Chemometric-assisted spectrophotometric methods and high performance liquid chromatography (HPLC) were developed for the simultaneous determination of the seven most commonly prescribed β-blockers (atenolol, sotalol, metoprolol, bisoprolol, propranolol, carvedilol and nebivolol). Principal component regression PCR, partial least square PLS and PLS with previous wavelength selection by genetic algorithm (GA-PLS) were used for chemometric analysis of spectral data of these drugs. The compositions of the mixtures used in the calibration set were varied to cover the linearity ranges 0.7-10 μg ml(-1) for AT, 1-15 μg ml(-1) for ST, 1-15 μg ml(-1) for MT, 0.3-5 μg ml(-1) for BS, 0.1-3 μg ml(-1) for PR, 0.1-3 μg ml(-1) for CV and 0.7-5 μg ml(-1) for NB. The analytical performances of these chemometric methods were characterized by relative prediction errors and were compared with each other. GA-PLS showed superiority over the other applied multivariate methods due to the wavelength selection. A new gradient HPLC method had been developed using statistical experimental design. Optimum conditions of separation were determined with the aid of central composite design. The developed HPLC method was found to be linear in the range of 0.2-20 μg ml(-1) for AT, 0.2-20 μg ml(-1) for ST, 0.1-15 μg ml(-1) for MT, 0.1-15 μg ml(-1) for BS, 0.1-13 μg ml(-1) for PR, 0.1-13 μg ml(-1) for CV and 0.4-20 μg ml(-1) for NB. No significant difference between the results of the proposed GA-PLS and HPLC methods with respect to accuracy and precision. The proposed analytical methods did not show any interference of the excipients when applied to pharmaceutical products.
Keywords: GA-PLS; HPLC; PCR; PLS; Statistical experimental design; β-Blockers.
Copyright © 2015 Elsevier B.V. All rights reserved.