Association of Cystatin C- and Creatinine-Based Estimated Glomerular Filtration Rate With Adverse Outcomes in Heart Failure With Preserved Ejection Fraction

Rationale and Objective: Cystatin C-based versus creatinine-based measures of estimated glomerular filtration rate (eGFR) are more predictive of worse outcomes in people with heart failure with reduced ejection fraction (HFrEF). It is unknown if cystatin C-based measures may also yield more accurate prognoses in heart failure with preserved ejection fraction (HFpEF). Study Design: A longitudinal analysis of people with HFpEF. Setting & Participants: 214 participants from the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist trial with available serum at baseline and 12 months in the National Heart, Lung, and Blood Institute’s Biologic Specimen and Data Repository Information Coordinating Center. Exposures: Baseline, time-varying, or change over 12 months in eGFRcr, eGFRcys, and eGFRcr-cys Outcomes: Composite of aborted cardiac arrest, heart failure hospitalization, or cardiovascular death. Analytical Approach: Fine-Gray and joint models. Results: Median follow-up was (IQR) 3.09 (2.07-4.44) years. All 3 baseline eGFR measures were independently associated with the risk of the composite endpoint, after adjustment for race, diabetes mellitus, and body mass index: HR (95% CI) per 5mL/min/1.73m² lower eGFR: eGFRcr, 1.16 (1.04-1.28); eGFRcys, 1.20 (1.08-1.34); eGFRcr-cys, 1.19 (1.08-1.32). After adjusting for NT-proBNP and age, no baseline eGFR measure was statistically significantly associated with the composite endpoint. Only a relative decline in eGFRcr-cys from baseline to 12 months was associated with the composite endpoint in fully adjusted models: eGFRcr-cys, HR 1.31 (1.02-1.67); per 10% eGFR decrease). Lower eGFR at any timepoint, using all 3 measures, was associated with the composite endpoint in fully-adjusted models. Limitations: small number of events, possible survival bias Conclusions: A decline in eGFRcr-cys from baseline to 12 months is associated with adverse cardiovascular outcomes in people wit reduced ejection fraction h HFrEF. Measures of eGFR that incorporate only cystatin C may not be of greater prognostic value in people with HFpEF than measures that also incorporate serum creatinine. Plain-Language Summary: Cystatin C-based measures of kidney function have been more predictive of adverse outcomes in people with heart failure with reduced ejection fraction (HFrEF), but their prognostic utility in heart failure with preserved ejection fraction (HFpEF) is unclear. In a longitudinal analysis of 214 participants from the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist trial, a HFpEF population, we found that a relative decline in eGFRcr-cys from baseline to 12 months was associated with adverse cardiovascular outcomes. However, cystatin C-based measures on their own were not associated with adverse cardiovascular outcomes. This suggests that cystatin C based measures may not be more prognostic that serum creatinine-based measures in HFpEF.