Multiple lipid scoring system for prediction of coronary heart disease risk: application to African Americans

BACKGROUND: Clinicians often obtain a panel of lipids but then only use low-density-lipoprotein (LDL) cholesterol to make clinical decisions. We previously described the multiple lipid measure, a strategy that integrates information about seven lipid measures. Our current inquiry uses the multiple lipid measure to create a scoring system and validates that system in a second cohort. METHODS AND RESULTS: A scoring system that uses total cholesterol, high-density lipoprotein (HDL) cholesterol, LDL cholesterol and triglycerides was developed and tested. African-American participants of the Atherosclerosis Risk in Communities (ARIC) Study were used to validate the multiple lipid measure score. For nonsmokers, scores > or = 2 had a hazard ratio of 4.25 (95% CI 1.92-9.40) compared to reference scores of < or = -3 in adjusted survival analysis predicting incident coronary heart disease risk in the ARIC. The best conventional single lipid measure for nonsmokers was LDL cholesterol. Compared to LDL cholesterol <100 mg/dl, those with LDL cholesterol > or = 160 mg/dl had a hazard ratio of 2.31 (95% CI 1.13-4.75). For current smokers, the best conventional lipid measure was the total cholesterol/HDL cholesterol ratio, which was similar in predictive ability to the multiple lipid measure score. However, the multiple lipid measure score predicted an additional 10% of the cohort at risk compared to the total cholesterol/HDL cholesterol ratio. CONCLUSIONS: The use of the multiple lipid scoring system improves the assessment of incident coronary heart disease risk and may have utility for clinicians in integrating lipid values.