Time-Updated Changes in Estimated GFR and Proteinuria and Major Adverse Cardiac Events: Findings from the Chronic Renal Insufficiency Cohort (CRIC) Study.

Abstract

RATIONALE AND OBJECTIVE\nEvaluating repeated measures of estimated glomerular filtration rate (eGFR) and urine protein-to-creatinine ratio (UPCR) over time may enhance our ability to understand the association between changes in kidney parameters and cardiovascular disease risk.\n\n\nSTUDY DESIGN\nProspective cohort study.\n\n\nSETTING AND PARTICIPANTS\nAnnual visit data from 2,438 participants in the Chronic Renal Insufficiency Cohort.\n\n\nEXPOSURES\nAverage and slope of eGFR and UPCR in time-updated, one-year exposure windows.\n\n\nOUTCOMES\nIncident heart failure, atherosclerotic cardiovascular disease events, death, and a composite of incident heart failure, atherosclerotic cardiovascular disease events, and death.\n\n\nANALYTIC APPROACH\nA landmark analysis, a dynamic approach to survival modeling that leverages longitudinal, iterative profiles of laboratory and clinical information to assess the time-updated three-year risk of adverse cardiovascular outcomes.\n\n\nRESULTS\nAdjusting for baseline and time-updated covariates, every standard deviation lower mean eGFR (19 mL/min/1.73m2) and declining slope of eGFR (8 mL/min/1.73m2/year) were independently associated with a higher risk of heart failure (mean eGFR hazard ratio [HR] 1.82, 95% confidence interval 1.39-2.44; slope HR 1.28, 1.12-1.45) and the composite outcome (mean HR 1.32, 1.11-1.54; slope HR 1.11, 1.03-1.20). Every standard deviation higher mean UPCR (136 mg/g) and rising UPCR (240 mg/g/year) were also independently associated with a higher risk of heart failure (mean HR 1.58, 1.28-1.97; slope HR 1.20, 1.10-1.29) and the composite outcome (mean HR 1.33, 1.17-1.50; slope HR 1.12, 1.06-1.18).\n\n\nLIMITATIONS\nLimited generalizability of annual eGFR and UPCR measurements; several biomarkers for cardiovascular disease risk were not available annually.\n\n\nCONCLUSIONS\nUsing the landmark approach to account for time-updated patterns of kidney function, average and slope of eGFR and proteinuria were independently associated with three-year cardiovascular risk. Short-term changes in kidney function provide information about cardiovascular risk incremental to level of kidney function, representing possible opportunities for more effective management of patients with chronic kidney disease.