Kidney function decline, heart failure hospitalization, and the effect of empagliflozin in heart failure: an EMPEROR-pooled analysis

Abstract

Abstract Background Kidney function decline in heart failure (HF) may accelerate during periods surrounding HF-related events, while sodium-glucose cotransporter 2 inhibitors slow kidney function decline. However, the temporal relationship between kidney function changes, HF-related events, and treatment effects has not been well characterized. Aims We aimed to explore kidney function trajectories around HF-related events and to assess whether empagliflozin modifies these changes. Methods and results In a pooled analysis of the EMPEROR-Reduced and EMPEROR-Preserved trials, we examined longitudinal changes in estimated glomerular filtration rate (eGFR) before and after HF-related events, defined as HF hospitalization or HF-related death. In the EMPEROR-Pooled analysis, HF-related events occurred in 12.7% of 9554 patients during a median follow-up of 20.9 months (14.7–28.6). Patients who experienced HF-related events had a steeper eGFR decline during the 12 months before the event than those without HF-related events: average −5.01 ml/min/1.73 m2/year vs −1.93 ml/min/1.73 m2/year; P < .001. During the 12 months after HF-related events, eGFR decline persisted at an average rate of −4.73 ml/min/1.73 m2/year. These trajectories were not materially altered after adjustment for concurrent NYHA class or natriuretic peptide levels. Compared with placebo, empagliflozin slowed eGFR decline in patients without events (between-group difference, 0.94 ml/min/1.73 m2/year) and before events (1.09 ml/min/1.73 m2/year). The post-event difference was smaller (0.29 ml/min/1.73 m2/year), without significant heterogeneity across periods (P > .10). Findings were consistent across trials and irrespective of prior HF hospitalization. Conclusion Across the LVEF spectrum, kidney function decline accelerated before HF-related events and remained rapid thereafter. Empagliflozin attenuated eGFR decline without evidence of heterogeneity across periods.