A Systematic Review Comparing the Prognostic Role of eGFR According to CKD-EPI and Older Age Validated Equations in Older Adults.
Elisa K Bongetti, Benjamin Lazarus, Rory Wolfe, Kevan R Polkinghorne
Abstract
Open AccessRationale & Objective: Assessing the clinical relevance of reduced estimated glomerular filtration rate (eGFR) in older adults is challenging because GFR naturally declines with age and not all equations, including the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations, have been validated for use in older adults. This systematic review compared the association between eGFR and health outcomes in older adults using eGFR equations validated for older age and the CKD-EPI eGFR. Study Design: Prognostic factor systematic review and meta-analysis. Setting & Study Populations: Community-dwelling adults aged ≥ 65 years. Selection Criteria for Studies: Studies using CKD-EPI and at least one of Berlin Initiative Study 1, Berlin Initiative Study 2, European Kidney Function Consortium, Full Age Spectrum, and Revised Lund-Malmo equations to calculate eGFR in association with mortality, cardiovascular outcomes, hospitalizations, or kidney failure. This study was registered on PROSPERO (ID: CRD42022359839). Analytical Approach: Meta-analysis was undertaken using random-effects models. Results: Thirteen studies met inclusion criteria with a total of 102,893 participants. The pooled mean age was 80 years (IQR: 76-84), and baseline CKD-EPI2009 eGFR was 63 mL/min/1.73m2 (IQR: 55-71). The overall quality of data was low, and most studies did not adjust for albuminuria. There was insufficient data assessing the outcomes of cardiovascular outcomes, hospitalizations, or kidney failure. There was no evidence that the association between eGFR and mortality varied according to whether an older age validated equation or CKD-EPI was used. Reduced eGFR according to cystatin C equations had stronger associations with mortality than creatinine-based equations. Limitations: Limited number of studies and overall low quality of data. Conclusions: There was no evidence that the association between eGFR and mortality in older adults differed between CKD-EPI and eGFR equations validated for older age. Studies comparing eGFR equations as predictors of hospitalizations, cardiovascular outcomes, or kidney failure were lacking.