A Non-Non-Answer in Airway Management in OHCA

In a lovely demonstration of the statistical inanity of non-inferiority trials, these authors present a simultaneously insightful and illogical data set examining airway management strategies during CPR in out-of-hospital arrest.

This is a clinical trial from France, randomizing patients out cardiac arrest to either bag-valve mask ventilation or placement of an advanced airway during CPR. Patients with significant challenges associated with BVM could cross over to ETI, and if patients achieved return of spontaneous circulation at any time, they were subsequently intubated as well. All emergency response and airway management was supervised by an emergency physician. Groups were fairly well matched by their randomization by center, and about 10% of the cohort crossed over to ETI from BVM due to failure of ventilation or gastric regurgitation.

Without wallowing too much in the statistical underpinnings, these authors defined a 1% absolute difference in favorable neurologic outcome at 30 days as their primary outcome measure for non-inferiority. Then, if non-inferiority was unable demonstrated, a test of difference would be performed for inferiority.

And, so, after all this, CPC 1-2 survival was: 4.2% in the BVM group and 4.3% in the ETI group, for a difference of 0.11% (1-sided 97.5% CI, −1.64% to infinity). It should be abundantly obvious that – considering the obviousness of their result and its gross failure to meet their statistical threshold – their sample size is completely inadequate. It then unsurprisingly follows their test of difference does not demonstrate inferiority of BMV as compared to ETI.

So, yes, BMV is not inferior, but not non-inferior, to ETI. This is why everyone hates Journal Club.

In the bigger picture, these data generally support deferring advanced airway placement during CPR. These authors did not observe any differences in low-flow time, even in their ETI group – but I would expect this might be a “best case scenario” with respect to minimal interruption and successful airway management.  Considering they have an EP assisting in the resuscitation and airway management, I would probably expect other settings are more prone to airway management failure and interruptions – and, really, the onus should be on those doing more to find solid data to back up their prehospital intervention.

“Effect of Bag-Mask Ventilation vs Endotracheal Intubation During Cardiopulmonary Resuscitation on Neurological Outcome After Out-of-Hospital Cardiorespiratory Arrest A Randomized Clinical Trial”
https://jamanetwork.com/journals/jama/article-abstract/2673550