Variability in cholesterol measurements: comparison of calculated and direct LDL cholesterol determinations

Calculated low-density lipoprotein cholesterol (LDL-C) concentrations determined from the Friedewald equation have a large intraindividual CV, in part because the calculation incorporates the variability of cholesterol, high-density lipoprotein cholesterol (HDL-C), and triglyceride measurements. We studied whether a new assay that measures LDL-C directly will reduce this variability and reduce the need for averaging serial specimens. Four blood samples were obtained 1 week apart from 35 mildly hypercholesterolemic subjects and analyzed for total cholesterol, triglycerides, and HDL-C. LDL-C was calculated by the Friedewald equation, and was also measured directly with a commercially available direct LDL-C assay. The intraindividual CV for the direct and calculated LDL-C assays were similar [CV of direct LDL-C assay (mean +/- SE): 6.8 +/- 0.5% vs calculated LDL-C: 7.3 +/- 0.6%; difference 0.44%, 95% confidence interval: -0.7-1.5%]. For both assays, at least two blood tests were required from each subject to reduce total variability of LDL-C to less than or equal to 5%. We conclude that the direct LDL-C assay did not reduce the variability in LDL-C compared with the conventional LDL-C calculation. However, it may have a specific role in lipid disorder evaluation and (or) monitoring when triglycerides are increased or the LDL-C value alone is needed.