Background: Glomerular filtration rate (GFR) is directly associated with survival. However, the prognostic significance of GFR might be different according to the formula used to estimate it. We aimed at comparing the association between GFR estimated using three different formulas and 1-year survival in elderly patients discharged from acute care hospitals.
Methods: Our series consisted of 439 patients aged 65 and older admitted to 11 acute care medical wards enrolled in a multicentre prospective observational study. GFR was estimated by body surface area-adjusted Cockcroft-Gault (CG-BSA), Modification of Diet in Renal Disease study (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formulas. The relative risk of mortality in patients with estimated GFR = 30-59.9 or < 30 mL/min/1.73 m(2) compared to people with estimated GFR ≥ 60 mL/min/1.73 m(2) was calculated using Cox regression analysis.
Results: Participants with reduced GFR showed an increased mortality, regardless of the equation used, and the highest one was associated with CG-BSA-estimated GFR < 30 mL/min/1.73 m(2). After adjusting for potential confounders, CKD-EPI-estimated GFR remained significantly associated with the outcome [30-59.9 mL/min/1.73 m(2), hazard ratio (HR) = 1.70, 95% confidence interval (95% CI) = 1.02-2.98; < 30 mL/min/1.73 m(2), HR = 2.60, 95% CI = 1.20-5.66], while the strength of the association was clearly reduced for MDRD (30-59.9 mL/min/1.73 m(2), HR = 1.47, 95% CI = 0.83-2.38; < 30 mL/min/1.73 m(2), HR = 2.07, 95% CI = 1.01-4.30) and CG-BSA (30-59.9 mL/min/1.73 m(2), HR = 1.79, 95% CI = 0.67-4.53; < 30 mL/min/1.73 m(2), HR = 2.68, 95% CI = 0.92-7.55).
Conclusion: GFR adds to the list of prognostic indicators in elderly and frail people, and CKD-EPI-derived GFR, which outperforms to some extent MDRD and CG-BSA-derived GFR in a multivariable predictive model, seems worthy of testing in wider populations.