Ann Intern Med 2006 Aug;145(4):247-54
Tufts-New England Medical Center, Boston, Massachusetts 02111, USA.
Background: Glomerular filtration rate (GFR) estimates facilitate detection of chronic kidney disease but require calibration of the serum creatinine assay to the laboratory that developed the equation. The 4-variable equation from the Modification of Diet in Renal Disease (MDRD) Study has been reexpressed for use with a standardized assay.
Objective: To describe the performance of the revised 4-variable MDRD Study equation and compare it with the performance of the 6-variable MDRD Study and Cockcroft-Gault equations.
Design: Comparison of estimated and measured GFR.
Setting: 15 clinical centers participating in a randomized, controlled trial.
Patients: 1628 patients with chronic kidney disease participating in the MDRD Study.
Measurements: Serum creatinine levels were calibrated to an assay traceable to isotope-dilution mass spectrometry. Glomerular filtration rate was measured as urinary clearance of 125I-iothalamate.
Results: Mean measured GFR was 39.8 mL/min per 1.73 m2 (SD, 21.2). Accuracy and precision of the revised 4-variable equation were similar to those of the original 6-variable equation and better than in the Cockcroft-Gault equation, even when the latter was corrected for bias, with 90%, 91%, 60%, and 83% of estimates within 30% of measured GFR, respectively. Differences between measured and estimated GFR were greater for all equations when the estimated GFR was 60 mL/min per 1.73 m2 or greater.
Limitations: The MDRD Study included few patients with a GFR greater than 90 mL/min per 1.73 m2. Equations were not compared in a separate study sample.
Conclusions: The 4-variable MDRD Study equation provides reasonably accurate GFR estimates in patients with chronic kidney disease and a measured GFR of less than 90 mL/min per 1.73 m2. By using the reexpressed MDRD Study equation with the standardized serum creatinine assay, clinical laboratories can report more accurate GFR estimates.