Evaluation of the creatinine assay on the Reflotron analyser under centralized and non-centralized conditions.

The creatinine assay, performed on Reflotron "dry chemistry" analysers in six offices of general practitioners, was evaluated as part of a project on the quality of laboratory determinations in decentralized situations. The method itself was evaluated before this investigation was started. Differences were noticed between Reflotron machines, batches of strips, and various types of sample, viz. heparin-treated blood or plasma gave higher results than serum of EDTA-treated samples. All Reflotron results were lower in comparison with wet chemistry methods. The "dry chemistry" technique was not influenced by the presence of glucose or bilirubin (below 170 mumol/l); haemoglobin interfered. By simultaneous sampling, all creatinine determinations performed on whole EDTA blood in the general practitioners offices were repeated on serum in centralized laboratories using "wet chemistry" techniques. In this comparison 3% of all creatinine determinations in general practitioners offices were outside the 95% confidence interval. We conclude that the creatinine determination on the Reflotron may be used in general practitioners office testing as long as the general practitioner is aware of its limitations and can rely on a "reference" laboratory for training, quality control and repetition of unexpected results.