Classification of renal disease status using estimated glomerular filtration rates in diabetes

The need for the incorporation of estimated glomerular filtration rates (eGFR) in diabetes renal risk assessment is increasingly recognised but the choice of equation to use is not clear. We evaluated the differential impact of eGFR, using the Cockcroft's & Gault's (C&G) and the Modification of Diet in Renal Disease (MDRD) study equations, on the prevalence of various stages of chronic renal disease in our diabetes population.

A cross sectional evaluation was conducted amongst 4548 individuals who attended our centre over an 18‐month period. SPSS was utilised for statistical analysis.

Of 4171 with complete data, the prevalence of individuals with eGFR >90, 90–60, 60–30 and <30ml/min/1.73m² were 25%, 46%, 27% and 2% respectively using the C&G equation and 9%, 62%, 27% and 2% respectively using the MDRD equation. The two equations were fully concordant in their classification of eGFR rank in 65%; in 20% of the cohort, the equations were discordant but not at an arbitrary eGFR threshold of 60ml/min/1.73m²; while in 15% of the population the two equations were discordant even at the threshold of 60ml/min/1.73m², the majority of whom had normal values of serum creatinine and urine albumin:creatinine ratio.

In conclusion, the prevalence of various stages of chronic renal disease in our diabetes cohort differed depending on the eGFR equation used, potentially impacting on service provision. To aid clarity and uniformity of practice, there is a need for organisations to decide on a single equation of choice before recommending it to routine diabetes care providers. Copyright © 2007 John Wiley & Sons.