SEATTLE – Patients with heart failure still benefit from spironolactone therapy even when they develop elevated potassium levels, a well-known and potentially life-threatening side effect of the medication, new data suggest.
Investigators led by Orly Vardeny, Pharm.D., conducted a secondary analysis of data from the RALES (Randomized Aldactone Evaluation Study) trial, in which more than 1,500 patients with heart failure were assigned to spironolactone (Aldactone) or placebo.
Orly Vardeny, Ph.D.
The spironolactone group had nearly quadruple the risk of hyperkalemia compared with the placebo group, she reported at the annual meeting of the Heart Failure Society of America. And hyperkalemia was indeed associated with a higher mortality rate.
But relative to their placebo-treated counterparts who did not have elevated potassium levels, spironolactone-treated patients who had elevated potassium levels, up to a serum potassium level of at least 5.5 mEq/L, were still less likely to die.
"The benefits of spironolactone are maintained in patients with mild to moderate hyperkalemia, and even in patients with potassium levels approaching 6.0 mEq/L," commented Dr. Vardeny, who is an associate professor of pharmacy at the University of Wisconsin in Madison.
"Our data argue for strategies that maintain spironolactone or aldosterone antagonist therapy even in the setting of mild to moderate hyperkalemia," she maintained.
"How many of these patients were on digoxin, and did you factor that into the analysis?" asked session comoderator Dr. Carl V. Leier of the Ohio State University Medical Center in Columbus.
Dr. Vardeny replied that analyses did not adjust for digoxin use, but they did adjust for age, estimated glomerular filtration rate, baseline potassium level, and diabetes.
"It seems that hypokalemia was more dangerous than hyperkalemia," commented attendee Dr. Edward Gilbert of the University of Utah in Salt Lake City. Might some of the benefit of spironolactone actually be related to reduced hypokalemia, he wondered.
"My personal bias is that there could be an association. This is really difficult to prove, only because potassium will vary so much, so the data will inform multiple categories of potassium," Dr. Vardeny replied. "But yes, we are investigating that as well."
The 1,663 patients in RALES had moderate to severe heart failure with a left ventricular ejection fraction of less than 35%, and were already receiving an angiotensin-converting enzyme inhibitor with or without a diuretic. They were randomized evenly to 25 mg of spironolactone daily, with optional titration up to 50 mg daily or placebo.
Serum potassium levels were monitored regularly. If patients developed hyperkalemia, investigators could reduce the dose of study medication but were encouraged to first adjust doses of other medications.
Main RALES results, previously reported, showed that spironolactone treatment was associated with a 30% reduction in the risk of all-cause mortality and a 35% reduction in the risk of hospitalization for worsening of heart failure (N. Engl. J. Med. 1999;341:709-17).
The new analysis showed that both treatment groups had mean potassium levels of 4.3 mEq/L at baseline. But levels increased within the first month in the spironolactone group and remained persistently higher than those in the placebo group, by 0.25 mEq/L on average, throughout the trial.
Relative to their peers in the placebo group, patients in the spironolactone group were significantly less likely to develop hypokalemia (6.5% vs. 16.2%) and more likely to develop hyperkalemia (14.5% vs. 4.2%), Dr. Vardeny reported. Hypokalemia was defined as a potassium level of less than 3.5 mEq/L; hyperkalemia equalled a potassium level of greater than 5.5 mEq/L.
In the spironolactone group, the rate of death was fairly stable – at about 17 per 100 patient-years – for patients with potassium levels ranging from 3.5 to 6.0 mEq/L. But it was higher among patients with lower levels (about 42 per 100 patient-years) or higher levels (about 25 per 100 patient-years).
In a multivariate analysis, patients were more likely to develop hypokalemia if they were black (hazard ratio, 2.0) or had a higher New York Heart Association functional class (HR, 1.5), whereas they were less likely if they had a higher baseline potassium level (HR, 0.25) or received spironolactone (HR, 0.36).
Patients were more likely to develop hyperkalemia if they received spironolactone (HR, 3.7), had a higher baseline potassium level (HR, 2.4), or had a history of diabetes (HR, 1.9).
Compared with placebo-treated patients having potassium levels of less than 5.0 mEq/L, spironolactone-treated patients with elevated potassium levels still had a significantly lower risk of death with potassium levels up to 5.5 mEq/L, and they had a nonsignificantly lower risk with levels up to about 6.0 mEq/L.