REMINDER: Upbeat Results for Early Post-MI Eplerenone, But Relevance Questioned

March 15, 2013

SAN FRANCISCO — At face value, the randomized REMINDER trial[1] suggests that starting an aldosterone blocker like eplerenone (Inspra, Pfizer) within the first day after an acute MI in patients without heart failure or LV dysfunction can improve "clinical outcomes." That conclusion fits well with several previous large trials that saw survival gains from aldosterone blockade in patients with severe heart failure, mild heart failure, and post-MI left-ventricular systolic dysfunction.

On closer inspection, the significant change in the trial's complex primary end point that accounted for nearly all of the benefit from eplerenone was not clinical at all. It was in levels of natriuretic peptides, a biomarker that may herald future LV dysfunction and heart failure but can go up in the absence of an overt clinical condition.

Dr Gilles Montalescot

"All the other components were not significant. There were, numerically speaking, trends in favor of eplerenone for heart-failure rehospitalization and [ventricular] arrhythmias, but nothing significant," according to Dr Gilles Montalescot (Pitié-Salpêtrière University Hospital, Paris, France) when presenting REMINDER earlier this week at the American College of Cardiology 2013 Scientific Sessions . Montalescot headed the trial's executive steering committee.

"This is also the first large study to demonstrate the good safety profile of eplerenone when it is administered early in patients presenting with heart attack," he said. Hyperkalemia was no more common in either the eplerenone or placebo groups, and the control patients had more hypokalemia (p=0.0002).

Dr Magnus Ohman

Commenting as a panelist after Montalescot's presentation, Dr Magnus Ohman (Duke University, Durham, NC) observed that since REMINDER looked at a low-risk population without heart failure or even poor LV function, it is something of a safety study for eplerenone instead of an efficacy study. That could be helpful, because aldosterone blockade, although recommended for heart failure, he noted, "represents the lowest-followed guideline in our guidelines. It is hardly ever used in practice." That's in part due to concerns about causing hyperkalemia.

So what the trial demonstrates, according to Ohman, "is that [eplerenone] is safe when you exclude patients with issues with renal failure."

The trial's hitch, Ohman observed, is the questionable clinical significance of a natriuretic-peptide elevation after 30 days. "What impact does that really have, given that it's more of a surrogate marker? Many patients have [such] elevations without overt heart failure." He speculated whether it might be a predictor of actual clinical outcomes further down the line.

Dr Miguel Quiñones

That was also the interpretation of Dr Miguel Quiñones (Methodist Hospital, Houston, TX), not part of REMINDER, in his presentation on the trial to the media. The trial was based on a solid hypothesis, he said, but it's hard to show clinical improvement in a low-risk population that is already well managed. Still, it showed a positive effect of eplerenone on a marker of subclinical heart failure. Had patients with natriuretic-peptide reductions been followed over the long term, the benefit potentially could "manifest as a lower incidence of heart failure in the years to come."

The trial, he said, is "really a potential game changer if we could demonstrate three to five years later a significant reduction in clinical outcomes."

REMINDER randomized 1012 patients with acute ST-segment-elevation MI (STEMI), without a history of heart failure or signs of current heart failure, an LVEF <40%, or renal insufficiency, to receive, "preferably before myocardial reperfusion," eplerenone at 25 to 50 mg/day or placebo on top of standard therapy. Treatment started "within the first 24 hours of symptom onset and preferably within the first 12 hours," Montalescot reported.

The diverse composite primary end point included CV mortality, rehospitalization, prolonged index hospitalization due to an HF diagnosis, sustained ventricular tachycardia or ventricular fibrillation, LVEF descending to <40% after one month, or levels of natriuretic peptides exceeding a predefined threshold after one month — all by intention to treat.

As a composite, the primary end point fell from 29.6% with placebo to 18.4% with eplerenone (p<0.0001) over a mean 10.5-month follow-up. But the biomarker component of the end point, the only to show a significant difference, accounted for most of the overall benefit. Natriuretic peptides elevated after a month in 25.9% of control patients and 16% of those on eplerenone (p<0.0002).

Hazard Ratios (95% CI) for the Primary End Pointa and Its Components in REMINDER

End point HR (95% CI)
Composite primary end point 0.57 (0.44–0.74)b
Primary end point components  
CV mortality 0.52 (0.05–5.99)
Rehospitalization/hospitalization extended due to HF 0.56 (0.20–1.54)
Sustained VT or VF [too few to estimate]
EF <40% after >1 mo 1.08 (0.58–2.00)
Natriuretic-peptide elevation >1 moa 0.58 (0.44–0.77)c

a. Brain-type natriuretic peptide (BNP) elevated to >200 pg/mL or N-terminal pro-BNP (NT-proBNP) elevated to >450 pg/mL for age <50, >900 pg/mL for age 50–75, and >1800 pg/mL for age >75

b. p<0.0001

c. p=0.0002

There were no significant differences in blood-pressure changes or renal laboratory markers, nor in adverse events leading to drug discontinuation, death, or gynecomastia, according to Montalescot.

Importantly, he observed, the two groups didn't differ significantly in prevalence of hyperkalemia, and hypokalemia was more common in the control group.

Indicators of Hyperkalemia and Hypokalemia in REMINDER

Potassium level parameter Eplerenone, n=506 Control, n=506 p
Change, baseline to 1 mo (mmol/L) 0.41 0.32 <0.0001
>6.0 mmol/L (%) 1.6 0.4 0.11
>5.5 mmol/L (%) 5.6 3.2 0.09
<4.0 mmol/L (%) 35.5 47.2 0.0002

Dr George Dangas

Speaking from the panel after Montalescot's presentation, Dr George Dangas (Mount Sinai School of Medicine, New York, NY) said, by his own informal calculations, the trial's primary-end point rate sans the biomarker was only 2% or 3%, "which means that 97% of the patients were essentially free of clinical events."

He said he wonders "how the healthcare system is going to go ahead with 97% of patients ultimately not having any event [but] being declared candidates for this therapy," when eplerenone is "rather expensive."

Dangas speculated that maybe aldosterone or natriuretic-peptide levels could be screened in such patients while they are still in the hospital so that the drug could be "appropriately targeted." That, or give the less expensive aldosterone blocker spironolactone to "all comers."

Montalescot had no disclosures.