Which creatinine-based estimated glomerular filtration rate equation best predicts all-cause mortality in Chinese subjects with type 2 diabetes?

Abstract

AIM: In Chinese, ethnicity-based and/or diabetes specific modifications of the Modification of Diet in Renal Disease (MDRD) and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations have been developed for determining estimated glomerular filtrate rate (eGFR). This study aimed to compare the performance of five different creatinine-based eGFR equations in predicting all-cause mortality among Chinese subjects with type 2 diabetes (T2DM). METHODS: A total of 6739 Chinese subjects with T2DM were included. Their eGFR was calculated using the MDRD, CKD-EPI, their respective modified equations for Chinese, and the diabetes specific CKD-EPI Chinese T2DM equations. Multiple Cox regression analysis was used to evaluate the associations of eGFR with all-cause mortality. C-statistics, net reclassification index (NRI) and integrated discrimination index (IDI) were applied to assess the discrimination and reclassification of each eGFR equation in predicting mortality outcome. RESULTS: Over a follow-up of 5.7years, the incidence of all-cause mortality was 12.9% (N=867). The CKD-EPI equation discriminated all-cause mortality better than the MDRD equation (C-statistics: 0.714 vs. 0.689, p<0.0001), and Chinese modification of their respective equations did not improve discrimination. Among the five eGFR equations evaluated, the CKD-EPI Chinese T2DM equation provided the best discrimination in predicting all-cause mortality among Chinese subjects with T2DM, and was the only equation providing a significantly positive NRI and IDI relative to the CKD-EPI equation. CONCLUSIONS: Among Chinese subjects with T2DM, our findings suggested that the CKD-EPI Chinese T2DM equation best predicted all-cause mortality, and relative to the CKD-EPI equation, conferred improved discrimination and reclassification.