The validation of estimated glomerular filtration rate (eGFR) equation for renal transplant recipients

Abstract

The re-expressed Modification of Diet in the Renal Disease (MDRD) equation and the Chronic Kidney Disease Epidemiology (CKD-EPI) equation were developed to estimate glomerular filtration rate in non-transplant chronic kidney disease (CKD) patients. The Nankivell equation was created to estimate GFR in transplant recipients. However, none of these formulas have been validated in Asian renal transplant patients. Several recently published studies have highlighted the need to adapt estimated glomerular filtration rate (eGFR) equations to the race of their patients. Although the eGFR equation for Thai CKD has been derived, it has not been validated for transplant recipients. Our study aimed validating the Nankivell equation, re-expressed MDRD equation, CKD-EPI, re-expressed MDRD equation with Thai racial factor correction, as well as the Thai eGFR equation in Thai renal transplant recipients. A total of 97 adult Thai renal transplant recipients were studied. The 99mTc-DTPA plasma clearance was used as a reference GFR. The serum creatinine was determined by IDMS reference enzymatic methods (CrEnz). The mean reference GFR and CrEnz were 67.86 ± 20.72 ml/min/1.73 m2 and 1.23 ± 0.59 mg/dl. The bias estimated by Bland-Altman analysis can be expressed as -12.11 ± 15.87 ml/ min/1.73 m2 for the Nankivell equation, 2.72 ± 13.90 ml/min/1.73 m2 for the re-expressed IDMS-traceable MDRD equation, -2.59 ± 14.16 ml/min/1.73 m2 for the CKD-EPI equation, -7.05 ± 17.34 ml/min/1.73 m2 for the Thai re-expressed MDRD with Thai racial factor, and -8.62 ± 16.00 ml/min/1.73 m2 for the Thai eGFR equation. The CKD-EPI equation provided the best accuracy and precision in terms of Pearson correlation coefficient, mean difference, error, and accuracy within 10%, 20%, and 30%. The equations derived mainly from Caucasian and/or non-transplant status can be applied to Thai transplantation recipients with some bias. The CKD-EPI had the least bias compared with other eGFR equations.