- Design
- Patient-level pooled cohort analysis and meta-analysis
- Population
- About 1,000 heart failure patients in four cohorts, plus EARLIER and EPHESUS trial data
- Primary outcome
- Clinical outcomes with MRA use by pre-treatment aldosterone
- Effect
- High aldosterone HR 0.63 (95% CI 0.42–0.92); low aldosterone HR 1.66 (1.12–2.45)
Investigators pooled patient-level data from heart failure cohorts with plasma aldosterone measured before mineralocorticoid receptor antagonist use (about 1,000 patients across four studies), then meta-analysed outcomes with the EARLIER and EPHESUS trials. Receptor activity was estimated from the urine sodium-to-potassium ratio.
MRAs roughly doubled aldosterone concentrations. In patients whose aldosterone was high before treatment, starting an MRA lowered receptor activity and was associated with better outcomes (HR 0.63). In those with low aldosterone, starting an MRA was associated with higher receptor activity and worse outcomes (HR 1.66). The difference between the two groups was statistically significant.
The authors call the findings hypothesis-generating, and they are: these are mostly observational associations, and the MRA trials that established benefit did not select by aldosterone. Guidelines still recommend MRAs across heart failure with reduced ejection fraction, and this study is not a reason to withhold one. It is a reason to watch for trials testing higher doses or aldosterone-guided therapy.
- Continue to prescribe MRAs as guideline-directed therapy in heart failure with reduced ejection fraction.
- Do not order plasma aldosterone to decide on MRA use; no threshold has been validated for this.
- Check potassium and creatinine at one week and one month after starting or increasing an MRA.
- Expect future trials to test whether low-dose MRAs under-block the receptor.
Why it matters
It challenges the assumption that the low MRA doses used in heart failure fully block the receptor.
Don't overread it
Associations from pooled cohorts cannot show that an MRA harms patients with low aldosterone.
The statistics, in plain English
The two hazard ratios point in opposite directions, and the interaction test (P=0.005) says that difference is unlikely to be chance. But these subgroups were not randomised by aldosterone level, so other differences between high- and low-aldosterone patients may explain part of it.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for cardiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free