The effect of serum creatinine method choice on estimated glomerular filtration rate determined by the abbreviated MDRD formula.

Background: The aim of this study was to measure the variability in estimated glomerular filtration rate (eGFR) calculated by laboratories in Northern Ireland where creatinine assays other than the Beckman CX3 assay used to derive the Modification of Diet in Renal Disease (MDRD) are utilized.

Methods: Fifty patient samples were analysed for serum creatinine by kinetic Jaffe assays on the Roche modular P-800 (compensated assay), the Beckman LX20 and the Abbott Aeroset analysers.

Results: The median (interquartile range) eGFR calculated by the abbreviated MDRD equation using the creatinine results obtained by each method were 45.4 (31.7-66.6), 49.2 (35.4-78.1) and 50.0 (35.1-71.2) mL/min/1.73 m2 for Beckman, Roche (compensated) and Abbott assays, respectively. Following mathematical alignment of creatinine methods to the Beckman method the median (interquartile range) eGFR for the Roche and Abbott methods were 44.0 (32.0-67.6) and 45.6 (31.2-64.8) mL/min/1.73 m2, respectively.

Conclusion: The regression equations published in this study can be used to align creatinine methods. However, method-related biases in eGFR of up to 10% are minimal in the context of the magnitude of variance reflected in the 90% confidence intervals.