Initiation of mineralocorticoid receptor antagonists in patients with cardiorenal diseases: potassium changes, hyperkalaemia risk and discontinuation in the CaRe study
S J Greene, T Kondo, S Adamsson Eryd, J Bodegard, M Thuresson, J Enxing, K Mullin, L Anderson, G SavareseAbstract
Background
A key barrier to using steroidal mineralocorticoid receptor antagonists (MRAs) to treat heart failure (HF) in clinical practice is the real or perceived risk of hyperkalaemia (HK), especially in patients with HF and chronic kidney disease (CKD).
Purpose
To describe serum potassium (K⁺) levels, HK risk and MRA drug utilization after MRA initiation using real-world data.
Methods
This was an observational longitudinal cohort study using data from Optum’s de-identified Clinformatics® Data Mart of medical and pharmacy claims in the USA. Patients with comorbid HF and CKD were indexed at first initiation of MRA (spironolactone or eplerenone) during 1 Jan 2021–30 Sep 2024 and grouped by starting dose (12.5, 25 or 50 mg). HF was defined as a recorded diagnosis code of HF. CKD was defined as a recorded diagnosis code of CKD or ≥2 estimated glomerular filtration rate (eGFR) measurements of <60 mL/min/1.73 m² taken ≥90 days apart. Baseline comorbidities were identified using all available data; patients had drug use and outpatient specialist visit data available in the year before indexing. Clinical outcomes were K⁺ level trajectories and HK events (K⁺ >5.5 mmol/L, diagnosis code of HK or potassium binder initiation). Drug utilization was described over 104 weeks or until death or discontinuation.
Results
Of 104 570 patients with HF and CKD initiating an MRA, 23.5% initiated on 12.5 mg, 64.2% on 25 mg and 12.2% on 50 mg. Patients initiating 50 mg were younger than those initiating the 12.5/25 mg doses (mean 73 vs 75–76 years). At baseline, median K⁺ was 4.2–4.3 mmol/L; 4.3–8.6% patients had K⁺ <3.5 and 5.8–6.1% had K⁺ >5.0. Median eGFR was 54–55 mL/min/1.73 m². At baseline, 5.4–6.9% patients were on sodium–glucose cotransporter 2 inhibitors and 60.7–66.1% on renin–angiotensin system inhibitors. After MRA initiation, early dose-dependent increases in K⁺ were observed at 2–8 weeks: mean change from baseline +0.23 (12.5 mg), +0.32 (25 mg), and +0.42 mmol/L (50 mg). After this initial increase, patients who discontinued MRA exhibited rapid K⁺ reduction while patients who remained on treatment had an overall plateau in K⁺ (Figure 1A). Two-year cumulative incidence of HK was high: 26.4% (12.5 mg), 29.7% (25 mg) and 39.6% (50 mg), corresponding to 18.1, 20.9 and 32.1 events per 100 patient-years, respectively (Figure 1B). MRA treatment persistence was low, with marked early discontinuation: over half of patients had discontinued treatment by 2 years (Figure 1C).
Conclusion
In patients with HF and CKD initiating MRA, early and dose-dependent K⁺ increases were observed alongside increasing HK incidence. Over half of patients discontinued treatment by 2 years, and risk of HK was high even with initiation of guideline-recommended lower starting doses. These findings highlight barriers to sustained MRA use in routine care and underscore the need for treatment strategies that enable improved long-term risk management in patients with cardiorenal diseases.For image description, please refer to the figure legend and surrounding text.