Consideration of the Application of Measurement Uncertainty as an Acceptable Criterion For Reagent Lot-to-lot Comparability Test in Nuclear Medicine

The reagent lot-to-lot comparability test evaluates the comparability between the reagents by comparing the differences in the measured quantity values obtained with different lot numbers of reagents. In the comparability test, however, there is no universal standard for setting the acceptable range. Measurement uncertainty is a parameter expressed as a numerical value obtained by quantitatively evaluating the uncertainty of a measurement result. This study examined whether measurement uncertainty can be used as a criterion to set acceptable ranges in comparability tests. Luteinizing hormone, insulin, thyroid stimulating hormone, CYFRA 21-1, C-peptide, free thyroxine, and cortisol were tested after manufacturing the test materials at low, medium, and high concentrations using quality control materials. The acceptable range was set by applying ±20% to the average measured quantity values for low-concentration substances and ±10% for high-concentration and medium-concentration substances. The estimated range of measurement uncertainty was wider than the acceptable range. The estimated range differed depending on the test item and the concentration of test material. Overall, the measurement uncertainty is calculated by reflecting the factors of uncertainty generated during laboratory testing and can be used as a reliable acceptance criterion in comparability tests.