Comparison of direct methods and HPLC for the measurement of HDL- and LDL-cholesterol with ultracentrifugation.

Although ultracentrifugation is the gold standard for lipoprotein analysis, inexpensive and easy direct methods for HDL- and LDL-cholesterol (C) have recently been developed. In this study, we compared representative methods of lipoprotein analysis, namely, ultracentrifugation, direct assay methods, and HPLC, to measure LDL- and HDL-C. A good correlation was observed between HDL-C by ultracentrifugation and HDL-C by direct methods or HPLC. A good correlation was also observed between LDL-C (d1.006-1.063) by ultracentrifugation and LDL-C by direct methods or HPLC. Although the correlation between LDL-C (d1.019-1.063) by ultracentrifugation and LDL-C by direct methods was also good, the correlation coefficient was significantly decreased, suggesting that 'LDL-C' by direct methods correlates better with LDL-C (d1.006-1.063) than LDL-C (d1.019-1.063) by ultracentrifugation. Although the correlation between IDL-C (d1.006-1.019) by ultracentrifugation and the difference in LDL-C by direct methods and LDL-C (d1.019-1.063) by ultracentrifugation was investigated, no significant correlation was observed. The IDL-C contained in LDL-C (d1.006-1.063) varied from 2-28%. In homozygous CETP-deficient and LCAT-deficient subjects, the dissociation was marked. It is crucial to understand that 'LDL-C' in the Guidelines for the Diagnosis and Treatment of Hyperlipidemias in Adults by the Japanese Atherosclerosis Society should be considered to be LDL-C (d1.006-1.063) and that 'LDL-C' by direct assay methods means LDL-C (d1.006-1.063) by ultracentrifugation.