Effects of sample aging on total cholesterol values determined by the automated ferric chloride-sulfuric acid and Liebermann-Burchard procedures.

To investigate the comparability of three commonly used methods for determination of total cholesterol in plasma in several studies, we used fresh plasma samples as well as plasmas and reference sera that had been stored frozen at -15 degrees C for as long as several years. Duplicate determinations by the manual method of Abell et al. (J. Biol. Chem. 195: 357, 1952) were compared with estimates from one to five continuous-flow analyzers by the ferric chloride-sulfuric acid procedure and also with estimates from five to 13 continuous-flow analyzers by the Liebermann-Burchard procedure with calibrator, as part of the laboratory standardization activities of the Lipid Research Clinics. The agreement among all three procedures was generally within acceptable limits (within 5% of the manual method) when plasmas or sera were fresh or had been frozen for less than one month. Results by the manual method of Abell et al. agreed well with those by the automated Liebermann-Burchard method for samples that had been stored at -15 degrees C for as long as two years. However, the automated ferric chloride-sulfuric acid procedure often showed unacceptably high values (as compared with those from the manual method) for samples that had been stored frozen for a year or more. With the ferric chloride-sulfuric acid method, measured cholesterol concentration increased about 2.5% per year of storage for at least two years. We conclude that reference sera of plasmas that have been kept in long-term frozen storage (-15 degrees C) are not suitable for ongoing standardization of the automated ferric chloride-sulfuric acid assay for cholesterol.