Multicentre evaluation of a new point-of-care test for the determination of CK-MB in whole blood

Background: The point-of-care (POC) test Roche CARDIAC CK-MB is a new assay which has been developed for the existing Roche Cardiac reader system. Methods: We performed a multicentre evaluation at six sites to assess the analytical performance of the POC CK-MB assay and to compare it with a quantitative laboratory CK-MB assay. Results: Within-series coefficients of variation (CV) resulting from 34 ten-fold measurements with patient samples ranged from 4.3% to 16.4%. Using quality control material, the mean CV values for day-to-day imprecision were 6.5% for the low level control and 8.4% for the high level control. Based upon 847 pairs of values, the mean relative bias of three independently calibrated lots of the POC CK-MB assay ranged from -6% to -11% in method comparisons with the lab CK-MB assay. The mean relative lot-to-lot differences of POC CK-MB were between -2% and +11%. No interference was observed with lipaemic blood (triglyceride concentrations up to 8.1 mmol/L), icteric blood (bilirubin concentrations up to 513 mu mol/L), haemolytic blood (haemoglobin concentrations up to 0.12 mmol/L), biotin (up to 30 mg/L) and rheumatoid factor (up to 119 IU/mL), or with 53 standard or cardiological drugs even in toxic concentrations. There was no influence on the results by varying haematocrit values in the range from 21% to 54%. A slight interference with human anti-mouse antibodies type 2 was found. No significant influence on the results with POC CK-MB was found by using sample volumes between 135 and 165 mu L. High CK-MB concentrations above the measuring range of POC CK-MB (1-40 mu g/L) did not lead to false low results due to potential high-dose hook effect. No significant effect of sample age on recovery occurred up to a sample age of 24 h. No cross-reactivity was found between the POC CK-MB assay and either CK-MM or CK-BB. A sub-study with healthy individuals confirmed the reference limits of 3.8 mu g/L for females and 6.7 mu g/L for males. Conclusions: The POC CK-MB assay showed a very good analytical performance with an excellent concordance with the calibration and reference laboratory method. It should be therefore suitable for its intended use in POC settings.