A comparison of methods for the estimation of serum cholesterol and values in random samples of populations in the 55-64 age group.

The increasing mortality from coronary heart disease has focused attention on its epidemiological and biological aspects, and in the latter connexion much work has been done on the relationship between blood cholesterol levels and heart disease. As a result many methods for the estimation of cholesterol levels in serum have been published and their multiplicity indicates the difficulties of the estimation. Reviews published by Zak and Ressler (1955) and Cook (1958) deal with the isolation and estimation, but few comparisons of more than two methods are available. The estimation has usually to be done on small amounts of material so that semi-micro colorimetric methods are normally used. The majority of these are based upon the LiebermannBurchard reaction. Liebermann described the colour changes which occur when concentrated sulphuric acid is added to a solution of cholesterol in acetic anhydride, and Burchard later developed this by applying the reaction in a nonaqueous solvent such as chloroform. The more sensitive method of Zlatkis, Zak, and Boyle (1953) makes use of the colour developed by cholesterol with concentrated sulphuric acid and ferric chloride. Several modifications of this method have been published, which suggests that it has some technical difficulties.