An ID-HPLC–MS/MS based candidate reference measurement procedure for the quantification of valproic acid in human serum

Valproic acid (VPA) is widely used for the treatment of epilepsy, bipolar disorder and other psychiatric diseases. Because of its narrow therapeutic window, monitoring the concentrations of VPA in patient’s serum is essential. In this paper, an isotope dilution-liquid chromatography-tandem mass spectrometry-(ID-HPLC–MS/MS) based candidate reference measurement procedure (RMP) for the quantification of VPA in human serum was established. After Samples pre-prepared by a simple protein precipitation method, the basic analytical performances of the candidate RMP was validated, including specificity, matrix effect, linearity, limit of quantitation (LOQ) and limit of detection (LOD), intra-and inter-batch precision, recovery rate, carryover, and stability. In order to make the measurement results of VPA traceable, we took all components that affect the accuracy of VPA measurement results into account to evaluate the corresponding uncertainties according to the GUM, including measurement imprecision, the purity of VPA standard, uncertainties caused by weighing during the preparation of calibrators and samples, matrix effect, recovery rate, carryover and sample stability. The candidate RMP separated VPA from potentially interferents in human serum and enabled measurement over a calibrated linear range from 2.60 to 203.57 μg/mL with a good correlation coefficient of 0.9995, and no significant matrix effects were observed. The intra-day and inter-day coefficients of variation (CV%) were 0.30–1.64% and 0.67–1.39%, respectively. The average recovery rates were between 94.55% and 96.93%. The LOD and LOQ were 0.23 μg/mL and 2.10 μg/mL, respectively. The expanded measurement uncertainties were 3.19% (k = 2) at 9.45 μg/mL, 3.14% (k = 2) at 48.86 μg/mL, and 2.95% (k = 2) at 97.23 μg/mL (k = 2), respectively.