The majority of clinical laboratories use the one stage factor assay for plasma factor measurements because it is simple and easy to automate. However, quality assessment scheme surveys on factors show that inter-laboratory agreement needs to be improved and the one stage assay needs to be standardized better. In this study, one stage factor assays were improved by optimizing assay calibrations. A serious drawback of current factor assay calibrations is that they can’t cover the clinically relevant range (from zero to high activities) by a reliable single calibration. To have measured activities in the calibrated range, high and low samples must be re-run, using higher and lower sample pre-dilutions, respectively. However, measuring samples diluted less than 1:10 should be avoided, because that violates the basic assumption of the one stage method that the deficient plasma reagent alone provides the other factors, and that the effect of other factors in the sample is negligible. Here, factor assays were improved by the introduction of a zero calibration point and the establishment of calibration curves with an optimization approach. In each of the deficient (<1%) plasmas used as a reagent, the remaining factor activity can be sufficient for measurable clotting time, even in saline or buffer samples. A zero calibration point anchors the calibration curve to a true baseline. In the optimization approach:
More than ten mathematical transformations can be used both for time/activity.
Two curves can create the calibration, by selecting two groups of calibration points.
Curve fittings can be linear, 2nd or 3rd order polynomials.
A cut-off point defines the end of one, and the beginning of the next curve on the combined calibration curve with a smooth transition.
Factor assays on ACL-TOP analyzers, which use calibrations established by an optimization approach: - Have a wide calibrated region, from zero up to 120–150% activity - Have excellent linearity in the calibrated region, and well beyond - Have increased tolerance for reagent variations - Give low activity results from 1:10 diluted samples with no need for reruns The software and parameter packages with improved HemosIL APTT-SynthASil and PT-RecombiPlasTin factor assays have been used in clinical laboratories in over eight different countries since February of 2007. The users are very satisfied with the robustness of the assays, excellent precision, accuracy, impressive linearity and the wide activity range of the calibrated region. The increased tolerance for reagent variations of the improved assays is well demostrated by the parameters (mathematical transformations, curve fits and cut-off points) identified as optimal for the HemosIL APTT-SynthASil-based FVIII, FIX, FXI and FXII assays, which ensure excellent performance for the corresponding HemosIL APTT-SP-based assays as well. Widespread use of zero calibration points (anchoring the calibration curve to a true baseline) and establishing calibration curves by the optimization approach would increase accuracy and lessen variability in plasma factor results among clinical laboratories.
Disclosure:Employment: All authors except C. Holland and I. Csapo are employees of Instrumentation Laboratory.