Abstract:

Objective To evaluate the accuracy of the measurement results in reference method using total protein standard reference material in frozen human serum as calibrator. Methods SRM927e，SRM927f and GBW09186-188 were used as calibrators to measure RELA 22A and 22B，appropriate calibrators were selected，and then RELA 23A，23B，18A，18B，17A，17B and 53 individual serum samples were measured，and the results were compared with the Clinical and Laboratory Standards Institute （CLSI） EP09-A3. Results The measurement results of RELA 22A and 22B using SRM927f as calibrator exceeded the International Federation of Clinical Chemistry and Laboratory Medicine（IFCC）equivalent limit，while the measurement results using SRM927e as calibrator were within the equivalent limit. The measurement results using GBW09186-188 as calibrators were all within the equivalent limit，and the bias of GBW09188 was the least. The bias range of the 3-year international comparison samples measured with GBW09188 as calibrator was -0.45%-0.45%，all of which were within the equivalent limit. The Bland-Altman results using SRM927e and GBW09188 as calibrators were all within 95% confidence interval （CI），indicating good consistency in the measurement results of the 2 calibrators. The bias at the medical decision level of the 4 regression analyses was within the clinically acceptable range. Conclusions The measurement results of serum total protein reference method with GBW09188 as calibrator are accurate and reliable.

Key words: Total protein standard reference material in frozen human serum, Reference method, Calibrator, Bland-Altman plot, Methodological comparison