CANDIDATE REFERENCE METHOD FOR DETERMINING SERUM CREATININE BY ISOCRATIC HPLC - VALIDATION WITH ISOTOPE-DILUTION GAS-CHROMATOGRAPHY MASS-SPECTROMETRY AND APPLICATION FOR ACCURACY ASSESSMENT OF ROUTINE TEST KITS

We present a candidate Reference Method for determining creatinine in serum, based on isocratic HPLC. The chromatographic column, with alkal ine-treated aluminum oxide as stationary phase, is eluted with an equi volume mixture of methanol/acetonitrile containing 70 mL/L aqueous NaO H (10 mmol/L), and ultraviolet absorbance is detected at 240 nm. We in vestigated the ruggedness of the method and validated its performance with isotope dilution gas chromatography-mass spectrometry (ID GC-MS) for a set of 22 patients' sera and 10 commercially available lyophiliz ed control materials. The mean deviation from ID GC-MS was +0.1% (rang e, -2.2% to +2.2%). The between-day CV, calculated from six independen t measurements performed on three different days, was 0.9% (range 0.2% to 1.9%); the within-run CV was 0.8%. The total error of the method w as <3%. These performance characteristics make the method suitable for target-setting of quality-control materials and accuracy assessment o f routine test kits. For the latter application, done with split-sampl e measurements of a panel of 83 patients' specimens, we used the Boehr inger Mannheim enzymatic creatinine PAP test on the Hitachi 911, the K odak Ektachem single-slide enzymatic creatinine test on the Ektachem 7 00, the Merck Jaffe kinetic assay on the Mega analyzer, and the Roche Jaffe kinetic test on the Cobas Mira.