The plasma lipoproteins as risk factors: comparison of electrophoretic and ultracentrifugation results.

Epidemiologic data from 1,859 men aged 45-65 were examined to explore which approach to analyzing plasma lipids and lipoproteins would be most useful for clinical medicine. In addition to analyzing fasting plasma lipids (cholesterol and triglyceride), we used two chemical techniques to measure the three major classes of lipoproteins: ultracentrifugation with heparin/manganese precipitation which yielded values for HDL-, LDL- and VLDL-cholesterol, and quantitative electrophoresis which yielded values for alpha-, beta- and pre-beta-lipoprotein. Both measures of the cholesterol-rich lipoprotein class (LDL-cholesterol and beta-lipoprotein) were related to CHD prevalence, but neither appeared to be superior to total serum cholesterol in the strength--approximately twofold--of the association. Both measures of the protein-rich lipoprotein class (HDL-cholesterol and alpha-lipoprotein) showed a two-fold inverse association with CHD, although the latter did not attain statistical significance. No member of the triglyceride-rich class (pre-beta-lipoprotein, VLDL-cholesterol, or serum triglyceride itself) was independently associated with CHD. We conclude that plasma total cholesterol remains a good single test for most clinical situations because of the prognostic information it contains, its widespread availability and its role in treatment decisions; also that either the plasma alpha-lipoprotein or HDL-cholesterol level can provide additional information on the likelihood of disease.