beta blockers not help if MI and normal LVEF

 Contrary to several current guidelines, prescribing beta-blockers in patients with acute MI and preserved ejection fraction does not seem to provide any benefit (see MI nl EF beta block not help NEJM2026 in dropbox, or DOI: 10.1056/NEJMoa2512686)

Details:

-- this study is a preplanned meta-analysis of five prior open-label trials that randomly assigned patients with a recent myocardial infarction, no other indications for beta-blocker therapy, and a left ventricular ejection fraction (LVEF) of at least 50% to receive either beta-blocker therapy or no beta-blocker therapy:

    -- the REBOOT study (Treatment with Beta-Blockers after Myocardial Infarction without Reduced Ejection Fraction): 6459 patients

    -- the REDUCE-AMI study (Randomized Evaluation of Decreased Usage of Beta-Blockers after Acute Myocardial Infarction): 4967 patients

    -- the BETAMI study (Norwegian Beta-Blocker Treatment after Acute Myocardial Infarction in Revascularized Patients without Reduced Left Ventricular Ejection Fraction): 2441 patients

    -- the DANBLOCK study (Danish Trial of Beta-Blocker Therapy after Myocardial Infarction without Heart Failure): 2277 patients

    -- the CAPITAL-RCT study (Carvedilol Post-Intervention Long-Term Administration in Large Scale Randomized Controlled Trial): 657 patients

-- overall, there was a total of 17,801 patients from these studies, of whom 8831 (49.6%) were assigned to a beta-blockers and 8970 (50.4%) to placebo

-- the demographics were very well-balanced between the groups assigned b-blocker or placebo, so these demographics below are for both groups:

    -- median age 62, 21% Female, 33% from Spain, 28% from Sweden/Estonia/New Zealand, 14% from Norway, 13% from Denmark, 9% from Italy, and 4% from Japan

    -- current smoker 33%, hypertension 48%, diabetes 17%, dyslipidemia (unspecified) 42%, previous MI 8%, stroke 2%

    -- STEMI (ST-elevated MI) in 46%; atrial fibrillation in 2%

    -- in-hospital treatment: percutaneous coronary intervention in 95%, coronary artery bypass grafting in 2%, no revascularization in 4%

    -- beta-blocker therapy before randomization 11%

    -- b-blocker type after randomization: bisoprolol 47%,metoprolol 46%, carvedilol 5%, other 2%

-- primary endpoint: a composite of death from any cause, myocardial infarction, or heart failure

-- secondary endpoints: the individual outcomes in the primary endpoint

    -- when the information was available in the individual studies, they assessed cardiac death, unplanned coronary revascularization, and malignant ventricular arrhythmias

-- the safety endpoints included ischemic stroke and advanced atrioventricular block

-- median follow-up of 3.6 years

Results:

-- less than 1% of participants were lost to follow-up or had missing data in the studies

-- the composite primary end-point event per assigned group:

    -- beta-blocker group: 717 patients (8.1%)

    -- placebo group: 748 patients (8.3%)

        -- hazard ratio, HR 0.97 (0.87 to 1.07), P = 0.54; ie not even close to being statistically significant

The following table also incorporates the results of the key secondary endpoints and the safety endpoints, none of them statistically significant

here is the visual comparison confirming that there was no evident benefit from beta-blockers for the primary composite endpoint and those of the individual components:

 

on review of the prespecified subgroups, there was no significant difference in outcomes whether male or female, >70yo or <70yo, whether MI was STEMI or NSTEMI, whether the individual had diabetes or hypertension or atrial fibrillation, or whether they had previously used b-blockers or not

Commentary:

-- the historical context of the beta-blocker post-MI being standard-of-care dates to small studies from the early 1980s, most based on ST-elevated MIs {STEMIs). But the treatment of MIs has changed oh-so-much, with vastly improved diagnostics, coronary artery reperfusion, revascularization techniques, and meds

    

-- the European Society of Cardiology in their 2023 guidelines note that "The added cardio-protective role of beta-blockers in post-ACS patients without reduced LVEF on otherwise optimal medical therapy needs to be clarified", and gives the use of b-blockers a class IIA recommendation in patients without a reduced LVEF: https://doi.org/10.1093/eurheartj/ehad191

-- the American Heart Association and American College of Cardiology in their 2025 guidelines still gives the use of beta-blockers a class I recommendation "in patients with ACS without contraindications, early (<24 hours) initiation of oral beta-blocker therapy is recommended to reduce risk of reinfarction and ventricular arrhythmias", with no mention of LVEF (though it does comment that we need more studies): https://doi.org/10.1161/CIR.0000000000001309

-- this current meta-analysis of the 5 studies was done because the results of the individual studies were not sufficiently statistically powered to have definitive assessments of the individual outcomes, and there were somewhat different conclusions in the different studies

-- this meta-analysis found that b-blockers were not associated with a lower composite of death from any cause, myocardial infarction, or heart failure; this result was consistent for each individual component of this composite, other secondary endpoints and safety endpoints, as well as across all prespecified subgroups

-- of note, another recent study in the Lancet of patients with mildly reduced LVEF of 40-49% that included 1885 patients from 4 of the above studies (REBOOT, BETAMI, DANBLOCK, CAPITAL-RCT trials), which had many fewer individuals who were in the specific LVEF range of 40-49% than in the current meta-analysis, did find a 25% reduction in the composite of all-cause death, new MI, or heart failure: (see Rossello X et al, the Lancet volume 406, Issue 10508, 13-19 September 2025, pages 1128-1137)

-- but the same researchers also published a study "Beta-blockers after myocardial infarction without reduced ejection fraction" (see MI LVEF down to 40 beta block not help NEJM2025 in dropbox, or DOI: 10.1056/NEJMoa2504735) found similar results to the current meta-analysis that beta-blockers post-MI in those with LVEF >40% (this was based on the REBOOT trial listed above). a few details on this study:

    -- open-label randomized trial in Spain and Italy in 8438 patients with acute MI with or without ST-elevation, including 4243 who received b-blockers and 4262 placebo, follow-up of 3.7 years

    -- all patients had coronary angiography (regardless of the final therapeutic strategy) during the hospitalization and had LVEF >40% before discharge

    -- those with heart failure during the hospitalization were excluded

    -- patients were randomized at the time of discharge or within 14 days after discharge

    -- all patients had standard care, and the managing physician determined the b-blocker used (85.6% were on bisoprolol, and the dose for the different b-blockers was less than the median for almost all, translating to 2.5mg bisoprolol for most)

    -- findings were that b-blockers had no effect on the incidence of death from any cause, reinfarction, or hospitalization for heart failure

-- the prior ABYSS study (https://pubmed.ncbi.nlm.nih.gov/39213187/) found that stopping b-blockers in patients with LVEF >40% after a median of 2.9 years had no effect on death from any cause, MI, heart failure, or stroke

-- overall the all-cause death rate in HFpEF (heart failure with preserved ejection fraction) is 32% lower than HFrEF (with preserved ejection fraction) when one adjusts for age, sex and heart failure etiology in both the clinical trials and the community studies However, without that adjustment there is not much difference in survival between patients having heart failure with preserved vs reduced ejection fraction: DOI: 10.1056/NEJMcp2305181

    -- this suggests, as is not surprising, that there seems to be a fundamental difference in the individuals with reduced ejection fraction, who have important, worse comorbidities (eg obesity, diabetes, hypertension, atrial fibrillation, chronic kidney disease,...) and likely more severe MIs (larger infarctions) that justifies the use of b-blockers (more prone to heart failure)

-- one concern of late is that b-blockers lower plasma renin levels. this does bring up a potentially important negative effect: low renin levels by themselves are associated with higher cardiovascular risk: https://gmodestmedblogs.blogspot.com/2025/12/hyperaldosteronism-targeting-renin-level.html. ie, if b-blockers are necessary, should renin levels be checked in patients on b-blockers, and should we be adding spironolactone if they are low???

Limitations:

-- as with meta-analyses overall, the authors are mixing together studies of individuals with differing demographics, comorbidities, overall medications taken, other cardiac meds taken, etc. Perhaps the different findings in the individual different studies reflects the fundamental differences in study design, participants involved, cultural issues such as diet quality, exercise, stress levels, occupation, occupational exposures, air quality (potentially a very important cardiovascular risk factor: https://gmodestmedblogs.blogspot.com/2016/06/air-pollution-and-heart-disease.html), microplastic exposure (https://gmodestmedblogs.blogspot.com/2024/03/plastics-too-many-and-too-bad-for-our.html), chronic inflammation... all of this might well affect the overall results and the subsequent merging of the 5 different/marginally compatible studies used in the meta-analysis. And the studies even had different primary endpoints. the authors did have access to individual-level data, but unclear how accurate the statistical combinations of these studies ultimately was

   -- for example, in the BETAMI and DANBLOCK trials, half the persons screened for eligibility were ineligible despite having LVEF>40% (largely because of having atrial fibrillation, uncontrolled hypertension, or heart failure). per the current authors: "thus, the absence of an overall benefit of beta-blockers in patients with a preserved LVEF after myocardial infarction in our study does not apply to all patients with myocardial infarction and a preserved ejection fraction"

    -- all trials were open-label, with 6- and 12-month crossover in 11-18% of the patients, which could bias the results

    -- all trials used the baseline characteristics of patients. did some patients start smoking? gain weight? develop kidney failure? have diabetes that got better or worse???

    -- there were differences in the trials, such as even the definition of heart failure, and different subgroup analyses

-- does the choice of b-blockers used matter? one of the studies limited the b-blocker to carvedilol. but there are pretty bid differences between the b-blockers, since some are cardioselective (eg metoprolol, bisoprolol, atenolol) and some non-cardioselective (carvedilol and propranolol, labetalol, timolol) and there could possibly be different cardiovascular outcomes depending on which b-blocker is used. overall there was strong bias to prescribing bisoprolol and metoprolol in the current meta-analysis.  the studies on b-blockers post-MI have typically used low doses, does that matter? would higher doses of some of them altered the conclusions?

-- their baseline patient characteristics are not very granular. they document that 33% were current smoker, 48% had hypertension, 17% had diabetes. what meds were they on? how controlled the hypertension and diabetes were (and both of these have some meds that are beneficial to the heart and both have some meds that make things worse: we really need to know what meds were used). 33% were current smokers, but how many others had quit smoking (in terms of smoking cessation, the usually cited timeframe where cardiovascular risk regresses to non-smokers is 15 years, but this has been challenged: a Korean study found that heavy smokers may require 25 years or so to reach the cardiovascular risk of never-smokers, so it is important to know if those quitting smoking still actually had an active cardiovascular risk factor: see smoking cessation increased CV risk to 25 yrs if heavy smoking in dropbox, or doi:10.1001/jamanetworkopen.2024.42639

so, despite pretty recent cardiology society recommendations in the US and Europe, the use of beta-blockers post-MI in those with LVEF >50% (and >40% in some studies) has been significantly challenged, including in the current meta-analysis of 17,801 patients. there have been significant shifts in the cardiovascular risk factors over time

    -- as a background, heart failure with reduced ejection fraction is a pretty different animal than with preserved ejection fraction: the former is based on major cardiovascular injury, the latter is based on metabolic differences (diabetes, hypertension, obesity...)

    -- some of the reasons for improvement in cardiovascular health is likely the quite large improvements in preventing, detecting, and treating cardiovascular risk factors and cardiovascular disease over the decades since the first studies found beta-blocker benefit

    -- some of this is likely due to the changes in the underlying conditions associated with MIs and heart failure (which clearly benefit from beta-blockers):

        -- on the one hand there has been well-documented increases in obesity, metabolic syndrome, diabetes, etc in the past few decades, many of these lead to more cardiovascular disease and particularly to more with reduced LVEF, which benefits from beta-blockers

        -- on the other hand, there have been important changes over the decades in lifestyle issues of diet, exercise, as well as effective medications and potential advances in adverse occupational and environmental exposures, that may decrease cardiovascular disease

    -- so it does seem to be a reasonable conclusion from the amalgam of above studies, despite their limitations and despite the fact that many of the analyses are based on the same individual studies, that beta-blockers do not seem to be necessary in those with MIs who do not have reduced ejection fraction (ie, <40%?). Of course, it would be great to have a single, large randomized controlled trial with lots of granular background data to be definitive

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

to get access to the dropbox: https://www.dropbox.com/scl/fo/pcf4vymtpo77f49wd2ljr/ANyJqB40CuBd-DMEr67XDoI?rlkey=5dsalmuhf8ekd0svsgvn93xm1&dl=0

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them