Bias evaluating of determination results of two kinds of cholesterol calibrator and problem application of calibrator

Objective:To evaluate the bias of two kinds of cholesterol calibrator determin the results of serum samples from patients to reflect the problem of application of clinical chemistry calibrator in China.Methods:The same reagents and instruments with two kinds of calibrator combined into two detection systems,one as a reference system,the other as a check system.With those two detection systems,8 patients' fresh serum samples of high,middle and low will be detected according to NCCLS EP9-A document everyday.Each specimen will be detected twice a day,5 days in all.The results are recorded to evaluate the bias of two detection systems and statistical analysis;with Roche calibration serum calibrate instrument,calibrator which from 4 kinds of kit,as sample to detect.The result was compared with the value of its marked.Results:Two kinds of calibrator for the same reagents and instrument,the mean relative bias of 40 determination results of serum samples was as high as 11.3%.Through statistical analysis,this difference owns statistical significance(P0.01).Compared the measured value of calibrator from 4 kinds of kit with the marked value,the relative bias was-14.1%~17.3%.Conclusion:Calibrator with a direct impact on the traceability of measurement results of serum samples from patients,and the feasibility of mutual recognition of test results of different hospitals.The calibrator interoperability is a significant problem urgently to be solved in China.