Comparative prognostic utility of conventional and novel lipid parameters for cardiovascular disease risk prediction: Do novel lipid parameters offer an advantage?

Abstract

Comparative data on the prognostic utility of novel lipid parameters vs. conventional lipid parameters in predicting coronary events are scant. We sought to compare the predictive value of various lipid measures for coronary events and to further examine the incremental value of novel lipid parameters over traditional cardiovascular risk factors in estimating cardiac risk. We performed a post-hoc analysis of the National Heart Lung and Blood Institute limited access dataset of Multi-Ethnic Study of Atherosclerosis subjects (n = 6693). The lipid measures considered in the estimation of coronary risk were conventional and novel lipid parameters, the latter included total low-density lipoprotein (LDL), high-density lipoprotein (HDL) and very low-density lipoprotein (VLDL)-particle concentrations (LDL-p, HDL-p and VLDL-p), LDL-p/HDL-p ratio, and LDL-p subfractions. The outcome measured was occurrence of any coronary event (CE) that included myocardial infarction, resuscitated cardiac arrest, cardiac death, and angina. During an average follow up of 4.5 years, 228 patients developed coronary events. In the multivariate Cox proportional hazards model, TC/HDL-c (HR: 3.27; 95% CI: 1.95 to 5.47, P < .0001) was a stronger predictor of CE. Among the novel lipid parameters, LDL-p/HDL-p (hazard ratio 2.84; 95% confidence interval 1.89 to 4.26; P < .0001) was a powerful independent predictor of CE. The c-statistics were similar for both LDL-p/HDL-p and TC/HDL-c ratios (0.60). The addition of LDL-p/HDL-p ratio to the Framingham risk score components resulted in a very small increase in the overall C statistic. In our large study cohort, a predictive model for future coronary events incorporating the best-available novel lipid parameter (LDL-p/HDL-p ratio) was comparable with the same model that incorporated conventional lipid ratios such as the TC/HDL-c ratio . The use of LDL-p/HDL-p ratio did not appear to offer incremental value over more traditional risk prediction models.