Association of Estimated GFR Slope and Heart Failure Progression in Older Adults

Abstract

Rationale & Objective

The impact of the estimated glomerular filtration rate (eGFR) slope on the progression of heart failure (HF) remains unclear. This study investigated the association of eGFR slope and the occurrence of incident HF in older adults either at elevated risk for HF or with cardiac abnormalities (structural, hemodynamic, or biomarker) (pre-HF).

Study Design

Prospective cohort study.

Setting & Participants

5,362 participants in the Atherosclerosis Risk in Communities at study visit 5 free of HF (at elevated risk [stage A]: n = 1,262; with cardiac abnormalities [stage B]: n = 4,100).

Exposure

eGFR slope, estimated from eGFR measurements between visits 1 and 5 using linear mixed-effects models.

Outcome

Incident clinical HF based on hospitalization ICD-9/10 codes.

Analytical Approach

Cause-specific hazard models.

Results

The mean age of study participants was 75.7 years, 58.4% were female, and the median follow-up was over 6 years. No significant association was found among individuals in stage A. In stage B, each 1 mL/min/1.73 m² per year decrease in eGFR slope was associated with a higher risk of HF (HR, 1.39 [95% CI, 1.11-1.73]). Multivariable-adjusted cubic spline analysis indicated that this association was nonlinear among individuals in stage B (P nonlinearity = 0.02). Compared with individuals with a moderate-to-high decline in eGFR (−1.87 to −1.25 mL/min/1.73 m² per year), individuals in stage B with a steeper eGFR slope (<−1.87 mL/min/1.73 m² per year) exhibited an increased risk of incident HF (HR, 1.58 [95% CI, 1.20-2.10]). A steeper eGFR slope in stage B was also significantly associated with HF with preserved ejection fraction (HFpEF) (HR, 1.54 [95% CI, 1.10-2.17]). Among individuals in stage B, there was no detectably elevated risk of incident clinical HF when the eGFR slope was greater than −1.25 mL/min/1.73 m² per year.

Limitations

The eGFR slope may not fully capture the complexity of kidney function decline because it may not account for the fluctuating patterns.

Conclusions

The eGFR slope may be a valuable measurement to identify individuals at high risk of developing HF, particularly HFpEF.

Plain-Language Summary

Heart failure (HF) is a prevalent condition in older people, and its occurrence may be affected by kidney function. This study examined the relationship between changes in kidney function assessed by the slope of estimated glomerular filtration rate (eGFR) and the incidence of clinically diagnosed HF. We monitored more than 5,000 older people who were either at elevated risk for HF (stage A) or had cardiac abnormalities of HF (stage B) but had not yet developed clinical HF. Our findings indicated that among individuals in stage B a more rapid decline in kidney function was associated with a greater risk of developing HF, in particular HF with preserved ejection fraction (HFpEF). These findings highlight the potential value of tracking changes in kidney function to identify people at elevated risk of HF.