GFR estimation by the MDRD equation using a standardized creatinine assay

Levey AS et al. (2006) Using standardized serum creatinine values in the Modification of Diet in Renal Disease Study equation for estimating glomerular filtration rate. Ann Intern Med 145: 247–254

The Modification of Diet in Renal Disease (MDRD) Study equation was developed to estimate glomerular filtration rate (GFR) from age, sex and race, and serum levels of creatinine, urea and albumin; a simplified four-variable form of the equation, which dispenses with the need for urea and albumin measurements, was subsequently formulated. The clinical utility of GFR estimates derived from these and other equations is compromised by the use of different creatinine assays in different laboratories. In response to an initiative implemented by the National Kidney Disease Education Program for creatinine standardization across the US, MDRD equations were recently revised and revalidated.

Levey et al. compared the revised MDRD equations with the Cockcroft–Gault equation, using data from the original MDRD study population of 1,628 patients with chronic kidney disease. GFR measured as urinary clearance of ¹²⁵I-iothalamate was used as a reference standard. The percentage of GFR estimates within 30% of measured GFR was 90% for the four-variable MDRD equation, 91% for the six-variable MDRD equation and 83% for the Cockcroft–Gault equation. Estimates derived from any of the equations were less accurate when measured GFR exceeded 60 ml/min/1.73 m².

The authors conclude that the simplicity of the four-variable MDRD equation and its comparable performance to its six-variable predecessor will make it a useful and reasonably reliable tool for estimating GFR in patients with chronic kidney disease once US standardization of serum creatinine measurement—expected in 2008—is achieved.