Aldosterone Antagonists

Sustained activation of aldosterone appears to play an important role in the pathophysiology of HF.62 Increased renin and angiotensin II levels contribute to the stimulation of aldosterone secretion. Elevated circulating levels of this hormone enhance sodium retention and potassium and magnesium loss. Aldosterone upsets autonomic balance by increasing sympathetic activation and parasympathetic inhibition and promotes cardiac and vascular structural remodeling through collagen synthesis.63-65

Although ACE inhibition may transiently decrease aldosterone secretion, there are diverse stimuli other than angiotensin II for the production of this hormone.66 Studies suggest a rapid return of aldosterone to levels similar to those before ACE inhibition.67 The potential pathophysiologic role of aldosterone and the results of a pilot study that suggested low doses of spironolactone were tolerated in HF, led to additional investigation of these agents in severe HF and subsequently in post-MI HF.68,69

The Randomized Aldactone Evaluation Study (RALES) was designed to determine the effect of low-dose spironolactone on survival in severely symptomatic (recent or current NYHA class IV) HF patients treated with an ACE inhibitor, loop diuretic, and, in many cases, digoxin.68 The study enrolled a total of 1663 patients with reduced LVEF (LVEF <=35%) resulting from ischemic and nonischemic etiologies. All-cause mortality was the prespecified primary endpoint. There were 386 (46%) deaths in the placebo group compared with 284 (35%) in the spironolactone group. The risks of sudden death or of death from progressive HF were both reduced. The frequency of hospitalization for HF was 35% lower in patients treated with spironolactone compared with placebo. Greater improvement was noted in NYHA functional class in those receiving spironolactone. Because deaths in class III patients were designated as a worsening in NYHA class, this functional improvement likely reflects the mortality benefit of the drug.

The inclusion and exclusion criteria for the RALES trial are important to consider when applying the study results to clinical practice. The yearly mortality rate in the placebo group was high, reflecting the advanced HF of study participants. The potential benefit of aldosterone antagonists in patients with milder HF and lower risk cannot be determined from RALES data. It should be noted that only 10% of placebo and 11% of spironolactone patients in the RALES trial were treated with beta blocker therapy. Patients with potassium levels >5.0 mmol/L were excluded, as were patients with abnormal renal function, defined as a creatinine >2.5 mg/dL. Patients recruited into the trial met the potassium inclusion criteria despite the frequent concomitant use of potassium supplementation at baseline (28%). Adhering to these patient characteristics may be necessary to avoid excessive hyperkalemia during spironolactone treatment. In clinical practice, a more conservative approach to serum creatinine may be warranted. The recommended serum creatinine cutoff of 2.5 mg/dL in this guideline is consistent with the eligibility criteria for the RALES trial. However, the majority of patients enrolled in RALES had a serum creatinine below this level. In addition, several groups including women, the elderly, or patients with low muscle mass may have a lower creatinine clearance for a given level of serum creatinine. For these patients, it may be reasonable to calculate an estimated creatinine clearance rather than relying solely on the serum creatinine value. Aldosterone antagonists are not recommended in patients with creatinine clearance <30 ml/min.

Spironolactone should be used in conjunction with standard therapy, including ACE inhibitors, digoxin, diuretics, and beta blockers. It should be initiated at a dose of 12.5 to 25 mg per day. Spironolactone can be titrated to 37.5 mg or 50 mg with careful monitoring in patients with refractory HF or persistent hypokalemia. Serum potassium and creatinine should be monitored closely in the first few weeks of therapy. If the serum potassium exceeds 5.0 mmol/L, then the dose of spironolactone should be decreased to 25 mg every other day and medications that could contribute to hyperkalemia should be adjusted. The risk of hyperkalemia with aldosterone antagonism is increased in patients with older age, diabetes, higher serum creatinine levels, and higher ACE inhibitor doses. In community settings the risk is far higher than documented during careful monitoring in trial settings, and may be as high as 20% .70 This risk should be taken into careful consideration when treating with an aldosterone antagonist, and remains present even after successful initiation of this therapy. Patients should continue to be monitored carefully and should be instructed not to take the aldosterone antagonist during any circumstances of volume loss such as gastroenteritis.

In addition to hyperkalemia, gynecomastia or breast pain may be important side effects of spironolactone, but not eplerenone. They were reported in 10% of the men randomized to spironolactone versus 1% of the males in the placebo group in the RALES trial. These side effects were more frequent in patients taking digoxin.

Clinical studies with the selective aldosterone antagonist, eplerenone, have demonstrated favorable results in patients with HF after acute MI. A multicenter, randomized, double-blind, placebo-controlled trial, Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS), tested the effect of eplerenone versus placebo in 6642 patients.69 Patients were enrolled after an acute MI if they had an LVEF <=40% and HF documented by signs and symptoms. HF signs and symptoms were not required if patients had diabetes. Exclusion criteria for the study included creatinine >2.5 mg/dL and serum potassium >5.0 mmol/L. Patients were generally receiving agents shown to be effective in reducing risk in patients after acute MI, including beta blockers, ACE inhibitors, aspirin and cholesterol-lowering agents. The hypothesis was that eplerenone would reduce overall mortality and cardiovascular mortality or hospitalization.

The results, after an average follow-up 16 months, revealed a statistically significant reduction in cardiovascular mortality or hospitalization and all-cause mortality and hospitalization in the group receiving eplerenone.69 The reduction in all-cause mortality was observed as early as 30 days after randomization.71 There was also a significant reduction in sudden cardiac death favoring eplerenone treatment.

Adverse reactions to eplerenone were uncommon. As with spironolactone, serious hyperkalemia was more prevalent with eplerenone treatment. It should be noted that baseline serum potassium concentration in both the eplerenone and placebo groups was 4.3 mmol/L. As outlined in the recommendation for use, it is important to monitor electrolytes, especially potassium. The major predictors of hyperkalemia in EPHESUS were estimated glomerular filtration rate (eGFR) <60 ml/min, baseline serum potassium above the median (4.3 mEq/L), diabetes mellitus, and prior use of antiarrhythmic drugs. The effect of eplerenone on all-cause mortality was not affected by baseline serum potassium or the change in serum potassium from baseline.72 Post-hoc analyses suggested that patients who were not on ACE inhibitors or ARBs and beta blockers had less benefit from the addition of eplerenone than those on these neurohormonal antagonists.69 A recent systematic review of post-MI and HF studies in subjects with reduced LVEF using the aldosterone antagonists spironolactone, eplerenone, and canrenoate confirmed benefits in all-cause mortality, hospitalizations, and LVEF.73

Remodeling Post MI. Another study randomized 134 patients post-anterior MI after revascularization to spironolactone versus placebo.74 All patients were on ACE inhibitors. After 1 month, LVEF was improved, end-diastolic dimension was reduced, and markers of collagen synthesis were reduced in the spironolactone group, indicating an improvement in LV remodeling after MI. One of the limitations of this study was that only 31% of patients were on beta blockers. A substudy of EPHESUS demonstrated lower levels of collagen biomarkers among patients randomized to eplerenone, suggesting that it suppresses post-MI remodeling.75

Aldosterone Antagonists in Mild to Moderate HF. Patients enrolled in RALES had chronic severe HF (NYHA IV at enrollment or in the past). EPHESUS studied patients who were post-MI. Aldosterone antagonists have not been proven effective in patients with mild to moderate HF in the absence of recent MI or in patients with HF and preserved LV systolic function.

Selective Versus Nonselective Aldosterone Antagonists. The efficacy of selective and nonselective aldosterone antagonists is generally considered to be equivalent. The potential advantage of a selective aldosterone blocker that blocks only the mineralocorticoid receptor is a reduction in side effects. A nonselective blocker, such as spironolactone, blocks the mineralocorticoid, glucocorticoid, androgen, and progesterone receptors, resulting in potential gynecomastia and sexual dysfunction. The incidence of gynecomastia with eplerenone in EPHESUS was 0.5%, whereas it was 10% with spironolactone in RALES.68,69

Hyperkalemia. Hyperkalemia is a life-threatening complication of aldosterone antagonists and is much more likely to occur in patients with diabetes or renal insufficiency or in those taking ACE inhibitors or ARBs. When more than one of these risk factors is present, the likelihood of hyperkalemia increases. In RALES and EPHESUS, aldosterone antagonists were not initiated if the creatinine was >2.5 mg/dL or serum potassium was >5.0 mmol/L. In RALES, the potassium was monitored every 4 weeks for 12 weeks, every 3 months up to a year, and every 6 months after the first year. In the EPHESUS trial, in which patients were taking a larger number of concomitant medications, potassium was measured at 48 hours, at 4-5 weeks, and then every 3 months. Potassium was measured 1 week after a dose increase of an aldosterone antagonist. Although patients with creatinine <2.5 mg/dL were enrolled in the clinical trials, very few patients actually had a creatinine >1.7 mg/dL. Thus additional monitoring should be considered in these patients.

Few patients will tolerate an aldosterone antagonist in the absence of concomitant therapy with a potassium-wasting diuretic. Potassium supplements and potassium-containing salt supplements should be reduced or, if possible, discontinued. Serum potassium monitoring should be at least as rigorous as in RALES and EPHESUS and more rigorous in patients with multiple risk factors. Nonsteroidal antiinflammatory agents, including cyclooxygenase-2 inhibitors, should be avoided because they may worsen renal insufficiency, increasing the risk of hyperkalemia.