The Application of Total Protein Reference Materials with Traceability and Commutability in Conventional Systems
Objective To explore the application value of total protein reference materials with traceability and commutability in conventional systems.Methods TP reference method and the conventional method were used to measure GBW09186-188,RELA 23A,23B,21B,20B,and 51 single serum samples simultaneously.Each sample was measured twice,and the theoretical and measured values of the standard substance were linearly fitted.The measured values of all individual serum samples were substituted into the fitting equation to obtain the calibrated results of the reference materials.Referring to CLSI EP9-A3,the results before and after calibration were compared with results of the reference method.Results The bias ranges of the conventional system for measuring international comparison samples before and after calibration were-1.88％--1.05％ and-0.43％-0.41％,respectively.After calibration with total protein in frozen human serum,the comparative bias was significantly reduced.The bias range between the measured values and the results of the reference method in Bland-Altman diagram was-4.98％--0.33％,both of which were negative bias,suggesting that systematic error was existing in the conventional system.The calibrated bias range was-3.59％-1.28％,with an average bias of-0.43％,which was close to zero.Calibration with reference materials was able to reduce the system error.The expected bias at the medical decision level of 45g/L in the four regression analyses of measured values was greater than the clinically acceptable range,and the expected bias after calibration by total protein in frozen human serum was within the clinically acceptable range.Conclusion Using total protein in frozen human serum with traceability and commutability as calibration standards for conventional systems can ensure the accuracy and comparability of detection results.
Total proteinReference methodReference materialsRoutine methodCalibration standardsMethodological comparison
NETL
NSTL
万方数据
