Analytical performance verification protocols and performance specifications of platelet-dependent von Willebrand factor activity testing

Abstract

Objective: To investigate the analytical performance verification protocols and performance specifications of platelet-dependent von Willebrand factor (VWF) activity testing (VWF:GPIbM) for clinical laboratories. Methods: According to Clinical Laboratory Standards Institute (CLSI) documents and National Health Standard of China, the performance verification of VWF:GPIbM was designed and implemented using Sysmex CS-5100 instrument and its corresponding reagents. (1) Precision verification: Two commercial quality control samples (with normal and low activity levels) and three plasma pools (with activity range from 5.0% to 150.0%) were prepared. Each sample was tested five times daily for five consecutive days. The coefficient of variation (CV) of intra-and inter-run precisions were calculated, and the precision evaluation criterion was set according to package inserts. (2) Trueness verification: The calibrator was diluted to five reference materials with activity values of 5.2%, 31.2%, 62.4%, 104.0% and 138.7%, and each reference material was tested five times daily for five consecutive days. The bias between the measured value and the reference value was calculated, and the trueness evaluation criterion was set according to the total allowable error. (3) Linearity verification: Ten pooled plasmas with theoretical value range from 3.6% to 160.4% were prepared for the linearity verification of two calibration curves set by the manufacturer (i.e. low range and normal range calibration curve). Each pooled plasma was tested three times in a single run. The slope and R2 of linear regression of mean of measured value and theoretical value, as well as bias, were calculated, and the linearity evaluation criterion was set according to National Health Standard of China and package inserts. (4) Limit of quantitation verification: The calibrator was diluted to two reference materials with activity values of 3.3% and 2.7%, and each material was tested twelve times. The limit of quantitation evaluation criterion was set according to CLSI document. Results: The CVs of intra-and inter-run were 1.0%-2.5% and 1.1%-2.6%, respectively. The biases of trueness verification were -0.4%, 1.0%, -2.6%, 0.3% and -2.7%, respectively. The linearity verification results of low range (3.6%-31.8%) and normal range (28.4%-160.4%) showed that the slopes of regression equation were 1.021 7 and 0.996 2, respectively, R2 were 0.993 5 and 0.993 9, respectively, and the biases with 0-1.8% and -10.1%-0 of plasmas met the criterion. The biases ranged from -0.4% to 0.3% of test results in the verification of limit of quantitation met the criterion. Conclusion: The verification results of VWF:GPIbM assay for precision, trueness, linearity and limit of quantification meet the performance requirements indicated in the package inserts and the criteria set in this study, which can be taken as a reference of performance verification for the clinical laboratories.