A Colorimetric Method for the Determination of Fructose in Blood and Urine*

In 1926 Campbell and Hanna (1) published a method for the determination of fructose based upon the reduction of molybdenum in phosphoric acid solution and the reoxidation of the reduced molybdenum with potassium permanganate. This method is not entirely specific for fructose; other substances in blood and urine have some reducing action upon the molybdic acid reagent. The method will not permit microapplications of the same magnitude as are possible with calorimetric methods, the regular procedure being developed to estimate 1 mg. quantities. The time required for reduction is 90 minutes, which limits the usefulness of the method. In 1927 a method based upon the Seliwanoff reaction was published by Kronenberger and Radt (2). These authors deproteinized blood with trichloroacetic acid and treated the filtrate with resorcinol and HCl. They stated that normal blood filtrates gave a brownish color with their reagents; and they experienced difficulty in determining fructose in urine because of interfering substances. They abandoned their procedure and adopted the method of Van Creveld (3), which was published the same year. Van Creveld’s method makes use of the blue color produced when fructose is dehydrated with HCl in the presence of diphenylamine, a reaction first observed by Ihl and Pechmann (4). In this method blood is deproteinized with HgCl,, the filtrate is mixed with HCl and diphenylamine in a tube and placed in a boiling water bath for 15 minutes, the color obtained is extracted with amyl alcohol, and comparison is made with standards pre-