Comparison of Different Estimating Equations for Prediction of Glomerular Filtration Rate in Living Kidney Donors

Abstract

Introduction: Assessment of renal function is a crucial step in evaluation of living kidney donors. The standard method for determining renal function is measurement of glomerular filtration rate (GFR) using I-123 iothalamate, Tc-99m Diethylene Triamine Pentaacetic Acid (DTPA) and 51Cr-Ethylene Diamine Tetraacetic Acid. As these methods are expensive and cannot be used in all clinical settings, it is common practice to estimate GFR by creatinine-based equations. The objective of this study is to compare commonly used estimating equations for the prediction of GFR in Living Kidney Donors.

Methods: In 75 healthy kidney donors, GFR estimated by Modification of Diet in Renal Disease Study equation (MDRD), Cockcroft-Gault formula(CG), Chronic Kidney Disease Epidemiology Collaboration(CKD- EPI) equation and 24 hour urinary creatinine clearance were compared to GFR measured by Tc-99m DTPA. Statistical analysis was done using Dunnett’s test and Bland-Altman plot. Similarly, accuracy, precision and bias of each equation were assessed.

Results: Mean GFR calculated by DTPA clearance, CG, MDRD, CKD-EPI equations and 24 hour urine creatinine clearance were 83.35±8.59, 78.99±17.17, 93.30±17.12, 96.34±13.36 and 137.96±43.65 ml/min/1.73m2 respectively. Applying Dunnett’s test, GFR by CG equation minimally underestimated GFR measured by DTPA (p=0.612) whereas GFR estimated by MDRD (p=0.034), CKD-EPI(p=0.03) and 24 hour urine creatinine clearance(p<0.001) were statistically significant. CG equation had the highest accuracy. Using Bland-Altman plot, the precision of CKD-EPI equation was the highest among all.

Conclusion: There is no single creatinine-based estimating equation to assess GFR with utmost accuracy and precision at the same time.