Non–HDL Cholesterol Shows Improved Accuracy for Cardiovascular Risk Score Classification Compared to Direct or Calculated LDL Cholesterol in a Dyslipidemic Population

Study Questions:

How accurate is a cardiovascular disease (CVD) risk score determined using three different measurements: direct low-density lipoprotein cholesterol (LDL-C), calculated LDL-C, and non–high-density lipoprotein cholesterol (non–HDL-C), when compared to classification by reference measurement?


A total of 175 individuals had lipids measured, including 138 with CVD or conditions that may affect LDL measurement. Direct LDL measurements using eight different reagents were compared with calculated LDL calculated by the Friedewald equation, using each manufacturer’s direct HDL assay measurements, and total cholesterol and triglyceride measurements by Roche and Siemens (Advia) assays, respectively. LDL was compared to the reference measurement performed at the Centers for Disease Control and Prevention.


For participants with triglycerides <200 mg/dl, the overall misclassification rate for the CVD risk score ranged from 5% to 17% for calculated LDL-C methods and 8% to 26% for direct LDL methods when compared to the reference measure. Non–HDL assays misclassified fewer patients than direct LDL for four of eight methods (p < 0.05). For participants with triglycerides ≥200 mg/dl and <400 mg/dl, direct LDL methods, in general, performed better than calculated LDL methods, and non–HDL methods showed better correspondence to the reference method for CVD risk score than either direct LDL or calculated LDL methods. ApoB correlated poorly with all direct LDL measurements (R2 from 0.47-0.61), but better with non–HDL (R2 = 0.83-0.84).


Except for hypertriglyceridemic individuals, seven of eight direct LDL methods failed to show improved CVD risk score classification over the corresponding calculated LDL methods. Non–HDL showed overall the best concordance with the reference method for CVD risk score classification of both normal and hypertriglyceridemic individuals.


Data are again moving towards the idea that LDL concentration or particle number might be better than calculated or direct LDL measurements. The major advantage of using non–HDL is that it is easily calculated, and correlates fairly well with concentration and particle number, but without additional cost. Some laboratories have been reporting this for years.

Clinical Topics: Diabetes and Cardiometabolic Disease, Dyslipidemia, Hypertriglyceridemia, Lipid Metabolism, Nonstatins

Keywords: Lipoproteins, LDL, Cholesterol, Dyslipidemias, Biological Markers, Cholesterol, LDL, Cardiovascular Diseases, Risk Factors, Cholesterol, HDL, Lipoproteins, HDL, Triglycerides, United States

< Back to Listings