EMCrit Wee – Abandon Epinephrine?

Two studies were mentioned:

Hagihara A, et al. Prehospital Epinephrine Use and Survival Among Patients With Out-of-Hospital Cardiac Arrest. JAMA. 2012;307(11):1161-1168

See Ryan’s blog for some great commentary.


The PACA Trial: Jacobs et al. Effect of adrenaline on survival in out-of-hospital cardiac arrest: A randomised double-blind placebo-controlled trial. Resuscitation. 2011 Sep;82(9):1138-43.


You finished the 'cast,
Now get CME credit

Not a subcriber yet? Why the heck not?
By subscribing, you can...

  • Get CME hours
  • Support the show
  • Write it off on your taxes or get reimbursed by your department

Sign Up Today!


Subscribe Now

If you enjoyed this post, you will almost certainly enjoy our others. Subscribe to our email list to keep informed on all of the ED Critical Care goodness.

This Post was by , MD, published 3 years ago. We never spam; we hate spammers! Spammers probably work for the Joint Commission.


  1. says

    I think the more important study, even though a post hoc analysis, it shows what happened to the patients who actually received epinephrine vs. those who did not receive epinephrine. One of the problems with studying such a traditional treatment is that some true believers will violate the protocol, if they have the opportunity. It is not proof, but it strongly suggests harm from epinephrine.

    Outcome when adrenaline (epinephrine) was actually given vs. not given – post hoc analysis of a randomized clinical trial.
    Olasveengen TM, Wik L, Sunde K, Steen PA.
    Resuscitation. 2011 Nov 22. [Epub ahead of print]
    PMID: 22115931 [PubMed – as supplied by publisher]


    We do not have any studies that have been done well enough to justify using epinephrine. We never should have let it become the standard of care without evidence. We need to be better than the homeopaths, acupuncturists, and other placebo practitioners.

    Since heart attack is the most common cause of cardiac arrest, is it wise to give epinephrine to everyone still in cardiac arrest after a shock?

    I think that there may be an indication(s) for epinephrine, but that we will not know what that indication(s) is until we do a study large enough to separate out any benefit in survival.

    ROSC is only a surrogate endpoint and therefore irrelevant when there is the possibility of having survival studies.

    We need to insist on an expiration date for any standard of care that is not based on evidence of improved survival.

    I completely agree with your recommended approach.


    • Braden says

      What you are saying makes sense. ROSC w/o neuro intact survival is worse than no ROSC at all. What I would be interested to know is if Epi + NEUROPROTECTION (a.k.a) hypothermia) would improve neuro outcomes while still allowing the increased rosc that you get with epi.

      Even more daring, yet sorta unrelated, , it would be interesting to see a study on Intra-Arrest hypothermia, to see if it works better than post-ROSC hypothermia, or what effect it would have.

      • says

        Braden, given the without hypothermia neuro-intact survival of v-fib (25%) there should be a signal even without the hypothermia unless epi causes harm. What I am saying is 1/4 of the additional ROSC patients should be a neuro-intact survival in the v-fib patients.

  2. ScottB says

    Rogue, I may be nit-picking, but I think you are a little off in your interpretation of the Olasveengen post-hoc study, which seemed to better demonstrate why retrospective studies of epinephrine may be inherently biased against epinephrine. Of importance, they found that early responders to CPR & defibrillation (i.e., those with the best chances of survival) are automatically placed in the no-epi cohort, skewing the results in favor of “placebo”. I think it helps build the case for an double-blind RCT, but not the case for epi causing harm (at least not any more than all the other retrospective studies).

    And just to be sure: I do agree with the assertion that epi NEEDS to be appropriately studied and that it most likely should be discontinued until there is adequate evidence to support its use…

    • says


      There were 85 who did not receive epinephrine. 6 did not receive epinephrine because of ROSC, but 55 were for unknown reasons. How many of those did not receive epinephrine because of ROSC prior to epinephrine? Nobody knows.

      Those could be eliminated and completely changed the results of the study.

      We will not know until there is a large double-blinded randomized placebo controlled study.


  3. Ben Hoffman says

    Adrenaline should be removed from the resuscitation of primary cardiac arrest; there is no evidence it increases survival to neurogenically intact hospital discharge and a growing body of evidence it does the opposite or has no effect whatsoever.

    Saying “BUT IT GETZ TEH ROSC!” means absolutely bloody nothing; resuscitating somebody to the point where they have a pulse is pointless if they’re only going to die a few days or a week later in ICU or if they’re going to be a vegetable, staying dead is far more dignified.

    Providing an intervention when there is no evidence it actually does anything beneficial is voodoo and charlatanism. As far back as 1994 it was stated as part of Intensive Care Officer (Paramedic) training that drugs had not been proven effective and that they might at best alter the environment a little to make defibrillation more effective. Somehow the message hasn’t quite sunk in.

    I think adrenaline might have more psychological benefit (“because I did all I could!”) to the person giving it than the patient.

    Olasveengen TM, Sunde K, Brunborg C, Thowsen J, Steen PA, Wik L. (2009) Intravenous drug administration during out of hospital cardiac arrest – a randomised trial; JAMA, 302(20):222-2229

    Hagihara, A., Hasegawa, M., Abe, T., Nagata, T., Wakata, Y., & Miyazaki, S. (2011) Prehospital Epinephrine Use and Survival Among Patients With Out-of-Hospital Cardiac Arrest. JAMA, 307(11), 1161-1168

    Jacobs IG, Finn JC, Jelinek GA, Oxer HF, Thompson PL. (2011) Effect of adrenaline on survival in out-of-hospital cardiac arrest: A randomised double-blind placebo-controlled trial. Resuscitation 82(9):1138-43

  4. Minh Le Cong says

    There is certain death without ROSC in cardiac arrest. Epinephrine is cheap and easily given. The prehospital research is suggestive both ways for harm and benefit but suffers from the same challenges in providing convincing results as does any other prehospital intervention like tracheal intubation in OHCA. We were sold a myth when Xigris entered the ICU market and wasted a lot of money splashing it around to septic patients. But we still gave it, as it gave us hope of benefit. Its ironic we now poke criticism at an old drug, dirt cheap , that DOES improve ROSC in cardiac arrest and survival to hospital admission. if it was an inhospital arrest, in ICU, and you knew the drug would improve ROSC, would you be really nOT giving it during your resuscitation? If you dont make it to first base, you will never make home. Sure thing you can get ROSC without epinephrine, but in my view its worth a dose during your resuscitation. If you can do bedside basic echo during CPR and see some cardiac motion, but have no shockable rhythm, then giving the epi is definitely worth a shot.No cardiac motion, no shockable rhythm and low ETCO2 despite good CPR..I agree no point giving drugs. but I would not condemn anyone who gave a dose of epi
    If you argue against using epi in cardiac arrest, you might as well throw out intubation as well as that has never been proven to improve survival to hospital discharge too. When evidence is lacking or of inadequate quality, we go with biologic plausability. that is why we still do intubation in arrest and give epi. Fundamentally I think they all have a role at some point in a cardiac arrest resuscitation. At what point and for whom, the truth is out there still!

    • says

      I believe the true objection is not merely that the drug increases ROSC, but not patient-important outcomes; but instead that if the drug does increase ROSC but not neurological outcome then it might actually be causing harm. Many of the post-arrest sequelae (post-arrest myocardial dysfunction and post-arrest neurological dysfunction) may eventually be traced to over-exuberant epi usage. The real answer is we don’t know. But every defense of epi as a standard of care that should not be abandoned makes it tougher to do a definitive study. Jacob’s PACA trial (which would have answered these questions) was destroyed by misguided defenders of an unproven treatment.

      Haven’t we thrown out intubation during arrest? The move to extra-glottic airways was an example of dogma challenge that may or may not be a good thing, but at least now we can study intubation vs. EGA; we could not have done that 5 years ago.

      • Minh Le Cong says

        Scott, I thought we had thrown out iintubation during arrest but recent posts seem to indicate providers are still intubating during chest compressions! Like trying to do a proper study of cricoid pressure, the definitive epi study in arrest is very challenging to organise. There is no financial incentive to do it. most providers think its biologically plausible to provide some benefit despite recognition it might cause some harm in some cases. its dirt cheap. Xigris was conclusively studied because it was filthy expensive and no one wanted to keep prescribing it until significant benefit was proven. PACA trial was suggestive of benefit to hospital discharge but did not have the numbers to conclusively prove it. 11 epi treated patients survived to hospital discharge vs 5 placebo treated patients. Thats pretty suggestive!

        • says

          Minh, you know the backstory behind that right? If the trial was conducted as Dr. Jacobs planned, I believe the numbers would have been definitive instead of suggestive.

          • Minh Le Cong says

            yeah I know. I have heard Jacobs talk on this to Mel on EMRAP as well as to my buddies in RFDS Western Australia section, where he lives, He actually gives great resuscitation lectures on the latest stuff. It was the closest we came to getting the textbook answers to our questions and he got ethics approval to do it, which I thought was impressive!
            The longest successful cardiac arrest resuscitation on record was about 90 minutes. that guy got everything. Bystander CPR, ETI, defib, epi, amiodarone, . One ER doc consulted over the phone and told the rescuers to stop. Another ER doc told them to try the amiodarone due to good ETCO2 reading..and it worked! Patient walked out of hospital three days later. Its hard to demonstrate harm of an intervention when the alternative is certain death. PACA results suggest the harm if any of epi is no more than saline placebo.

      • Minh Le Cong says

        It seems an odd and circular arguement to make, by saying that epinephrine may harm patients who are technically dead already. The supposition of course is that in the ten to twenty percent of cardiac arrest patients who you manage to get ROSC , then giving epi to achieve that, will harm them in the long run.The PACA trial indicated that it was not statistically significantly better than placebo but certainly not worse.i.e harmed patients by lowering survival to discharge, compared with placebo. In fact there was a trend to benefit.

        • says

          Yep, PACA maybe good–not bad; Japanese trial maybe bad-def. not good. What now? I say take epi out of routine use, keep it as an add-on drug like calcium or bicarb. Then study it for real in blinded RCT and then we will know.

          • Minh Le Cong says

            The Japanese study was not placebo controlled like PACA was. I agree. You cannot condemn those who choose or not choose to give epi in cardiac arrest at this juncture in our knowledge time line. interesting proposal to change current standard resucitation guidelines and remove epi from routine use. perhaps interview Jacobs on that one?

        • says

          PACA did not conclude that epinephrine was not worse.

          The OR of improved survival was 2.2 (0.7–6.3). Yes, that is a trend toward better outcomes, but it is not proof of a lack of harm.

          We can harm patients by giving epinephrine to patients who would otherwise be resuscitated with less neurological deficits – if the epinephrine in causing neurological deficits. This neurological harm is certainly biologically plausible.

          We can harm patients by giving epinephrine to patients who would otherwise be resuscitated – if the epinephrine in causing decreased survival. This decreased survival is certainly biologically plausible.

          We know that epinephrine produces more ROSC.

          We do not know if epinephrine produces long-term harm in order to produce that increase in ROSC.


          • Minh Le Cong says

            Rogue, love your fighting spirit! folks, We just did an interview about this and hope to get it out soon as the next prehospital podcast!

            Certainly its a stalemate in my view. Yes we can harm with epi. nothing is benign. its all in the timing and the dose! We dont know yet is what it boils down to. But if you are doing good CPR, got the airway sorted, are having a good eTCO2 trace and there is refractory VF arrest, what choice do you have when all your other therapies have failed? Do you call it a day, with no trial of a single dose of epi? Knowing that it does improve ROSC…we give epi drips to patients in shocked states all the time with no concern of neurologica damage..because the alternative is fairly obvious.
            this is not the best we can do..I agree. we need to improve. but this like the paradox of Buridan’s ass. Between two uncertain or ill defined choices, waiting for a clear answer may harm the ASS.

    • says

      Minh Le Cong,

      I apologize for the length of this, but . . .

      If Xigris were cheap, would that change anything?

      The cost, or age, of the drug is not important in determining if the drug works. We expect the patient to end up with much larger medical bills when epinephrine is given, than when epinephrine is not given.

      The cost of the epinephrine is not what is expensive. The resultant ICU stay is what is expensive.

      We need a study comparing the patients who do receive epinephrine vs. patients who receive a placebo. Then we will know how much more expensive it is to treat cardiac arrest with epinephrine, than without epinephrine. I expect that the cost of Xigris will be much less than the added costs of treatment with epinephrine.

      The real problem with Xigris and epinephrine is the same – no evidence of improved survival, in spite of similar biological plausibility.

      At least, the biological plausibility was similar right up until we found out that Xigris is over-hyped. What could be a better description of the effects of epinephrine, than over-hyped?

      Until there is evidence of survival benefit, we should not be routinely using anything outside of controlled studies.

      HDE (High-Dose Epinephrine) produces more ROSC than SED (Standard-Dose Epinephrine).

      If we are going to be logically consistent, we should be insisting on using HDE and using it aggressively. After all, If you dont make it to first base, you will never make home.

      HDE produces more ROSC than SDE and there is still not proof that HDE produces worse outcomes.

      How can we justify depriving out patients of this ROSC-saving drug?

      Since we are using baseball analogies to treat patients, rather than evidence, how are we getting on base?

      It does matter.

      A walk is a great way to get on base. So is a single. Being hit by a pitch is not great, but if we are only interested in scoring runs and we have pinch runner, that may be OK. However, we may lose a valuable playing to injuries. What if we do not have the possibility of a pinch runner.

      Certainly, our patients cannot be replaced at first base by pinch ROSC patients who have not been hit with the epinephrine effects. If the player had been hit in the knee and a pinch runner is not available, then the player is now a liability trying to hobble around the bases (except in the case of a ball hit over the wall). This may be the appropriate baseball analogy for the use of epinephrine in cardiac arrest. We do not know which is the best analogy without both research and a creative writing consultation.

      Surrogate endpoints, such as ROSC, are great for creating hypotheses for survival studies. Biological plausibility is also useful for creating testable hypotheses for survival studies.

      Research is how we find out what works.

      Surrogate endpoints are also great for coming up with treatments that are abandoned after being found to cause much more harm than benefit.

      We still have no evidence that epinephrine does anything more than improve the documented vital signs for EMS and in the ED.

      We are just treating the monitor. We are not treating the patient.

      Should we focus on what happens in the ICU? These are not the patients CPR and defibrillation were supposed to treat. The population having cardiac arrest at home, or in public, is not the same as the population in the ICU. We should not treat them as if they have the same medical histories.

      Should we throw out intubation in cardiac arrest?


      Ventilations have never been shown to improve outcomes in the initial management of cardiac arrest. This is just a traditional treatment. Intubation in cardiac at has evidence of harm, but not benefit. Using intubation in cardiac arrest outside of controlled trials is just as bad as using epinephrine. This is wishful thinking, not medicine.

      When evidence is lacking or of inadequate quality, we go with biologic plausability.

      That is what keeps the alternative medicine practitioners going.

      They do not need evidence. They only need plausibility.

      Homeopathy might work, so there is no reason to oppose this treatment that consistently fails to produce outcomes that are any better than could be produced with placebo.

      Acupuncture might work, so there is no reason to oppose this treatment that consistently fails to produce outcomes that are any better than could be produced with placebo.

      Voodoo might work, so there is no reason to oppose this treatment that consistently fails to produce outcomes that are any better than could be produced with placebo.

      A rubber chicken might work . . . OK, I won’t go there. I like your rubber chicken.

      Too much of what we do is based on our idea of plausibility that is limited by what we think we know about the way the body works. We frequently find that the explanations we though worked, do not work – or do not work in the way that we thought they did. When we stop questioning what we know, we stop improving the care we provide to patients.

      Biological plausibility and epinephrine?

      What is plausible about giving a drug that increases afterload to a patient in cardiac arrest?

      The most common cause of cardiac arrest is heart attack. What is plausible about giving epinephrine to a patient with a heart attack, just because the heart attack has produced cardiac arrest?

      If epinephrine works for heart attacks in cardiac arrest, shouldn’t we be giving epinephrine to heart attack patients not yet in cardiac arrest. Don’t we need to use epinephrine prophylaxis because of this obvious biological plausibility.

      When we look at biological plausibility, we also need to look at the biologically plausible harms, not just the proven harms. We cannot state that we will consider the plausible benefits, but only the proven harms.

      The reason we need evidence is that biological plausibility most often results in treatments that are harmful.

      There are plenty of plausible reasons for giving epinephrine and for opposing giving epinephrine. There is only one way to find out which of these plausible reasons, if any, are valid.

      That is why so many people commenting here, regardless of what their feelings on what the evidence will show, understand that we need evidence to continue treatment with this unproven treatment.

      I am not condemning anyone for giving epinephrine, but for opposing a requirement that we study epinephrine in a large enough double-blinded placebo controlled randomized trial to find out whether there are any benefits to epinephrine in cardiac arrest.

      I wouldn’t choose the word condemn, but I strongly oppose continuing the status quo.


  5. Minh Le Cong says

    Hey Rogue
    love the response! no need to apologise. this is how we improve what we do! passion for excellence.
    PACA is our best available evidence to date. it was a placebo controlled trial so compared epi to homeopathic salty water. no increased harm was demonstrated. but increased ROSC for whatever worth you might want to credit that. Agree Another trial is helpful . disagree we need to change international guidelines at this point. Take the ResQPod and ITD debate. One trial is suggestive of benefit but I have not gone out and bought my pair and changed my practice of Cardiac arrest resuscitation. We should not be evidence based zealots in clinical medicine and health care. Where definitive evidence is lacking or incomplete we still need to make a clinical opinion based decision.
    between ROSC and death, its a hard choice but I am hedging most of the general public choose ROSC! As for the expensive care unit stay post ROSC and that wasting money, I dont think thats a reasonable arguement point against using epi in arrest. epi is plausible to work by increasing coronary artery perfusion . Its just not plausible, the PACA and past work , shows it does increase ROSC. its a real effect!
    Scott, we should interview Rogue and get this passion on tape!

    • says

      Minh Le Cong,

      disagree we need to change international guidelines at this point.

      We should not be evidence based zealots in clinical medicine and health care. Where definitive evidence is lacking or incomplete we still need to make a clinical opinion based decision.

      While not all treatment requires evidence of improved survival, I don’t think that we should be making treatments standards of care until after we have evidence of improved survival, or similar appropriate endpoint.

      Epinephrine is the standard of care, so there has been opposition to depriving patients of the standard of care, because is is presumed that it is the best available treatment. We should not create a situation that limits our ability to find out what is best for our patients because of too much regard for a traditional treatment.

      This is more of a criticism of the way we approach standards of care, than it is specific to epinephrine. We have plenty of other standards of care that do not have good evidence to support their use. They are based on anecdote, tradition, and inertia.

      Without evidence, it is easy to allow our biases to mislead us, and we are all biased. Not all evidence needs to be randomized, placebo controlled trials, since some treatments just do not lend themselves to that kind of study, but we should do our best to study that which can be studied.

      Epinephrine is an easy study to set up along the lines of PACA – as long as the politicians are kept out of it.

      Without evidence, we have anecdote-based medicine, tradition-based medicine, and inertia-based medicine.

      Anecdotes are great for presenting information about hypotheses to study, but also for presenting information about the possible harm from treatments. This is something where I do not want to adopt the precautionary principle, but evidence suggestive of harm ought to be taken seriously.

      Droperidol should have received a thorough examination, rather than a black box warning. The FDA needs to reconsider, even though caution was indicated at the time. The FDA made the mistake of acting as if droperidol had the same evidence of lack of harm that a recently approved drug would have – almost none.

      Banning something because of suspicion of harm is no better than making something a standard of care because of suspicion of benefit.

      between ROSC and death, its a hard choice but I am hedging most of the general public choose ROSC!

      The same general public that was led in opposition to PACA?

      I do not think that the opinion of the general public is important in determining what should be the standard of care or what should be the standard of evidence.

      Medicine should not be a popularity contest.

      I agree that ROSC is several times higher with epinephrine. Studies have consistently shown that.

      Where are the extra survivors due to this magnification of ROSC?

      I see ROSC as the equivalent of winning on a technicality, rather than on merits. For example, the death row inmate who is pardoned by the Governor, but is executed because the pardon is not delivered on time. He received the pardon (technical win), but that technicality isn’t doing him any good.

      Until we can show that epinephrine is improving survival, we are hoping that the benefit of epinephrine induced ROSC is the most important effect of epinephrine.

      After 50 years, why can’t we show any clear benefit in survival from epinephrine?


  6. Minh Le Cong says

    okay Rogue
    Your baseball analogies kick ass..much better than mine. I concede on that front!

  7. Ben Hoffman says

    There is no evidence that adrenaline or intubation in cardiac arrest are helpful and evidence they are more than likely of harm.

    There is evidence adrenaline increases ROSC but not survival to neurologically intact discharge from hospital (1, 2) but rather the opposite (3). ROSC by itself doesn’t mean anything and is not a metric by which to measure “success” in cardiac arrest resuscitation. It’s quite equivalent to winning the battle but loosing the war.

    There is evidence intubation attempts interrupt CPR (4) and counter-evidence showing high quality, minimally interrupted CPR improves survival to hospital discharge (5). Additional evidence shows that there is no difference in survival to hospital discharge in patients who are intubated vs. patients who have an LMA placed (6) and perhaps most compelling is a study from Arizona that shows improved survival to hospital discharge in patients who are passively oxygenated and do not receive positive pressure ventilations or so called “cardiocerebral resuscitation” (7)

    Intervention without evidence should be dismissed as nothing more than witch doctor voodoo and charlatanistic quackery offering psychological comfort to those dispensing said quackery because “I did all I could!”. This is especially true in the case of interventions where evidence points to harm.

    (1) Jacobs, Ian G, Finn, Judith C.. Jelinek, George A.. Oxer, Harry F.. Thompson, Peter L (2011): Effect of adrenaline on survival in out-of-hospital cardiac arrest: A randomised double-blind placebo-controlled trial. Resuscitation – September 2011 (Vol. 82, Issue 9, Pages 1138-1143

    (2) Olasveengen TM, Sunde K, Brunborg C, Thowsen J, Steen PA, Wik L (2009): Intravenous drug administration during out-of-hospital cardiac arrest: a randomized trial: JAMA. 2009 Nov 25;302(20):2222-9.

    (3) Hagihara, A., Hasegawa, M., Abe, T., Nagata, T., Wakata, Y., & Miyazaki, S. (2011) Prehospital Epinephrine Use and Survival Among Patients With Out-of-Hospital Cardiac Arrest. JAMA, 307(11), 1161-1168

    (4) Wang HE, Simeone SJ, Weaver MD, et al. Interruptions in Cardiopulmonary Resuscitation from Paramedic Endotracheal Intubation. Ann of Emerg Med. 2009:54(5):645 652.

    (5) Christenson J, Andrusiek D, Everson-Stewart S, Kudenchuk P, Hostler D, Powell J et al (2009) Chest compression fraction determines survival in patients with out-of-hospital ventricular fibrillation. Circulation. 2009 Sep 29;120(13):1241-7

    (6) Kajino K, Iwami T, Kitamura T, Daya M, Ong ME, Nishiuchi T. et al (2011): Comparison of supraglottic airway versus endotracheal intubation for the pre-hospital treatment of out-of-hospital cardiac arrest. Crit Care;15(5)

    (7) Bobrow BJ, Ewy GA. (2009): Ventilation during resuscitation efforts for out-of-hospital primary cardiac arrest. Curr Opin Crit Care;15(3):228-33.

    • says

      Hello Scott & your group of active debaters. I’m relatively new as a listener to EMCrit – and I think this “Wee” (perhaps the biggest discussion ever from a small “Wee”) was posted after I began responding to your podcasts. I’ve “met” Rogue Medic (long back-and-forth with him and myself on his blog with the only conclusion from that the certainty that he and I will both respectfully agree to disagree).

      Anyway- My reply will be short because I wrote it all up in detailed on my ACLS COMMENTS- Issue #10 (https://www.kg-ekgpress.com/acls_comments-_issue_10/ ). To my reading – the studies being cited (by Hagihara, Jacobs, Olasveengen) are all fatally flawed (discussed in full at the above link). It is BECAUSE Epi increases ROSC that when given to a group of patients with OHCA who are for the most part already dead – that Epi doesn’t improve meaningful survival – and perhaps (because it increases chance of ROSC) its use may result in more patients who survive only to persist in vegetative state …. It’s NOT necessarily Epi’s fault – it’s the fault of the fact that the drug is given to a group of patients who are unlikely to recover with intact neurologic status …

      There are a number of issues to resolve in the future – not the least of which (as discussed in Scott’s excellent recent podcast on cooling post-arrest) is whether therapeutic hypothermia might improve survival and neurologic function in these non-VF-OHCA patients – though the problem still being (as discussed on Scott’s podcast) that we don’t yet know how to determine which non-VF-OHCA patients are going to favorably respond to hypothermia (some do – but will this treatment also render some with severe permanent neurologic impairment … ).

      Finally – although I hear loud and clear request by your commenters to discontinue Epi immediately because of no proof of benefit – this is just a lot more difficult to do than if there wasn’t such a long history of giving Epi. NOTE – I am NOT saying (!!!!) that we should continue Epi “because its always been given” (Honest Rogue Medic – I’m really NOT saying that). What I AM saying is that to my mind there is yet NO PROOF OF HARM from giving Epi to patients in cardiac arrest not otherwise responding. I’m OPEN to doing well-run randomized controlled clinical trials of Epi vs no Epi – IF methodology is appropriate (by that I mean you can’t just restudy patients who are for practical purposes already dead and say Epi “caused” neurologic impairment … ). But outside of those to-be-done-in-the-future well-run prospective trials – I don’t think Epi should be completed abandoned for all cases of cardiac arrest at this time (my opinion). Instead, as I state in my “Bottom Line-My Answer” (at the above link in my ACLS Comments) – I think judgement should be used by the providers on the scene as to relative likelihood of benefit vs harm from using Epi for OHCA (with the “harm” NOT being “because Epi caused harm” – but rather because Epi resulted in ROSC of a patient who was already brain dead ….).

      Sorry I told an “untruth” and wasn’t so “brief”. I don’t expect my answers will change the opinion of some of the other commenters (and I wish Rogue Medic ALL the BEST regardless) – but I felt a need to put in another opinion.

      FINAL THOUGHT: Perhaps there IS one thing that ALL of us commenters CAN agree on – Namely that Scott Weingart does a SENSATIONAL JOB !!!

  8. Ben Hoffman says

    Further to what I wrote above, there is no evidence showing that placement of an advanced airway improves neurologically intact survival in patients who suffer a cardiac arrest and evidence showing no difference in survival between intubation and the LMA; see (6) above. There is also evidence showing that delayed ventilation and avoidance of hyperventilation improve neurogenic outcomes in cardiac arrest patient (7, 8)

    I know I’m getting a little away from the subject of adrenaline but that (along with shoving an endotracheal tube down the patients gob) is nothing more than tradition and superstition wrapped up in physiologic theory and in terms of intubation, extrapolation of what works in other settings to the setting of cardiac arrest.

    One of the best explanations I ever heard on this subject is “you can’t take that stuff away from me, it’s what makes me a Paramedic!”. I know we are not exclusively discussing pre hospital cardiac arrest but that’s probably hit the nail on the head; these interventions provide psychological benefit to those dishing them out because “I did all I could!”.

    It’s bloody snake oil and voodoo and nothing more; holding hands and chanting pagan around the patient would likely have more benefit, especially if it didn’t interrupt CPR, subject the patient to the raised ICP of laryngoscopy or provide a conduit for massively supra physiologic tidal volumes to be forced down the patients gob which increases intra thoracic pressure and lower venous return to the heart.

    I’ve been banging on about oxygenation not being the same as ventilation and myocardial ischaemia not being the same as hypoxaemia for a few years now; maybe it’s time I started chanting about ROSC not being the same as neurogenically intact survival.

    (7) Kellum MJ, Kennedy KW, Ewy GA. Cardiocerebral resuscitation improves survival of patients with out-of-hospital cardiac arrest. Am J Med 119:335–40, 2006.

    (8) Aufderheide TP, Lurie KG. Death by hyperventilation: A common and life-threatening problem during cardiopulmonary resuscitation. Crit Care Med 32:S345–51, 2004.

    • Minh Le Cong says

      thanks Ben. that is useful. I would suggest caution in assuming cardiac arrest is one entity with one pathhysiology. Reversible causes of arrest do occur and may not respond to purely chest compressions and defibrillation. no one has done a randomised controlled trial of giving potassium in hypokalaemic arrest but I suspect our peers would consider that a reasonable decision despite lack of controlled evidence in the event of refractory arrest not responding to initial measures. and no one has done a long term hospital discharge survival study on treating hyperkalaemic arrest with calcium yet our peers would consider that standard of care. Why? because ROSC is an appropriate initial goal of cardiac arrest management, whether that be from whatever cause, reversible or not. If I take your line of arguement correctly, you would not administer epinephrine in a prehospital arrest from a betablocker overdose? Perhaps you would administer glucqgon or insulin dextrose, despite those therapies having just as little controlled data for long term survival than epi? chest compressions and defib might bring ROSC in a poisoning but if you were to call the code after doing only CPR and defib, peer review might auggest more could and should have been done. Intubation wise I agree with you. there is too much glottic fascination in the prehospital setting and overventilation does cause harm in my opinion, too.

      • says

        Just to throw more wrenches in the works, the AHA has removed potassium supplementation during arrest from the recs for hypokalemic arrest if I remember correctly.

        • Minh Le Cong says

          when dont you remember anything correctly, Scott!
          this is the reference

          read carefully! it advises against empiric bolus admin of KCL in suspected hypokalaemic arrest. Logical, totally. My suggestion is that in known hypokalaemic arrest ..someone intubated and ventilated, being transported IHT , you do an iSTAT for lytes, woah K is 1.5,they go into torsades arrest…we are not going to just do CPR and defib are we?
          does the fact we have no controlled data for long term survival stop us from giving a drug that is biologically plausible to improve ROSC?

      • says


        I would suggest caution in assuming cardiac arrest is one entity with one pathhysiology.

        One of my objections to the routine use of epinephrine in cardiac arrest is that it is treating cardiac arrest as if it is one entity with one pathhysiology.

        Everybody dead gets epi!

        The exception is those who are resuscitated before epinephrine can be given.

        Where is the attempt to avoid treating the cardiac arrests due to MI? Heart attack is supposed to be the most common cause of cardiac arrest. Epinephrine may be the worst legal drug we could give to a heart attack patient, so where is the biological plausibility in that?

        If an ACS patient has a palpable pulse and I attempt to give a bolus of epinephrine to him, I would expect the response to be almost violent.

        If there is no palpable pulse, the same patient is expected to be given a bolus of epinephrine.

        The difference is the ability to palpate a pulse. Does the ability to palpate a pulse change biological plausibility?

        If I give the MI patient 1 mg epinephrine, obtain ROSC, but expand the MI to the point of irreversible damage, am I making things better?

        If I continue to treat the patient without epinephrine and obtain ROSC later, is the patient worse off?

        If I resuscitate the patient with nitrates, is that a bad thing?

        ROSC is only an acceptable goal, when other goals are impractical.

        There is nothing impractical about demanding evidence of improved survival with good neurological function for patients treated with epinephrine in cardiac arrest.

        Specific cardiac arrest etiologies become much more difficult to study to the point of improved survival, but that does not mean that we are doing anything more than guessing at what we are doing for these patients when we provide any sort of antidote.

        Here is a good article on the treatment of toxicologic causes of cardiac arrest.

        Dissecting the ACLS Guidelines on Cardiac Arrest from Toxic Ingestions
        Emergency Medicine News:
        October 2011 – Volume 33 – Issue 10 – pp 16-18
        doi: 10.1097/01.EEM.0000406945.05619.ca
        Roberts, James R. MD


        • Minh Le Cong says

          I think we found another point of agreement! Yes , there should be no routine use of epi in cardiac arrest.

  9. says

    ACLS Modifications in Management of Severe Cardiotoxicity Due to Hypokalemia
    Life-threatening hypokalemia is uncommon but can occur in the setting of gastrointestinal and renal losses and is associated with hypomagnesemia. Severe hypokalemia will alter cardiac tissue excitability and conduction. Hypokalemia can produce ECG changes such as U waves, T-wave flattening, and arrhythmias (especially if the patient is taking digoxin), particularly ventricular arrhythmias,205,206 which, if left untreated, deteriorate to PEA or asystole.

    Several studies reported an association with hypokalemia and development of ventricular fibrillation,207–210 whereas a single animal study reported that hypokalemia lowered the ventricular fibrillation threshold.211 However, the management of hypokalemia in the setting of cardiotoxicity, specifically torsades de pointes, is largely based on historical case reports that report slow infusion of potassium over hours.212 The effect of bolus administration of potassium for cardiac arrest suspected to be secondary to hypokalemia is unknown and ill advised (Class III, LOE C).

    2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science
    Part 12: Cardiac Arrest in Special Situations
    Part 12.6: Cardiac Arrest Associated With Life-Threatening Electrolyte Disturbances

    • Minh Le Cong says

      lets try a thought experiment.
      The President of the USA drops with a VF arrest right in front of you during an election rally to EMS and ED critical care providers(he was promising a major boost to funding for EMS and providing more ED critical care beds with shiny new vents for each bed)

      This is possibly the best chance the President has to survive his VF arrest, being in a roomful of emergency folk with all their gear (which was available to show to the President)

      The Secret service staff demand you assist before more help arrives.

      You start CPR, you defib. nothing is working. a Colleague visiting from Australia suggests after five minutes as per international guidelines to try a one mg dose of epinephrine(he calls it adrenaline but you forgive him for that ). The Secret Service staff a looking very upset and some are crying now…they do not want to be the crew who lost the President during their shift.

      Are you going to not give epi at this point because you believe it might lead to long term harm? even though the Secret service have done a quick google search on their iphone and found that epi does lead to improved ROSC?

      • says


        That is an interesting situation. I have debated this with Dr. Ken Grauer. He thinks that the reason that epinephrine may produce bad outcomes is that it is not given early enough. If I understand his position, it would be the earliest that epinephrine/adrenaline could be given that would be most likely to improve the outcome.

        I do not get to make these decisions, since I am a paramedic. I need to follow my protocols, which state to give epinephrine following defibrillation with no ROSC. I would prefer to work on improving the quality of the CPR, but that should have been my goal from the beginning.

        I do not see a reason to believe that epinephrine will improve the outcome. I do not know of any evidence that shows that patients have a better outcome when epinephrine is given during cardiac arrest, but I do know of a bunch of studies which strongly suggest that patients have a bad outcome when epinephrine is given during cardiac arrest.

        I also do not know of any controlled trials that show epinephrine stops crying.

        I would point the Secret Service agents to the many posts I have written on the ways that surrogate endpoints, such as ROSC, mislead us to harm patients.

        Look how the therapeutic bleeding has made the patient much calmer.

        Look how the therapeutic bleeding has stopped the seizures.

        Look how the therapeutic bleeding has reduced inflammation.

        Look how the therapeutic bleeding has . . . .

        Maybe I should remove a pint. His humours are clearly interfering with the resuscitation. 😉


  10. Ben Hoffman says

    God all this new high tech medicine is so complex I say we go back to the days when whatever was wrong with the patient was hacked off with a rusty saw by a bootleg surgeon in his street clothes then the wound was cauterised in boiling oil and the patient quickly died from either hypovolaemia or infection; Pare, Semmelweis and Baron Lister be damned! (oh and adrenaline too right?)

    Cardiac arrest has multiple aetiologies I will not dispute that; but for a standard primary cardiac arrest I wouldn’t give adrenaline, I wouldn’t intubate the patient prior to ROSC and as for amiodarone I’d still give them some but only because the guidelines say so; if I remember correctly it has a higher rate of ROSC than patients who received lidocaine but no difference to survival.

    If somebody has a hypokalaemic arrest I’d give them some potassium, for somebody who scoffs down too many beta blockers give them some glucagon and so on …

    Tim brings up a very good point (one I raised elsewhere a few days ago) that the leading cause of cardiac arrest is dysrhythmia; the leading cause of dysrhythmia is myocardial infarction. Giving adrenaline to somebody with a myocardial infarction that had a cardiac arrest is probably not a good idea. If you got an MI patient with right coronary artery occlusion and was very bradycardic and hypotensive because their SA node is dysfunctional; would you give them an adrenaline drip to flog their ticker along and lift the blood pressure? (that is an option here in New Zealand seeing as how dopamine is not carried pre-hospitally) I wouldn’t be too keen on doing so because I am concerned about increasing his myocardial oxygenation mismatch, aggravating his infarction and possibly putting him into cardiac arrest.

    I wouldn’t give the President any adrenaline, nor would I intubate him and I’d only put an LMA in because it’s hands-free bag and mask and you don’t have to have somebody tied up continually providing a jaw thrust and holding the mask on his face.

    For the post-ROSC patient it is appropriate to anaesthetise, paralyse and intubate the patient so enable a faster path to the cath lab; after a quick stop off in ED to plug him into the hospital sedation/paralysis mix so he doesn’t wake up halfway through PCI because the vecuronium the ambo’s gave him wore off.

    • Minh Le Cong says

      thanks Ben. I dont understand why you say you would still give amiodarone because its in the guidelines, yet you would not give adrenaline, which is still in all the international , including NZ, resuscitation guidelines? Was that a mistype on your part?
      aMiodarone has never been shown to improve survival to hospital discharge from OHCA.

    • Minh Le Cong says

      I agree, Ben. Adrenaline in myocardial infarction is not a decision to be taken lightly! But we are not talking about the same situation, are we? We are talking about OHCA, usually from MI but not always. These patients are dead, no pulse, no signs of life. The heart has stopped pumping blood. Technically its a reversible cause , MI causing arrest. If you can clear the thrombus and reperfuse the heart then might get it going again. The point is if you dont get the heart going again or provide alternative means of perfusing the brain and organs, then there is certain death. Adrenaline is one thing you can do to try restarting the heart. Its not the only thing and should not be the priority.
      At the next OHCA you attend, you have a perfectly reasonable decision to make. Dont give adrenaline! If you are worried about getting into trouble by not following protocol, then just pretend to get IV or IO access and deliberately fail to gain access! So you can say you tried but could not get access to give adrenaline…thats as long as you believe that drugs are ineffective in OHCA management.
      I would suggest though that condemning others who choose to follow guidelines, is not reasonable, when you can choose not to.

  11. Ben Hoffman says

    I don’t think vasopressin (ADH) has any advantage over adrenaline biologically and my brief search of the evidence seems to support this (9,10).

    I initially thought it not beyond the realms of biologic possibility that if adrenaline has harmful effects on the heart through beta stimulation increasing myocardial workload and oxygen consumption then as V1/V2 agonist vasopressin may not have this same effect as it is not stimulating these receptors. I am most likely however not correct.

    (9) Volker Wenzel, M.D., Anette C. Krismer, M.D., H. Richard Arntz, M.D., Helmut Sitter, Ph.D., Karl H. Stadlbauer, M.D., and Karl H. Lindner, M.D. (2004): Vasopressor during Cardiopulmonary Resuscitation Study Group N Engl J Med; 350:105-11

    (10) Aung K, Htay T. (2005): Vasopressin for cardiac arrest: a systematic review and meta-analysis. Arch Intern Med;165(1):17-24.

  12. DocXology says

    Let us assume that a 10% improvement in survival to discharge is significant. How big a study would you need to show this occurs with adrenaline (acknowledging the great heterogeneity of OHCA patients)? Has that study been done? And if that study has not been done, what end-points should we use instead to evaluate its effectiveness?

  13. DocXology says

    Interesting that both the PACA and JAMA studies have one of the lowest reported OHCA modern survival figures (105) and less asystolic rhythms.


    Could it be possible that you aren’t going to show many intervention will benefit (including adrenaline) if you are predominantly treating dead people?

    The VF/VT subgroup had better odds with adrenaline but CI too wide (probably due to lack of power in the analysis).

    Maybe the lesson here is adrenaline is for resuscitation – not reanimation.

  14. DocXology says

    Edit above:

    Interestingly that both the PACA and JAMA studies have one of the lowest reported OHCA modern survival figures (10%) and higher proportion of non-shockable (especially asystolic rhythms) compared to another recently published paper;


    which had survival figures of > 10%.

    Could it be possible that you aren’t going to show many intervention will benefit (including adrenaline) if you are predominantly treating dead people?

    The VF/VT subgroup had better odds with adrenaline but CI too wide (probably due to lack of power in the analysis).

    Maybe the lesson here is adrenaline is for resuscitation – not reanimation.

  15. Brian Steiner says

    I am currently an MS-4 and going to be starting my EM residency at Yale in June. I’ve been listening to these pod casts and think they are amazing!

    Quick question about the practicalities of designing an RCT to study this question. The one thing that jumps out in my mind is how would you even consent subjects? I would think that this would be especially challenging as this study is challenging such a (for better or for worse) time-honored standard of care.

    I feel like I might be ignorant of some way around this issue so any help with this question would be awesome!

    • says

      It was all set up in Australia includind consent, waivers, IRB, etc. Then the EMS services that agreed to participate pulled out and the study became under-powered. I would say a trial like this could easily pass IRB. All of the trials have been negative to date, yet people are still doing it. This defines equipoise.


Speak Your Mind (Along with your name, job, and affiliation)