MI preserved EF: beta-blockers not help post-MI

 a recent article found that patients sustaining a myocardial infarction who have preserved LV ejection fraction did not benefit from beta-blockers (see MI bblockers longterm not help NEJM2024 in dropbox, or DOI: 10.1056/NEJMoa2401479), in the REDUCE-AMI trial

Details:

-- 5020 patients from 45 centers in Sweden, Estonia, and New Zealand (95% from Sweden) who had an acute MI who had coronary angiography revealing obstructive coronary artery disease with a left ventricular ejection fraction (LVEF) of at least 50% were randomized to metoprolol or bisoprolol vs placebo for 3.5 years, in a study from 2017-2023

    -- metoprolol was first choice with goal of at least 100 mg/d; bisoprolol was second choice with goal of  at least 5 mg/d

    -- hypertension control: recommendation that no b-blockers be used in the placebo group. those already on b-blockers were encouraged to be tapeed off them over 2-4 weeks

-- median age 65, 23% women, 

-- hypertension 46%, diabetes 14%, prior MI 7%, prior heart failure 7%

-- prior to study: 12% on beta-blocker

-- ST-segment elevation MI in 35%

-- coronary angiography: 1-vessel disease in 55%, 2-vessel in 27%, 3-vessel or left main in 17%

-- percutaneous interventions in 96%, CABG in 4%

-- discharge meds: aspirin in 96%, P2Y12 receptor blocker (eg clopidogrel, ticlopidine, ticagrelor, prasugrel) in 96%, ACE-I or ARB in 80%, statin in 99%

-- primary endpoint: composite of death from any cause or new MI

-- secondary endpoint: death from any cause, death from cardiovascular causes, subsequent MI, hospitalization for atrial fibrillation, hospitalization of heart failure

-- safety endpoints: hospitalization for bradycardia, second- or third-degree atrioventricular block, hypotension, syncope, implantation of pacemaker; hospitalization for asthma or COPD; hospitalization for stroke

Results:

--primary endpoint:

    -- b-blocker: 199 events in 2508 patients (7.9%)

    -- placebo: 208 events in 2512 patients (8.3%)

        -- HR 0.96 (0.79-1.16), p=0.64, no statistically significant difference

-- secondary endpoint (death from any cause):

    -- b-blocker: 97 events in 2508 patients (3.9%)

    -- placebo: 103 events in 2512 patients (4.1%)

        -- HR 0.94 (0.71-1.24), no statistically significant difference

-- secondary endpoint (death from cardiovascular causes):

    -- b-blocker: 38 events in 2508 patients (1.5%)

    -- placebo: 33 events in 2512 patients (1.3%)

        -- HR 1.15 (0.72-1.84), no statistically significant difference

-- secondary endpoint (MI):

    -- b-blocker: 112 events in 2508 patients (4.5%)

    -- placebo:117 events in 2512 patients (4.7%)

        -- HR 0.96 (0.74-1.24), no statistically significant difference

-- secondary endpoint (hospitalization for atrial fibrillation):

    -- b-blocker: 27 events in 2508 patients (1.1%)

    -- placebo: 34 events in 2512 patients (1.4%)

        -- HR 0.79 (0.48-1.31), no statistically significant difference

-- secondary endpoint (hospitalization of heart failure):

    -- b-blocker: 20 events in 2508 patients (0.8%)

    -- placebo: 22 events in 2512 patients (0.9%)

        -- HR 0.91(0.50-1.166), no statistically significant difference

-- safety endpoints (hospitalization for bradycardia, second- or third-degree atrioventricular block, hypotension, syncope, implantation of pacemaker):

    -- b-blocker: 86 events in 2508 patients (3.4%)

    -- placebo: 80 events in 2512 patients (3.2%)

        -- HR 1.08 (0.79-1.46), no statistically significant difference

 -- safety endpoints (hospitalization for asthma or COPD)

    -- b-blocker: 15 events in 2508 patients (0.6%)

    -- placebo: 16 events in 2512 patients (0.6%)

        -- HR 0.94 (0.46-1.89), no statistically significant difference

-- safety endpoints (hospitalization for stroke)

    -- b-blocker: 36 events in 2508 patients (1.4%)

    -- placebo: 46 events in 2512 patients (1.8%)

        -- HR 6.80 (-7.11 to 20.72), no statistically significant difference [the math here seems to be wrong, but conclusion unchanged...]

Commentary:

 -- prior trials of beta-blockers have been done many years ago (mostly in the 1980s) finding benefit in those with HFrEF (heart failure with reduced ejection fraction) as well as those with mostly large MIs and left ventricular systolic dysfunction (20% decrease in mortality)

    -- at that prehistoric time, not much intervention was being done (no troponins, percutaneous coronary interventions, antithrombotics, statins, RAS antagonists,...)

-- a 2014 meta-analysis found that b-blockers in the coronary reperfusion era did not seem to have the benefit post-MI as in the  pre-reperfusion era, this new meta-analysis finding a 13% decrease in MI, and a 20% decrease in angina, but a 10% increase in heart failure, a 29% increase in cardiogenic shock and a 64% increase in drug discontinuation, with no change in cardiovascular mortality and overall mortality: see chf bblockers not help much post-reperfusion era AmJMed2014 in dropbox, or http://dx.doi.org/10.1016/j.amjmed.2014.05.032 )

-- other studies have also found that the benefit in those with reduced EF is much greater than in those with preserved EF post-MI. A few studies in patients with midrange LV ejection fraction (40-50%) did find some benefit from b-blockers

    – however, a Cochrane review of beta-blockers in patients with MI but without heart failure found moderate- to low-quality evidence (only one study included patients in the "reperfusion era") of a 28% decrease in major cardiovascular events and a 27% decrease in cardiovascular mortality. This study (DOI: 10.1002/14651858.CD012565.pub2.) included patients with EF >40%. Were their conclusions different because of the lack of more state-of-the-art current studies or because they used a 40% instead of a 50% LVEF cutpoint?

--this current trial found that there was no difference in all of the primary and secondary outcomes between those with MI and preserved ejection fraction if given b-blockers; and there was no evident difference noted in the graph above even within the first year of therapy

    -- a point of reference: beta-blockers seem to have a relatively short duration of therapeutic effectiveness in those with decreased LVEF, maximally in the first year but still evident for up to 2-3 years: Beta Blocker Use After Acute Myocardial Infarction in the Patient with Normal Systolic Function: When is it “Ok” to Discontinue? - PMC (nih.gov)

-- the patients in the above trial were well-treated in terms of having percutaneous cardiovascular interventions and receiving current state-of-the-art care

Limitations:

-- this was an open-label parallel-group study, which allowed for some of the manipulations above such as decreasing b-blockers in those assigned to the placebo group. this type of study is less statistically rigorous than a truly randomized one: were those who had been on b-blockers but were assigned to placebo different from those who had not on b-blockers in ways that might distort the overall results (ie, a selection bias)?

-- as a registry-based study, there was no ability to assess the quality of some of the contained information: there is more control over data quality when dedicated research staff are eliciting the patient information in a consistent manner

-- this was largely a study of a uniform population in Sweden, which might limit the results to others. were there different percentages of patients with LVEF>50% having underlying amyloid cardiomyopathies in Sweden than in other countries (and beta-blockers should be avoided in those with cardiac amyloid, since there may be more bradycardia, A/V block or hypotension)?  any difference in this type of underlying condition (or other genetic ones) might affect the response to the b-blocker intervention in Sweden and limit generalizability to other areas

-- the safety endpoints included only hospitalizations and not all relevant endpoints that may well be associated with b-blocker usage (eg, heart failure treated as outpatient), thereby understating their occurrence

-- patients with other indications for beta-blockers were excluded from the study, which might also limit the generalizability of the results to other populations. Perhaps that group having an MI with preserved LVEF might have cardiovascular benefit by continuing b-blockers?? (and we do not know why they were on b-blockers: patients with migraine, who might have been on b-blockers for migraine prophylaxis, have a 33% increased risk of MI)

-- there were several patients who were in the placebo group who were on b-blockers (14%) for unclear reasons and unclear duration, also muddying the waters a bit

so, 

-- the big issue here is that there is no clear benefit to placing patients on beta-blockers post-MI if they have a preserved ejection fraction (LVEF>50%)

-- there are no clear recent US guidelines on the use of beta-blockers post-MI, though the European Society of Cardiology's 2023 guidelines on acute coronary syndromes do note that information on beta-blockers after ACS  in patients with LVEF >40% "is less well established" than in those with lower LVEFs, and that there are currently several ongoing trials to assess this issue

-- this question of beta-blocker use does raise the general issue of "moving targets": many treatments in medicine, including the goals of treatments (eg, what is the appropriate LDL target in high-risk patients?), change pretty frequently depending on technology developments and newer studies. And many of the guidelines are older and/or based on older studies. so, when a study like this one comes out, it probably is reasonable to strongly reconsider the older approach of using b-blockers for at least 1 year post-MI. in this case, that conclusion is supported by some of the more recent of the older studies. Ongoing studies will provide clearer guidance.

    -- one pretty striking example of a fix to this issue was done by the AASLD (American Association for the Study of Liver Diseases), when the early assessments and treatments of hepatitis C infection, for example, was changing so quickly that their guidelines very  included the new studies right away; or, similarly with UpToDate, where the authors quickly incorporate new information into their suggested assessment and approach to diseases. This would seem to be a great step forward for all guidelines..... 

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

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- 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

if you would the article, please email me.

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list