The FELLOW Trial: An End to Apneic Oxygenation?

Background: Every year there are a handful of “game changing” publications that truly change how we care for our patients. One such paper was a paper by Scott Weingart and Richard Levitan in the Annals of Emergency Medicine in 2011 on the topics of preoxygenation and apneic oxygenation (This paper was already reviewed on REBEL EM – Preoxygenation and Apneic Oxygenation). As many of us know, one of the most common and feared complications dealt with in critically ill patients requiring endotracheal intubation is hypoxemia. Hypoxemia can subsequently lead to cardiac arrest and death. Since the advent of apneic oxygenation this common complication seems to have decreased in occurrence, but is apneic oxygenation effective in all settings?

 The Facilitating EndotrachealL intubation by Laryngoscopy technique and apneic Oxygenation Within the intensive care unit (FELLOW) Trial

What They Did:

  • Randomized, Open-label, Pragmatic trial of 150 adults undergoing endotracheal intubation in a medical ICU at Vanderbilt
  • Determine if apneic oxygenation actually increases arterial oxygen saturations in patients undergoing endotracheal intubation in an ICU
  • Randomized to:
    • Apneic Oxygenation = 15L/min of 100% O2 via high-flow nasal cannula during laryngoscopy
    • Usual Care = No supplemental oxygen during laryngoscopy


  • Primary: Lowest arterial oxygen saturation between induction and two minutes after completion of intubation measured by pulse oximetry (SpO2)
  • Secondary Efficacy Outcomes:
    • Incidence of hypoxemia (SpO2 < 90%)
    • Incidence of Severe hypoxemia (SpO2 <80%)
    • Desaturation (decrease in SpO2 >3%)
    • Change in saturation from baseline
  • Secondary Safety Outcomes:
    • Cormack-Lehane Grade of Glottic View
    • Incidence of Successful Intubation on first Laryngoscopy Attempt (Placement of ETT in trachea)
    • Number of Laryngoscopy Attempts
    • Time from Induction to Intubation
    • Need for Additional Airway Equipment or Operators
    • Incidence of Non-Hypoxemia Complications
  • Tertiary Outcomes:
    • Duration of Mechanical Ventilation
    • ICU Length of Stay
    • In-Hospital Mortality


  • 150 patients randomized
    • 77 patients apneic oxygenation
    • 73 patients usual care
  • Median lowest arterial oxygen saturation:
    • 92% with apneic oxygenation
    • 90% with usual care
    • 95% CI 1.6 – 7.5%; p = 0.16
  • No difference in incidence of oxygen saturation <90%
    • 44.7% vs 47.2% (p = 0.87)
  • No difference in incidence of oxygen saturation <80%
    • 15.8% vs 25.0% (p = 0.22)
  • No difference in incidence of decrease in oxygen saturation >3%
    • 53.9% vs 55.6% (p = 0.87)
  • There was no statistical difference between duration of mechanical ventilation, ICU length of stay, and in-hospital mortality between groups


  • 1st randomized controlled trial to evaluate apneic oxygenation vs usual care
  • Largest trial to date evaluating apneic oxygenation vs usual care
  • Trial occurred outside the operating room setting where intubations are elective and patients may be healthier than patients in the ED or ICU
  • Patients randomized in a 1:1 fashion for intervention and control in sealed envelopes
  • Data collection during intubation was performed by independent observers not aware of the study hypothesis or involved in the procedure itself
  • To confirm accuracy of data collection by independent observes, the primary investigators also collected data for 10% of intubations


  • This is a single center study of one medical ICU of an academic center and may not generalize to community settings or ED settings
  • Excluded patients requiring emergent intubation due to inability to randomize patients. Also patients were excluded if providers wanted a specific approach to oxygenation. Both of these lead to a selection bias with the sickest patients being removed from the study.
  • In the apneic oxygenation group HFNC was used prior to induction, but unclear for how long before the procedure
  • After enrollment clinicians and study personnel no longer blinded to study group assignments
  • This study was powered to detect 5% differences in lowest arterial oxygen saturation, so smaller differences may have been missed
  • Patient requiring prolonged intubation times or abnormal upper airway anatomy were excluded from this trial
  • Patientts were randomized to DL vs VL. There is no real data published on this for us to review, but time to intubation could play a role in Ap Ox.


  • Pre-oxygenation and apneic oxygenation is fairly cheap and easy way to maintain oxygen saturation. Additionally, nasal canula is widely used in preoxygenation so it’s already on the patient at the time of induction. This study did not show increase in harm by using apneic oxygenation.
  • Looking at the supplemental material some interesting stats:
    • Induction agent was etomidate in >90% of cases in both arms
    • Paralytic agent was Rocuronium >50% of cases in both arms
    • The difficulty of the intubation was rated easy in >75% of intubations
  • A majority of patients were not left apneic prior to intubation. For example BPAP was used in approximately 1/3 of patients and another 40% of patients BVM was used prior to and as induction medications were pushed up to laryngoscopy which defeats the purpose of the apneic portion of apneic oxygenation.
  • 1/3 of patients had BiPAP so tidal volumes could be observed for airway patency, but in the other 2/3 of patients airway patency could not be confirmed which is essential for apneic oxygenation to work.

Thoughts from the Twitterverse:

The FELLOW Trial

The FELLOW Trial 1

The FELLOW Trial 2

Author Conclusion: Apneic Oxygenation does not appear to increase lowest arterial oxygen saturation during endotracheal intubation of critically ill patients compared to usual care. These findings do not support routine use of apneic oxygenation during endotracheal intubation of critically ill adults

Clinical Take Home Point: In critically ill patients requiring intubation, if  you use BPAP or BVM during the pre-oxygenation period, you may not get as much benefit from apneic oxygenation.  However, it is important to remember that apneic oxygenation is cheap, low risk (no risk) intervention, with no proven harms to date.


  1. Semler MW et al. Randomized Trial of Apneic Oxygenation During Endotracheal Intubation of the Critically Ill. Am J Respir Crit Care Med 2015 [Epub ahead of print] PMID: 26426458

For More Thoughts on This Topic Checkout:

Post Peer Reviewed By: Anand Swaminathan (Twitter: @EMSwami) and Matt Astin (Twitter: @mastinmd)

Cite this article as: Salim Rezaie, "The FELLOW Trial: An End to Apneic Oxygenation?", REBEL EM blog, October 12, 2015. Available at:
The following two tabs change content below.

Salim Rezaie

Emergency Physician at Greater San Antonio Emergency Physicians (GSEP)
Creator & Founder of REBEL EM

Like this article?

Share on facebook
Share on Facebook
Share on twitter
Share on Twitter
Share on linkedin
Share on Linkdin
Share on email
Share via Email

Want to support rebelem?

13 thoughts on “The FELLOW Trial: An End to Apneic Oxygenation?”

  1. Salim,
    This is a great review.
    One take-home from this study is that in patients with significant shunt physiology apneic oxygenation will not be as effective as it is in previously studied populations (anesthesia studies on elective cases with normal lungs). Apneic oxygenation seems to work phenomenally well in patients with normal lungs, but this study reinforces the times I have seen apneic oxygenation mysteriously not work in sick pneumonia patients who have desaturated precipitously despite an aggressive pre oxygenation strategy.
    It would be wonderful to see this study repeated in an ED setting, however with little to no downside to using ap ox. I don’t know that it absolutely needs to be done.

    • Hey Rob,
      TY for reading and leaving your comment. You make an excellent point with ApOx in shunt physiology. Readers should remember that in shunts (i.e. Perfusion to non-ventilated lung) no matter how much oxygen you give a patient you will minimally or maybe even won’t help improve oxygen saturation. These patients need PEEP. Common conditions that can cause shunt physiology include pneumonia, pulmonary edema, and ARDS. ApOx therefore will not help in these patients. TY for bringing that point up.


  2. Gotta chip in 2 cents:
    1) These patients were tubed within 1-2 minutes and we may not see benefit of ap-ox until further out.
    2) They only tubed 67% on the first pass and didn’t describe what they did in between attempts. – to me this is huge. Piggybacking off the discrepancy with regards to the induction phase, we need to know if they were bagged between attempts or if they were left completely apneic (as this may be where we see the benefit)

    Hard to say exactly which patient population will benefit from apneic oxygenation but cheap intervention that probably works for some folks (i.e. I’m not ready to change my practice yet)

    • Hello Tyler,

      Spot on…the whole point of ApOx is to delay time to delay…we may not see that benefit in 2 min time. I found the lack of description between attempts to be difficult to interpret as well…TY for reading and appreciate you continuing the conversation


  3. Nice post Salim
    The big weakness of this study for me is this. How much does your average intubation desat during the process? I reckon I could do dozens before I get a meaningful desaturation. Apnoeic oxygenation is not guarding against a desat in the average intubation. It is to buy time when things are tougher than expected. This trial is analogous to randomizing 150 people to seatbelt or no seatbelt for a drive to the shops. You are simply not going to find a difference in serious injury in such a trial and nor was FELLOW going to find a difference in outcome from Apnoeic oxygenation

    I’m sticking with it for now.

    • Hello Mark,
      100% agree….the whole point of ApOx is to prevent feast in prolonged intubation but remember in patients with shunt physiology this may not work but then again what’s the harm in doing it….worth a shot I would say. If you are already using BVM or BiPAP may just add minimal benefit but then again what’s the harm? Appreciate u reading and leaving ur comments….I agree at this point it is a bit premature to change practice of not using ApOx.



Leave a Comment

Time limit is exhausted. Please reload CAPTCHA.