While direct measurements of glomerular filtration rate (GFR) provide the most accurate evaluation of pre-donation kidney function, guidelines do not systematically require the use of a reference method. We evaluated whether and to what extent relying upon creatinine-based estimating equations (eGFR) rather than direct measurement of GFR (mGFR) alters the selection of potential living donors. We compared the impact of 4 equations (the MDRD study equation, the CKD-EPI equation, the revised Lund-Malmö equation, and the full age spectrum [FAS] equation) on the evaluation of 2733 potential donors with GFR measured by reference methods. We also considered the impact of using either absolute or age-adapted GFR thresholds. The CKD-EPI and FAS equations had the best performances (P10 of 50.6% and 47.8%; P30 of 94.4% and 93.1%, respectively) and led to the lowest proportion of improperly evaluated candidates. Misclassification was more frequent when GFR adequacy was defined as an absolute threshold of 90 ml/min/1.73m2 as compared to an age-adapted definition (26% and 5%, respectively). Interpretation of eGFR according to an absolute threshold of 90 ml/min/1.73m2 identified 1804 candidates eligible to donate, compared to 2648 when mGFR was interpreted with age-adapted thresholds. We conclude that creatinine-based estimates cannot substitute for direct GFR measurement to evaluate candidates for kidney donation. When reference methods for direct GFR measurement are not available, our data suggest that a strategy based on age-adapted eGFR values estimated with either the CKD-EPI or FAS equation should be preferred.
