Application of net analyte signal and principal component regression for rapid simultaneous determination of Levodopa and carbidopa in commercial pharmaceutical formulation and breast (human) milk sample using spectrophotometric method

Abstract

In this study, a UV–vis spectrophotometric method coupled with net analyte signal (NAS) and principal component regression (PCR) as multivariate calibration methods were used for the simultaneous determination of levodopa (LEV) and carbidopa (CBD) in prepared mixtures, pharmaceutical formulation, and breast milk sample. The mean recovery of the NAS model was 98.10% and 99.60% for LEV and CBD, respectively. Also, the relative standard deviation (RSD%) values were found to be lower than 5.5% and 4% for LEV and CBD, respectively. On the other hand, the mean recovery of LEV and CBD related to the PCR method was obtained at 96.86% and 92.43%, respectively. K-Fold cross-validation was used to estimate the number of components, which was 7 and 3 with a mean square error prediction (MSEP) of 1.50 and 7.14 for LEV and CBD, respectively. The results revealed that the NAS model was better than the PCR model. Additionally, the proposed NAS-based calibration method was successfully developed for the simultaneous analyses of LEV and CBD in a commercial tablet and breast milk.