Beyond ACLS: Cognitively Offloading During a Cardiac Arrest

22 Sep
September 22, 2016

Beyond ACLSToday I am giving a talk at the 25th National Emergency Medicine Symposium by Kaiser Permanente in Maui, HI.  The focus of this talk was on how to cognitively offload our minds as we are running a resuscitation. ACLS provides us with a framework in treating adult victims of Cardiac Arrest (CA) or other cardiopulmonary emergencies. This helps get providers who don’t commonly deal with CA, to improve things, such as the quality of CPR, minimizing interruptions during CPR for pulse checks, and the timing/dosing of epinephrine. Emergency Medicine (EM) and the prehospital world are different than many environments in medicine. We get minimal information at the time of patient arrival while at the same time the disease process that is taking place has not quite defined itself.  We are constantly expected to acutely manage and resuscitate anyone who comes in our doors 24-7-365, many times without crucial information. Our job therefore should be to ensure coronary and cerebral perfusion are at their highest quality, but also simultaneously putting the pieces of the puzzle together to figure out why our patient is in CA. It can be very difficult to do both and many times we sacrifice one for the other. It is therefore important to cognitively offload ourselves during the resuscitation of our patients in CA and focus our attention on why they are in CA. As a disclosure for this lecture I did state that some of the recommendations made have evidence to support them and others are more theoretical and certainly up for discussion.

Lets Start with a Case:

A 64 year old caucasian male has a witnessed, out of hospital cardiac arrest (OHCA).  His past medical history is unknown. He receives bystander initiated CPR, intubated prehospital by EMS and brought to the ED with no pulse, last rhythm was PEA, and IV access was lost en route.

The Remainder of this Post Will Focus on going beyond ACLS and Cognitively Offloading During the Arrest in Regards to:

  • CPR
  • Access
  • Epinephrine
  • PEA Workup
  • Pulse Checks

Cardiopulmonary Resuscitation (CPR)

The 2015 American Heart Association/American College of Cardiology (AHA/ACC) published a new update for CPR quality in Circulation [1]. The key summary points for were:

  • Rate: 100 – 120 compressions/min
  • Depth: 2 – 2.4 inches
  • Recoil: Allow Full Recoil
  • Interruptions: Minimize Pauses

A review article by Cunningham LM et al [2], showed that even brief pauses in CPR can cause decreased coronary perfusion pressure (See figure below).   The decreased perfusion pressure does not immediately return to adequate levels at the reinitiation of CPR. Decreased coronary perfusion pressure actually lasts longer than the pause itself and takes time to gradually build pressure back up to a baseline level, which means that cardiac tissue is actually ischemic and not being perfused for a longer period than the pause itself. This is important because ischemic cardiac tissue is less likely to be successfully defibrillated in shockable rhythms.

CPR Pauses

It takes up quite a bit of mental bandwidth to ensure that CPR is performed well and there is already enough going on during the resuscitation, so is there a way that we can cognitively offload our minds?

There was a systematic review and meta-analysis of randomized clinical trials trying to answer the question of mechanical CPR vs manual CPR in OHCA by Gates S et al in Resuscitation 2015 [3]. The trial included 5 randomized clinical trials (RCTs) with over 10,000 patients with OHCA. Ultimately, the study found no difference in ROSC, Survival, or Survival with Good Neurologic Outcomes. None of the studies showed superiority of mechanical CPR, but also none of the studies showed inferiority to manual CPR.  In other words, mechanical CPR to date is no better or worse than manual CPR.

Clinical Bottom Line: Doing high quality uninterrupted CPR is important, but this is not the endpoint of resuscitation. Mechanical CPR is a great way to off load our minds.  We now don’t have to focus on the rate of compressions, depth of compressions, interruptions in compressions and instead focus on why our patient has had a CA.

Our Patient:  Our patient gets mechanical CPR initiated, but remember EMS lost IV access…


I recently asked the twitterverse, in a patient with CA and no IV Access, what is your go to access?  The results are below:

Access Twitter Poll

In my mind there are really only 2 sites for intraosseous (IO) access during CA: Humeral Head or Proximal Tibia.  Sternal IO has been shown to give a faster infusion rate than other sites, but may get in the way of CPR in CA [10]. The proximal humerus is closer to the heart and thus gets medications circulating faster than the proximal tibia.  On the other hand the proximal tibia is away from the head of the bed which is where most people are typically congregated for airway management and CPR.  Another way to think about the IO sites of access are:

  • Humeral Head and Sternal IO = Central Venous Catheter (Closer to the Heart)
  • Tibial IO = Peripheral Intravenous Access (More Distal to the Heart)

There was a RCT done by Reades et al in Ann Emerg Med 2011 [4], which is the largest RCT to date comparing humeral IO vs tibial IO vs peripheral IV (PIV).  This study included 183 patients with OHCA. Their primary outcome was 1st attempt success at access and a secondary outcome was time to access.

  • First Attempt Success:
    • Tibial IO: 91%
    • Humeral IO: 51%
    • PIV: 43%
  • Time to Access (Min)
    • Tibial IO: 4.6
    • Humeral IO: 7.0
    • PIV: 5.8
  • A Huge caveat to this trial, is that this was a prehospital study, and not a study in the ED setting.  Secondly there may have been some issue with familiarity in the placement of humeral IOs as is evidenced by the time to access (7 minutes?!?).

Clinical Bottom Line: When having difficulty getting IV access during CA, humeral IO has a higher success rate and faster time to access being achieved vs IV

Our Patient: Receives two 15 gauge Proximal Humerus IOs and several people in the room begin to call for epinephrine (adrenaline)

Epinephrine Dosing/Timing

What does the twitterverse feel is the correct dosing and timing of epinephrine in CA:

Epi Twitter Poll

The 2015 Adult Advanced Cardiac Life Support (ACLS) Guidelines  recommend that it MAYBE reasonable to give epinephrine 1mg IV q3 – 5 minutes maybe reasonable [5].  They give this a Class IIb recommendation, which is another way of saying weak evidence (Benefit ≥ Risk).

Many studies looking at outcomes with dosing epinephrine as recommended by ACLS look at surrogate outcomes, specifically ROSC.  In CA, ROSC is the first step, but what we really care about is neurologically intact survival.  To date, there is no study that has shown epinephrine in CA improves neurologically intact survival.

Evidence for Epi

Epinephrine does produce some beneficial effects in CA of increased coronary perfusion pressure and cerebral perfusion pressure during CPR, but in bolus doses may increase myocardial oxygen consumption and therefore increase subendocardial ischemia.

Now I will be the first to admit, it can cause undue unrest in the dynamics of your team during the resuscitation if you recommend skipping the epinephrine entirely without a very lengthy explanation.  Instead, I have two other options to consider that will remove the distraction of remembering epinephrine every 3 – 5 minutes:

  • Option 1: The Dirty Epi Drip – Push 1mg (i.e. 1 ampule) of code cart epinephrine into 1,000mL of NS which gives you a final concentration of 1mcg/mL.  You can run this wide open with a pressure bag until, a) your patients hemodynamics stabilize, or b)an epinephrine drip is ready for use.
  • Option 2: Quarter Dose Epi Drip: Run epinephrine at 0.5mcg/kg/min.  In a 100kg patient this would be the equivalent of 250mcg over 5 minutes (i.e. quarter dose)

Running an epinephrine drip allows for several things to occur:

  1. Cognitive Offload: No one has to remember, “is time for the next epi,” every 3 – 5 minutes
  2. Minimizing Physiological Lability: Continuous dose of epinephrine allows for smoother coronary and cerebral perfusion pressure without increased myocardial oxygen consumption and subendocardial ischemia (This is my contention, but no studies comparing epi drip vs standard 1mg epi dosing q3 – 5min in CA)
  3. Avoidance of post ROSC Drop in Blood Pressure: Titrate the drip to the dose needed

Clinical Bottom Line: No study to date has shown that the ACLS recommended 1mg epinephrine q3 – 5 minutes improves neurologically intact survival in CA.  Maybe a better way would be to run a continuous infusion of lower dose epinephrine.

Our Patient: Starts with a Dirty Epi Drip and is transitioned to a Quarter Dose Epi Drip

PEA Workup

Approximately 1/3 of patients in CA have PEA and it is well known that patients with PEA as an initial rhythm have a worse survival than patients with a shockable rhythm [6].  If you are like me you learned all about the H’s and T’s that could potentially cause PEA, but even now try and list all of them.  Now imagine yourself trying to go through this differential while resuscitating a patient in PEA CA.  Pittman et al published an interesting article in Med Princ Pract 2014 [6] which focused on the few causes of PEA that we can rapidly and effectively correct instead of memorizing the traditional H’s and T’s.  The steps in their algorithm are simple:

  1. Determine if the PEA is narrow (QRS Duration <0.12) or wide (QRS Duration ≥0.12)
  2. Narrow-complex PEA is generally due to mechanical problems caused by right ventricular inflow or outflow obstruction
  3. Wide-complex PEA is generally due to metabolic problems

Narrow Complex PEA

Wide Complex PEA


  1. Instead of randomly listing 10 – 13 causes of PEA by memorizing the H’s and T’s this new algorithm categorizes PEA into the etiologies with the highest likelihood and clinical relevance.  Even better it provides specific treatment recommendations for those etiologies.
  2. 4/5 causes of narrow-complex PEA can be ruled out with bedside ultrasound.


  1. This algorithm has not been tested the trauma setting
  2. This algorithm has not been systematically tested for resuscitation outcomes vs more traditional H’s and T’s
  3. Pulmonary Embolism and Acute Myocardial Infarction can cause wide-complex PEA (i.e. RBBB and LBBB)

A more in-depth review of this new algorithm for PEA can be found HERE.

Clinical Bottom Line:Using the new classification system of PEA simplifies the working differential and initial treatment approach in conjunction with bedside ultrasound, but still needs to be tested for resuscitation outcomes.

Our Patient: Our patient has a narrow complex PEA.  Bedside ultrasound finds a dilated RV > LV and concern for a massive PE.  The patient receives systemic tPA and has a change in the rhythm on the monitor.  Someone calls out, can you feel a pulse?

Pulse Checks

What does the twitterverse feel is the best way to check for ROSC in CA? The results are below:

ROSC Twitter Poll

Ochoa FJ et al published a trial in Resuscitation 1998 [7] in which ED physicians, ICU physicians, and nurses tried to identify a carotid pulse in a healthy male volunteer with normal blood pressure.  43.1% of health professionals required >5secs to detect the carotid pulse and 4.3% required >10secs.

So is the manual pulse check dead?  Maybe.  I certainly think we have better technologies to assist in determining ROSC that don’t require pauses in CPR:

  1. Arterial Line:  The benefit of the arterial line is we get direct feedback on cardiac output, but the downside is it takes time to place
  2. Bedside Transthoracic Echocardiogram (TTE): This gives the provider real time feedback of what the heart is actually doing (i.e. Not feeling a pulse manually, when in reality the heart is weakly contracting – Pseudo PEA), can also give real time CPR quality feedback, and potentially has the ability to predict the chances of ROSC [8].  The downside is we still have to stop CPR to look for cardiac activity, but if done correctly should not take >10secs to see if the LV is contracting or not. Blyth L published a systematic review of patients in CA in Academic Emergency Medicine 2012 [8]. This included 8 studies with 568 patients with OHCA and who had a bedside echocardiogram performed. At 20 minutes if there was cardiac motion on echo 51.6% of patients achieved ROSC and 48.4% did not achieve ROSC.  On the other hand, if there was no cardiac motion at 20 minutes on bedside echo only 2.4% achieved ROSC while 97.6% did not. One major issue with this study is ROSC was used as a surrogate marker, and neurologic outcomes were not assessed.
  3. End Tidal CO2 (EtCO2):  For me, this is where the money is at. Ahrens T et al looked at EtCO2 in OHCA as a prognostic indicator of outcomes [9] and found that a EtCO2 <10 – 15mmHg at 20 minutes predicted a 0.8% chance of survival. The studies evaluating the use of EtCO2 in CA have been in intubated patients, observational studies, and to date none have investigated survival with intact neurologic outcome. The 2015 AHA/ACC guidelines actually state that in intubated patients, failure to achieve an EtCO2 of >10mmHg by waveform capnography after 20 minutes of CPR maybe considered as one component of a multimodal approach to decide when to end resuscitative efforts, but should not be used in isolation (ClassIIb)[5]. Currently there is no optimal duration of resuscitation before the termination of resuscitation (TOR).
    • Lungs require ventilation and perfusion to work.  If we keep ventilation constant with a bag valve mask or ventilator then perfusion is the only factor not accounted for.  Therefore an increase in cardiac output should increase the EtCO2 reading.  However, anything that interrupts ventilation will cause false readings of EtCO2 (i.e. Dislodged Endotracheal Tube, Respiratory Causes of CA, and Pulmonary Embolism)

Clinical Bottom Line: Stopping CPR for pulse checks is not accurate at predicting ROSC and potentially detrimental to neurologic outcomes of patients. Adjuncts such as bedside echocardiography and EtCO2 add to a multimodal approach in predicting ROSC as well as giving useful information in decisions of termination of resuscitation.

Our Patient: Had EtCO2 placed and after thrombolytics had a jump in his EtCO2 to 30mmHg and found to have ROSC on bedside Echo

Critical Take Home Points to go Beyond ACLS and Cognitively Offload During Resuscitation Efforts of Cardiac Arrest:

  • CPR: Mechanical CPR > Manual CPR
  • Access: IO Access > IV Access
  • Epinephrine: Epinephrine Drip > Epinephrine Bolus
  • PEA Workup: Ultrasound > H’s & T’s
  • Pulse Checks: EtCO2 + Bedside TTE > Manual Pulse Checks

For More on This Checkout:


  1. Kleinman ME et al. Adult Basic Life Support and Cardiopulmonary Resuscitation Quality: 2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2015; 132 (18 Suppl 2): S414 – 25. PMID: 264722993
  2. Cunningham LM et al. Cardiopulmonary Resuscitation for Cardiac arrest: The Importance of Uninterrupted Chest Compressions in Cardiac Arrest Resuscitation. Am J Emerg Med 2012; 30(8): 1630 – 8. PMID: 22633716
  3. Gates S et al. Mechanical Chest Compression for Out of Hospital Cardiac Arrest: Systematic Review and Meta-Analysis. Resuscitation 2015; 94: 91 – 7. PMID: 26190673
  4. Reades R et al. Intraosseous Versus Intravenous Vascular Access During Out-of-Hospital Cardiac Arrest: A Randomized Controlled Trial. Ann Emerg Med 2011; 58 (6): 509 – 16. PMID: 21856044
  5. Link MS et al. Part 7: Adult Advanced Cardiovascular Life Support: 2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2015; 132 (18 Suppl 2): S444 – 64. PMID: 26472995
  6. Littmann et al. A Simplified and Structured Teaching Tool for the Evaluation and Management of Pulseless Electrical Activity. Med Princ Pract 2014; 23 (1): 1 – 6. PMID: 23949188
  7. Ochoa FJ et al. Competence of Health Professionals to Check the Carotid Pulse. Resuscitation 1998; 37: 173 – 75. PMID: 9715777
  8. Blyth L et al. Bedside Focused Echocardiography as Predictor of Survival in Cardiac Arrest Patients: A Systematic Review. Academic Emergency Medicine 2012; 19 (10): 1119 – 26. PMID: 23039118
  9. Ahrens T et al. End-tidal carbon dioxide measurements as a prognostic indicator of outcome in cardiac arrest. Am J Crit Care 2001; 10(6): 391 – 8. PMID: 11688606
  10. Paley J et al. Introsseous Infusion Rates Under High Pressure: A Cadaveric Comparison of Anatomic Sites. J Trauma Acute Care Surg 2015; 78 (2): 295 – 9. PMID: 25757113

Post Peer Reviewed By: Anand Swaminathan (Twitter: @EMSwami)

The following two tabs change content below.

Salim Rezaie

Emergency Physician at Greater San Antonio Emergency Physicians (GSEP)
Creator & Founder of R.E.B.E.L. EM
15 replies
  1. Rahul says:

    Good one.
    If you want even more offloading – just tell poeple: Adrenaline every 4 minutes (every other cycle). Why stick with the confusing 3-5 mins?

  2. Willem Jooste says:

    Hi Salim,

    Excellent post, as always. Just a question for clarification. You speak of using TTR as alternative to pulse checks. I understand the usefulness when assessing the differential for a PEA, but TTE cannot differentiate between PEA and ROSC, can it? Which pulse checks are ultimately for.

    Thanks again for all the great work!


    • Salim Rezaie says:

      Hey Willem,
      TTE helps differentiate between pulse and no pulse. You can clearly see LV function on TTE, so yes if you see LV function then you have ROSC and that patient needs inotropic support, and if no LV function, that patient needs CPR. Appreciate your reading and your question.


  3. Scott Carpenter says:

    Nice write-up.

    It’s my understanding that hypoxia is at least a contributing factor in 50-60% of in-hospital refractory codes (PEA especially.) Based on this, I would like to see hypoxia incorporated into the “new” PEA algorithm. My initial thought (total guess) is that this would tend to cause a wide complex PEA. Your thoughts appreciated!


    • Scott Carpenter says:

      Sorry for multiple posts-
      On another note, I completely agree that epi every 3 to 5 minutes tends to be confusing. I usually suggest the following sequence-
      shock, no drug (The patient’s best shot at survival)
      if no ROSC, shock, epi, then shock, amio, then shock, repeat amio150 and it’s been 4 minutes since the last epi, so double down on drugs during that sequence. If they havent started thinking H’s and T’s by then, it’s high time!


      • Salim Rezaie says:

        Hello Again Scott,
        No worries on multiple comments on the post. I have to humbly disagree here…the patients best chance at neurologically intact survival, is Early, High Quality CPR and Defibrillation. These are the only two things that have been shown to improve neuro intact survival. Epi in bolus form of 1mg IV has not been shown to be beneficial for this endpoint, and this may be due to increased arterial vasoconstriction to the cerebral arteries, and coronary arteries. Instead a smoother drip that is titrated seems to make more physiological sense, but as stated in the post, this is just my contention with no evidence to support it. Hope this helps.


    • Salim Rezaie says:

      Hello Scott,
      Hopefully one of two things will happen during the code:

      1. The patient gets intubated which should resolve hypoxemia with FiO2 100% & PEEP
      2. The patient gets NC at 15LPM and BVM at 15LPM which should also help resolve hypoxemia with FiO2 100%

      Either way, you will have a pulse ox on the patient, so if you know they are hypoxemic, the initial vital signs would show this, but even if you didn’t, the patient is getting O2 regardless.


  4. Mario Rugna says:

    Thanks for the great resource. I have some questions. What you think about refractory fv/tv cardiac arrest. Does beta blockers and dual sequential def have a role in that? Do you think VA ECMO is a chance in selected patients or just a waste of time and respirces. Thanks again. Mario Rugna

  5. RW says:

    Wow. 7 minutes to obtain proximal humerus IO access is an insane amount of time. Setting up should take 1 or 2 minutes and the actual procedure can be done easily in less than 60 seconds. Literally! Practitioners need some major training and practice if it’s taking 7 minutes. It’s hard for me to imagine someone taking that long. The proximal humerus site allows medications to get to the right atrium of the heart in less than 3 seconds. It’s central line performance. The proximal tibia site isn’t preferred in an arrest because by comparison medications will take 1- 2 minutes to get to the heart. IV in the EJ or proximal humerus IO in an arrest for best performance.

    • Salim Rezaie says:

      Hey Rusty,
      Agreed. My thoughts are two things….providers are not trained and too many people at the head of the bed. Literally, takes <2min to do and is my go to in tough IV access situations in an arrest.


  6. Ariel says:

    Great post.
    I am a PEM provider, so I’m always trying to pay a bit closer attention to dosing.
    Is there any reasoning/explanation you can provide about how you came up with the amount of epi to put in the dirty epi drip, so that I might come up with a way to translate the dosing for a peds patient?

    • Salim Rezaie says:

      Hello Ariel,
      First let me start off by saying, I have no data basis for this, but also know the data for 1mg epi q3-5 min is not robust in its favor. The rollercoaster of give 1mg and watch it wear off only to give another mg never made physiologic sense to me. So, in adults we generally give 1mg epinephrine q3-5min and start with a 1L bolus of crystalloids, so it seemed to make sense to me to mix the two together and run wide open, knowing that I would not be giving more than 1mg of epinephrine in 3 – 5 min, but also knowing that I would be giving some epi to hep peripherally vasoconstrict without increasing afterload too much and increasing oxygen demand on the heart. In pediatric terms I guess this would be the equivalent of 0.01mg/kg of 1:10,000 epi with a max dose of 1mg in 20cc/kg of crystalloid with a max of 1L. Hope this helps.



Trackbacks & Pingbacks

  1. […] Salim Rezaie has some interesting thoughts on cognitive offloading during ACLS. […]

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.