Critical Care, Critical Care Alert

Critical Care Alert: Bag-Mask Ventilation During Tracheal Intubation of Critically Ill Adults

Critical Care Alert

Casey JD, Janz DR, Russell DW, et al. Bag-Mask Ventilation during Tracheal Intubation of Critically Ill AdultsN Engl J Med. 2019;380:811-821.

To determine if the application of positive pressure ventilation by bag valve mask during tracheal intubation decreases hypoxemia and improves patient outcomes.

Hypoxemia is a common and dangerous complication during intubation of critically ill adults. As hypoxemia progresses it increases the risk of cardiac arrest. A healthy adult can remain apneic for an extended period of time after proper preoxygenation and denitrification. Critically ill patients, however, can desaturate rapidly from the time of induction to the initiation of laryngoscopy. Many critically ill patients undergoing intubation have a physiologic shunt such as severe atelectasis, pneumonia, or ARDS. As shunt progresses, the amount of oxygenated blood decreases as blood bypasses the ventilated lung. Maximal preoxygenation and denitrification cannot overcome the large amount of pulmonary circulation bypassing the lungs and entering the systemic circulation. The approach to overcoming shunt is the application of PEEP. This trial by Casey et al. looked to assess if the application of PEEP and ventilation during endotracheal intubation is safe and improved rates of hypoxemia.


Study Design: Multi-centered, parallel-group, unblinded, pragmatic, randomized trial. The trial enrolled 401 patients with 199 patients assigned to the trial group and 202 patients assigned to the control group.

                ▪       Sites – 7 academic ICUs in the U.S.

                ▪       Time – March 15, 2017, to May 6, 2018

Intervention: Compared bag-mask ventilation versus no ventilation between induction and laryngoscopy during endotracheal intubation. The BVM group was attached to an oxygen flow rate of at least 15 liters per minute, had a PEEP valve attached set between 5-10 cm of water, had an oropharyngeal airway placed, with a two-handed mask seal by the intubating clinician. The BVM was ventilated at 10 breaths per minute with the smallest volume to generate a visible chest rise.

Randomization: Patients were randomized in permuted blocks of 2, 4, and 6. Trial-group assignments were placed in non-transparent envelopes. The groups remained concealed until after enrollment.

Inclusion Criteria: Adults (age >17) undergoing endotracheal intubation in the ICU

Exclusion Criteria

  • Pregnancy
  • Incarceration
  • Immediate need for intubation that randomization was precluded
  • Clinicians thought that ventilation was necessary between induction and laryngoscopy due to high risk of hypoxemia
  • Clinicians thought that the risk of aspiration was too high, eg from active emesis, hematemesis or hemoptysis

Primary Outcome: Lowest oxygen saturation, measured by Sp02, observed between induction and 2 minutes after intubation

Secondary Outcome: The incidence of severe hypoxemia, defined as an Sp02 < 80%, in the interval described above

Total N=401
BVM group N=199
No ventilation group N=202

Primary Outcome

  • Median Sp02 was 96% in the BVM group vs 93%in the control group. P=0.01. The mean difference in the lowest Sp02 was 5.2% (CI 95%, 2.8-7.5).

Secondary Outcomes

  • Relative risk 0.48 to have severe hypoxemia in BVM compared to control group. (95% CI, 0.3-0.77)

Additional Outcomes

  • Sp02 <90% RR 0.74 in the BVM group vs no-ventilation group. (95% CI, 0.56-0.97)
  • Sp02 <70% RR 0.41 in the BVM group vs no-ventilation group. (95% CI, 0.18-0.90). P=0.01.
  • No significant differences in aspiration or new opacity on chest x-ray 48 hours after intubation.
  • No significant change in in-hospital mortality, number of ventilator-free days, or ICU free days.


  • Randomized, multi-center, pragmatic trial with an intention to treat analysis and group concealment.
  • Performed on critically ill patients in ICU as opposed to healthy patients in the operating room.
  • Similar demographic data between the groups


  • Non-blinded
  • Preoxygenation was not uniform between the two groups. However, the BVM group started with a lower Sp02 and a post-hoc analysis accounting for preoxygenation methods did not change the findings.
  • The BVM group had 100% apneic oxygenation compared to the no ventilation group at 77%. Due to the conflicting data regarding apneic oxygenation and its inability to reverse shunt this probably doesn’t have too much effect on the primary and secondary outcomes.1,2
  • Only looked at BVM ventilation. Did not assess non-invasive ventilation.
  • BVM ventilation is a skill. To obtain a proper mask seal and apply safe, adequate ventilation is operator dependent and cannot necessarily be extrapolated out to the general community.

This was a well-designed trial providing us with high-quality data examining BVM ventilation in the time between induction and laryngoscopy. The BVM group had a 5.2% higher Sp02 compared to the no-ventilation group in the post-hoc analysis. This is further compounded by the BVM group having 12% less hypoxemia with NNT= 9 to prevent severe hypoxemia in one patient. There were no safety outcome differences or significant side effects.

ED Take-Away
This will be practice-changing for many clinicians. This article shows that BVM ventilation between induction and laryngoscopy is safe and helps prevent severe hypoxemia. Even with maximal preoxygenation therapy, there is still a substantial risk of severe hypoxemia during intubation. Generally, this is secondary to shunt physiology in critically ill patients. This study suggests a way to avoid hypoxemia in intubating critically ill patients which may lead to a reduction in risk of cardiac arrest and hypoxic brain injury.

The application of apneic oxygenation by nasal cannula combined with BVM ventilation can reduce the risk of mask leak and provide the benefit of both apneic oxygenation and mitigating the effect of shunt. These concepts can be combined with the use of non-invasive ventilation to continue ventilation during paralysis. The patient can be ventilated with the head of the bed up position until the start of laryngoscopy in order to prevent lung derecruitment and reduce the risk of aspiration.

1. Semler M, Janz D, Lentz R, et al. Randomized Trial of Apneic Oxygenation during Endotracheal Intubation of the Critically Ill. Am J Respir Crit Care Med. 2016; 193(3):273-280
2. Pavlov I, Medrano S, Weingart S. Apneic oxygenation reduces the incidence of hypoxemia during emergency intubation: A systematic review and meta-analysis. Am J Emerg Med. 2017;35(8):1184–1189.
3. Casey JD, Janz DR, Russell DW, et al. Bag-Mask Ventilation during Tracheal Intubation of Critically Ill Adults. N Engl J Med. 2019;380:811-821.



Related Articles

Critical Care Alert: Hydrocortisone plus Fludrocortisone for Adults with Septic Shock (APROCCHSS)

Can the use of hydrocortisone plus fludrocortisone and/or drotrecogin alfa improve clinical outcomes in patients with septic shock?

Critical Care Alert: Fluid Overload and Mortality in Adult Critical Care Patients

Fluid overload has not been clearly defined in the critical care arena. In situations where patients have renal, cardiac, and pulmonary dysfunction, commonly seen in the ICU, large volume fluid therap