition of CRP to a model with FRS and CAC only improves discrimination in all-cause mortality (p=0.001) ( Table 1 ). Conclusion: In participants with T2DM, measures of CAC and CRP are significantly associated with CVD (CAC only) and all-cause (CAC and CRP) mortality independent of the FRS; however, the improvement in CVD and all-cause mortality prediction was predominantly driven by CAC. Table 1 C-Statistics for the Combined Assessment of Framingham Risk Factors, CAC and CRP in Predicting Cardiovascular and All-Cause Mortality CVD Mortality All-Cause Mortality C-Statistic p-value C-Statistic p-value Model 1 FRS Only 0.715 -- 0.701 -- Model 2 vs. Model 1 Adding CRP to FRS Adding CAC to FRS 0.715 0.768 P=0.737 P=0.001 0.709 0.760 p=0.002 P=0.001 Model 3 vs. Model 2 Adding CAC to FRS + CRP Adding CRP to FRS + CAC 0.769 0.769 P=0.001 P=0.981 0.765 0.765 P=0.001 P=0.001