Rapid, Accurate, and Quantitative Detection of Propranolol in Multiple Human Biofluids via Surface-Enhanced Raman Scattering

There has been an increasing demand for rapid and sensitive techniques for the identification and quantification of pharmaceutical compounds in human biofluids during the past few decades, and surface enhanced Raman scattering (SERS) is one of a number of physicochemical techniques with the potential to meet these demands. In this study we have developed a SERS-based analytical approach for the assessment of human biofluids in combination with chemometrics. This novel approach has enabled the detection and quantification of the beta-blocker propranolol spiked into human serum, plasma, and urine at physiologically relevant concentrations. A range of multivariate statistical analysis techniques, including principal component analysis (PCA), principal component discriminant function analysis (PC-DFA) and partial least-squares regression (PLSR) were employed to investigate the relationship between the full SERS spectral data and the level of propranolol. The SERS spectra when combined with PCA and PC-DFA demonstrated clear differentiation of neat biofluids and biofluids spiked with varying concentrations of propranolol ranging from 0 to 120 mu M, and clear trends in ordination scores space could be correlated with the level of propranolol. Since PCA and PC-DFA are categorical classifiers, PLSR modeling was subsequently used to provide accurate propranolol quantification within all biofluids with high prediction accuracy (expressed as root-mean-square error of predictions) of 0.58, 9.68, and 1.69 for serum, plasma, and urine respectively, and these models also had excellent linearity for the training and test sets between 0 and 120 mu M. The limit of detection as calculated from the area under the naphthalene ring vibration from propranolol was 133.1 ng/mL (0.45 mu M), 156.8 ng/mL (0.53 mu M), and 168.6 ng/mL (0.57 mu M) for serum, plasma, and urine, respectively. This result shows a consistent signal irrespective of biofluid, and all are well within the expected physiological level of this drug during therapy. The results of this study demonstrate the potential of SERS application as a diagnostic screening method, following further validation and optimization to improve detection of pharmaceutical compounds and quantification in human biofluids, which may open up new exciting opportunities for future use in various biomedical and forensic applications.