Analysis of plasma lipoproteins by ultracentrifugation in a new fixed angle rotor: evaluation of a phosphotungstic acid/MgCl2 precipitation and a quantitative lipoprotein electrophoresis assay.

A routine procedure for the ultracentrifugal analysis of human plasma lipoproteins by means of a fixed angle rotor was developed. It was applied as reference method for the evaluation of two widely used procedures of lipoprotein analysis, i.e. measurement of HDL-cholesterol after precipitation of the apolipoprotein B containing lipoproteins with a new phosphotungstic acid/MgCl2 reagent, and a quantitative lipoprotein electrophoresis system. A new statistical approach to the multivariate comparison of analytical methods, the linear structural relationship model, has been applied. The HDL-cholesterol levels measured after phosphotungstic acid/MgCl2 precipitation agreed well to those determined by ultracentrifugation, whereas electrophoretically quantified alpha-lipoprotein-cholesterol exceeded the ultracentrifugal HDL-cholesterol. The Friedewald formula obviously underestimated LDL-cholesterol due to an overestimation of VLDL-cholesterol, whereas the electrophoretically quantified pre-beta- and beta-lipoprotein-cholesterol levels fairly coincided with the respective ultracentrifugal measurements. The inter-assay reproducibility of the HDL-cholesterol determination after phosphotungstic acid/MgCl2 precipitation and subsequent LDL-cholesterol quantification according to Friedewald was statistically equivalent to that of ultracentrifugation, whereas the quantitative lipoprotein electrophoresis proved less precise.