A novel method for quantitative analysis of noradrenaline in human plasma samples by combining kinetic fluorescence spectra with second-order calibration based on the alternating least-squares principle was proposed. Noradrenaline, a weak fluorescent substance, can be transformed into a highly fluorescent product by oxidation in alkaline solution. This paper studied the oxidation of noradrenaline by reaction with potassium hexacyanoferrate(III) in the presence of boric buffer (pH 9.06) and ascorbic acid. Fluorescence spectra were measured at consecutive time points in an emission range of 439～550 nm for every sample when a constant 390 nm excitation wavelength was used, hence, creating a three-way response data array which was then analyzed by trilinear decomposition method. Second-order calibration methods based on parallel factor analysis (PARAFAC) and full rank parallel factor analysis (FRA-PARAFAC) algorithms were employed for the quantification of noradrenaline in human plasma samples. The results for PARAFAC and FRA-PARAFAC were very similar with average recovery (AR) and root-mean-squared error of prediction (RMSEP). When the component number was chosen to 3, the obtained average recoveries were (102.0±4.1)% for PARAFAC and (102.4±4.0)% for FRA-PARAFAC, respectively. The root-mean-squared errors of prediction were 0.0197 for PARAFAC and 0.0207 for FRA-PARAFAC, respectively.

Key words: noradrenaline, kinetic fluorometric method, trilinear decomposition, second-order calibration, parallel factor analysis (PARAFAC), full rank parallel factor analysis (FRA-PARAFAC)