The Modification of Diet in Renal Disease (MDRD) equations provide valid estimations of glomerular filtration rates in patients with advanced heart failure

Background: Glomerular filtration rate (GFR) has major prognostic implications in heart failure. Our objective was to validate the MDRD prediction equations for GFR in patients with advanced heart failure, and to compare their predictive performance to that of the Cockcroft-Gault (CG) equation. Methods: We analysed GFR in 45 patients referred for heart transplantation evaluation. (51)Cr-EDTA-measured GFR was compared to GFR estimates obtained by MDRD1 and MDRD2 equations, CG equation using actual body weight, and ideal body weight. Regression analyses and Pearson correlations were performed, and Bland and Altman plots were drawn. ROC curves were obtained to illustrate each equation's ability to predict a GFR less than 60 ml/min/1.73 m(2) (moderate renal impairment). Results: Patients had a mean age of 52 years, and 69% were in NYHA class III. The mean EDTA-measured GFR was 46.9 +/- 17.2 ml/min/ 1.73 m(2). The MDRD1 equation provided the best predictive model (narrowest limits of agreement; r=0.766, p < 0.001), and the highest performance in predicting a GFR less than 60 ml/min/1.73 m(2) (area under curve: 0.901). Conclusions: MDRD equations, especially MDRD1, adequately predict GFR in advanced heart failure, with higher accuracy than the CG equation. MDRD1 also has higher performance in predicting a GFR less than 60 ml/min/1.73 m(2). (c) 2005 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.