Secrets of Spironolactone: Continuing Insights from TOPCAT Americas

‘All things are poisons, for there is nothing without poisonous qualities. It is only the dose which makes a thing poison.’ The story of spironolactone use in patients with heart failure (HF) is an interesting one. Initially used for ascites and cardiac oedema as an antagonist of aldosterone first in large doses (e.g. 400–600 mg daily) and subsequently with slightly lower doses (e.g. 100–200 mg daily),1, 2 spironolactone was eventually demonstrated to improve the prognosis of patients with HF and reduced ejection fraction (HFrEF) when used in relatively smaller doses (25–50 mg daily).3 The diuretic effect of spironolactone at these doses is small, as the synergistic diuretic benefit in combination with loop diuretics is usually observed at higher doses (50–200 mg daily).4 Mineralocorticoid receptor antagonists possess antifibrotic and antiarrhythmic properties,5, 6 and substantially reduce mortality with early administration in patients with left ventricular dysfunction and HF,7, 8 a benefit that is unlikely to be mediated by diuretic effects. The anticipation that spironolactone could counteract the myocardial and vascular pro-fibrotic activity of aldosterone, known to be elevated in both HF with preserved ejection fraction (HFpEF) as in HFrEF,9 led to the hypothesis that this agent might be beneficial in patients with HFpEF.10 The Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) was an international, randomized, double-blind, placebo-controlled trial of spironolactone (initially 15 mg once daily with attempted up-titration to a maximum of 30 or 45 mg daily during the first 4 months after randomization) vs. placebo in patients with HF and ejection fraction ≥45% with either a HF hospitalization within 12 months before randomization or an elevated B-type natriuretic peptide (BNP) level (BNP ≥100 pg/mL or N-terminal pro-BNP ≥360 pg/mL) within 60 days.11 Patients were recruited from North and South America, Russia, and Georgia. Despite the neutral overall effect of spironolactone on the primary composite of cardiovascular death, aborted cardiac arrest, or HF hospitalization in the entire TOPCAT cohort,11 post-hoc analyses demonstrated significant regional differences in event rates and trial conduct.12, 13 In the Americas, where a higher percentage of patients had detectable blood levels of the active metabolite canrenone compared to Russia (97% vs. 70%, P < 0.001),13 the rates of the primary composite endpoint, cardiovascular death, and HF hospitalization were reduced by spironolactone, with a mean daily dose of only 21.7 and 25.9 mg for spironolactone and placebo, respectively, at 8 months.12 In a recent analysis from the TOPCAT Americas, based on surrogates such as weight and loop diuretic dose, the benefit of spironolactone on outcomes could not be solely attributed to potential diuretic effects, suggesting the presence of non-diuretic mechanisms.14 In this issue of the Journal, Ferreira and colleagues, in a post-hoc analysis of the TOPCAT Americas cohort of patients with HFpEF, provide additional insights on the titration and discontinuation of spironolactone and the dose–response association between spironolactone and clinical events.15 Ferreira et al. report that, during the first year of the study, patients over age 75 and those with renal dysfunction (estimated glomerular filtration rate ≤60 mL/min) and higher potassium levels (>4.5 mmol/L) received lower doses of spironolactone (median daily dose ≤20 mg). For the entire follow-up period, achieved spironolactone doses were even lower (median dose approximately 15 mg) in these patients. Discontinuation rates reached 25.5% for active drug and 18.3% for placebo during the first year. Up to 30% of patients discontinued spironolactone if they had baseline renal dysfunction or hyperkalaemia. Additional factors that had an adverse impact on up-titration were lower body mass index and lower systolic blood pressure. Interestingly, failure to up-titrate the study drug did not seem to affect the benefit of spironolactone on outcomes. However, the need to discontinue the study drug was associated with higher event rates, with a trend towards a more prominent ‘rebound’ effect in the spironolactone arm. Several observations by Ferreira et al. merit further discussion. First, a substantial number of patients (one out of four in the active arm in the first year) had to discontinue study drug. This was associated with an increase in the number of clinical events, suggesting the need to discontinue spironolactone may be a marker of incipient clinical fragility. Second, Ferreira et al. reported that even lower-than-target doses of spironolactone in TOPCAT were associated with clinical benefit, comparable to that observed in patients receiving target doses.15 Despite the statistical caveats discussed below, the clinical implications of this finding are important. In contrast to the case with angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and beta-blockers, where achieving guideline-recommended target doses improves clinical outcomes in terms of mortality and hospitalizations,16 the findings by Ferreira et al. suggest that benefit with spironolactone may be seen at an individual's maximally tolerated dose and encourage use of even low doses over ‘no dose’ of spironolactone. On the other hand, hyperkalaemia is clearly a dose-dependent effect of spironolactone.17 Still, there is substantial evidence gap regarding the pharmacology of spironolactone in contemporary patients with HF. For example, in the Aldosterone Targeted Neurohormonal Combined with Natriuresis Therapy in Heart Failure (ATHENA-HF) trial,18 the addition of a moderately high (with current standards) dose of spironolactone (100 mg) to usual care in patients with acute HF did not result in more prominent diuresis, suggesting that higher doses are not necessarily better. A subsequent pharmacologic substudy demonstrated that concentrations of spironolactone active metabolites were lower-than-anticipated for at least 48 h in the high-dose spironolactone group and may have contributed to the absence of pharmacological effects of spironolactone in the ATHENA-HF trial.19 In all, there is still a lot to learn about the pharmacology and physiologic activity of spironolactone in HF. Despite these important clinical implications, the analysis of Ferreira et al.15 has also several caveats that merit further discussion. First, there was no a priori stratification according to study drug doses and patient risk profile in TOPCAT. Therefore, subsequent dose groups are inherently biased, as both failure to up-titrate and drug discontinuation were ‘informative’ events (i.e. these events designated groups with different characteristics compared to the entire trial cohort) and therefore the findings in these groups may not apply back to the entire study population. Second, the use of a single measure of location (mean/median) for study drug dose during the first year of follow-up may have concealed temporal trends in drug dosage that could affect the outcome, as drug dose is inherently a time-dependent covariate and may change based on past data but it cannot be predicted in the future. Since the drug dose was averaged over time, early deaths or hospitalizations may have been erroneously attributed to lower doses. Third, a recurrent event method for hospitalizations would have been more informative and provide more statistical over a time-to-first event approach. Finally, since all these analyses were not a priori powered, ‘neutral’ findings (e.g. persistent benefit with lower drug doses) should only be viewed as hypothesis-generating. Despite these caveats, the study of Ferreira et al. adds to the mounting evidence suggesting that the long-term beneficial effects of maximally tolerated doses of spironolactone (and mineralocorticoid receptor antagonism in general) are present for both HFrEF and HFpEF patients and that the effects are not necessarily mediated by the induction of diuresis. As there is evidence-practice gap, with large-scale real-world data showing that mineralocorticoid receptor antagonists are largely underprescribed and often discontinued, even in patients with HFrEF,20 the study of Ferreira et al. encourages providers to persist with even smaller doses when higher doses are not tolerated. Finally, the doses of spironolactone used in TOPCAT are not commercially available, requiring clinicians be willing to adapt. Until further prospective data are available, it is reasonable to assume that low-dose mineralocorticoid receptor antagonism is better than no antagonism at all. Conflict of interest: A.P.K. has received research support from the National Heart, Lung, and Blood Institute, American Heart Association, and Critical Diagnostics. H.A.S. is a speaker for Novartis and Pfizer. All other authors have nothing to disclose.