Analytical reproducibility evaluated from duplicate measurements of authentic ante- and postmortem blood samples by...

Precision (or imprecision) is one of the central performance parameters of all analytical methods. It is often evaluated during method validation in spiked samples, with a relatively low number of measurements typically during a short period. Validation results are well documented in the literature; however, evaluation of imprecision in authentic cases compared with long-term imprecision from quality control samples has not often been reported on. The aim of this study was to investigate imprecision estimated from duplicate measurements of ante- and postmortem blood samples and long-term imprecision estimates from quality control samples and to compare variations between them. Data were analyzed by using robust statistics, where results for the 29 analytes most frequently quantified by liquid chromatography-tandem mass spectrometry in ante- and postmortem blood samples were included. A total of 41,460 positive findings in authentic whole blood and 51,522 measurements from quality controls were investigated. Analysis was performed in duplicate on independent batches on two separate days. Overall, the imprecision estimated in quality control and the authentic samples were akin for most analytes. Ante- and postmortem blood samples showed similar imprecision for the majority of the analytes and were approximately the same level as long-term imprecision estimated from the quality control samples at low level, whereas relative imprecision of the quality control samples at high level were slightly lower than ante- and postmortem blood samples. The methods we evaluated showed satisfactory reproducibility and robustness for the investigated analytes.