Background: In routine clinical laboratory testing and numerous epidemiological studies, LDL-cholesterol (LDL-C) has been estimated commonly using the Friedewald equation. We investigated the relationship between the Friedewald equation and 4 homogeneous assays for LDL-C.
Methods: LDL-C was determined by 4 homogeneous assays [liquid selective detergent method: LDL-C (L), selective solubilization method: LDL-C (S), elimination method: LDL-C (E), and enzyme selective protecting method: LDL-C (P)]. Samples with discrepancies between the Friedewald equation and the 4 homogeneous assays for LDL-C were subjected to polyacrylamide gel electrophoresis and the beta-quantification method.
Results: The correlations between the Friedewald equation and the 4 homogeneous LDL-C assays were as follows: LDL-C (L) (r=0.962), LDL-C (S) (r=0.986), LDL-C (E) (r=0.946) and LDL-C (P) (r=0.963). Discrepancies were observed in sera from type III hyperlipoproteinemia patients and in sera containing large amounts of midband and small dense LDL on polyacrylamide gel electrophoresis. LDL-C (S) was most strongly correlated with the beta-quantification method even in sera from patients with type III hyperlipoproteinemia.
Conclusions: Of the 4 homogeneous assays for LDL-C, LDL-C (S) exhibited the closest correlation with the Friedewald equation and the beta-quantification method, thus reflecting the current clinical databases for coronary heart disease.