Objectives: This study screened current GFR evaluation equations that showed high accuracy for elderly populations and evaluated the applicability of these equations for an elderly Chinese population.
Measurements: A standard dual plasma sampling method (DPSM) of estimating 99mTc-diethylene triamine penta-acetic acid clearance was used to determine measured GFR (mGFR).
Results: Comprehensive information was received for a total of 151 elderly individuals, with a mean mGFR of 65.39 ± 24.19 ml/min/1.73 m2. For the overall samples, the accuracy (P30), bias, absolute bias and interquartile ranges (IQRs) of the CKD2 (cystatin C(CysC)-serum creatinine(SCr)), CKD-EPI(CysC-SCr), Cockcroft-Gault(CG), CKD2(CysC), CKD-EPI(CysC), and Hoek equations were superior to c-aGFR3, c-aGFR4 and Grubb equation, Bland-Altman analysis also demonstrated a consistent result. Among elderly subjects with mGFR≥60 ml/min/1.73 m2, the CKD2 (CysC-SCr) and CKD-EPI (CysC-SCr) equations showed significantly higher correlations and accuracy than the other examined equations. Among elderly subjects with mGFR<60 ml/min/1.73 m2, only the CG equation showed an accuracy (P30) of greater than 70% and demonstrated higher precision than the other examined equations.
Conclusion: For the elderly population, the CG, CKD2, CKD-EPI, and Hoek equations exhibited good accuracy. The CKD2(CysC-SCr) equation and CKD-EPI(CysC-SCr) equation demonstrated relatively high accuracy for evaluating elderly subjects with mGFR≥60 ml/min/1.73 m2, whereas the CG equation was more suitable for evaluating elderly subjects with mGFR<60 ml/min/1.73 m2.