Glomerular filtration rate estimated by differing measures and risk of all‐cause mortality among Chinese individuals without or with diabetes: A nationwide prospective study

BACKGROUND: Whether estimated glomerular filtration rates (eGFRs) by differing biomarkers are differentially associated with mortality or whether the associations differ by diabetes status remains unclear, especially in Chinese population. METHODS: We included 6995 participants without diabetes (mean age: 60.4 years) and 1543 with diabetes (mean age: 61.8 years). Each eGFR measure was divided into normal (≥90 mL/min/1.73 m(2)), modestly declined (60 to <90 mL/min/1.73 m(2)), and chronic kidney disease (CKD) (<60 mL/min/1.73 m(2)) groups. Cox proportional hazards models were used to estimate hazard ratio (HR) of all‐cause mortality associated with each eGFR. RESULTS: Over a follow‐up of 7 years, 677 and 215 deaths occurred among individuals without or with diabetes, respectively. Among those without diabetes, all measures of modestly declined eGFR were not associated with mortality, whereas CKD defined by eGFR cystatin C (eGFRcys) and eGFR creatinine (eGFRcr)‐cys (HRs were 1.71 and 1.55, respectively) but not by eGFRcr were associated with higher risk of mortality. Among diabetes, all measures of modestly declined eGFR (HRs: 1.53, 1.56, and 2.09 for eGFRcr, eGFRcys, and eGFRcr‐cys, respectively) and CKD (HRs: 2.57, 2.99, and 3.92 for eGFRcr, eGFRcys, and eGFRcr‐cys, respectively) were associated with higher risk of mortality. Regardless of diabetes status, an addition of eGFRcys or eGFRcr‐cys to traditional risk factors lead to a larger improvement in the prediction of all‐cause mortality risk than adding eGFRcr. CONCLUSIONS: The association of eGFR with mortality risk appeared to be varied by its measures and by diabetes status among middle‐aged and older Chinese, which needs to be considered in clinical practice.