An Evaluation of Comparability Between Conventional and Reference Methods for Serum Total Protein According to EP9-A3
Objective To evaluate the consistency of serum total protein routine methods with the reference method,in order to provide more accurate results for the clinical practice.Methods Serum total protein reference method and conventional method were applied to measure GBW09186-188,RELA22A,22B,18A,18B,17A,17B and 40 individual serum samples.According to EP9-A3,the detection results of the two methods were compared.An interchangeability analysis of the calibrators used by conventional methods was performed with reference to EP14-A3.Results The bias of the reference method for measuring standard materials and comparison samples was-1.66％-0.51％,while the bias of the conventional method was-2.34％-0.30％.The bias of the reference method was smaller than that of the conventional method.All the data points in the Bland-Altman plot were uniformly distributed around the mean value and very high percentage of them(41/43)were within 95％confidence interval,indicating a good consistency between the two methods.Based on results of the three regression analyses,the bias at the 45g/L medical decision level was greater than the clinically acceptable range,and the bias at the other medical decision levels was smaller than the clinically acceptable range.The bias between the measured value and the predicted value of the calibrators used in the conventional method was 1.23％and-1.65％,respectively.Both calibrators exceeded 95％CI,and the calibrators were not interchangeable.Conclusion The conventional method for serum total protein shows systematic errors,and there is a slight deviation in accuracy at low values.
Total proteinReference methodRoutine methodBland-AltmanMethodological comparison
NETL
NSTL
万方数据
