Abstract
Measurement of apolipoprotein A-I in native human sera by endpoint or rate nephelometry may give falsely high results for sera that have a high triglyceride content. Such erroneous determinations of apolipoprotein A-I can be avoided if one instead assays supernates of sera from which apolipoprotein B-containing particles have been precipitated with phosphotungstic acid and MgCl2. A high triglyceride concentration in serum does not impair the completeness of the precipitation of apolipoprotein B-containing lipoproteins or the analytical recovery of apolipoprotein A-I. In a prospective epidemiologic study, by using rate nephelometry we found that there were different mean concentrations for apolipoprotein A-I in supernates of sera from men (1.36 g/L) and women (1.49 g/L). Furthermore, the known risk factors for atherosclerosis showed different correlations with apolipoprotein A-I than with high-density lipoprotein cholesterol.