Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

What is the Best Way to Manage Cardiac Arrest According to the Evidence?

There is an excellent review article by Dr. Bentley Bobrow and Dr. Gordon Ewy on the best management of sudden cardiac arrest from the bystander to the ICU (Intensive Care Unit).

They point out something that we tend to resist learning. Cardiac arrest that is not due to respiratory causes does not need respiratory treatment. A person who is unresponsive and gasping is exhibiting signs of cardiac arrest, not signs of respiratory problems.

Except in newborns, gasping or agonal breathing is a common sign of cardiac arrest, occurring in slightly more than 50% of patients with primary cardiac arrest.22-25 [1]


Gasping does not mean alive and well. Gasping means dead and having a good chance at resuscitation. Unresponsive and gasping means there is a need for compressions.

If adequate chest compressions are promptly initiated, the patient will continue to gasp.23 [1]



Of interest is that only a minority of individuals with noncardiac arrest received CO-CPR.35 In Arizona, the public was generally capable of recognizing respiratory arrest, where chest compressions and assisted ventilations were recommended.[1]



It probably has less to do with taking away the ventilation, than with making the compressions continuous and high quality, but ventilations do decrease blood return to the chest and increase the likelihood of vomiting (regardless of what has been eaten), so there are benefits from removing the ventilations.


Passive oxygen insufflation means just putting a mask over the patient’s mouth and nose and allowing oxygen to be delivered passively. The rise and fall of the chest, due to compressions, and diffusion will allow for all of the oxygenation the patient will need.

Standard CPR (Std CPR) means alternating compressions with two ventilations every 30 compressions. Standard CPR is clearly not what we want to do, unless we want to keep patients from being resuscitated.

The problem is that the vast majority of physicians have no idea what the survival rate of patients with OHCA is in their area. This needs to change if major progress is to be made.[1]


Many of us do not know the results of what we do, so it is not surprising that a lot of EMS treatment is mythological.

Medicine is a field that encourages superstition. Patients provide intermittent reinforcement, which may be the most effect means of creating superstitions. Intermittent reinforcement?[2]

The only way to know the effectiveness of your Emergency Medical System is to know the survival of patients with OHCA and a shockable rhythm. If it is less than 38%,they should be encouraged to institute CCR and reevaluate the results.[1]


Maybe you are already doing better than 38% walking out of the hospital, then you are probably already using continuous compressions and passive oxygen insufflation. If you are not, then you need to improve your patient care.


[1] Cardiocerebral Resuscitation: An Approach to Improving Survival of Patients With Primary Cardiac Arrest.
Ewy GA, Bobrow BJ.
J Intensive Care Med. 2014 Jul 30. pii: 0885066614544450. [Epub ahead of print]
PMID: 25077491 [PubMed – as supplied by publisher]

[2] Intermittent reinforcements

Pigeons experimented on in a scientific study were more responsive to intermittent reinforcements, than positive reinforcements.[16] In other words, pigeons were more prone to act when they only sometimes could get what they wanted. This effect was such that behavioral responses were maximized when the reward rate was at 50% (in other words, when the uncertainty was maximized), and would gradually decline toward values on either side of 50%.[17] R.B Sparkman, a journalist specialized on what motivates human behavior, claims this is also true for humans, and may in part explain human tendencies such as gambling addiction.[18]


Ewy, G., & Bobrow, B. (2014). Cardiocerebral Resuscitation: An Approach to Improving Survival of Patients With Primary Cardiac Arrest Journal of Intensive Care Medicine DOI: 10.1177/0885066614544450



  1. Adenosine


    Love this blog, stop by from Research Blogging regularly

  2. Dearest Rogue,

    I’ve been an avid reader of your blog (as well as plenty of others), for quite some time, and I have a question for you, oh-great-and-experienced-wise-one.

    I’m a relatively new EMT in a rural area and just had my first cardiac arrest. I was the driver. 50 yo black male, unexplained collapse outside McDonald’s, CPR in progress upon arrival. Police officers are doing 30:2. We arrive and switch to Pit Crew. Pads are placed, rhythm is not shockable. Upon opening the airway, we find it filled with vomitus. My partner suctions, and then begins bagging, o2 cranked all the way up. The patient is backboarded and loaded. In the back, a Lucas is placed. The medic places a line. Our fire department assistance, also one of our most experienced drivers, takes over driving to the hospital. The medic places a line and we go en route. En route, the medic intubates. somewhere during this process, the patient is suctioned again but BVM is interrupted for about fifteen minutes. We are scolded for walking into the ER without continuous BVM. Time of death is called about ten minutes after our arrival.

    Did we eff up? According to your article, more ventilation than passive is unnecessary. So, since the patient was getting 15 liters through the BVM, even if it wasn’t squeezed, that was okay, right? Or should we have been bagging, especially considering the airway obstruction? Should he have been intubated sooner, like on scene? Should an oral airway have been placed?

    The general consensus was that we did well, though we could have handled annoying bystanders better and we could have gotten there quicker. But talking with everyone on the call, including our incredibly experienced medic, there really wasn’t anything we could have done differently that would have changed the outcome. I still think we should have been bagging, since we spent the remainder of the ride twiddling our thumbs and completely forgetting to bag, which is very noobish and unprofessional of us. But I get this eerie feeling that we didn’t do everything we could have to save his life. My department doesn’t have a good setup for after-action reports (as in, it’s nonexistent), so I thought since your post here kind of speaks to the situation, I’d ask your opinion.

    Thanks in advance for any wisdom,

    • Carly,

      Thank you for your kind words.

      Any time you need to choose between ventilations and compressions, the compressions should be chosen, even with the current CPR guidelines.

      So, since the patient was getting 15 liters through the BVM, even if it wasn’t squeezed, that was okay, right?

      With a nonrebreather mask, 15 liters is providing passive oxygenation, but with a BVM there is a valve between the patient and the exhaust port, so the oxygen is not going to the patient – unless the patient is breathing spontaneously and creating enough negative pressure to draw oxygen past the valve or the bag is being squeezed. Placing the BVM over the mouth and nose with a good seal would be depriving the patient of oxygen if the patient is not breathing and the bag is not being squeezed. Do the chest compressions cause enough negative pressure to draw oxygen past the valave? I don’t know.

      This video should help to show how the valve works.


      Should he have been intubated sooner, like on scene?

      I am not a fan of intubation in cardiac arrest, so I do not think that intubating sooner is the answer, but it is in the protocols. With vomit in the airway, it may be worthwhile to intubate.

      Unless there is some unusual reason to get off scene quickly (threat of violence, patient likely to need treatment only available at the ED, . . .) everything should be done on scene. If the patient does not respond, then we should be pronouncing patients on scene.

      Pennsylvania protocols do not encourage moving patients or ventilating patients early.

      2013 PA BLS protocol 331A General Cardiac Arrest – Adult pages 54-57 – PDF of PA BLS protocols.

      B. For efficient “pit crew” style care, the EMS agency medical director should establish whether any ventilation is given during initial compression cycles. If BVM ventilation is used, compressions should not be interrupted when giving a ventilation every 15 compressions.

      2013 PA ALS protocol 3031A General Cardiac Arrest – Adult pages 18-22 – PDF of PA ALS protocols.

      There is not much different, but the medic should be making sure that care, such as BVM bagging, is delivered according to protocol, when there is a medic on scene. I have seen intubated patients who have been given paralytic medicine, and not able to breathe spontaneously, not being bagged even in university hospitals and trauma centers – until someone speaks up and mentions that the patient is not breathing (and not supposed to be, but is supposed to have someone breathing for them with paralytics on board).

      Footnote 2 discourages transport before 4 cycles of 200 compressions.

      2. Do not move or package patient for transport at this time. Chest compressions are much less effective during patient transportation/movement, and any possible interventions by medical command will be less effective without optimal CPR.

      Field termination would not be permitted by BLS protocols in this case because the arrest appears to have been witnessed, but ALS can call medical command for permission to terminate on scene.

      7. AHA Guidelines suggest that the following are reliable and valid criteria for BLS termination of resuscitation. Before moving the patient to the ambulance, consider contact with medical command for orders to terminate CPR in the field if ALL of the following apply:
      a. Arrest not witnessed by EMS personnel, AND
      b. No return of spontaneous circulation/ pulse (prior to transport), AND
      c. No AED shock was delivered (prior to transport).

      My department doesn’t have a good setup for after-action reports (as in, it’s nonexistent), so I thought since your post here kind of speaks to the situation, I’d ask your opinion.

      Pennsylvania requires some sort of QI, which should include debriefing/run review, but future protocol revisions may add a requirement for

      C. The EMS agency, overseen by the agency medical director, must perform a QI review of care and outcome for every patient that receives CPR.
      1. The QI should be coordinated with involved ALS agency and receiving hospital to include hospital admission, discharge, and condition information. This EMS agency QI can be accomplished by participation in the Cardiac Arrest Registry for Enhanced Survival (CARES) program through the ALS agency.
      2. The QI should be coordinated with local PSAP/dispatch centers to review opportunities to assure optimal recognition of possible cardiac arrest cases and provision of dispatch-assisted CPR (including hands-only CPR when appropriate).

      An important part of improving quality is reviewing what was done on the call.

      Think about how things could have been done differently.

      Discuss things with other people on the call without blaming anyone for anything.

      Many AEDs come with audio recorders that are supposed to be used for purely quality improvement and not for discipline. After use, the tapes are erased, so they are not available as evidence in any kind of law suit. This is one way for management to provide feedback, but only if they do not attempt to use the recordings as an excuse for any kind of punishment, write-ups, or anything negative.

      When we review everything that actually happened, we are provided with evidence that our memories are not as good as we thought.