Summary: In this secondary analysis of the EARLIER randomized trial (n=300, acute heart failure, Japan), the combination of diabetes and chronic kidney disease was associated with a markedly higher risk of cardiovascular death or hospitalisation (adjusted HR 2.33, 95% CI 1.12-4.84). The effects of eplerenone on heart-failure outcomes and adverse events did not differ significantly by diabetes or CKD status (all P-for-interaction >0.05), although the trial was small and underpowered for subgroup efficacy.
PICO Summary
| Element | Detail |
|---|---|
| Population | 300 patients hospitalised for acute heart failure (mean age 67 years, 73% male); 39% with diabetes, 58% with CKD, 26% with both; multicentre randomised trial, Japan. |
| Intervention | Eplerenone (a steroidal mineralocorticoid receptor antagonist) for 6 months, with subgroups defined by diabetes and/or CKD status (CKD = eGFR <45 ml/min/1.73 m2 or UACR ≥30 mg/g). |
| Comparison | Placebo for 6 months; and a within-cohort comparison of patients with both diabetes and CKD versus those with neither. |
| Outcome | Diabetes plus CKD (26%) was associated with higher risk of cardiovascular death and/or hospitalisation versus neither condition (HR 2.57, 95% CI 1.29-5.12, P=0.007; P-for-interaction=0.049; adjusted HR 2.33, 95% CI 1.12-4.84, P=0.02). The effects of eplerenone on heart-failure outcomes (cardiovascular death, HF hospitalisation, worsening HF, or out-of-hospital diuretic intensification) and on adverse events were not significantly modified by diabetes or CKD status (all P-for-interaction >0.05). No subgroup-specific treatment efficacy estimate (effect size, ARR, or NNT) was reported. |
Expert Commentary
This secondary analysis of the EARLIER trial delivers two distinct messages that should not be conflated. The robust signal is prognostic: among patients hospitalised for acute heart failure, the coexistence of diabetes and CKD more than doubled the adjusted risk of cardiovascular death or hospitalisation. That association is consistent with a large body of evidence and is clinically credible, though it is observational within the cohort and cannot establish causation. The second message, that eplerenone behaved consistently across diabetes and CKD subgroups, is far weaker than it may appear. A non-significant interaction term in a trial of only 300 patients is an absence of evidence for effect modification, not evidence of uniform benefit; the subgroup analyses are underpowered, and no subgroup-specific effect size, absolute risk reduction, or number-needed-to-treat is provided. The single weighed limitation is therefore statistical power: with few events, the consistency claim rests on confidence intervals too wide to exclude meaningful heterogeneity. Can I use this with my patients? Use the prognostic half now, to flag the acute-HF patient who has both diabetes and CKD as high-risk and worthy of closer follow-up. Do not use it as fresh proof that eplerenone is specifically efficacious in that subgroup. Clinicians should anchor mineralocorticoid receptor antagonist decisions to the larger outcome trials and watch potassium and renal function closely. A larger, adequately powered analysis is needed before subgroup efficacy can be asserted.
References
Kobayashi M, Yamashina A, Satomi K, Watanabe M, Takagi R, Tezuka A, et al. Eplerenone, diabetes, and chronic kidney disease in patients hospitalized for acute heart failure: findings from the EARLIER trial. Cardiovasc Diabetol. 2025;24(1):136. doi:10.1186/s12933-025-02659-y