Related Experiment Videos
Performance of four homogeneous direct methods for LDL-cholesterol
W Greg Miller1, Parvin P Waymack, F Philip Anderson
1Department of Pathology, Virginia Commonwealth University Health System, Richmond, VA 23298, USA. millerg@hsc.vcu.edu
Clinical Chemistry
|February 28, 2002
Summary
Four homogeneous LDL-cholesterol methods showed nonspecificity toward abnormal lipoproteins, failing to meet accuracy goals. These LDL-cholesterol results offer no improvement over the Friedewald equation for triglyceride levels under 4000 mg/L.
Area of Science:
- Clinical Chemistry
- Lipid Metabolism
- Diagnostic Accuracy
Background:
- Evaluation of homogeneous LDL-cholesterol (Low-Density Lipoprotein cholesterol) methods is crucial for accurate cardiovascular risk assessment.
- Existing methods must be assessed for precision, accuracy, and specificity, particularly in patients with abnormal lipoprotein profiles.
Purpose of the Study:
- To evaluate the precision, accuracy, and specificity of four homogeneous LDL-cholesterol methods (Genzyme, Reference Diagnostics, Roche, Sigma).
- To compare the performance of these methods against the beta-quantification reference method.
- To determine if these methods meet the National Cholesterol Education Program (NCEP) goal for total error (<12%).
Main Methods:
- Four homogeneous LDL-cholesterol assays were performed on a Roche/Hitachi 911 analyzer.
- Precision was assessed over 20 days using quality-control and frozen serum specimens.
- Accuracy and specificity were evaluated using sera from 100 participants (including 60 with hyperlipidemias) by comparing results to beta-quantification, analyzing regression, total error, and bias versus triglyceride concentration.
Main Results:
- All methods demonstrated total coefficients of variation (CV) below 2%.
- Significant biases and total errors were observed, with Genzyme (12.6%), Reference Diagnostics (16.5%), Roche (41.6%), and Sigma (38.3%) exceeding the NCEP goal of <12% total error.
- Three methods showed a significant slope of bias versus triglyceride concentration (P <0.001), indicating issues with specificity in hyperlipidemic samples.
Conclusions:
- The evaluated homogeneous LDL-cholesterol methods exhibit nonspecificity towards abnormal lipoproteins, compromising their accuracy.
- None of the tested methods consistently met the NCEP total error goal of <12%.
- These homogeneous LDL-cholesterol assays do not offer improved performance compared to the Friedewald equation for LDL-cholesterol estimation when triglyceride levels are below 4000 mg/L.