Methods

Results

Baseline characteristics were similar between treatment groups (Table 3). Significantly more eplerenone-treated patients had serum K⁺ >5.5 mEq/L (15.6% versus 11.2%, P<0.001) and serum K⁺ ≥6.0 mEq/L (5.4% versus 3.8%, P=0.002) than the placebo group (Figure 1); however, hypokalemia (<3.5 mEq/L) was significantly less frequent in the eplerenone group (8.4% versus 13.1%, P<0.001; Figure 1). Withdrawal from the study because of hyperkalemia was <1% in both treatment groups (placebo 0.3%, eplerenone 0.7%; P≤0.05).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Frequency of serum potassium >5.5, ≥6.0, or <3.5 mEq/L in the eplerenone and placebo groups in EPHESUS. *P<0.001; †P=0.002.

Time to Hyperkalemia

The mean time to reach maximum serum K⁺ values after randomization was similar in both the eplerenone- and placebo-treated patients (mean [SE] 168 [3.4] days for eplerenone versus 159 [3.4] days for placebo; P=0.058). The maximum values occurred within 90 days of study in 59% of eplerenone-treated patients and in 62% of placebo-treated patients. Figure 2 illustrates time to maximum K⁺ values for all EPHESUS patients by treatment group. Figure 3 shows the time to reach the first occurrence of serum K⁺ ≥6.0 mEq/L in eplerenone- and placebo-treated patients for any patients who experienced this level of serum K⁺ at any time during the study.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Time in days to maximum potassium values for the eplerenone (A) and placebo (B) groups.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. For patients who experienced serum potassium ≥6.0 mEq/L at any time during the study, time in days to the occurrence of serum potassium ≥6.0 mEq/L for the eplerenone (A) and placebo (B) groups.

Discussion

These post hoc analyses demonstrate that the selective aldosterone blocker eplerenone, when administered at a dose of 25 to 50 mg/d, is associated with a 4.4% absolute increase in the incidence of hyperkalemia (>5.5 mEq/L) and a 1.6% absolute increase in the incidence of more marked hyperkalemia (≥6.0 mEq/L). The major predictors of hyperkalemia (≥6.0 mEq/L) in the present study population, in which >80% of patients were administered either an ACE inhibitor or an angiotensin receptor blocker and 75% were administered a β-blocker, were an eGFR <60 · min⁻¹ · 1.73 m⁻², a baseline serum K⁺ above the median (4.3 mEq/L), the presence of diabetes mellitus, and prior use of antiarrhythmic drugs. Antiarrhythmic drug use was an unexpected risk factor. Although the use of an ACE inhibitor or an angiotensin receptor blocker in combination with an aldosterone blocker has previously been shown to increase the risk of hyperkalemia in similar patients,^10,11 in the present analysis, the use of these agents in addition to eplerenone did not contribute to the risk independent of elevated baseline serum K⁺, low eGFR, or the presence of diabetes mellitus. Similarly, the use of a diuretic did not independently affect the risk for hyperkalemia (≥6.0 mEq/L) or the effectiveness of eplerenone in reducing either of the primary end points, all-cause mortality and the composite of cardiovascular mortality/cardiovascular hospitalization. Approximately one half of the episodes of hyperkalemia (≥6.0 mEq/L) occurred within 30 days of initiation of therapy, although sporadic episodes of hyperkalemia were observed throughout the 2-year follow-up period (Figure 3), which likely reflects changes in renal function or the use of pharmacological agents that affect K⁺ excretion, such as nonsteroidal antiinflammatory drugs. Overall, less than 1% of patients who were randomized to eplerenone had to discontinue therapy because of hyperkalemia. No relationship was found between either baseline serum K⁺ or the change in serum K⁺ from baseline and the benefit of eplerenone on all-cause mortality overall or in any subgroup, including those at risk for developing hyperkalemia.

Similar to the results from the Randomized Aldactone Evaluation Study (RALES) of spironolactone in patients with chronic severe HF due to left ventricular systolic dysfunction,¹² no death was attributable to hyperkalemia in patients randomized to eplerenone in EPHESUS. These experiences in clinical trials contrast with several reports of serious hyperkalemia in patients given spironolactone for chronic HF in clinical practice, some of which were associated with increased hospital admissions and death.^4–6,13,14 The occurrence of hyperkalemia in these trials may have been related to the higher doses used than what had been studied previously in the reference trials, EPHESUS and RALES. In clinical practice, patients are often older than in randomized trials such as EPHESUS and RALES and may have a reduced eGFR that is not reflected by serum creatinine. (Serum creatinine is an inaccurate metric of true of kidney function in the elderly.) Such patients may be given an aldosterone blocker regardless of baseline serum K⁺, serum creatinine, or estimated renal function. For example, in 1 report on the use of spironolactone in patients with chronic HF, a 15% incidence of hyperkalemia was found.⁶ However, in one third of these patients, no single measurement of serum K⁺ or creatinine occurred within the first 3 months of initiation of spironolactone. The lower incidence and risk of hyperkalemia associated with aldosterone blockade with eplerenone in EPHESUS and with spironolactone in RALES compared with that observed in clinical practice can be attributed to the fact that patients were excluded from randomization to the aldosterone blocker if they had a baseline serum K⁺ >5.0 mEq/L, a serum creatinine >2.5 mg/dL, or both in these clinical trials. Furthermore, in both EPHESUS and RALES, serum K⁺ was measured within 1 week of starting the aldosterone blocker and again at 4 weeks, as well as every 3 to 6 months thereafter. In addition to periodic K⁺ monitoring, it is recommended that whenever a change in electrolyte status is suspected, such as during an episode of vomiting or diarrhea or at initiation or dosage adjustment of concomitant diuretic therapy, serum K⁺ should be remeasured or the aldosterone blocker withheld until measurement of serum K⁺ is possible. The dose of aldosterone blocker should be adjusted according to serum K⁺ measurements as follows: The dose of the aldosterone blocker should be halved if serum K⁺ is >5.5 mEq/L, and if serum K⁺ is ≥6.0 mEq/L in a nonhemolyzed sample, the aldosterone blocker should be discontinued until serum K⁺ falls below 5.5 mEq/L (Table 1).^1,12

Patients with risk factors for hyperkalemia, such as those with renal dysfunction or diabetes mellitus, may benefit by more frequent monitoring of serum K⁺. A recent study has suggested that the risk of hyperkalemia associated with aldosterone blockade in patients with HF and renal dysfunction, defined as creatinine clearance <60 mL/min, can be managed safely by the use of a low-K⁺ diet, adjustment of the dose of aldosterone blocker to renal function, and careful monitoring of serum K⁺.¹⁵ In the present small study, no deaths or hospitalizations related to hyperkalemia occurred. Although in EPHESUS and RALES, exclusion criteria were based on serum creatinine, serum creatinine may not adequately reflect renal function in the elderly, and one should determine an eGFR for these patients.

A recent subanalysis from EPHESUS¹⁶ showed that although the benefit of eplerenone on all-cause mortality decreased with worsening levels of kidney function, no evidence was found of an increased risk of all-cause mortality even in patients with stage 4 nephropathy (ie, creatinine clearance ≤30 mL/min). Moreover, in the present analysis, no significant interaction of eGFR with the ability of eplerenone to reduce all-cause mortality was present, with a hazard ratio of 0.52 for eplerenone compared with placebo in patients with eGFR <60 · min⁻¹ · 1.73 m⁻² (P for interaction=0.13). This suggests that eplerenone can be used safely in patients with this degree of renal dysfunction provided they are monitored appropriately, as was done in the present trial.

Although we recommend discontinuing an aldosterone blocker in patients with serum K⁺ ≥6.0 mEq/L, recent studies have pointed out that serum K⁺ may not accurately reflect tissue K⁺ in patients with HF. For example, patients with HF have been found to have a serum K⁺ >7.0 mEq/L without any ECG changes or clinical manifestations of hyperkalemia because their tissue K⁺ was normal or low, as reflected by the red blood cell K⁺ concentration.¹⁷ In patients with HF, an increase in oxidative stress may result in a defect in Na/K ATPase and transport of K⁺ into tissues.

In EPHESUS, a significant absolute decrease of 4.7% was found in the incidence of hypokalemia (serum K⁺ <3.5 mEq/L), and the incidence of hypokalemia was greater than that of hyperkalemia (serum K⁺ ≥6.0 mEq/L; Figure 1). Although many clinicians become concerned about the risk of hypokalemia only when serum K⁺ <3.5 mEq/L, a recent analysis of the National Heart, Lung, and Blood Institute’s Digitalis Investigation Group trial showed an increased risk of both sudden cardiac death and progressive HF in patients with a serum K⁺ <4.5 mEq/L regardless of the use of K⁺ supplements.¹⁸ This relationship of decreased serum K⁺ (<4.4 mEq/L) as an independent predictor of sudden cardiac death in patients with HF has been demonstrated previously in other studies,¹⁹ and thus, it has been suggested that serum K⁺ should be maintained between 4.5 and 5.5 mEq/L in patients with HF.²⁰ As mentioned previously, patients with HF may have difficulty in transporting K⁺ into the tissues, and therefore, tissue levels of K⁺ may be considerably less than serum levels, which would place them at increased risk for myocardial fibrosis, hypertrophy, and sudden cardiac death. However, although eplerenone reduced the incidence of hypokalemia (<3.5 mEq/L), no significant relationship was found between the change in serum K⁺ from baseline and the effectiveness of eplerenone in reducing total mortality in the present study.

In conclusion, these results demonstrate a beneficial effect of eplerenone on all-cause mortality regardless of baseline risk factors for the development of hyperkalemia (≥6.0 mEq/L). However, to obtain this beneficial effect on mortality, it is necessary to ensure the safe use of the agent, which involves measuring both serum K⁺ and either eGFR or estimated creatinine clearance before therapy is instituted; excluding patients with serum K⁺ greater than 5.0 mEq/L, serum creatinine >2.5 mg/dL, and/or an estimated creatinine clearance ≤30 mL/min or below; periodically monitoring serum K⁺; and adjusting the dose of eplerenone as necessary, according to serum K⁺ and changes in renal function. This view is reflected in major therapeutic guidelines in the United States² and Europe,³ which, based on the results from EPHESUS, recommend that all eligible post-AMI patients with concomitant HF be treated with an aldosterone blocker in addition to an ACE inhibitor (or an angiotensin receptor blocker) and a β-blocker.