Impact of mineralocorticoid receptor antagonists on the risk of sudden cardiac death across cardio-kidney-metabolic diseases: A meta-analysis of randomized clinical trials

Abstract

Mineralocorticoid receptor antagonists (MRAs) are well established treatments across different cardio-kidney-metabolic (CKM) conditions, from heart failure (HF) to diabetic chronic kidney disease (CKD).^{1, 2} Spironolactone, eplerenone, and more recently finerenone have demonstrated significant reductions in cardiovascular events in landmark randomized clinical trials (RCTs) across these populations.

Sudden cardiac death (SCD) represents an important proportion of cardiovascular death mode; still, SCD risk varies across different CKM conditions. For example, SCD rates range from 5.8 per 100 person-years (py) in HF with reduced ejection fraction (HFrEF) to 0.4/100 py in those with non-end stage CKD.^{3, 4}

Prior individual patient data meta-analysis of RCTs showed that MRAs (vs. placebo) significantly reduced the relative risk of SCD in patients with HFrEF by 23%.³ However, it remains unclear whether similar effects may be seen in patients with preserved/mildly reduced ejection fraction (HFpEF/HFmrEF) or in those with type 2 diabetes (T2D) with albuminuric CKD.

The aim of this study was to assess the impact of MRAs on the occurrence of SCD across different CKM conditions: HFrEF, myocardial infarction (MI) with left ventricular systolic dysfunction, HFpEF/HFmrEF, and albuminuric CKD in T2D.

In this updated trial-level meta-analysis, we included data from RCTs which tested the impact of MRAs (vs. placebo) on SCD. PubMed/MEDLINE was searched from inception to 10 September 2024, and a total of five trials met the eligibility criteria and were included in this meta-analysis.

Beyond the recently published Finerenone in Heart Failure and Chronic Kidney Disease with Type 2 Diabetes (FINE-HEART) individual patient level meta-analysis of trials testing finerenone in diabetic CKD and HFpEF/HFmrEF (PROSPERO registry CRD42024570467), we included the Randomized Aldactone Evaluation Study (RALES), the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) and the Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF) in patients with HFrEF, and a post-hoc analysis of SCD events in HFpEF patients included from the Americas region (United States, Canada, Brazil and Argentina) of the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT).^{1, 2, 5-7} EPHESUS was a post-MI trial, but patients had to have HF or left ventricular systolic dysfunction to be included.⁶ All trials reported SCD as adjudicated by independent committees. SCD events collected from TOPCAT-Americas also included patients with aborted cardiac arrest.

A random-effect meta-analysis was used. Heterogeneity across trials was assessed using the Thompson I² and Cochrane's Q test. Data on number of SCD events in the MRA and placebo groups were retrieved from each trial and the respective odds ratio (OR) and 95% confidence intervals (95% CI) calculated. Publication bias was assessed by visual inspection of funnel plots, with the ascertainment for potential asymmetry of published results by Egger's regression test and Duval and Tweedie's trim-and-fill method. Statistical analyses were performed using Stata® (StataCorp. 2021, Stata Statistical Software: Release 17; StataCorp LLC, College Station, TX, USA).

Five eligible trials included a total of 31 790 patients. Trials differed in some important aspects. Different patient populations were included, and different MRAs tested across trials: spironolactone in RALES and TOPCAT, eplerenone in EPHESUS and EMPHASIS-HF, and finerenone in the FINE-HEART pooled data. Follow-up time also differed, from a mean follow-up of 16 months in the EPHESUS trial to a median 36 months in TOPCAT-Americas. SCD incidence rates were also markedly different: from 6.5 events/100 py in the RALES trial, 4.2 events/100 py in the EPHESUS trial, 2.8 events/100 py in the EMPHASIS-HF trial, 1.4 events/100 py in the TOPCAT-Americas, and 0.8 events/100 py in the FINE-HEART pooled data.

Overall, MRAs (vs. placebo) significantly reduced the relative risk of SCD by 21% (OR 0.79, 95% CI 0.70–0.89; p < 0.001) (Figure 1). No significant heterogeneity was found across trials (I² = 0%; p = 0.98).

Data retrieved from funnel plots, and analysis of Egger's regression test and Duval and Tweedie's trim-and-fill method did not suggest publication bias (data not shown).

This meta-analysis including over 31 000 patients across CKM conditions showed that patients randomized to MRAs had a 21% lower relative risk of SCD compared with placebo. The consistent effect across populations suggests that the reduction in SCD may extend beyond HFrEF and be generalized across CKM populations. Particularly, the homogeneous results seen in the TOPCAT-Americas and the FINE-HEART meta-analysis further suggest that the beneficial effects of MRAs in SCD risk reduction may be generalized to both diabetic CKD and HFpEF/HFmrEF.

Traditionally considered a prevalent outcome in patients with HFrEF accounting for nearly half of deaths from cardiovascular causes, patients with other CKM conditions may experience relatively high rates of SCD. For example, SCD accounted for ~20% of all deaths over approximately 3 years in patients with HFpEF recruited from the Americas region in the TOPCAT trial.⁷ Although the risk of SCD varies greatly across CKM populations, being the highest (6.5/100 py) in patients with severely symptomatic HFrEF (as those enrolled in RALES), lowest among patients with T2D and albuminuric CKD (0.4–0.8/100 py), and intermediate among those with mildly symptomatic HFrEF and HFmrEF/HFpEF (1.4–2.8/100 py), MRAs reduce such risk regardless of the population being studied and irrespective of ejection fraction.

Beyond the magnitude of risk, SCD causes may also vary across CKM populations. Tachyarrhythmias are frequently implicated in SCD in HFrEF patients, but these events were rare in those with HFpEF/HFmrEF and CKD.⁸ In these populations, the higher prevalence of bradyarrhythmias and asystole potentially contributes to the SCD burden. Furthermore, although usually considered a synonym of arrhythmic death, SCD may be secondary to other prevalent pathological processes in CKM populations, including MI, pulmonary embolism, stroke, aortic dissection, or ruptured aortic aneurysms.⁹ The established effect of MRAs on reducing fibrosis, inflammation, neurohormonal activation and impacting HF and CKD disease progression provide the pathophysiological rationale for SCD reduction across these conditions, supporting their broader applicability beyond HFrEF.¹⁰

Important limitations should be acknowledged in this meta-analysis. The absolute number of events in older HFrEF RCTs were markedly higher compared with contemporary studies. A decrease in SCD over time has been described in correlation with better HF treatments.¹¹ Nevertheless, the relative reduction in SCD events was similar across trials and the protective effect of MRAs was consistent across populations, extending the benefits of MRAs to prevent SCD across a wide range of CKM conditions. Finally, the individual trial data from the FINE-HEART meta-analysis were not available to us.

In conclusion, our updated meta-analysis suggests that the benefit of MRAs on SCD risk reduction is consistent across CKM populations, adding to their established benefits on cardiovascular and kidney outcomes.

Conflict of interest: P.M. received salary support from an educational grant by Janssen. F.Z. reports personal fees from Applied Therapeutics, Bayer, Betagenon, Biopeutics, Boehringer, BMS, CVRx, Cardior, Cereno pharmaceutical, Cellprothera, Merck, Northsea, NovoNordisk, Otsuka, Owkin, having stock options at G3Pharmaceutical and equities at Cereno pharmaceutical, and being the founder of Cardiovascular Clinical Trialists (CVCT). J.P.F. has received research support from Boehringer Ingelheim, AstraZeneca, Bayer, Novartis, Bial, and Salamandra.