Author + information
- Received March 21, 2016
- Revision received June 6, 2016
- Accepted June 14, 2016
- Published online September 1, 2016.
- Javed Butler, MD, MPH, MBAa,∗ (, )
- Adrian F. Hernandez, MDb,
- Kevin J. Anstrom, PhDb,
- Andreas Kalogeropoulos, MD, PhDc,
- Margaret M. Redfield, MDd,
- Marvin A. Konstam, MDe,
- W.H. Wilson Tang, MDf,
- G. Michael Felker, MDb,
- Monica R. Shah, MDg and
- Eugene Braunwald, MDh
- aDepartment of Medicine, Stony Brook University, Stony Brook, New York
- bDepartment of Medicine, Duke University, Durham, North Carolina
- cDepartment of Medicine, Emory University, Atlanta, Georgia
- dDepartment of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
- eDepartment of Medicine, Tufts University, Boston, Massachusetts
- fDepartment of Cardiovascular Medicine, Cleveland Clinic, Cleveland, Ohio
- gDivision of Cardiovascular Sciences, National Heart, Lung, and Blood Institute, Baltimore, Maryland
- hDepartment of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. Javed Butler, Cardiology Division, Stony Brook University, Health Sciences Center, T-16, Room 080, Stony Brook, New York 11794.
Although therapy with mineralocorticoid receptor antagonists (MRAs) is recommended for patients with chronic heart failure (HF) with reduced ejection fraction and in post-infarction HF, it has not been studied well in acute HF (AHF) despite being commonly used in this setting. At high doses, MRA therapy in AHF may relieve congestion through its natriuretic properties and mitigate the effects of adverse neurohormonal activation associated with intravenous loop diuretics. The ATHENA-HF (Aldosterone Targeted Neurohormonal Combined with Natriuresis Therapy in Heart Failure) trial is a randomized, double-blind, placebo-controlled study of the safety and efficacy of 100 mg/day spironolactone versus placebo (or continued low-dose spironolactone use in participants who are already receiving spironolactone at baseline) in 360 patients hospitalized for AHF. Patients are randomized within 24 h of receiving the first dose of intravenous diuretics. The primary objective is to determine if high-dose spironolactone, compared with standard care, will lead to greater reductions in N-terminal pro−B-type natriuretic peptide levels from randomization to 96 h. The secondary endpoints include changes in the clinical congestion score, dyspnea relief, urine output, weight change, loop diuretic dose, and in-hospital worsening HF. Index hospital length of stay and 30-day clinical outcomes will be assessed. Safety endpoints include risk of hyperkalemia and renal function. Differences among patients with reduced versus preserved ejection fraction will be determined. (Study of High-dose Spironolactone vs. Placebo Therapy in Acute Heart Failure [ATHENA-HF]; NCT02235077)
- acute heart failure
- heart failure
- mineralocorticoid receptor antagonist
- natriuretic peptides
Heart failure (HF) accounts for more than 1 million hospitalizations in the United States annually (1,2). Hospitalizations for HF are associated with a significantly elevated risk for post-discharge mortality and recurrent hospitalizations. Mortality or readmission risk at 60 days post-discharge is ∼30% and may be as high as 50% by 6 months in these patients (3–7). Although therapy for chronic HF with reduced ejection fraction has evolved over time, favorably affecting survival, outcomes for patients with acute heart failure (AHF) have not changed much in the past 2 decades. Thus far, none of the trials have shown an improvement in post-discharge outcomes in patients with AHF (8,9). Thus, there remains a pressing need to develop interventions that can improve outcomes safely in this high-risk group of patients.
Persistent Congestion and Outcomes in Acute Heart Failure
Worsening congestion is the main reason for hospitalization for most AHF patients, and diuretics remain the mainstay of therapy. Even with the use of intravenous diuretics, more than one-half of patients lose ≤5 lbs, and up to 20% may actually gain weight during hospitalization (10). Despite improvement in symptoms, a large proportion of these patients continue to have persistent congestion at discharge (11), which whether measured clinically (12,13) with right heart catheterization or by natriuretic peptide levels (14,15), predict post-discharge outcome. Persistent congestion may be subclinical, related to diuretic resistance, or when further diuretic use is difficult in the face of worsening renal function.
Importance of Aldosterone in Acute Heart Failure
The renin-angiotensin-aldosterone-system (RAAS) is activated in HF. In normal subjects, aldosterone levels range between 2 and 9 ng/dl, but in 1 AHF trial, median levels were 11.0 ng/dl and were more than the upper normal range in 33.2% of patients (16). Aldosterone levels are only transiently suppressed with angiotensin-converting enzyme inhibition. Intravenous loop diuretic use in AHF further intensifies RAAS activation and secondary hyperaldosteronism (17,18), enhancing proximal tubular sodium absorption and decreased distal sodium delivery, which impairs the normal escape mechanism from the sodium-retaining effect of aldosterone. Therefore, beyond myocardial and vascular adverse effects, hyperaldosteronism directly contributes to diuretic resistance (19). Loop diuretics also block sodium chloride transport at the macula densa, which stimulates the RAAS independent of renal sodium loss (17) (Figure 1).
Mineralocorticoid Receptor Antagonist Use in Acute Heart Failure
The low-dose mineralocorticoid receptor antagonists (MRAs) used in chronic HF are believed to benefit patients by antifibrotic, but not natriuretic, effects (20). Inhibition of mineralocorticoid receptors at higher MRA doses may cause significant natriuresis. Resistance to loop diuretics in AHF may be overcome by natriuretic doses of spironolactone (>50 mg/day) (19). In a study of 6 HF patients, 200 mg twice a day of spironolactone caused a marked increase in sodium excretion, which led to a negative sodium balance (21). In another study, patients with severe HF who were resistant to high-dose loop diuretics responded with increased natriuresis using 100 mg/day of spironolactone (22).
The clinical benefit and safety of high-dose MRA use in AHF was recently supported by a single-center, single-blind trial of 100 patients treated with standard therapy alone or with the addition of spironolactone initiated within 24 h (23). Spironolactone dose was 94.5 ± 23.3 mg on day 1 and 62.7 ± 24.3 mg on day 3. Increase in creatinine by ≥0.3 mg/dl from day 1 to day 3 was more likely to occur in the standard of care arm (20% vs. 4%; p = 0.038). Plasma N-terminal pro−B-type natriuretic peptide (NT-proBNP) levels were comparable at baseline but were lower in the spironolactone group at day 3 (2,488 ± 4,579 pg/ml in control subjects vs. 1,555 ± 1,832 pg/ml in the spironolactone group; p = 0.05). A greater proportion of patients in the spironolactone group were free of congestion at day 3, and a higher proportion had transitioned from intravenous to oral furosemide (82% vs. 44%; p < 0.001). These findings support the safety and potential efficacy of a high-dose spironolactone strategy in AHF. However, this was a single-center nonrandomized study, and the investigators were not blinded to the intervention, which raises concerns for potential bias. Although high-dose MRA may be effective in the setting of congestion, increased risks of hyperkalemia or elevation in creatinine also need to be studied further (24).
In the EVEREST (Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan) trial, which enrolled patients with left ventricular ejection fractions of <40% who were hospitalized for AHF and who received standard therapy, the median baseline aldosterone blood level was 11.0 ng/dl (25th to 75th percentile: 2 to 21 ng/dl) and was more than the upper normal range in 33.2% of patients. Median aldosterone levels increased during hospital stay from 11 ng/dl at baseline to 15 ng/dl at discharge (p < 0.001) and remained increased 6 months after discharge (16 ng/dl; p < 0.001 vs. baseline). Higher serum aldosterone levels were correlated with worse post-discharge outcomes. After a median follow-up of 9.9 months, higher baseline aldosterone levels were associated with an increased risk of mortality and the combined endpoint of cardiovascular mortality plus HF readmission in adjusted models (hazard ratio [HR]: 1.49; 95% confidence interval [CI]: 1.11 to 1.99; and HR: 1.40; 95% CI: 1.11 to 1.78, respectively) in the highest quartile compared with the lowest quartile (16).
Risk for Hyperkalemia
Most of the data on hyperkalemia with MRAs is on chronic HF. In AHF, hypokalemia is more common and is often due to a defect in sodium/potassium adenosine triphosphatase activity and an intracellular potassium shift caused by oxidative stress and neurohormonal activation in combination with loop diuretic use (25). Spironolactone is rapidly metabolized to several metabolites that produce natriuretic and antikaliuretic effects (26). Although the natriuretic effects decline over 48 to 72 h, the antikaliuretic effects may be observed for several days following discontinuation, underscoring the importance of careful monitoring of potassium levels, especially in patients with chronic kidney disease (27). In a study in 6 chronic HF patients, 200 mg twice daily of spironolactone led to negative sodium balance without significant increases in potassium levels (3.9 to 4.1 mmol/l) or changes in creatinine clearance after 4 days (21). In 18 patients with advanced HF who received 50 to 200 mg of spironolactone in addition to standard treatment, there was no significant increase in serum potassium (4.0 mEq/l vs. 4.2 mEq/l) or creatinine (1.3 mg/dl vs. 1.4 mg/dl) during an average follow up of 41 weeks (28). In an AHF study that assessed 50- to 100-mg spironolactone use, there was no difference in serum potassium between the control and spironolactone groups (3.9 mmol/l vs. 4.1 mmol/l, respectively) at day 3 (p = 0.15) (23).
Endpoints in Acute Heart Failure
Congestion is the most common manifestation of AHF and is related to both symptoms and prognosis. Persistent congestion at discharge is associated with worse outcomes. The effectiveness of decongestion can be measured in multiple ways.
Natriuretic peptide levels
Natriuretic peptide are markers of wall stretch and have advantages over clinical signs for assessment of congestion due to the lack of sensitivity and inter-rater reliability of examination. Even with symptom relief, readmission and mortality risk remains high if natriuretic peptides stay elevated in patients with AHF. Baseline levels and changes in-hospital are both associated with filling pressures and outcomes, with best outcomes seen when natriuretic peptide levels decrease by >30% in-hospital (29). Discharge natriuretic peptide levels are superior to admission levels or changes in levels during hospitalization for predicting risk (30).
In several studies, dyspnea improvement in AHF was associated with improved outcomes post-discharge. In the ASCEND-HF (Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure) trial, early dyspnea relief measured on a 7-point Likert scale was associated with lower 30-day mortality or HF hospitalization risk (31). In the RELAX-AHF (Efficacy and Safety of Relaxin for the Treatment of Acute Heart Failure) trial, serelaxin was associated with both improvement in dyspnea as assessed by visual analog scale and lower 6-month mortality risk (32). Thus, significant and early improvement in dyspnea may predict improved long-term outcomes, besides being a therapeutic goal by itself.
Clinical congestion score
In the EVEREST trial, a modified composite congestion score at discharge calculated by summing the individual scores for orthopnea, jugular venous distension, and pedal edema (on a standardized 4-point scale ranging from 0 to 3) was predictive of 30-day all-cause mortality (HR: 1.34; 95% CI: 1.14 to 1.58) and combined morality and HF readmission risk (HR: 1.13; 95% CI; 1.03 to 1.25) (12).
Mortality reduction is the gold standard outcome for clinical trials in chronic HF. Although several previous trials focused primarily on symptom improvement in clinical trials for AHF, it is now well recognized that symptom improvement during hospitalization is an inconsistent predictor of post-discharge mortality (33,34). Therapies like inotropes may improve symptoms and worsen the risk for mortality. In-hospital mortality for patients with AHF is low, but the main concern is the high (up to 30% at 1 year) post-discharge mortality (35). Thus, assessing post-discharge mortality as both an efficacy and safety outcome in this high-risk group of AHF patients is imperative.
Similar to mortality, readmission post-discharge from an AHF hospitalization also remains a critical endpoint, both from a clinical and an economic perspective (36). From a clinical perspective, recurrent hospitalizations are associated with a change in the natural history trajectory of the disease process, with each hospitalization portending a worse prognosis subsequently (37). From an economic perspective, recurrent hospitalizations account for the largest proportion of direct cost of care for HF. Considering that HF is the number 1 discharge diagnosis in the United States for Medicare beneficiaries, there are now financial penalties imposed by the Centers for Medicare and Medicaid Services for hospitals that have a high rate of readmission post-discharge after an AHF hospitalization. Thus, effective therapies for these patients should preferably translate into a reduction in risk for readmissions.
In-hospital worsening heart failure
Acute worsening heart failure (WHF) is reported in a sizable portion of patients hospitalized for HF and is increasingly being recognized as an entity that is associated with an adverse in-hospital course (35). WHF is generally defined as WHF symptoms and signs that require an intensification of therapy, and is reported to be seen from 5% to 42% of HF admissions (37). Recent data suggest that some experimental therapies may reduce the risk of development of WHF among hospitalized HF patients, and this is associated with a reduction in the risk of subsequent post-discharge cardiovascular mortality (38). In this respect, WHF holds promise as an endpoint for AHF clinical trials. However, a better understanding of the pathophysiology and a consensus on the definition of WHF is still needed, and until then, it serves as a potential secondary endpoint.
The ATHENA-HF Trial
The ATHENA-HF (Aldosterone Targeted Neurohormonal Combined With Natriuresis Therapy in Heart Failure) trial is a randomized, double-blind, placebo-controlled trial assessing the impact of high-dose spironolactone versus placebo (or continued lose-dose spironolactone use) on natriuretic peptide levels among patients hospitalized for AHF. The study schema is presented in the Central Illustration.
The study will determine if high-dose spironolactone administered to patients with AHF will lead to greater reductions in NT-proBNP levels from randomization to 96 h compared with standard of care. To assess the early use of natriuretic doses of spironolactone in AHF and its impact on preventing diuretic resistance, the study intervention is initiated within 24 h of the first dose of intravenous diuretics.
Eligibility and intervention
Eligibility criteria are listed in Table 1. Patients hospitalized with at least 1 sign and 1 symptom of AHF with an NT-proBNP level ≥1,000 pg/ml or BNP ≥250 pg/ml measured within 24 h from randomization are eligible. Patients who are either not on spironolactone therapy at home or those who are on low-dose spironolactone (12.5 or 25 mg/day) are eligible. Patients must have a serum potassium concentration of ≤5.0 mmol/l, an estimated glomerular filtration rate of ≥30 ml/min/1.73 m2, and systolic blood pressure of >90 mm Hg.
Patients who are not on spironolactone at home are randomized to 100-mg spironolactone or placebo. Those already on 12.5 or 25 mg/day spironolactone are randomized to 100 or 25 mg/day but not to placebo, to avoid ethical concerns with discontinuing long-term therapy. Patients on eplerenone will not be included because in the acute setting it may not be easily known if the patient has been previously intolerant of spironolactone. Patients already taking >25 mg of spironolactone will be excluded due to potential overlap with the natriuretic potential of intermediate-dose spironolactone use.
Baseline evaluation includes history, physical examination, vital signs and body weight assessment, review of medications, measurement of renal function and electrolytes, dyspnea assessments (7-point Likert and visual analog scale measured off oxygen for >3 min), pregnancy test for women of childbearing potential, and collection of blood for core laboratory measurement of NT-proBNP levels.
Duration of intervention
The median length of stay for HF hospitalization in the United States is 4.3 days (39). To have comparable efficacy assessment within the 2 arms and avoid comparing outcomes among patients with potentially different lengths of stay, and to conform with the prevalent norms of the duration of hospitalization, the duration of intervention and the primary endpoint assessment was chosen to be 96 hours.
Volume assessment and dose adjustment
All other medications, including diuretics, are left at the discretion of the treating physician. The study drug is discontinued after 96 h, and further MRA use is left to the treating physician’s discretion. If the patient is clinically euvolemic in <96 h, the investigators may consider changing loop diuretics to oral dosing.
Ejection fraction measured within 6 months before randomization is obtained. Those without this information undergo ejection fraction assessment by any modality during hospitalization. Patients will be eligible regardless of ejection fraction; therefore, this determination does not need to precede randomization. Ejection fraction assessment is performed for a pre-specified secondary analysis to ascertain possible differential effects of intervention in patients with reduced versus preserved ejection fraction.
Renal function and hyperkalemia
A modest increase in serum creatinine with diureses is seen in many patients with AHF that usually reverses over time. However, in some patients, a rise in serum creatinine portends a poor prognosis (40). There is a clinical concern about acute kidney injury with high-dose MRA use in AHF. The decision regarding management of patients with worsening creatinine is left to the discretion of the treating physicians. It is recommended that: 1) if serum creatinine increases ≤0.5 mg/dl, and the patient is diuresing and improving, and is still fluid overloaded, the study drug should be continued per protocol; 2) if serum creatinine increases >0.5 mg/dl, and the patient is diuresing and clinically improving but is still fluid overloaded, consider decreasing study drug to 50 mg; and 3) if the patient becomes oliguric, with worsening serum creatinine, and develops acute kidney injury criteria, hold the study drug and reassess in 24 h.
To closely follow the patients to mitigate the risk of hyperkalemia, electrolytes are measured at least every 24 h until 96 h and at discharge. Potassium supplementation and potassium-containing salt substitutes are discontinued, and high-potassium-containing foods are avoided during the study protocol. Study drug dose is adjusted based on serum potassium level as shown in Table 2.
The primary endpoint of ATHENA-HF is the proportional change in NT-proBNP from randomization to 96 h. Multiple secondary endpoints from randomization to 96 h are also assessed, including: 1) clinical congestion score; 2) dyspnea relief by Likert and by visual analog scales; 3) net urine output; 4) weight change; 5) loop diuretic dose need in furosemide dose equivalents; and 6) development of in-hospital WHF, defined as WHF signs and symptoms that require additional therapy.
Exploratory analyses include a day 30 post-randomization telephone call to ascertain: 1) all-cause mortality; 2) all-cause readmissions; 3) outpatient worsening HF, defined as HF readmission or emergency department visits or need for outpatient IV diuretics; 4) MRA use and loop diuretic dose requirement at day 30; and 5) length of stay for index hospitalization. All participants are also contacted by telephone at 60 ± 3 days to assess vital status.
Safety endpoints include change in serum creatinine and incidence of hyperkalemia (>5.5 or >6.0 mmol/l) from randomization to 96 h post-randomization.
Sample size and power calculations
Previous HF network data suggest that the SD for the proportional change (on the log scale) in NT-proBNP from randomization to 96 hours is approximately 0.55 to 0.60 (41). It is anticipated that 25% of the subjects enrolled will be on low-dose MRA at randomization. Assuming a 20% reduction in NT-proBNP from enrollment in the MRA group compared with placebo for the subset of patients not on an MRA at enrollment and a 10% improvement in the subset on low-dose MRA at baseline yields an overall benefit of 17.5% for the study population. With a 1:1 randomization and a 2-sided type I error rate of 0.05, a total sample size of 360 subjects will provide ∼85% power. These calculations are based on the 2-sample t-test. For the sensitivity analysis using the worst-rank approach for missing values due to death, the total sample size of 360 subjects provides 90% power to detect a difference in the setting in which a randomly selected subject on high-dose spironolactone has a 60% chance of having a better response than a randomly selected subject in the placebo/low-dose arm. Both calculations allow for a consent withdrawal rate of ∼2%. For continuous secondary endpoints, the study will have ≈90% power to detect differences of 0.35 SDs between treatment groups. These calculations assume a common variance and normally distributed errors for the 2-sample t-test, with a 2-sided type I error rate of 0.05.
There is potential for greater reductions in natriuretic peptide levels in patients in the placebo arm of the study based on the special expertise for HF management at the sites where patients are being enrolled, although this is a concern with most HF trials in which the placebo arm outcomes are better than those seen in real-life based on similar concerns.
The primary analysis will be based on a regression model using an outcome variable based on the log of the proportional change in NT-proBNP from randomization to 96 h. The primary analysis will use a linear regression model with an indicator variable for treatment assignment, an indicator for MRA use before admission, and the log of the baseline NT-proBNP level. Missing values of the 96-h NT-proBNP levels will be imputed using a multiple imputation algorithm (42). In a sensitivity analysis, values missing due to death will be imputed to the worst possible value. This analysis will account for low-dose MRA at enrollment using a stratified version of the Wilcoxon Mann-Whitney test.
General linear models and nonparametric approaches will be used to analyze the continuous outcomes. For binary outcomes, chi-square tests and Fisher exact test will be used for unadjusted comparisons. For adjusted comparisons, logistic regression analysis will be used to compare the estimated odds ratios and 95% confidence intervals. Unadjusted time-to-event comparisons will be conducted using Kaplan-Meier survival estimates and log-rank tests. For adjusted analyses, Cox proportional hazards regression models will be used to estimate hazard ratios. Sensitivity analyses will be used to assess the influence of informatively missing values on the results. Subgroup analyses will be conducted based on baseline factors, including MRA use before hospitalization, sex, preserved versus reduced ejection fraction, and age younger or older than or equal to 65 years. Interim data analysis for efficacy and futility will not be conducted due to the small size and short duration of this trial. The safety analyses will be based on the entire randomized population.
The first patient in ATHENA HF trial was enrolled on December 13, 2014, and the last patient was enrolled in May 2016. Follow-up of the patients is ongoing.
There is a need to find novel interventions that effectively and safely promote diuresis and improve outcome in AHF. MRA therapy has shown benefit across the spectrum of chronic HF with reduced ejection fraction and a suggestion of benefit in those with preserved ejection fraction, as well in certain regions (43). There are sound theoretical reasons to expect benefit with high-dose MRA therapy in AHF as well. If the ATHENA-HF trial shows promising results, it will lay the groundwork for a more definite outcome trial in AHF. This is even more intriguing considering the development of the novel selective MRA, finerenone, which has shown benefit in early phase trials in patients with WHF and chronic kidney disease (44), as well as the growing literature on the efficacy and safety of long-term use of novel potassium binders to help mitigate the complications related to the use of RAAS inhibitors.
The Heart Failure Clinical Research Network is supported by the National Heart Lung, and Blood Institute and the National Institutes of Health (U10HL084904 for the coordinating center; and U10HL084861, U10HL084875, U10HL084877, U10HL084889, U10HL084890, U10HL084891, U10HL084899, U10HL084907, and U10HL084931 for the clinical centers).
Dr. Hernandez has received grants from Amgen, Bayer, Bristol-Myers Squibb, Merck, and Novartis; and has been a consultant for Bayer, Merck, and Novartis. Dr. Felker has received grants from Novartis, Amgen, Otsuka, and Roche Diagnostics; and has been a consultant for Novartis, Amgen, Bristol-Myers Squibb, Cytokinetics, and Trevena. Dr. Braunwald has received a grant for his Chair of the NHLBI Heat Failure at Duke University; a research grant from Novartis; and has been an uncompensated lecturer and consultant for Novartis. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute heart failure
- confidence interval
- heart failure
- hazard ratio
- mineralocorticoid receptor antagonist
- N-terminal pro−B-type natriuretic peptide
- renin-angiotensin-aldosterone system
- worsening heart failure
- Received March 21, 2016.
- Revision received June 6, 2016.
- Accepted June 14, 2016.
- American College of Cardiology Foundation
- Lloyd-Jones D.,
- Adams R.J.,
- Brown T.M.,
- et al.
- Lloyd-Jones D.,
- Adams R.,
- Carnethon M.,
- et al.
- Krumholz H.M.,
- Merrill A.R.,
- Schone E.M.,
- et al.
- Felker G.M.,
- Pang P.S.,
- Adams K.F.,
- et al.
- Gheorghiade M.,
- Filippatos G.
- Gheorghiade M.,
- Vaduganathan M.,
- Fonarow G.C.,
- Bonow R.O.
- Ambrosy A.P.,
- Pang P.S.,
- Khan S.,
- et al.
- Lala A.,
- McNulty S.E.,
- Mentz R.J.,
- et al.
- Drazner M.H.,
- Hellkamp A.S.,
- Leier C.V.,
- et al.
- Abdallah J.G.,
- Schrier R.W.,
- Edelstein C.,
- Jennings S.D.,
- Wyse B.,
- Ellison D.H.
- Mentz R.J.,
- Stevens S.R.,
- DeVore A.D.,
- et al.
- Bansal S.,
- Lindenfeld J.,
- Schrier R.W.
- Bettencourt P.,
- Azevedo A.,
- Pimenta J.,
- Frioes F.,
- Ferreira S.,
- Ferreira A.
- Filippatos G.,
- Teerlink J.R.,
- Farmakis D.,
- et al.
- Packer M.,
- Colucci W.,
- Fisher L.,
- et al.
- Mozaffarian D.,
- Benjamin E.J.,
- Go A.S.,
- et al.
- Adams K.F. Jr..,
- Fonarow G.C.,
- Emerman C.L.,
- et al.
- Schafer J.L.
- Pfeffer M.A.,
- Claggett B.,
- Assmann S.F.,
- et al.
- Pitt B.,
- Anker S.D.,
- Bohm M.,
- et al.
- Persistent Congestion and Outcomes in Acute Heart Failure
- Importance of Aldosterone in Acute Heart Failure
- Mineralocorticoid Receptor Antagonist Use in Acute Heart Failure
- Risk for Hyperkalemia
- Endpoints in Acute Heart Failure
- The ATHENA-HF Trial
- Statistical Considerations