Should We COMMIT To This?


I try to keep up on current events in medicine, but sometimes discover that I miss things that may affect the manner in which I practice medicine.

I attended a lecture on new treatments for STEMI and had a brief discussion with the lecturer afterwards regarding “Quality Measures” as determined by CMS.

He mentioned a study that I had not read that goes by the acronym “COMMIT” – the Clopidogrel and Metoprolol in Myocardial Infarction Trial. This was a HUGE study that enrolled more that 45,000 patients in 1250 different hospitals. Patients suspected of having an acute MI were randomized to receive clopridogrel, metoprolol or placebo and were then continued on the medication until discharge or until they had been in the hospital for a week.

Interesting point of the study was that patients who are given beta blockers “on arrival” had no difference in the primary endpoints of the study – death, reinfarction, or cardiac arrest. There was a 0.005 probability that the people receiving metoprolol would not experience ventricular fibrillation or reinfarction, but more than 1% of the patients receiving early beta blockers went into cardiogenic shock! As an aside, those on clopridogrel showed a significant improvement in the primary study endpoints.

By getting early beta blockers, initially the patients did worse than those getting placebos. Over time, the reduction in reinfarctions and in ventricular fibrillation caused study participants to break just about even in terms of adverse events. The conclusion of the COMMIT study was that “it might generally be prudent to consider starting beta-blocker therapy in hospital only when the haemodynamic condition after MI has stabilised.”

The COMMIT study results somewhat conflict with other studies cited by the American College of Cardiology and the American Heart Association in their joint 2004 guidelines for management of patients with acute ST-Elevation MI.

Yet one of the “Hospital Compare” indicators for “quality care” in heart attack patients that the patient receive a “beta-blocker at arrival.” Even though the ACC/AHA recommendations are for patients with ST elevation MI, the Hospital Compare site does not make that distinction.

Hospitals all over the nation brag about their statistics giving beta blockers at arrival. Cedars-Siani Hospital, Henry Ford Hospital, the Mayo Clinic, the Indiana Community Health Network, St. Luke’s Hospital in Missouri, and the University Health Care System in Georgia are just a few of the hospitals that I found posting their beta-blocker stats online. Is there conclusive evidence that we are helping, and not harming, patients by giving them beta blockers at arrival?

I spent an hour online researching on PubMed, eMedicine, and MD Consult. I couldn’t find any studies showing a significant improvement in outcomes with “beta blockers at admission” for acute MI patients. I’m hoping that someone out there can post a link or two. The Beta Blocker Heart Attack Trial showed a significant improvement in outcomes when beta blockers were started 5-21 days after hospital admission, not “at arrival.”

Even if there are some studies out there that do show a benefit, how do we reconcile the potential harm shown in the COMMIT study? If there is disagreement in the literature about the possible harm of a therapy, should hospitals get a bad “grade” for not giving that therapy?

I would love to see data comparing the rates of death, reinfarction, etc. AFTER all of the CMS quality care measures were instituted. Couldn’t find that online, either.

CMS is already making things worse for patients with pneumonia care. Now a huge study shows it may also be wrong with recommending early beta blocker administration.

And there is no quality care measure for  giving clopridogrel (Plavix) — a medicine that is proven to reduce morbidity and mortality in acute MIs.

Isn’t one of the tenets of medicine to “first do no harm“?

What gives?

Picture credit here


  1. once you realize that P4P measures and quality rankings are more about creating a tiered payment system to save money for payors (private or govt) than they are about improving lives of patients, it will all make sense.

  2. Bear in mind that the codes for the P4P measures are three-fold: Beta-blocker given, beta-blocker not given for patient factors, beta-blocker not given for hospital factors, or more simply: yes, contraindicated, no. Only the third code counts as a deficiency when they are crunching the numbers for analysis.

    While your point is EXTREMELY valid that the measures are dubious at best, there is at least this mitigating factor, that you can “opt out” of a given measure if in your judgment it is clinically appropriate. But you need to clearly and explicitly document it.

    Does it count if we say that the medical studies don’t support it? 😉

  3. This may have reference to some studies on the goodness of β-blockers. It was too medical for me to understand, and besides, I find cardiology more boring than just about anything. But it is mercifully short, and it wouldn’t take you doctors more than a minute to read and understand it.

  4. Our protocol is to give 600 mg Plavix on arrival to STEMIs, and I give beta blockers to patients who I think will benefit from them, namely patients whose BP and pulse suggest they will benefit from decreasing cardiac workload (and therefore cardiac oxygen consumption). Some STEMI patients with hypertension and pulses in the 60 range are more likely to benefit from early ACEI.

  5. I believe that this study (don’t have the complete article handy right now) used very high dose beta-blockers at fixed doses w/o consideration for titration of HR/BP.
    My collaborating MD felt that this high dose of beta-blocker likely explained some of the worse outcomes in those on BB.

  6. We have stopped routinely giving STEMI and ACS patients Beta blockers on arrival – and our cardiologists are on board with it. I personally only give it to people who are really tachycardic with a decent BP – and almost never to people with IWMI.

    So what does your hospital administration say to your “Hospital Compare” grades?
    When competing hospitals show that they are 100% and you hospital is at 20%, it doesn’t look good even if the metrics are bogus.

  7. Dr. Greenbbs on

    I never give b-blocker to IWMI patients…usually the right sided component is too prohibitive for me to do so.

    Last night I had a guy with an anginal equivalent (new onset SOB) who had SBP of like 180…definitely benefited from the b-blocker and ntg.

    Just goes to show you….when non-doctors and the government get in the business of medicine, things just go to hell.

  8. How did the researchers convince patients to actually enroll in this study? I can’t imagine that such a high number actually agreed to possibly receiving a placebo for treatment of an acute MI. Did they promise them Starbucks gift-cards?

  9. The motivation to participate in a study is not something that would easily fit into a comment. I am not the person to write that, anyway. The patients were receiving the other standard treatments for heart attacks based on their presentation.

    On the other hand, why do we assume that the standard treatment is beneficial to patients?

    So much of the ACLS treatments are in the algorithms because, in the 1980s, some experts in 1980s medicine felt that these were the best treatments available and would lead to the best outcomes. The more we learn through research, the more we realize that a lot of these recommendations were not good for the patients.

    Is there any good reason to suppose that the standard treatment is better than placebo? If a study is done well, we should find out. If not, then your guess is as good as mine.

Leave A Reply