Evaluation of the Effect of Decreased Surrogate Concentrations on the Calculation Value of Recovery Rates Using Multiple Regression Analysis and Calibration Curves on LC/MS Analysis

The effect of varying the surrogate concentration on the calculation value of the recovery rate was investigated by LC/MS and data analysis methods using <i>o</i>-toluidine as a model compound. The low-concentration mixture samples, which simulated the decrease in concentrations of analyte (<i>o</i>-toluidine, <i>m</i>/<i>z</i> 108) and surrogate (<i>o</i>-toluidine-<i>d</i> ₃, <i>m</i>/<i>z</i> 111), were prepared by diluting the standard mixture. Furthermore, the relationship between the prepared and detected analyte/surrogate concentrations in these model samples was examined, both in the presence and absence of the internal standard (IS, <i>o</i>-toluidine-4,6-<i>d</i> ₂, <i>m</i>/<i>z</i> 110). Although the same proportion of components was present at each concentration level, the changes in the analyte/surrogate peak area proportions were more significant at lower sample concentrations. Moreover, this effect was more pronounced in samples containing both IS and surrogate than in those containing only the surrogate, with the proportion decreasing to approximately 60% owing to the <i>m</i>/<i>z</i> 111 peak from the IS. Among the methods for calculating the recovery rate, multiple regression analysis using analyte/IS and surrogate/IS could calculate the recovery rate at low concentrations, whereas the method using two calibration curves with analyte/IS and surrogate/IS performed the same or slightly worse. Furthermore, upon applying the methods to solid-phase extraction samples, the results were the same. The accuracy of the recovery rate at each concentration level was evaluated by checking whether the detected (measured) analyte/surrogate ratio matched the prepared (theoretical) ratio in the standard mixture.