Comparison of two creatinine-based estimating equations in predicting all-cause and cardiovascular mortality in patients with type 2 diabetes.

OBJECTIVEdTo compare the performance of two glomerular filtration rate (GFR)-estimatingequations in predicting the risk of all-cause and cardiovascular mortality in type 2 diabetic patients.RESEARCH DESIGNANDMETHODSdWe followed 2,823 type 2 diabetic outpatientsfor a period of 6 years for the occurrence of all-cause and cardiovascular mortality. GFR wasestimated using the four-variable Modification of Diet in Renal Disease (MDRD) study equationand the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.RESULTSdAt baseline, an estimated GFR (eGFR) ,60 mL/min/1.73 m2 was present in 22.0and 20.2% of patients using the MDRD study equation and the CKD-EPI equation, respectively.A total of 309 patients died during the follow-up (152 patients from cardiovascular causes). Bothcreatinine-based equations were associated with an increased risk of all-cause and cardiovascularmortality. However, the CKD-EPI equation provided amore accurate risk prediction ofmortalitythan the MDRD study equation. Receiving operating characteristic curves showed that the areasunder the curve (AUCs) for all-cause mortality (AUC 0.712 [95% CI 0.682–0.741]) and cardiovascularmortality (0.771 [0.734–0.808]) using eGFRCKD-EPI were significantly greater (P ,0.0001 by the z statistic) than those obtained by using eGFRMDRD (0.679 [0.647–0.711] forall-cause mortality and 0.739 [0.698–0.783] for cardiovascular mortality).CONCLUSIONSdOur findings suggest that the estimation of GFR using the CKD-EPI equationmore appropriately stratifies patients with type 2 diabetes according to the risk of all-causeand cardiovascular mortality compared with the MDRD study equation.