Novel lipid parameters no better than conventional measures for CVD risk prediction
MedWire News: Results of a US study do not support the routine use of novel lipid parameters to assess future cardiovascular disease (CVD) risk.
Luis Afonso, from Wayne State University in Detroit, Michigan, and colleagues found that the prognostic value of particle concentration and low-density lipoprotein (LDL) cholesterol subfractions were comparable to that of conventional lipid parameters in predicting future coronary events.
Furthermore, the use of novel lipid particle ratios in risk prediction models did not appear to offer incremental value over more traditional measures.
The team compared the prognostic value of conventional lipid parameters (total cholesterol, LDL cholesterol, high-density lipoprotein [HDL] cholesterol, triglycerides, and nonHDL cholesterol), conventional lipid ratios (total/HDL cholesterol, LDL/HDL cholesterol, and triglycerides/HDL cholesterol), and nuclear magnetic resonance (NMR)-measured novel lipid parameters (total very-LDL [VLDL], LDL and HDL particle concentrations [VLDL-p, LDL-p, and HDL-p], LDL-p/HDL-p ratio, and all the LDL-p subfractions) in 6693 men and women, aged 45-84 years, using data from the Multi-Ethnic Study of Atherosclerosis.
In total, 228 patients experienced coronary events over the median follow-up period of 4.5 years.
Afonso and team report that all three categories of lipid parameters were significantly predictive of future coronary events, with the exception of HDL cholesterol and large LDL-p concentration.
The total/HDL cholesterol ratio was the strongest predictor overall; the incidence of coronary events increased 3.27-fold for every 1-unit increase, while LDL-p/HDL-p ratio was the strongest independent predictor among the novel lipid parameters, increasing coronary risk 2.84 fold for every 1-unit increase.
However, c-statistics were similar for both ratios, at 0.60, indicating a comparable predictive value for LDL-p/HDL-p and total/HDL cholesterol ratios.
Furthermore, addition of LDL-p/HDL-p ratio to the Framingham risk score components only added minimal incremental value to the model compared with the existing traditional risk prediction model; the respective c-statistics were 0.74 versus 0.73.
Commenting on the implications of these findings in the Journal of Clinical Lipidology, the researchers say that the cost involved in NMR spectroscopy would demand a large increase in the predictive utility of novel lipid parameters.
Therefore, the team concludes that the expense of measuring lipid particles "may not be cost-effective or justified, given the marginal incremental value to risk prediction."
MedWire (www.medwire-news.md) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2011
By Nikki Withers