Abstract
RATIONALE & OBJECTIVE: Chronic kidney disease (CKD) is a prevalent condition with high mortality rates. Cardiovascular disease (CVD) is accepted as the leading cause of death in CKD, but data are limited, and no study has evaluated the cause of death in those with progressive CKD versus stable kidney function. STUDY DESIGN: Retrospective cohort. SETTING & PARTICIPANTS: Adults receiving primary care at M Health Fairview (MHFV) after December 31, 2012, with linked Minnesota Death Index data before December 31, 2019, were included. A second cohort was created from adult participants in the 1996-2006 National Health and Nutrition Examination Survey (NHANES) linked with the National Death Index through 2015. Individuals with kidney replacement therapy at baseline were excluded. EXPOSURES: Estimated glomerular filtration rate (eGFR) and proteinuria assessed at baseline defined the exposure categories for MHFV and NHANES. CKD progression in MHFV was also defined as an eGFR decrease ≥30% from baseline or incident kidney replacement therapy. OUTCOME: CVD-, malignancy-, and dementia-attributed death. ANALYTICAL APPROACH: Multinomial logistic regression. RESULTS: For both cohorts, CVD death was more common than malignancy death for those with eGFR <60 mL/min/1.73 m(2), whereas the converse was true for those with higher eGFR without proteinuria. In NHANES, CVD deaths were higher in those with proteinuria and eGFR ≥60 mL/min/1.73 m(2). CKD progression in MHFV had a limited impact on the association with the cause of death except on dementia deaths, which were less common with progression at several stages of CKD. Proteinuria had limited impact on the association with the cause of death across a range of eGFR levels. LIMITATIONS: Limited follow-up and, for MHFV, nonprotocolized measures of kidney function were limitations, as were the intrinsic accuracy limitations for death certificates. CONCLUSIONS: CVD death is the most significant cause of death observed for those with a reduced eGFR irrespective of CKD progression.