Simultaneous Quantification of Metronidazole and Spiramycin by Ultraviolet-Visible Method: Comparison of Wavelength Selection Techniques

Abstract

The accurate quantification of analytes in samples with overlapping spectra remains a critical challenge in analytical chemistry, particularly for complex pharmaceutical formulations. Developing a method for simultaneous quantification of metronidazole (MTZ) and spiramycin (SPI) that is rapid, accurate, reliable, and cost-effective is imperative. This study aims to apply an ultraviolet-visible (UV-Vis) method integrating wavelength selection algorithms with partial least squares (PLS) regression for the simultaneous quantification of MTZ and SPI in tablets. Concurrently, it clarifies the steps to compute the limit of detection range, specifically the minimum and maximum values, in PLS methods. UV-Vis spectra of MTZ and SPI mixtures in standard concentration sets were scanned from 200–400 nm with a scan interval of 0.5 nm. The backward variable elimination–partial least square (BVE-PLS), and genetic algorithm–partial least square (GA-PLS) methods were employed for optimal wavelength selection. Analyte concentrations were computed via PLS methods using acquired spectral data. Multivariate regression models were evaluated via statistical parameters: the coefficient of determination (R²), root mean square error, and residual analysis. Methods were assessed following Association of Official Analytical Chemists guidelines. The results demonstrate that the BVE-PLS and GA-PLS wavelength selection methods produced superior computational outcomes compared to full spectral data PLS methods. Ultimately, simultaneous quantification methods for MTZ and SPI were successfully developed, validated, and shown to be accurate, straightforward, rapid, cost-effective, and stable.