Abstract
In this study, spectrophotometric determination of donepezil and rivastigmine in healthy human urine samples was carried out by the statistical method. Partial least squares (PLS) and principal component regression (PCR) from multivariate calibration methods were used to evaluate the data obtained from the UV-Vis spectroscopy analysis of the urine sample. Mixtures of each early substance were prepared prior to urine sample analysis, and simultaneous determination of donepezil and rivastigmine was performed on the established chemometric model without any prior separation. The calibration curves of each drug were analyzed, and linearity values were also analyzed. For donepezil and rivastigmine, they were 0.9989 and 0.9997, respectively, and were linear over the concentration range of the synthetic mixture. When both chemometric methods (PLS and PCR) were evaluated in terms of accuracy and reproducibility, very high recoveries and small standard deviations were determined. In the PLS method, the standard error of prediction (SEC), the sum of the prediction residual errors (PRESS), the limit of quantitation (LOQ), and the limit of detection (LOD) values were 0.015, 0.0030, 0.067, 0.24, 0.018, 0.0042, 0.089, and 0.301 for donepezil and rivastigmine, respectively. In the PCR method, SEC, PRESS, LOD, and LOQ values are 0.016, 0.0054, 0.066, and 0.23 for donepezil and 0.022, 0.0062, 0.091, and 0.300 for rivastigmine. Chemometrics is used for speed, simplicity, and reliability. The proposed methods have been successfully applied to a sample of urine.