# Cardiovascular Outcomes and Hyperkalemia Risk in Patients With Diabetes, Chronic Kidney Disease and Heart Failure: A Real-World Comparison of Non-steroidal versus Steroidal Mineralocorticoid Receptor Antagonists

## TL;DR

Non-steroidal MRAs like finerenone may offer better safety and outcomes than steroidal MRAs in patients with diabetes, kidney disease, and heart failure.

## Contribution

Real-world evidence comparing non-steroidal and steroidal MRAs in patients with T2DM, CKD, and HF, showing improved safety and outcomes with non-steroidal MRAs.

## Key findings

- Non-steroidal MRAs were linked to lower risks of hyperkalemia and severe hyperkalemia.
- Non-steroidal MRAs were associated with reduced arrhythmia, stroke, and mortality rates.
- Non-steroidal MRAs showed a trend toward lower MI rates, though not statistically significant.

## Abstract

Background: Steroidal mineralocorticoid receptor antagonists (MRAs) reduce morbidity in heart failure (HF) but frequently cause hyperkalemia, limiting long-term use. Non-steroidal MRAs such as finerenone offer improved receptor selectivity, but real-world comparative data remain scarce.

Methods: We used the TriNetX Global Collaborative Network to identify adults with type 2 diabetes mellitus (T2DM), chronic kidney disease (CKD), and HF initiated on a non-steroidal (finerenone) or steroidal MRA (spironolactone, eplerenone) between January 2020 and December 2024. Propensity score matching (1:1) yielded 780 patients per cohort. Outcomes included hyperkalemia, arrhythmia, stroke, myocardial infarction (MI), and all-cause mortality.

Results: Non-steroidal MRAs were associated with significantly lower risks of hyperkalemia >5.5 mmol/L (16.0% vs 21.8%; HR 0.683, 95% CI 0.542-0.861; p=0.001) and severe hyperkalemia >6.0 mmol/L (7.6% vs 10.4%; HR 0.690, 95% CI 0.493-0.965; p=0.029). They were also linked to a lower arrhythmia incidence (31.2% vs 44.7%; HR 0.596, 95% CI 0.506-0.703; p<0.001), stroke (6.2% vs 9.9%; HR 0.606, 95% CI 0.422-0.869; p=0.006), and all-cause mortality (6.3% vs 16.5%; HR 0.359, 95% CI 0.258-0.499; p<0.001). MI was numerically lower (10.9% vs 13.5%; HR 0.762, 95% CI 0.572-1.014; p=0.061) but not statistically significant.

Conclusion: In patients with T2DM, CKD, and HF, non-steroidal MRAs were associated with improved safety and cardiovascular outcomes compared with steroidal MRAs. However, the observational study design and, hence, the limitation in data that can be collected warrant cautious interpretation. Prospective randomized head-to-head trials are warranted before adaptation in clinical practice.