Background: Dried blood spot (DBS) sampling offers a minimally invasive sampling method for therapeutic drug monitoring of antipsychotics. To facilitate implementation in clinical practice, the aim of this study was to perform a clinical validation study of a DBS method for quantification of risperidone, aripiprazole, pipamperone, and their major metabolites 9-OH risperidone and dehydro-aripiprazole in a real-life, clinical setting.
Methods: Paired DBS and venous plasma samples were analyzed (n = 35 for risperidone, n = 21 for aripiprazole, n = 21 for pipamperone). Estimated plasma concentrations were calculated from DBS concentrations based on hematocrit and/or Deming regression formulas. Deming regression and Bland-Altman analyses were used to determine the agreement between the calculated and measured plasma concentrations. For Bland-Altman analysis, the following acceptance limit was used: for a minimum of 67% of the samples, the difference of the 2 measurements should be within 20% of their mean.
Results: The median venous plasma levels were 0.9 mcg/L for risperidone, 14.8 mcg/L for 9-OH risperidone, 135.4 mcg/L for aripiprazole, 54.9 mcg/L for dehydro-aripiprazole, and 56.4 mcg/L for pipamperone. All antipsychotics required different correction formulas of DBS concentrations for best agreement. Subsequently, no constant or proportional bias was observed using Deming regression analysis. With Bland-Altman analyses, for risperidone, 45% of the samples were within the 20% limits; for 9-OH risperidone, 36%; for aripiprazole, 45%; for dehydro-aripiprazole, 35%; and for pipamperone, 43%.
Conclusions: The DBS method to quantify risperidone, aripiprazole, pipamperone, and their major metabolites did not meet the acceptance criteria in the Bland-Altman analyses. Therefore, this DBS method was not clinically valid. This study shows the importance of a clinical validation study with use of Bland-Altman plots before clinical implementation.