Assessing the Reliability of Creatinine-Estimated Glomerular Filtration Rate in Living Kidney Donor Candidates: Does it Measure up to the Measured Rate?

Abstract

Abstract A living donor kidney transplant is the best therapy option for patients with end stage renal disease (ESRD); however, it is associated with an increased risk of ESRD development in the donor. Thorough evaluation of the donor candidate’s kidney function status is essential for the appropriate selection of candidates with minimal risk of post-donation kidney dysfunction. In this study, glomerular filtration rate (GFR) values measured by iohexol plasma clearance (mGFR) and estimated by the Chronic Kidney Disease Epidemiology Collaboration (CKD-Epi) creatinine equations (eGFR), with and without the race factor (2009 and 2021 CKD-Epi equations, respectively), were compared in a cohort of 303 kidney donor candidates. GFR values were analyzed using Bland-Altman plots, linear regression, Mann-Whitney U tests, and p10 and p30 values. Bland-Altman analysis revealed a lack of agreement between the 2009 and 2021 CKD-Epi eGFR and corresponding mGFR values, with both equations showing a negative bias of 10 – 11 mL/min/1.73m2 and indicating an underestimation of creatinine-eGFR. eGFR underestimation was confirmed by Mann-Whitney U analysis; median GFR values were significantly lower for both CKD-Epi equations as compared to corresponding iohexol-measured values (2009: 85.86; 2021: 87.29; mGFR: 96.58). Despite the overall underestimation, over 14% of donor candidates with eGFR >90 mL/min/1.73m2 had an mGFR value <90 mL/min/1.73m2. Conversely, 36% - 38% of donors with normal kidney function of >90 mL/min/1.73m2 as determined by iohexol-mGFR had corresponding eGFR values <90 mL/min/1.73m2. eGFR showed a significant but poor correlation with mGFR, with R2 values of 0.2775 and 0.2803, respectively. While the 2021 eGFR equation had better accuracy as compared to the 2009 equation (p30 of 81% versus 76%), it displayed low overall accuracy with a p10 of <43%. Separate analysis of Black donor candidates’ eGFR and mGFR values revealed similar trends as observed for the total cohort. While exclusion of the race factor in the 2021 CKD-Epi equation slightly improved accuracy (p30 of 77% versus 72%), the correlation with mGFR remained poor for both equations (R2: 0.1455 and 0.1377 for the 2009 and 2021 equations, respectively). The results of this study confirmed that creatinine-eGFR values, regardless of the race factor, differ significantly from iohexol-mGFR. Our data shows that even potential kidney donors with eGFR >90 mL/min/1.73m2 cannot be presumed to have normal kidney function, while eGFR <90 mL/min/1.73m2 cannot accurately exclude donor candidates. We conclude that creatinine-eGFR cannot reliably serve as a substitute for mGFR in the context of living donor kidney function evaluation.