A.M. Vasilevsky, G.A. Konoplev, O.S. Lopatenko

Uric acid is a final product of purine metabolism in humans. Changes of it-s concentration in biological media are observed in various morbid conditions, e.g. a gout, an atherosclerosis, coronary heart disease, tumor processes. In clinical practice determination of uric acid concentration is a routine laboratory test which is carried out quite often. The series of uric acid solutions with concentrations 0.05, 0.10, 0.15, 0.20 mmol/l was prepared. Then UV absorption spectra of these samples in the relevant wavelengths range of 190 - 350 nm were measured. The spectra were simulated mathematically as a combination of Lorentzian and Gaussian curves. The procedure is based on the maximization of the similarity coefficient (normalized scalar multiplication of two vectors representing spectra) between the measured spectrum of uric acid and the spectrum recovered as a result of simulation. Values of similarity coefficient exceeding 999.0 were achieved. Concentration calculated from simulated spectra differed from the real values no more than 5%. The offered mathematical model of uric acid spectral absorption can be used in the multicomponent spectral analysis of low-molecular weight media, e.g. serum, dialyzate and urine.