Comparative analysis and acceptability evaluation on results of four cardiac muscle enzymes with different detection system

Objective The comparability level for the results of four cardiac muscle enzymes,including AST,LDH,CK,CK-MB,measured by different detection system in a same laboratory department was investigated through bias estimation and method comparison in order to obtain experimental data for standardization of serum enzyme mensuration and medical laboratory accreditation.Methods Referring to EP9-A file of NCCLS,the comparison methods were accredited by ISO/IEC 17025,which includes Hitachi 7170 biochemical analyzer,reagent of Roche,c.f.a.s.calibrator and controller of Roche.Other systems were used as the experimental system.Regarding half of the error range allowed by CLIA'88 as standard,after detecting bias and system error of four cardiac enzymes in fresh serum we judged clinical acceptability between experimental methods(Y) and comparison method(X) using t test and regress analysis.Results Except AST of system 4 and CK-MB of system 1,2,3,which bias were not acceptable,the rest items were clinical accepted.Conclusion If there are more than two systems for the same test are used in a laboratory,method comparison is necessary and bias estimation for judgement of clinical acceptability in order to insure the comparability of experimental results.