Are Extraglottic Airways Harmful in Cardiac Arrest?

Are we creating a blockage of blood flow to the brain?

This article has created quite a stir in the resuscitation community:

Impairment of carotid artery blood flow by supraglottic airway use in a swine model of cardiac arrest.  Segal N, Yannopoulos D, Mahoney BD, Frascone RJ, Matsuura T, Cowles CG, McKnite SH, Chase DG.  Resuscitation. 2012 Mar 28.


Are EGAs harming carotid blood flow during CPR and therefore making neurological outcomes worse? At least in pigs, this is worrisome. Human data to follow.

What am I going to do with this? At least for now, keep using LMAs (the device associated with the least problems), but now I will check cuff pressure to make sure it stays below 40 cm H20

Additional Reading:

Andy Neill has an amazing Anatomy for Emergency Medicine Post on this very issue

Now, on to the wee…

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. minh le cong says

    awesome wee! Thankyou. no where else would you get such cutting edge international resuscitation opinion.

  2. says

    Great interview! Interesting paper. Makes you think. What I’m going to do is continue using igel. Not sure about the seal, but It kind of makes sense. It doesn’t have a balloon, and I can put it in much quicker. It is a question of balance.

    • says

      I think with no proof whatsoever, that the Igel is probably safe as I think the seal would be at minimum pressure. Dan Cook is also a developing an LMA that deflates during exhalation.

  3. says

    One question for whomever… Which type of LMA device were they using for the EGA study in regards to CBF? My service is currently trailing the LMA Supreme and Fastrach against the King LTSD. I did notice that it dose take more air to inflate the LMS Supreme over the LMS Fastrach due to balloon sizes. Our trial has ended and I am about to write a pros and cons comparison regarding the devices mentioned above. So this can possibly influence final decisions.


    Matt King FP-C

  4. Curtis Farrell says

    Dr. Weingart,

    I saw this presented at NAEMSP. What was most convincing about this article was that the next speaker was Dr. Henry Wang, who presented his paper from the ROC data comparing ETT vs SGA and survival. ETT wins by a long shot. Coincidence, I think not.

    • says

      I don’t know the answer to this one, but on the reanalysis to the ROC data paper: this was a observational trial for airway choice. The medics chose SGA for a reason as opposed to intubating them. Despite multi-variate analysis, I don’t think this paper really answers the question.

      • Chris Wearmouth says

        Quite a few of these observational studies seem to show no airway intervention>ETI>SGA. I think this is most likely due to the nature of EMS responding to a cardiac arrest. The ‘one shock into rhythm’ patients won’t receive any advanced airway interventions and will obviously do better, but not because someone explicitly chose not to stick a tube down their throat. Likewise, most paramedics will attempt ETI first, only progressing to SGA as a ‘rescue device’ or if they predict a difficult intubation. This means that SGAs will be reserved for those patients who have had a longer down time, or are more difficult to mange e.g. unable to intubate. It’s therefore unsurprising that in an observational study SGAs will have a poorer survival to hospital discharge! We need a RCT comparing SGA to ETI in the same group of paramedics in the same population, it’s the only way to conclusively prove one way or the other.

  5. says

    We still do not have any evidence of survival benefit from any type of ventilation in the initial management of cardiac arrest, but we debate the best way to provide this questionable treatment.

    This paper is just more evidence of how little we know about the harms we are causing and the benefits we unjustifiably claim we are causing.

    We should be limiting ourselves to treatments with demonstrable benefit and only adding other treatments in ways that control for the variables.

    Currently, our uncontrolled variables include –

    ventilations (and all of the variations of ventilating),

    various drug delivery routes (various IVs, IOs, and central lines, or just squirting the drugs down the tube based on what we do with some people who have healthy circulation to their lungs),

    epinephrine (and all of the possible doses and dosing intervals),

    vasopression (the same),

    all of the possible antiarrhythmics, their doses, and their possible combinations and intervals,

    whatever might be close to its expiration date,

    whatever might be a personal favorite based on anecdote,

    whatever somebody claims “couldn’t hurt,”

    and whatever else just seems to fit the mood.

    Modern resuscitation research is about as controlled as can be expected with the current state of unreasonable optimism-based medicine.


  6. Don Diakow says

    Good info. We were in discussion of this yesterday as we trained new recruits on Supra-glottic devices. One question…….if you don’t have access to a cuff pressure measurement device and one is simply inflating i.e.: King tube according to the coloured 100ml syringe (colour markings on the syringe synonymous to the colour of the 15 ml adapter on the King) anyone know if that is close to 40cm/H2O pressure?

  7. stefan mifsud says

    i believe till conclusive results regarding safety use of supraglottic airway during cardiac arrest are published , we shall refrain from risking the brains of our patients. Most bvm studies did better that other devices in respect to long term survival.

    • says

      If you truly want to go in this direction (and you v. well may be right), Ben Bobrow’s protocols in AZ of using just a NRB (I would sub in a nasal cannula at 15 lpm) may be the way to go. If you do use the NC, I would add in a few BVM breaths a minute.

  8. Lungs says

    Is there any indication that it decreases positive neuro outcome? Ok so it looks like it decreases blood flow in a pig. Are we sure flooding oxygen rich blood into a hypoxic brain is a good thing?? Or should we cool the brain then get some oxygen rich blood back into it?

    • says

      At this stage, the amount of blood hitting the brain with conventional CPR is quite low. To cut it off entirely, based on all we know, seems to be very bad for neuro outcome–but this is extrapolated, so your question is a good one.

  9. says

    I feel like this tells us little. First let me say I’m all about using ET tube. Having said that most of what has been found in the last two year (practical application of AHA standards) shows that prolonged intubation time are know to have a direct impact on survival. As we now know it is because of the effect it has on compressions. I think the information in this study is badly timed. As we are working very hard to keep EMS focused on aggressive compressions this seems to raise doubt. I know we are saving more lives with pump hard fast and deep. We are just starting to get everyone onboard and it is make a difference in the field. I find this information no more that curious. I have read all of the study I can find. And in this PIG study variables are important. As we know from combat wounds head angle effects carotid artery position and blood flow. What was head angle and does it translate. What size and brand of each airway. This warrant more study, but that’s all. It took 50 years of use and 25 years of careful study to decide we were spending too much time on airway. I would say nothing should change with this study. Lets see human study and get more detailed angled and airways used. Dave

    • says

      nobody wants to return to the days where we stop compressions for even 1 second to allow intubation. If there are intubation advocates after this study, then the tube must be passed without interruption of compressions. Airways used are listed above.

  10. minh le cong says

    folks, its challenging to keep up with this all is it not!
    The evidence base is poor and not all that helpful to honest.
    Even Dr Wangs research is contradictory at times, earlier work indicating worse outcome with prehospital ETI, now this latest work that indicates a benefit.Segals work with the pigs is disturbing but insufficient to be practice changing at this moment. Who knows, maybe all the epinephrine studies in OHCA did not show a benefit because of all the SGAs placed limiting blood flow to an ischaemic brain? Paradoxically in a possible future we may return full circle to ETI and epinephrine when evidence emerges that when we combined therapeutic hypothermia, brain injury was reduced post ROSC?
    however you need to remain pragmatic. if you cannot BVM or ETI, dropping a SGA in during resus is reasonable. the latest resuscitation science indicates chest compression only CPR maybe not as good as we think

    the longest recorded successful cardiac resuscitation

  11. says

    Great podcast & discussion above. I think this study is certainly thought provoking, but I think it should be far from practice changing. Even in the discussion in the comments here, multiple knowledgable people are dismissing the results from a huge observational real world study (which I do as well- we don’t know why some got tubed and some got SGAs) while in the same breath considering changing management of real patients based on a physiology-based outcome measure on 9 animals of a different species. And while their main study graph is scary looking, the big dips are for the combitube; the dips for LMA are pretty small. Given the time it takes to place an ETT vs LMA, the impossiblity of getting good positioning to tube an arrest, the near necessity of breaks in compression to get a good view, I’m gonna stick with LMA with as little cuff air as possible (which is what I had been trained for good seals with AirQs regardless).

  12. minh le cong says

    I did not finish sorry.
    the longest recorded successful resuscitation of OHCA was here

    Note the pivotal role of capnopgraphy. For those who delve into the story in detail, it was winter time and snowing I believe so an element of hypothermia may have divinely assisted this mans near full recovery.
    Note BVM was performed mainly then converted to ETI by HEMS crew.
    MY Lord, even epinephrine was used then amiodarone!
    He is just lucky you shout back? Well if someone had not kept up to date with the role of capnography in cardiac arrest resusctiation, in other words, pay attention to the ebb and flows of the resuscitation science, this man would have been pronounced dead long before 96 minutes were up,
    Now if I can measure capnography via LMA, thats good enough for me, but pay respect to the pressure in the cuff. thats what I hae gleaned from the interview with Segal, If I can convert to ETI without stopping compressions, even better. Thats where the stuff that Jim DuCanto and tubing via LMA can be helpful here.

  13. Nicolas SEGAL says

    Hello guys
    Thanks for listening the podcast
    Here is the answer to some comment.
    This is a provocative study, it is not done to change our practice for the moment. We must keep doing what we are doing for the moment, good CPR, short no flows and low flows, cooling. A human study is actually being done that will give more information. This study shows that we should not take for granted none prove science. We know that we can insert SGD faster and without stopping CPR, it does not prove that it improves long term survival.
    I continue to use the LMA here in France. I have the advantage that the King tube and the Combi are not use here so I don’t have to choose. However here in France we do not have paramedics, we have Emergency MDs in the tier two ambulances, MDs with (more training and) more practice (the key element). Most of us insert several ETT per week; I have used a LMA only twice in 3 years.
    Answer to Matt King: We use an old model of LMA because of the anatomy of the pig, however in all the other model of LMA the cuff have the same design which means probably equivalent results.
    Answer to Curtis Farrell: we knew the result of Dr Henry Wang and yes it is convincing. We don’t have a clear proof but several small elements in favor of the results of our study.
    Answer to Ivor : We have not tested the Igel, I will no extrapolate for the result with this device.
    Answer to Rogue Medic : I completely agree with you, we must debate, do more study and try to find a final answer to the ventilation question during CPR.
    Answer to Lungs : we do not have final answer for long term outcome, however Dr Chase (senior author from this paper) have some partial result (monocentric, small number) which MAY look bad. Of course those results cannot give any type of final answer and are not publish.
    Answer to David Shrader : The idea of this study is to be provocative. Stopping CPR for any reason is bad. But we should not do the opposite and say we can insert a SGD without stopping CPR so they are good. We do not have any proof that they increase long term survival. We should keep using them because they look good and we must prove that they are good or bad. I do not have any conflict of interest, if we prove that SGD save neuro outcome I am happy, if we prove the opposite I am happy, I just want/need proofs. This study change nothing for the moment except that will try to look for an answer.
    Answer to minh le cong: I personnaly do not believe that epi is the answer to CPR (check my papers on pubmed). I do agree if you cannot BVM or ETI, dropping a SGA in during resus is reasonable.

    • Minh Le Cong says

      Nicolas Segal, merci Beaucoups! I enjoyed your podcast and comments here. well written and argued. COuld I ask your views /opinion on the Impedance Threshold Device and the Active Compression/Decompression device in CPR?

      • Nicolas Segal says

        Before giving my opinion I prefer to remember you that I do not have any financial conflict of interest however I have done several paper with Doctors Yannopoulos and Lurie and Dr Plaisance is my PhD director so you may consider that I have a scientific conflict of interest.

        I am strongly for the ITV and the ACD in particular when use together. But you need to remember that they must inside a perfect chain of survival.

        I have seen the results of both the ITD and the ACD on hundreds of pigs, the hemodynamic effect is really amazing. And I saw a lot of our piggies alive and well at 24 and 48h.

        I have discuss with several EMS directors (including RJ Frascone, Brian Mahone and Tom Aufderheide which was first authors for both the ROC-PRIMED and the ResQTRIAL) and I am more then convinced that they do save life.

        If I have a cardiac arrest, I hope both of them will be use on me.

        Dr Lurie do several time a year a lab demonstration of the ACD + ITD (and several other stuff including the one presented in EMCrit Podcast 69 – The Future of CPR with Keith Lurie and Demetris Yannopoulos; that you have already listen and comment), if you are not convinced, you should ask to attend one and see by yourself.

        best regards

    • Jimmy D says

      Hi Dr Segal, thank you for your podcast! It’s Jim DuCanto from Milwaukee (across town from Aufderheide and Lurie).

      You received some of my questions through Scott and addressed them in your podcast (from the Society for Airway Management Web Forum), specifically, appropriateness of the selection of SGA size for 32 kg animals, whether the head and neck anatomy was relevant to humans and were the SGA’s appropriately inflated with air, etc..

      Let me first begin by congratulating you and thanking you for bringing the potential of SGA induced cerebral ischemia to everyone’s attention.

      I have a few simple questions to ask,

      1. Can you run any more pigs through your experimental protocol? The influence of SGA size and design is the principal issue many of us would like to investigate. The iGel and the Air-Q SP (an LMA style SGA that adjusts its cuff pressure based on ambient airway pressure) are the two most important candidates for further research with your model.

      2. I read your paper, but I don’t remember the exact pig breed you used (I only remember the weight). A colleague and I with a lot of interest in this topic have considered “sizing” these pigs appropriately in the same fashion that Archie Brain utilized when he invented the LMA: Pouring a quick set plaster-type material into the pharynx of a cadaver (pig), and obtaining the size and shape of their pharynx. We could produce a pig-LMA for testing in your model.

      Just thoughts and considerations–again thank you. I know a little bit about how onerous the process of animal research can be, and I do not intend to cast aspersions on your study—it’s simply that you really shook the newly evolving paradigm of SGA based rescue a bit, and many of us are going to squirm a bit as a result. Thanks again!

      • Nicolas Segal says

        1 unfortunately for the moment No. We do not have any more money for this study and we have use all the pigs that the ethics commitee allow us to.

        2 you are correct we forget to put the breed, it was domestic crossbreed
        Pouring anything in a dead pig is going to be ultrafun. You can always contact Keith Lurie (Tom Aufderheide can give you his email) and ask hiim if he is interested.
        The study may be interesting .

        • Jimmy D says

          Thank you–I’ll drop Aufderheide a note. I think we do need to standardize on the pig breed so we can get the pig-SGA right.

          As we say in Wisconsin, “you-betcha’!”

  14. Andrew DeWolf says

    Good evening everyone, thank you for your insights and continued teaching. In summarizing what was posted and mentioned in the podcast, it sounds like the following plans may be reasonable:: 1)BVM and OPA with continuous chest compression cpr and minimal interruptions. 2) If advanced airway needed, ETT appears superior choice at moment, but SGA such as LMA is appropriate and LMA seems to have less carotid compression than King and Combitubes? 3) Therapeutic Hypothermia 4) Titrate ventilations to ETCo2. Does this seem to be the essence of your current thoughts and suggestions or am I missing anything?
    Thanks Dr Weingart and Team.

    • says


      Pretty good, I would reword to:
      Early arrest-probably best to put on nasal cannula and maintain jaw thrust-perhaps a few breaths per minute with BVM and OPA
      Prolonged Arrest-LMA with minimal inflation pressures or ET, placed without interruption of compression.
      Ther hypothermia
      You can’t titrate ventilations to ETCO2, in these cases ETCO2 is a perfusion marker, not a ventilation marker. Can look at the ETCO2 resp rate to make sure you are bagging slowly.

      • Kim says

        Hi Scott,
        I had a patient that came in after at least 30 min of asystole/PEA. I did not intubate as I thought the patient likely needed compression versus intubation. Given that the patient seem to be a difficult intubation (overweight, beard, short neck), I felt compression would help more. My question, when do you decide to intubate? that is, how many minutes of down time (for reasonable change of neurologic recovery)? Do you intubate every patient regardless of how long they have been down? What is they never had a pulse or a shockable rhythm? My colleague recommended that the patient be intubated because the patient had emesis and facial cyanosis and said he may have choked. Is facial cyanosis a sure sign of choking, or is it a sequalae of cardiac arrest as well (venous congestion or stasis)? We ended up intubating but no ROSC.

  15. Chris Stanton says

    Hope this thread is still being read. As a new Paramedic (less than one year), I really have no experience pro/con ETT vs SGA. Doesn’t Segal’s study have a glaring control problem in that all the pigs were initially started with ETT and THEN switched to a random SGA? I understand the reasoning, but in such a small model could we not have seen a few which had ETT done throughout or SGA throughout to assure that these results are not part of some progressive pathology due to length of resuscitation? Hope that didn’t sound too stupid. Thanks to all the folks who do the research and those like you who put it out there.
    Chris Stanton
    West Georgia


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