OBJECTIVETo analyze the sources and distribution decision limits on reference values of myocardial injury markers based on the feedback results of 150 laboratories which participated the 2012 external quality assessment (EQA) programs in China.

METHODSDecision limits on myocardial injury markers from EQA programs were analyzed. Data from maternal and child care service center, children's hospital and corporations were excluded. Abnormal values and errors were eliminated. Mean, median, maximum value, minimum value, P2.5 and P97.5 were calculated. Sources of decision limits were summarized and data were filtered and only values obtained according to the reagent manufacturer's instructions as a reference source were used and grouped based on reagent characteristics.

RESULTSAccording to the surveys on reference interval sources, most of reference sources were derived from reagent manufacturer's instructions and the ratio of each item was: CK-MB (µg/L) 91.30%, Mb (µg/L) 88.44%, cTn-I (µg/L) 86.29%, cTn-T (µg/L) 92.00%, CK-MB (U/L) 70.65%. According to the surveys, the mean and median values of each test parameter were close to each other (CK-MB:3.74, 4.94 µg/L;Mb:76.09, 72.00 µg/L; CTn-I:0.06,0.09 µg/L;cTn-T:0.01, 0.02 µg/L;CK-MB:23.00, 25.00 U/L) . There were significant variations on P2.5 and P97.5 distribution ranges: CK-MB 2.48-7.09 µg/L, Mb 46.00-140.03 µg/L, cTn-I 0.01-1.68 µg/L, cTn-T 0.01-12.61 µg/L, CK-MB 6.00-30.60 U/L. According to the surveys on using various reagents, there were also significant variations on P2.5 and P97.5 reagent distributions.

CONCLUSIONSThe decision limit of the mean and median for each test item is close to each other while the P2.5 and P97.5 distribution of each test item consistency is poor. There is therefore an urgent need to establish an uniform reference values for cardiac markers in China.