Comparison of MESA of and Framingham risk scores in the prediction of coronary artery disease severity.

BACKGROUND: The correlation between cardiovascular risk scoring systems and the severity of coronary artery diseases (CAD) is not clear. The present research aimed to evaluate the Multi-Ethnic Study of Atherosclerosis (MESA) risk score and Framingham risk score (FRS), using the Gensini score (GS) system as reference, so as to determine which model is better for the prediction of CAD severity. METHODS: This research was a single-center and cross-sectional observational study. In total, 1423 patients were included in our study. Three different groups were formed according to GS: 0 < GS ≤ 22 (low GS group, n = 484); 22 < GS ≤ 42 (intermediate GS group, n = 468); GS > 42 (high GS group, n = 471). Logistic and linear regression analyses were carried out to explore the relationship between the risk score models and the GS. The performance of the risk models was determined by receiver operating characteristic curve (ROC) analysis. RESULTS: The MESA risk score and the FRS both had a statistically significant power for the prediction of CAD severity (MESA area under curve: 0.630; FRS area under curve: 0.613). Furthermore, the MESA had a better performance in predicting the severity (p < 0.05) of CAD compared with the FRS. In the subgroup analysis, the MESA showed a better performance in the male (p < 0.05), diabetes mellitus (p < 0.05), and smoking subgroups (p < 0.05) compared with the FRS. CONCLUSION: The MESA and FRS predicted the severity of CAD in the Chinese population of this study. Moreover, the MESA had a better performance than the FRS model in predicting the severity of CAD in the overall population as well as in the male, smoking, diabetes, and non-diabetes subgroups.