The Last Gasp

It is undoubted that effective airway management is a critical link in the care of patients with both in-hospital cardiac arrest and out-of-hospital cardiac arrest.  But how exactly should one manage the airway?  What results in the best outcomes for our patients? Should we be aiming to intubate every patient? Or, are extraglottic devices as effective (or more effective)? What about the good old bag-valve mask?  In our most recent Journal Club we explored the evidence surrounding airway management in cardiac arrest, covering 3 high impact articles.  We also touch on an abstract presented at the 2018 SAEM Academic Assembly which should add significantly to the body of literature when it is published in full.  Take a listen to our recap podcast below and/or read on for the summaries and links to the articles.

Andersen LW, Granfeldt A, Callaway CW, et al. Association Between Tracheal Intubation During Adult In-Hospital Cardiac Arrest and Survival. JAMA 2017;317(5):494–13.

This study was a large, propensity-matched cohort study in which Andersen and colleagues reported a strong association between worse survival in patients intubated during in-hospital cardiac arrest compared to those who were not. The study had multiple strengths, including: large size, data from a well-documented registry from hundreds of hospitals around the country, and detailed minute-by-minute data. It is also notable because of the diversity that study results are generalizable and applicable to many patients with in-hospital cardiac arrest. There were several sensitivity analyses performed, and the time-based propensity score allowed for matching patients down to the minute which provided a far more detailed analysis than had previous been performed.

However, there were still limitations to consider. Any study which is observational in nature is limited and we are unable to identify what exactly occurred at the time of intubation. Were there any significant complications during the intubation, and if there were did they potentially delay intubation? Was CPR potentially held for prolonged periods of time? Who was the person intubating (physician (attending/resident), respiratory therapist, etc) and what was their experience level? It is also notable that because patients were matched by the minute, some controls (non-intubated patients) became cases (intubated patients) minutes later. Also, if one considers the study period is over 14 years (2000-2014) it is reasonable to assume equipment and protocols during that time have changed. Finally, despite propensity matching, there is likely some degree of residual confounding that remains.  

Despite these limitations, I do believe the results are valuable and should cause practitioners at the bedside who are faced with flash moment decisions during in-hospital cardiac arrest to prioritize care. While the ABCs of resuscitation are important for survival, focusing on endotracheal intubation above all else does not appear to be the most critical step.

Jabre P, Penaloza A, Pinero D, et al. Effect of Bag-Mask Ventilation vs Endotracheal Intubation During Cardiopulmonary Resuscitation on Neurological Outcome After Out-of-Hospital Cardiorespiratory Arrest. JAMA 2018;319(8):779–9.

This study was a randomized, multicenter, noninferiority trial published in JAMA earlier this year. The purpose of this study was to compare bag mask ventilation and endotracheal intubation in the treatment of patients with OHCA.  The investigators’ hypothesis was that bag mask ventilation was not inferior to endotracheal intubation with respect to 28-day favorable neurological outcome.

This study involved 20 prehospital EMS centers in two countries, France and Belgium.  Patients with OHCA were randomized into either receiving bag mask ventilation or endotracheal intubation as their airway management during compressions.  If ROSC was obtained, patients in both groups were then intubated by the medical team. Inclusion criteria for the study were adults aged over 18 with OHCA.  Exclusion criteria were patients with suspected massive aspiration before resuscitation, presence of a DNR, known pregnancy, or imprisonment.  It is important to recognize that the EMS teams in this study were made up of an ambulance driver, a nurse, and an EM Attending Physician.  The EM physician was the sole provider performing the intubations.  

The primary outcome of the study was survival at 28 days with favorable neurologic outcome which was defined as a Glasgow-Pittsburgh Cerebral Performance Category of 2 or less. Other endpoints of the study included rate of survival to hospital admission, rate of survival at 28 days, rate of ROSC, intubation and bag mask ventilation difficulty, and rate of bag mask ventilation or intubation failure.  

2043 patients were enrolled in the study, with 1020 patients in the bag mask ventilation group and 1023 in the intubation group.  In regards to the primary outcome, there was a favorable functional survival rate of 4.3% in the bag mask ventilation group and 4.2% in the intubation group.  However, since the confidence interval crossed the non-inferiority margin, they could not demonstrate non-inferiority.  

For secondary outcomes, they demonstrated that the rate of ROSC was significantly greater in the ETI group (38.9% vs 34.2%; p = .03).  However, the differences in survival to hospital admission and survival at day 28 were not significant between the two groups.  They did have slightly higher rates of airway management difficulty, failure, and regurgitation of gastric contents in the bag mask ventilation group (18.1%, 6.7%, 15.2%) when compared to the endotracheal intubation group (13.4%, 2.1%, 7.5%).

The authors in their discussion and conclusion reiterated that the study failed to demonstrate non-inferiority or inferiority and that further research was needed.  They did mention that the study was likely underpowered which contributed to their failure in demonstrating non-inferiority. Another important consideration is that the EMS team in the study included both a nurse and physician. This makes generalizability to the US a bit more difficult.  

While this study did not demonstrate non-inferiority, I think there are several take-aways from the paper.  Despite a very large sample size, there was little difference in 28 day favorable neurologic outcomes with bag mask ventilation.  However, there did seem to be more difficulties with bag mask ventilation and more episodes of regurgitation.  With this, what I am going to tell my EMS squads is that if bag mask ventilation is going well, there is no need to automatically move towards intubation. They should focus on good chest compressions and ACLS.  However, if there are difficulties with bag mask ventilation for whatever reason, then moving towards intubation is a reasonable option depending on the training of the squad.    

Benoit JL, Gerecht RB, Steuerwald MT, McMullan JT. Endotracheal intubation versus supraglottic airway placement in out-of-hospital cardiac arrest: A meta-analysis. Resuscitation 2015;93:20–6.

This was a systematic review and meta-analysis published in Resuscitation in 2015.  The authors hypothesized that ETI would have worse outcomes than supraglottic airways (SGA).  To investigate their hypothesis, the authors conducted a systematic search of PubMed, Scopus, and Cochrane databases.  Including articles that studied out-of-hospital cardiac arrest (OHCA) where the airway was managed by the responding EMS providers and comparing ETI to any SGA (King Laryngeal Tube, Combitube, LMA, or iGel).  They excluded traumatic arrests, pediatric patients, physician or nurse intubators, EMS agencies using RSI, video laryngoscopy, and overlapping databases.  The authors reviewed 3,454 titles, 325 abstracts, and 96 full texts to ultimately find 10 articles that met their inclusion and exclusion criteria.

All 10 of the included articles were observational cohorts were all "low" or "very low" quality of evidence by GRADE methodology.  In total the meta-analysis included 34, 533 ETI patients, 41,116 SGA patients, for a total of 75,649 patients.

What did they find?

Odds Ratios (OR) for ETI as compared to SGA

  • ROSC – OR 1.28, 95% CI 1.05–1.55 
  • Survival to Hospital Admission – OR 1.34, CI 1.03–1.75
  • Neurologically Intact Survival to Hospital Discharge – OR 1.33, CI 1.09–1.61
  • Survival to Hospital Discharge – OR 1.15, CI 0.97–1.37

The authors conclusion was that patients with OHCA who receive ETI by EMS are more likely to have ROSC, survive to admission, and survive neurologically intact

There are a number of theoretical reasons why SGA may be worse than ETI including:

  • Aspiration risk
  • Upper airway bleeding
  • Esophageal laceration
  • Tongue upper airway edema
  • SGA causing tracheal injury
  • Leak of air from poor seal, ineffective ventilation
  • Decreased carotid blood flow observed in one porcine study

There are a number of potential weaknesses to this study.  Of the included studies none were randomized control trials.  A lack of control for confounders in the included studies (shockable rhythm, witnessed arrest, bystander CPR) induces a risk of bias in the results.

  • In the largest trial included, Tanabe et al, the ETI group had 5% higher rate of witnessed arrests, and 4% higher rate of bystander CPR.  
  • In the 4th largest trial included, Cady et al, the ETI group had a 7.3% higher rate of witnessed arrest.

Additionally, some patients may have received a SGA after failed ETI

Overall this is the most robust meta-analysis available on the subject.  The authors did a good job of addressing the question of EMS intubators specifically. There were however several weaknesses that may have contributed to worse outcomes in SGA group.

  • Lack of control for cofounders favoring the ETI group (shockable rhythm, witnessed arrest, bystander CPR)
  • Lack of control for SGA placement after multiple failed ETI attempts
  • Usage of older SGA devices thoughts (Combitube)

Wang, H., Schmicker, R., Stephens, S., Daya, M., Idris, A., Carlson, J., Colella, M., Herren, H., Hansen, M., Richmond, N., Puyana, JC., Aufderheide, T., Gray, R., Gray, P., Verkest, M., Owens, P., Brienza, A., Sternig, K., May, S., Sopko, G., Weisfeldt, M., & Nichol, G. (2018) Laryngeal Tube vs Endotracheal Intubation in Adult Out-of-Hospital Cardiac Arrest: The Resuscitation Outcomes Consortium Pragmatic Airway Resuscitation Trial. SAEM Annual Meeting Abstracts. Acad Emerg Med, 25: S8-S284. doi:10.1111/acem.13424

This abstract was presented at the 2018 SAEM Academic Assembly.  This is the first randomized control trial comparing EGD to intubation for out of hospital cardiac arrest.  As with any abstract, it is difficult to judge fully and with certainty the validity of the results.  The authors describe a multi-center randomized control trial with cluster randomization and a cross-over at 3-5 months.  They looked at the impact of EGD vs endotracheal intubation on 72-hour survival (primary outcome).  Secondary outcomes included return of spontaneous circulation, survival to hospital discharge, favorable neurological status at hospital discharge, and adverse events (oropharyngeal or hypopharyngeal injury, airway swelling, and pneumonia or pneumonitis).  They were able to enroll 3005 patients into the trial and found well-matched cohorts in the ETI vs EGD groups.  They found significantly higher rates of 72-hour survival (18.2% vs 15.3%), ROSC (27.9% vs 24.1%), hospital survival (10.8% vs 8.0%), and favorable neurologic outcome at discharge (7.0% vs 5.0%) for the EGD group as compared to the ETI group.  

Ultimately these results are intriguing and we look forward to reading the paper in its entirety on its release and we look forward to seeing the results potentially duplicated in future studies.


  • Anderson, et al (2017) - Nicole Soria, MD
  • Jabre, et al (2018) - Tim Murphy, MD
  • Benoit, et al (2015) - Robert Whitford, MD
  • Wang, et al (2018) - Jeffery Hill, MD MEd
  • Intro, Peer Review, Editing, Posting - Jeffery Hill, MD MEd