Today is another look at some new trials presented at the recent European Society of Cardiology Meeting. This post is another look at one of the most prescribed classes of drugs in all of cardiology - beta blockers - and a dive into how the evidence base on them is changing in the modern era.
A year ago, I wrote about how all research should have an expiration date. The premise was simple: medical evidence ages, and what made sense in 1985 might not make sense today. The REDUCE-AMI trial had just challenged four decades of clinical practice around beta blockers after heart attacks.
Now we have even more data, with two massive trials - REBOOT-CNIC and BETAMI-DANBLOCK - recently presented at ESC and published in the New England Journal of Medicine.
Together with REDUCE-AMI, they're telling us something important about how medicine should evolve and reinforcing a paradigm change in the practice of post heart attack care.
The bottom line up front: Beta blockers shouldn't be standard practice for every patient after a heart attack.1
What the New Trials Found
Both REBOOT CNIC and BETAMI-DANBLOCK were pretty similar to REDUCE-AMI.
They randomized patients after a heart attack with a normal ejection fraction2 and without evidence of heart failure to two groups: beta blockers and no beta blockers.3
REBOOT-CNIC enrolled about 8,500 patients across Spain and Italy with heart attacks and ejection fractions above 40%. After nearly four years of follow-up, beta blockers showed no benefit for death, recurrent heart attack, or heart failure admission. In fact, there was a non-significant signal for increased stroke risk (which we’ll come back to later).
BETAMI-DANBLOCK told a slightly different story. This combined Danish-Norwegian trial of about 5,500 patients actually showed a small benefit from beta blockers in a composite endpoint.
That difference was driven by a reduction non-fatal heart attacks. There was no difference in death or heart failure.
And if you dig into the subgroup analyses4 from both of these trials, there seems to be a suggestion that sicker patients (those with an ejection fraction in the 40-50% range rather than over 50%) were the ones who had a benefit from beta blocker therapy.
Together with REDUCE-AMI's findings, a clear picture emerges: routine beta blocker use after a heart attack isn’t a life saving intervention if your ejection fraction is normal.
There are a few lessons that I think we should take from these data.
Lesson 1: The Individualized Medicine Lesson
This isn't about abandoning beta blockers entirely. It's about precision.
Remember the subgroup analyses showing a trend toward benefit in patients with lower ejection fractions. And also remember that beta blockers remain the standard of care for chronic heart failure with a reduced ejection fraction.
The take home I have is that patients who aren’t that sick really aren’t that sick. But those with more myocardial damage who need more from us.5
This should change practice for the patients that were included in these studies: low-risk MIs with prompt cath lab trip, good revascularization results, and preserved cardiac function. We can probably skip the beta blocker unless there’s another compelling reason to use one.
But if the patient has a late presentation with pretty high troponin and significant myocardial damage, I’m not ready to throw out the beta blocker. And if there’s a significant wall motion abnormality or the ejection fraction doesn’t really look normal, I think it’s pretty reasonable to give someone a beta blocker.
You shouldn’t blindly apply trial results to all patients.
Medicine is about seeing the person in front of you, not the population average.
Lesson 2: Goodhart's Law Strikes Again
Here's where this gets institutionally interesting. Beta blocker prescription rates after MI have been a quality metric that hospitals get measured on. And that means potential financial penalties for missing.
When a measure becomes a target, it ceases to be a good measure.
If quality metrics can’t change at the same pace that the medical evidence does, we’re going to be stuck with some really bad incentives in medical care. Right now that incentive structures pushes doctors to prescribe beta blockers on the margins, even when the evidence was already shaky.
Now we have definitive data showing that broad-brush beta blocker prescribing doesn't help most patients. But changing quality metrics is harder than changing prescribing habits. The infrastructure of healthcare measurement lags behind the science.
And in a world where more health care is metricized and we are going to see more and more input from AI in medical decision making, we should have a higher bar for the standards that we set.
If we don’t require a higher bar of evidence for the medical care that we incentivize, ChatGPT is going to turn into a pretty mediocre physician.
Lesson 3: Modern Cardiology, Modern Evidence
The nature of heart attack care has changed since beta blockers came on the scene.
We have amazing revascularization technology, better stents, more potent blood thinners, improved lipid lowering options, a greater understanding of the importance of blood pressure control, and a revolution in care for metabolic syndrome.
REDUCE-AMI, REBOOT, and BETAMI-DANBLOCK all demonstrate that in the contemporary era, with modern revascularization and secondary prevention, beta blockers add limited value to most patients with heart attacks.
That's not a failure of the old trials. It's a sign that medicine has evolved.
And this reinforces my firm belief that all clinical trials should have expiration dates.
Lesson 4: Beta Blockers Are Bad High Blood Pressure Medications
Beta blockers have fallen out of favor as blood pressure medications because they don’t reduce stroke as well as other antihypertensive medications.
And so that stroke signal in REBOOT-CNIC deserves some serious attention. It builds on what we already know about beta blockers and hypertension, which is that they aren’t as good as other antihypertensive agents at reducing central aortic pressure, which is the pressure that actually matters for stroke prevention.
Thiazide diuretics, ACE inhibitors, ARBs, and calcium channel blockers are much better at reducing central aortic pressure, which is why they’re considered to be first line treatments in the guidelines.6 But beta blockers reduce your peripheral pressure, and they create the false assurance that your hypertension is at goal.
This creates a phenomenon that I’m calling pseudonormotension.
Your brachial blood pressure (what we measure in clinic) looks fine, but your central aortic pressure remains elevated. You appear to have controlled hypertension, but your brain and other organs are still experiencing higher pressures and the resultant .
This could explain why beta blockers consistently underperform for stroke prevention compared to other blood pressure medications, and why we're seeing this concerning signal in the post-MI trials.
If beta blockers are giving us false reassurance about blood pressure control while leaving central pressures elevated, that's a recipe for increased cerebrovascular risk.
It makes me wonder to what extent beta blockers' historical benefit after MI was really just treating undertreated hypertension? And now in an era where we have better tools for blood pressure management, lipid control, and antiplatelet therapy, beta blockers might just not be as good as the other options that we have for prevention.
What This Means for Patients
Post heart attack treatment might mean a slightly smaller cocktail of medications.
Please let me be clear - this newsletter does not establish a doctor/patient relationship, so if you've had a heart attack and are on beta blockers, don't stop them without talking to your doctor.
The evidence now supports a more nuanced approach. Your cardiologist should be able to look at your individual risk profile and make a recommendation based on your specific situation, not just check a box on a quality metric.
These trials aren't just about beta blockers. They're a case study in how medical evidence should work:
Evidence should have expiration dates. As standard of care changes, we need to continuously reevaluate whether old practices still make sense.
Quality metrics must evolve with the science. The lag between new evidence and updated benchmarks creates perverse incentives.
AI and algorithmic medicine make this more urgent. As healthcare becomes more automated and metric-driven, outdated targets will become more entrenched, not less.
We're likely to see guidelines change over the next year as these results get incorporated into clinical practice recommendations. The European Society of Cardiology and American College of Cardiology will need to wrestle with how to translate this evidence into practical guidance.
The next challenge isn't just updating the science, it's updating the systems. Quality metrics, electronic health record alerts, and clinical decision support tools all need to catch up to what the evidence actually shows.
Beta blockers after heart attacks have been clinical canon for four decades. These trials suggest it's time to retire that canon and replace it with something more sophisticated: individualized medicine based on contemporary evidence.
That's how medical progress is supposed to work.
Please don’t throw the baby out with the bathwater. Just because it shouldn’t be the standard for every patient doesn’t mean that no patient should get a beta blocker after a heart attack. These are still some of the most widely used drugs in cardiology, even if they become slightly less widely used after these trials.
Ejection fraction is probably the best validated metric in figuring out whether beta blockers should be given to a patient. If you’re not in medicine, ejection fraction is basically the percent of blood that the heart pumps out with each beat. Normal is 50-70%, under 40% is considered significantly reduced, and 40-49% is in a gray area, termed “mildly reduced.”
These were open label trials, which normally would be concerning. People know what they’re getting and you miss out on the placebo effect. The reason that this isn’t all that concerning is that when you measure outcomes like death, heart attack, or stroke, you don’t really need to worry about the placebo effect.
Subgroup analyses are hypothesis generating and not hypothesis testing. But that doesn’t mean we ignore them completely as they’re a layer of information that I think is worth integrating into our overall interpretation of the data.
John Mandrola made the really important point that the difference between an ejection fraction of 45% and 40% is just not something that cardiologists can reliably distinguish, so some of these differences are somewhat subjective. And ejection fraction is dynamic with changes in response to loading conditions and to medical therapy.
That doesn’t mean all doctors follow the current guidelines. I can’t tell you how many patients I’ve seen who are prescribed metoprolol or atenolol as a first line antihypertensive medication, which is almost certainly the wrong thing to do.
Thank you for this. Lesson 2 is so important for many quality metrics, not just BB therapy post MI. Lesson 4 was fascinating. I inherited many patients from retired providers who have been on beta blockade for their HTN for decades and I haven’t changed the meds because their BP in the office is good. Now I’m rethinking that and have the knowledge needed to explain why I should change to an ACE-I or ARB. Thanks!
Thanks for covering this. I’d stopped prescribing BB at discharge with anyone post MI, post revasc, with preserved LV fxn (ie EF>50), based on Reduce-AMI. As the trial was silent on EF 40-50, I had continued with BB use in those patients. Given the latest subgroup meta-analysis of Reboot, Betami-Danblock, and Capital RCT, I will continue with that practice pattern. As you also note, none of this should obscure the fact that BB remain foundational in anyone post MI with HF or EF<40….none of these trials apply to those patients. I would submit that those with incomplete revascularization and residual ischemic substrate post MI would also merit BB continuation, irrespective of EF.
For a detailed summary of the evidence for post MI BB to begin with (starting in the 1980s), the landmark studies are reviewed in detail on the Cardiology Trials substack (for which Dr. Mandrola is also a contributor). Those trials pre-date my training era and were “settled science” at the time which I did not question….but it was certainly an eye-opener to see those details with a 21st century lens. And that exercise definitely sustains your contention that evidence should have an expiration date….or at least should be re-examined periodically in the face of new evidence and/or evolution of practice patterns.