Chemometric-assisted UV spectrophotometric methods for determination of miconazole nitrate and lidocaine hydrochloride along with potential impurity and dosage from preservatives

Abstract

Three accurate, simple, and precise chemometric techniques, principal component regression (PCR), partial least squares (PLS), and backward interval partial least squares (biPLS) were used to resolve the severely overlapped UV spectra of miconazole nitrate (MIC) and Lidocaine hydrochloride (LDC) along with the toxic impurity of LDC; dimethyl aniline (DMA) and the two inactive ingredients; methyl paraben (MTP) and saccharin sodium (SAC). The concentration ranges of the developed models were found to be (2.40–12.00 µg/mL) for LDC and MIC, (1.50–7.50 µg/mL) for DMA and MTP, and (2.00–6.00 µg/mL) for SAC. The proposed methods were found to be green, rapid, and were effectively used to analyze the studied compounds in both laboratory-prepared mixtures and antifungal oral gel, where no impurity was detected. The obtained results revealed that PLS algorithm was superior to PCR depending on the lowest root mean square error of prediction (RMSEP) and correlation coefficient values (r). The biPLS model, constructed with [3, 4, 5, 6, 8, and 9] subintervals, is considered the most efficient model with the lowest number of latent variables. biPLS is ideal for data analysis and enhancing model performance and robustness by focusing on the most relevant spectral regions. When compared to a reported HPLC method, the proposed methods showed non-significant difference regarding accuracy and precision. The developed models often yield faster results than HPLC. Once the model is built, it takes no time to predict multiple samples without requiring reconstruction, in addition, the proposed models minimize the costs of solvents and equipment compared to HPLC, making them a valuable option for quality control laboratories.