Calibration of Technicon Chem 1 multitest analysers

Recently, two Technicon Chem 1 multitest clinical chemistry analysers were introduced after an evaluation in our laboratory. Calibration according to the instructions of the manufacturer resulted in significant differences between the assay results of the two Chem 1 analysers for twelve of twenty-two methods. To remove these differences we used an additional procedure based on the NCCLS EP-9P protocol for method comparison. Four reference methods were used as comparison methods. Our results indicate that, for three of these methods, the results on both Chem 1 analysers are biased in slope and intercept, and corrections of these parameters are necessary to produce results comparable to the reference method values. For the remaining eighteen assays, methods in routine use in our laboratory were used as comparison methods. In fifteen of eighteen assays, corrections of slope and (or) intercept were necessary to prevent changes in the cumulative average patient data produced by our laboratory. An adjustment of the slopes and intercepts, according to the multi-point calibration procedure presented here, resulted in significant differences between system-specific values measured in our laboratory and those assigned by the manufacturer for ten of thirteen methods. We conclude that the use of the multi-point calibration procedure presented in this article (based on calibration according to the instructions of the manufacturer and NCCLS EP-9P) greatly improves the intra-laboratory comparability and therefore should be part of multi-centre evaluations.