Comparison of drug analysis in whole blood and dried blood spots

Abstract

Aim: Analysis of dried blood spots (DBS) becomes increasingly accepted in therapeutic drug monitoring whereas its application by analogy to forensic samples has not been further studied. Contrary to whole blood, DBS sampling is easier, allows storage without additional cooling and decreases the risk of infections with blood-borne viruses. The aim of our study was to investigate whether determination of alprazolam, risperidone, 9-hydroxyrisperidone, zopiclone, 3,4-methylenedioxymethamphetamine (MDMA), 3,4-methylenedioxyamphetamine (MDA) and dexamphetamine from DBS is as reliable as from whole blood. Methods: DBS and whole blood analysis was performed using 100 µL-specimens. Analysis was performed by LC/MS/MS following liquid-liquid extraction. Results from respective samples were compared using Bland-Altman difference plots. Results and Discussion: The number of corresponding specimens, the mean concentration ratio (r), the mean difference between the 2 methods (d) and the limits of agreement (l, mean difference ± 1.96 SD) for each analyte were as follows: Alprazolam: n=22, r=0.99, d= -0.09 ng/mL, l: -1.11 to 0.92 ng/mL; risperidone: n=10, r=1.07, d=0.83 ng/mL, l: -0.67 to 2,32 ng/mL; 9-hydroxyrisperidone: n=14, r=1.04, d=0.64 ng/mL, l: -1.13 to 2.40 ng/mL; zopiclone: n=45, r=1.19, d=3.99 ng/mL, l: -3.62 to 11.59 ng/mL; MDMA: n=35, r=0.99, d= -3.55 ng/mL, l: -14.34 to 7.25 ng/mL; MDA: n=30, r=0.99, d=0.02 ng/mL, l: -1.36 to 1.40 ng/mL; dexamphetamine: n=29, r=0.95, d= -1.03 ng/mL, l: -3.32 to 1.25 ng/mL. Variability of differences between methods was fairly constant across the range of measurement for all analytes. At least 95 % of all differences were within the limits of agreement. Conclusion: For all analytes except zopiclone results from DBS exactly matched those from whole blood. The blood/DBS-ratio of zopiclone significantly differed from 1.00; the BlandAltman difference plot showed 3 outliers, 2 of them were close to the limits of agreement. This may be due to zopiclone’s lability, which is currently under investigation.