What Now for Beta-Blockers Post-MI? Reconciling REDUCE-AMI and ABYSS

This transcript has been edited for clarity. 
Michelle L. O’Donoghue, MD: Hi. I’m Michelle O’Donoghue reporting for Medscape. I’m at the European Society of Cardiology (ESC) Congress in London, and I’m joined by Dr Gilles Montalescot. He’s a professor of cardiology at Sorbonne University. One of the more interesting, and perhaps surprising to some, results that were presented at this congress were the results of the ABYSS trial. This was a trial of beta-blocker discontinuation. 
Before we get into the results of ABYSS, let’s take a step back and set the stage for the beta-blocker research, because for many years, we hadn’t seen much in this space. What do we know about the benefit of beta-blockers after myocardial infarction (MI) or for patients with stable coronary disease?
Gilles Montalescot, MD, PhD: It’s a very good question. Clearly, the studies that we have with beta-blockers are old studies — old randomized studies done even before the era of mechanical reperfusion. What we have seen more recently is big registries suggesting that in this kind of patient, very much a stable patient, you may question the role of beta-blockers for secondary prevention.
It’s very tempting for the physicians in everyday practice to say, “You’re doing, well, nothing has happened in 1-3 years since your MI, we can probably remove the beta-blocker and have a better quality of life.” But we needed a randomized study to be sure that safety was also okay.
O’Donoghue: You’re right. We had a lot of observational studies that supported the idea that stopping beta-blockers, especially years after having an MI, may be a safe strategy. And a lot of patients are happy to be rid of some of their medications. Now in 2024, it’s been a big year for beta-blocker research. Earlier this year, what did we learn from the REDUCE-AMI trial?
Montalescot: REDUCE-AMI looked at the introduction of beta-blockers in patients hospitalized for an acute MI when there is no left ventricular (LV) dysfunction, no heart failure — let’s say a simple MI, the patient is doing well. Their hypothesis was that beta-blockers might have no benefit, and this is what they found. 
Our study is very different. We are looking at patients who had an MI 3 years earlier (on average) and who were still on beta-blockers. We randomized 3700 patients to either beta-blocker removal or continuation. We wanted to show that there was no harm and that we may also improve the quality of life, which is a big issue for the patients. 
O’Donoghue: Ultimately, what did the results of ABYSS show? 
Montalescot: The exact opposite of our initial hypothesis. We were unable to show noninferiority of interruption of beta-blocker vs continuation. We had an excess of events, numerically speaking, when we stopped the beta-blocker. That was a surprise. That was a safety signal. 
Of course, the study was open label. We had all sorts of beta-blockers with different doses, so as you can imagine, it was very difficult to have a blinded trial. That’s a limitation of the study, but at the end of the day, when you look at death, MI, stroke, and rehospitalization for any cardiovascular reason, the curves tend to diverge after 3 years. We see an excess of hospitalizations, for coronary reasons mostly, when we stopped beta-blockers.
O’Donoghue: You make an important point there, that it was a broad composite outcome. For death, MI, and stroke, there was not much of a signal toward any increase in those outcomes with beta-blocker discontinuation. But there was a profound effect on hospitalization for cardiovascular reasons. So do we think that it’s anginal symptoms driving those hospitalizations?
Montalescot: That was clearly angina, and it led to more coronary angiograms, more percutaneous coronary interventions. We also observed poor control of blood pressure and heart rate. It was a 4–mm Hg difference in diastolic blood pressure and a 10-beats/min difference for heart rate with beta-blocker discontinuation. We know that blood pressure and heart rate are prognostic markers in these patients, I’m not saying that it clearly explains what happened. The study stopped at 3, 4, or 5 years of follow-up for some patients, but the curves diverge. What would we have seen if we had 6 or 7 years of follow-up? I don’t know, but we probably would see a significant difference. 
O’Donoghue: Were those differences in blood pressure and heart rate sustained over time? Because you’re talking about how those curves continue to diverge. And do you think that that was driven by that persistent difference in blood pressure?
Montalescot: It’s very interesting to look at, because the rise in blood pressure and heart rate happened very quickly after the interruption of beta-blockers. And of course, investigators had the option to adjust medications, and they did so more frequently in the interruption group than the continuation group. Despite the increasing doses of antihypertensive drugs, we still saw a persistent difference in blood pressure and heart rate.
O’Donoghue: That’s interesting, right? Because I think in many ways, that actually reflects real-world practice. Do you think there’s any magical property of the beta-blocker itself? If the blood pressure had been the same between the two groups, do you think that ultimately it would have been a neutral or noninferior trial, or is there an argument to be made that there is something about beta-blockers inherently that would be different from, say, calcium-channel blockers? 
Montalescot: I don’t know the answer to your question. What we can say is that there was this effect on heart rate and blood pressure and absolutely no effect on quality of life when we expected to show a benefit for the interruption. Knowing this, it’s very difficult to say you can stop beta-blockers in stable patients 3 years after MI and nothing is going to happen.
O’Donoghue: The natural question is how to reconcile the findings of REDUCE-AMI with those of ABYSS. I think that there are ways in which the two studies are consistent. Because REDUCE-AMI is about beta-blockers early after MI, it showed no benefit on death, MI or stroke, which is similar to what you saw as well. You didn’t see that those particular outcomes were modified.
But do you think ABYSS makes an argument for the idea that there is benefit for beta-blockers early post-MI, despite the neutral findings of REDUCE-AMI? If they had perhaps as broad an outcome as you did, that there might have been modification of other types of anginal, symptoms, or early revascularization event. 
Or do the findings of ABYSS argue that you had you never started beta-blockers in the first place, you might have been better at blood pressure control rather than a discontinuation strategy. What’s the takeaway?
Montalescot: It’s all speculative, but we have to keep in mind that REDUCE-AMI is a randomized study that looked at an administrative database. So they had death, MI, and stroke, but they didn’t look at the other endpoints, such as rehospitalization for cardiovascular reasons — which we did, with adjudication of all the events. 
If we compare the two studies, it’s correct to say that the two populations are very similar. If you apply the message of REDUCE-AMI and say there is no need for beta-blockers in these patients, there is no more a question about removal of the beta-blockers 3 years later. 
The removal of beta-blockers from patients who got them early on (we have plenty of these patients in our outpatient clinics) may be different. We cannot fully reconcile these two studies.
O’Donoghue: That’s the real question: Is it an issue of discontinuation, or was there no benefit of initiation in the first place? That’s, of course, the clinical question. So maybe that should be our final question for you. What are you going to do in your practice if you have a post-MI patient? Are you routinely starting beta-blockers?
Montalescot: Even in a simple MI patient going to the critical care unit after laboratory tests, I would find it difficult not to give beta-blockers on the basis of REDUCE-AMI only. We still continue to give beta-blockers to these patients. Now, 2 or 3 years later, we have to decide what to do. ABYSS tells us that if the patient tolerates the drug and is not asking for beta-blocker removal, we should probably keep the drug on board for the control of blood pressure heart rate, angina, but the hard endpoints are not impacted. 
O’Donoghue: Thank you for discussing this complicated topic, and I’m glad that there are still ongoing studies in this space that will hopefully further inform upon what we should do as clinicians. Thank you again for joining me today. 
Montalescot: Thank you for having me. 
O’Donoghue: Signing off for Medscape, this is Dr Michelle O’Donoghue.
Michelle O’Donoghue is a cardiologist at Brigham and Women’s Hospital and senior investigator with the TIMI Study Group. A strong believer in evidence-based medicine, she relishes discussions about the published literature. A native Canadian, Michelle loves spending time outdoors with her family but admits with shame that she’s never strapped on hockey skates.