Direct analysis of amiloride and triamterene mixtures by fluorescence spectrometry using partial-least squares calibration

The simultaneous determination of amiloride and triamterene in urine samples using fluorimetry in combination with partial-least squares (PLS) multivariate calibration is proposed. Triamterene and amiloride exhibit overlapped spectra and urine produces background fluorescence that precludes the direct determination of these diuretics by conventional fluorimetry. Although, the qualitative composition of the fluorescent metabolites in urine from healthy people is virtually invariable, their quantitative composition exhibits some differences. Thus, the shape of the spectrum and the position of the fluorescence maxima change as the urine is diluted. The determination was performed in a 1:1 (v/v) ethanol /water medium at an apparent pH of 6.3 provided by 0.01M sodium citrate/citric acid buffer. An excitation wavelength of 365nm was used for both amiloride and triamterene. The corresponding emission maxima were located at 413 and 437nm for amiloride and triamterene, respectively. A calibration set consisting of samples standards was used in combination with a factor design; two levels per factor and a central star design (i.e. a central composite design) were used. In order to ensure accurate results, the calibration matrix was implement an urine sample containing no triamterene or amiloride (i.e. urine blank). The components of the calibration matrix were triamterene, amiloride and urine. The concentration of amiloride was varied from 64 to 320ng/ml and that of triamterene from 20 to 100ng/ml. Urine dilution was varied from 1:35 to 1:65. Urine was used as the third component of the calibration matrix in order to include the information inherent in changes in the fluorescence spectrum for urine upon dilution.