Blood aldosterone concentrations and effects of mineralocorticoid receptor antagonists in heart failure

BACKGROUND AND AIMS: Low-dose mineralocorticoid receptor antagonists (MRAs) are guideline-recommended heart failure (HF) therapies. However, MRAs increase aldosterone production, and the sub-saturating doses utilized may allow continued mineralocorticoid receptor (MR) stimulation. The aim of the current analysis was to understand how baseline and change in aldosterone concentrations during MRA therapy impacts MR activity and clinical outcomes. METHODS: HF cohorts with MRA exposure and plasma aldosterone concentrations available were included in patient-level, pooled cohort analyses [DOSE, CARRESS-HF, MDR, and TOPCAT (n = 1019)]. In the MDR cohort (n = 136), urine sodium to potassium ratio was utilized to quantitate MR activity. The relationship of pre-MRA aldosterone concentration, MRA use, and clinical outcomes were meta-analysed utilizing the above pooled cohorts in addition to the EARLIER (n = 300) and EPHESUS (n = 453) trials. RESULTS: MRA use was associated with significantly higher median aldosterone concentrations [MRA = 310 (interquartile range, IQR 180, 533) pg/mL vs no MRA = 174 (IQR 106, 299) pg/mL, P < .001] in the pooled cohort. In the MDR cohort, higher aldosterone was correlated with higher MR activity, with a similar relationship on MRA (r = -.52, P < .001) vs off MRA (r = -.44, P < .001, P interaction = .65), differing only in that higher aldosterone concentrations were required on MRA to achieve the same level of MR activity. In patients with high pre-MRA aldosterone, new MRA initiation reduced MR activity, but MRA initiation increased MR activity in patients with low pre-MRA aldosterone [median change in urine Na/K with high aldosterone =1.4 (IQR 1.1, 2.9) vs low aldosterone = -.9 (IQR -2.1, -.1), P < .001]. In the pooled cohort, the association between MRA use and clinical outcomes was dependent on pre-MRA aldosterone concentration (P interaction = .005). In patients with high pre-MRA aldosterone, MRA was associated with substantially improved clinical outcomes [hazard ratio (HR) .63, 95% confidence interval (CI) .42-.92, P = .02]. However, in patients with low pre-MRA aldosterone, MRA use was associated with worse clinical outcomes (HR 1.66, 95% CI 1.12-2.45, P = .01). CONCLUSIONS: Low-dose MRAs significantly increase aldosterone but inadequately block MR activity (measured by urine Na/K) at these new higher aldosterone concentrations. In patients with high pre-MRA aldosterone, MRA use is associated with improved MR activity and clinical outcomes. However, in patients with lower pre-MRA aldosterone concentrations, MRA use is associated with worsened MR activity and clinical outcomes. MRAs remain guideline-directed HF medications with established population level benefit, but these hypothesis-generating findings indicate additional research is warranted to understand if outcomes can be further improved.