Reduced kidney function has been associated with an increased risk for adverse outcomes. Accurate assessment of glomerular filtration rate (GFR) is key to diagnosis and management of kidney disfunction. Debate exists on the best GFR estimation equation for elderly people. This study aimed to compare the predictive validity and discriminative ability of four GFR equations in relation to 2-year and 6-year mortality in exceptional longevity (EL) (those over 95 years old with intact health) individuals and is an ideal model to address factors relating to life span and age-related diseases.
This study used 6 years' data of 278 EL from the Rugao longevity cohort. Baseline GFR was estimated using four equations: Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, Modification of Diet in Renal Disease Study (MDRD) equation, Berlin Initiative Study-1 (BIS) equation, and modified MDRD equation. Predictive validity was tested using Cox proportional hazards analysis. Overall improvement in reclassification based on estimated GFR (eGFR) was assessed applying net reclassification improvement (NRI).
Mean age of participants was 97±2 years with median follow-up of 2.6 years. Median (IQR) eGFR by CKD-EPI, MDRD, BIS, and modified MDRD equations were 73.9 (62.2-77.6), 82.3 (67.4-98.6), 56.4 (47.9-63.9), and 101.5 (83.1-121.6) mL/min per 1.73 m2, respectively. Higher eGFREPI was associated with lower mortality after multivariate adjustment (for continuous eGFREPI, HRtwo-year 1.018, 95% CI 1.002-1.033, P=0.023; HRsix-year 1.013, 95% CI 1.002-1.025, P=0.022), while eGFR from other equations did not show any associations with mortality. NRI for two-year mortality was 0.14 and approximately significant, which may favor the CKD-EPI when compared to BIS equation (P=0.052).
The CKD-EPI equation showed more accurate estimation of kidney function in the elderly with respect to GFR distribution and predictability of mortality risk.