Abstract
BACKGROUND: The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) and Modification of Diet in Renal Disease (MDRD) equations are common for calculating estimated glomerular filtration rate (eGFR). Unlike CKD, the key pathological change of diabetic kidney disease (DKD) is glomerulosclerosis. METHODS: To conduct a meta-analysis of the diagnostic performance of the CKD-EPI and MDRD equations in diabetic patients, we searched PubMed, Embase and the Cochrane Library for studies comparing standard GFR (sGFR) with eGFR using these two equations. RESULTS: Thirteen studies of 7192 diabetic patients reporting data on bias or accuracy were included. At the study level, both equations underestimated eGFR. CKD-EPI was more accurate in studies with mean GFR ≥60 mL/minute/1.73 m(2). At the individual level, both equations overestimated GFR by 6.38 mL/minute/1.73 m(2) (95% confidence interval [CI] 2.67-10.1) and 7.65 mL/minute/1.73 m(2) (95% CI 2.78-12.52), respectively, for sGFR < 90 mL/minute/1.73 m(2). The CKD-EPI equation was 7.61% (95% CI 4.66-10.56) more accurate in subjects with sGFR > 90 mL/minute/1.73 m(2). The CKD-EPI equation performed poorly in diabetic patients. CONCLUSIONS: The CKD-EPI equation can be used to estimate GFR in patients with incipient DKD, but has drawbacks. Improved eGFR equations suitable for diabetic populations are needed.