The Expiration Date for the Beta-Blockers is REDUCE-AMI

Precise News

I modified the verb reduced with still because the trials that established the benefit of beta-blockers post-MI were done before acute revascularization was standard care.
Yet, it wasn’t just theory that stirred doubt about beta-blockers post-MI.
A meta-analysis of 60 trials of post-MI beta-blockers reported that the later trials (vs the early trials) no longer found any benefit from beta-blockers after MI.
The REDUCE-AMI Trial REDUCE-AMI has special features.
Consider a typical randomized trial of post-MI beta-blockers.
But understanding why this happened actually increases my confidence in the lack of benefit of post-MI beta-blockers.
We should remember that their data apply only to patients with preserved LV function — probably the majority of modern-day post-MI patients.
When ongoing trials confirm the results of REDUCE-AMI, the care of post-MI will have been greatly advanced.


Although the primary endpoint of the REDUCE-AMI trial, which examined the effects of beta-blockers after myocardial infarction (MI), was not statistically significant, it is likely to rank among the most significant trials conducted in cardiology today.

After reviewing the findings, let’s discuss three key lessons.

Researchers from Lund University/Skane University Hospital in Sweden, led by Troels Yndigegn, MD, posed a straightforward question: Do beta-blockers still lower mortality and recurrent MI in patients who recently experienced an acute MI but have preserved left ventricular (LV) ejection fraction?

The reason I changed the verb reduced with still was because the studies that proved beta-blockers were beneficial after MI weren’t conducted when acute revascularization was the norm. Modern MI treatment differs greatly from that of the 1980s. Adrenergic blockade may not theoretically provide much benefit when an urgent percutaneous coronary intervention (PCI) opens an artery and reduces myocardial injury.

Still, doubts regarding beta-blockers after MI were not limited to theory. Furthermore, there is proof. The later trials (as opposed to the early trials) no longer found any benefit from beta-blockers after MI, according to a meta-analysis of 60 trials of these medications.

The Trial of REDUCE-AMI.

REDUCE-AMI has unique characteristics. The trial was pragmatic, open-label, and registry-based. The majority of the patients were drawn at random from Sweden. The results could be counted thanks to the Swedish registry. In this trial, patients were randomized to receive either no beta-blocker or a beta-blocker (bisoprolol or metoprolol) at random. There was also no placebo arm.

Seventy-nine percent of the beta-blocker arm and eighty-three percent of the control arm experienced the primary endpoint of either death or MI after three and a half years of follow-up (HR, 0.96; 95 percent CI, 0.79-1.16; P = .64). Secondary endpoints that did not vary between the two groups included heart failure, hospital admission for atrial fibrillation, and cardiovascular death. Safety indicators like hypotension, bradycardia, or lung disease hospitalization didn’t either.

Most evidence ought to have a date of expiration.

The pharmacologic effect of a beta-blocker tablet is the same today as it was in the 1980s. However, almost all aspects of MI care have evolved. Over 95% of patients in REDUCE-AMI received treatment with PCI, dual antiplatelet therapy, and a statin. Approximately 80% of patients were prescribed an angiotensin-converting enzyme inhibitor or an angiotensin-receptor blocker. In the 1980s, not many of these therapies were available.

The prevention of arrhythmic death and complications related to wall stress was likely the primary reason for the advantages of adrenergic blockade prior to revascularization. It follows that beta-blocker therapy might not add much, since current therapy has significantly decreased these complications.

Standard vs. Pragmatic Trials.

Its culture of randomization is one of the many remarkable features of Scandinavian medicine. Medical care is routinely infused with trials. Researchers are able to track clinical endpoints thanks to the large registries where each citizen is assigned a unique identification number.

Because the trial is similar to routine care, the benefit of such pragmatic trials is that applying the results to routine care is made easier. The effect estimates could be less reliable, which is a drawback.

Take a look at a standard randomized trial involving beta-blockers after a MI. This would entail multiple follow-up appointments linked to the trial as well as a placebo medication to preserve blinding. An experiment of this kind would require more money and manpower, but it might be less biased.

Given its pragmatic nature, the REDUCE-AMI results need to be interpreted with caution. One of the main reasons for the result’s confidence is the low likelihood of an open-label design influencing or miscounting hard endpoints like death or CV death in the registries.

Proof of Non-existence versus Non-existence of Proof.

In both arms, the annual rate of primary outcome events was almost the same (2.4% vs. 2.5% for the control). Additionally, the .64 P value suggests that these results would not be surprising if there was no difference between the two groups.

Statists would contend that this does not allow us to draw the conclusion that there is no evidence of a beta-blocker effect because the confidence intervals range from 0.79 (21 percent benefit) to 1.16 (16 percent harm). They contend that REDUCE-AMI lacks sufficient power.

However, realizing why this occurred strengthens my belief that post-MI beta-blockers are not beneficial. Because the researchers optimistically targeted a 25 percent benefit and overestimated the event rates, REDUCE-AMI had wide confidence intervals.

It was less than 3 percent, compared to the investigators’ expectations of a 7% annual rate of events in the control group. To anticipate a 25% decrease in the primary endpoint with beta blockers was also unduly optimistic. If they had powered the trial for a benefit of less than 25% and/or anticipated a lower event rate, they would have needed a lot more participants. Larger trials naturally present challenges related to cost, viability, and the possibility of over-sampling patients. Setting a strict benchmark for trialists to meet is enrolling the ideal number of patients.

This is a weak purist argument, in my opinion. The lack of a significant effect makes sense when you consider the pessimistic pretrial knowledge (no recent positive beta-blocker trial) in conjunction with the low actual event rates. In addition, the secondary outcomes showed no signals at all. When a patient has atypical chest pain and no risk factors, the REDUCE-AMI results are comparable to a negative stress test.

The good news is that when LV function is preserved after MI, at least three trials are still being conducted to determine whether beta-blockade is helpful. The outcomes of DANBLOCK, BETAMI, and REBOOT will most likely increase the confidence in the REDUCE-AMI results.

Takeaways for Reduced-AMI.

The multidecade dogma surrounding the use of beta-blockers following MI is starting to be disproved by the REDUCE-AMI investigators. It’s important to keep in mind that their findings only apply to patients who have preserved left ventricular function, which is most likely the case for most post-MI patients today.

Specific clinical and general evidentiary lessons can be learned from this trial. Clinically, all of our efforts are directed toward preventing further events once an artery is opened and an acute MI is terminated. Removing a potentially harmful drug free medication frees up resources for more research into evidence-based treatments. The post-MI care will have advanced significantly when ongoing trials validate the findings of REDUCE-AMI.

To me, this trial highlights a new development in contemporary cardiology: the tremendous benefit of reexamining established beliefs. The question in contemporary cardiology going forward will not be so much whether a therapy is available as it will be whether we should use it.

Additionally, the researchers have demonstrated the benefits of incorporating randomization into standard medical care. While there is always room for improvement, I contend that we learn more the more we randomize. Scandinavia is a great example for evidence-based cardiology to emulate.

John Mandrola is a podcaster and writer for Medscape in addition to practicing cardiac electrophysiology in Louisville, Kentucky. In terms of medical practice, he supports conservatism. He frequently writes about the state of medical evidence and takes part in clinical research.

scroll to top