Updates in the Airway

The year of 2019 ushered in an abundance of interesting, practice-informing, and excellently conducted research regarding airway and peri-intubation management that is potentially applicable to the emergency physician caring for critically ill patients.

Some of these studies investigated the efficacy of commonly practiced interventions, others challenged long-standing dogma, while some embodied novelty and innovation foreshadowing the possible future standards in airway management to come. A selection of articles we felt particularly fit these characteristics are outlined below, with several other great article suggestions published in 2019 to also read. Trials are presented in "PICO" format — population, intervention, comparison, and outcome — followed by observational studies, which are descriptively summarized.

PrePARE: peri-intubation crystalloid bolus to prevent hypotension¹
Hypotension is common in critically ill patients undergoing emergent intubation and is associated with worse outcomes such as significantly higher mortality². Administration of a fluid bolus has been proposed to combat this peri-intubation hypotension. PrePARE sought to determine the effectiveness of fluid boluses in preventing peri-intubation hypotension.

• P: Critically ill adults (≥18 years old) undergoing urgent/emergent intubation
• I: 500 mL intravenous (IV) crystalloid bolus prior to induction
• C: No fluid bolus
• O: Composite primary outcome: cardiovascular collapse, defined as systolic blood pressure (SBP) less than 65 mmHg or new/increased vasopressor use in the period between induction and 2 minutes post-intubation, cardiac arrest within one hour after intubation, or death within one hour after intubation

Study design: Multicenter (nine sites: six medical intensive care units [ICU], one trauma ICU, one neurological ICU, and one ED), parallel-group (with PreVent³, see below), non-blinded, pragmatic randomized controlled trial (RCT)

Results: 337 patients were enrolled; 168 randomized to the fluid bolus (intervention) group and 169 randomized to the no fluid bolus (control) group. The trial was stopped early due to medical futility after a planned interim analysis (originally planned for 500 patient enrollment). The two groups were balanced well in baseline characteristics and protocol compliance was high. The majority of patients were intubated for hypoxemic respiratory failure (51% of patients in the intervention group and 41% in the control group), followed then by altered mental status (28% in the intervention group vs 31% in the control group). There was strong compliance with the protocol for each group. The primary composite outcome of cardiovascular collapse occurred in 20% of patients in the fluid bolus group as compared to 18% in the no fluid bolus group (absolute difference: 1.3% [95% confidence interval (CI): -7.1 to 9.7], p=0.76). Interestingly, in the subgroup of patients receiving positive pressure ventilation by non-invasive positive pressure ventilation (NIPPV) for preoxygenation or bag valve mask (BVM) after induction, there was a trend towards less incidence of cardiovascular collapse in the fluid bolus group.

Takeaway: In critically ill adults being intubated, the routine use of a peri-intubation crystalloid bolus does not significantly decrease rates of cardiovascular collapse. However, in patients who have an indication for crystalloid (e.g., hypovolemic patients) or patients receiving positive pressure ventilation for preoxygenation with subsequent reduced preload, a peri-intubation crystalloid bolus could be considered to aid in preventing cardiovascular collapse.

PreVent: BVM after induction to prevent hypoxemia³
Rapid sequence intubation (RSI) has been used primarily for critically ill patients undergoing emergent intubation, to allow the shortest apnea time, to abate hypoxemia, and to theoretically lower the risk of aspiration. Given the risk of aspiration induced by BVM is likely much lower than the potential for preventing hypoxemia with BVM use after induction, the PreVent trial aimed to assess the effectiveness of BVM after induction in preventing hypoxemia as compared to a more traditional RSI technique.

• P: Adults undergoing urgent/emergent intubation in the ICU
• I: BVM (≥15 L/min oxygen, with positive end expiratory pressure [PEEP] valve at 5-10 cm H₂O, for 10 breaths per minute) from time of induction to laryngoscopy
• C: No BVM use, unless after a failed intubation attempt, for treatment of hypoxemia, or if deemed clinically necessary by the treating clinician
• O: Primary outcome: lowest peripheral oxygen saturation (SpO₂) between induction and two minutes post-intubation. Secondary outcome: severe hypoxemia, defined as SpO₂<80% during the same interval

Study design: Multicenter (7 ICUs: 5 MICUs, 1 trauma ICU, and 1 neurological ICU), parallel-group (with PrePARE¹), non-blinded, pragmatic RCT

Results: 401 patients were enrolled; 199 were randomized to the BVM (intervention) group and 202 to the no BVM (control) group. Baseline characteristics were well-balanced between groups, and protocol compliance was high. The primary indication for intubation in both groups was hypoxemic respiratory failure (59% of patients in the BVM group vs 57% in the no BVM group) followed by altered mental status (37% vs 38%, respectively). Of note, preoxygenation with BVM was significantly more common in the BVM group, and preoxygenation with heated high flow nasal cannula (HFNC) or NIPPV was more common in the no BVM group. Importantly, the SpO₂ at the time of induction (99%) was not different between groups. For the primary outcome, the median lowest SpO₂ was 96% in the BVM group and 93% in the no BVM group (mean difference: 3.9% [95% CI: 1.4-6.5], p<0.05). Results were similar, showing significantly higher lowest SpO₂ in the BVM group, in post-hoc analysis adjusting for method used for preoxygenation. For the secondary outcome, 10.9% of the BVM group vs 22.8% of no BVM group (relative risk [RR]: 0.48 [95% CI: 0.30-0.77]) had severe hypoxemia. Operator-reported aspiration occurred in 2.5% of the BVM group vs 4% in the no BVM group (RR: 0.63 [95% CI: 0.21-1.91]), and the incidence of new post-intubation opacity on chest x-ray was 16.4% vs 14.8% (RR: 1.11 [95% CI: 0.7-1.77]), respectively. 

Takeaway: In critically ill adults undergoing intubation, use of BVM after induction, as compared to no BVM use, was associated with statistically significant higher lowest SpO₂ and statistically significant lower rates of severe hypoxemia, with no difference in aspiration rates. In these patients, and specifically, patients with pulmonary shunt physiology, one may consider the use of BVM to prevent hypoxemia after induction during urgent/emergent intubation.

Caveat: While this study is very well-designed and carried out, with intriguing data, the application of its results to an exclusively ED patient population—whom virtually none are nil per os (NPO)—must be carried out with caution as ED patients may be at higher risk of aspiration than the patient population in this study, which may be potentiated with BVM use.

FLORALI-2: NIPPV vs HFNC for preoxygenation⁴
Severe hypoxemia and subsequent cardiovascular collapse are prevalent in critically ill adults undergoing emergent intubation, and preoxygenation is the primary strategy to prevent this. HFNC and NIPPV can provide higher inspired oxygen than conventional devices and may be more effective in preoxygenating patients. Furthermore, HFNC can remain in place during laryngoscopy (read: apneic oxygenation) and thereby may better prevent hypoxemia associated with intubation. FLORALI-2 aimed to determine if NIPPV was more effective than HFNC for preoxygenation.

• P: Adults in the ICU undergoing intubation for acute hypoxemic respiratory failure
• I: NIPPV preoxygenation for 3-5 minutes: facemask connected to the ventilator with pressure support mode to obtain tidal volumes of 6-8 mL/kg predicted bodyweight with PEEP of 5 cm H₂O and FiO₂ of 100%, removed just prior to laryngoscopy
• C: HFNC (60 L/min and 100% FiO₂) for preoxygenation for 3-5 minutes prior to induction and continued during laryngoscopy
• O: Primary outcome: severe hypoxemia, as defined as SpO₂<80% for at least 5 seconds in duration, during the period from induction to five minutes post-intubation. Pre-study defined subgroups included patients with moderate-severe hypoxemia (P/F≤200 mmHg)

Study design: Multicenter (28 ICUs in France), parallel-group, non-blinded RCT 

Results: 313 patients were analyzed; 142 randomized to NIPPV (intervention group) and 171 to HFNC (control group). Baseline characteristics were similar between groups. Protocol compliance was high. The primary etiology of respiratory failure was pneumonia in both groups (35% of patients in both groups). The P/F was similar in both groups at enrollment (142 mmHg in the NIPPV group vs 148 mmHg in the HFNC group). The SpO₂ at the end of preoxygenation was similar between groups (97% vs 96%, respectively, absolute difference: 1.0 [95% CI: 0 to 2.0], p=0.08). The primary outcome of severe hypoxemia occurred in 23% of patients in the NIPPV group vs 27% in the HFNC group (absolute difference: -4.2% [95% CI: -13.7 to 5.5], p=0.39). In the subgroup analysis of moderate-severe hypoxemic patients (n=117 in NIPPV group and n=125 in HFNC group), the primary outcome occurred in 24% of the NIPPV group vs 35% in HFNC group (absolute difference: -11.3% [95% CI: -22.3 to 0.3], p=0.0553). Of note, the SpO₂ at the end of preoxygenation in this subgroup was 97% in the NIPPV group vs 96% (absolute difference: 1.0 [95% CI: -0 to 2.0], p=0.02). 

Takeaway: NIPPV, as compared to HFNC (continued throughout laryngoscopy), for the preoxygenation of patients intubated for acute hypoxemic respiratory failure was not associated with less episodes of severe hypoxemia. However, there were trends towards less episodes of severe hypoxemia with the use of NIPPV, and especially in patients being intubated with moderate-severe hypoxemia. This suggests that NIPPV may be most beneficial in preoxygenating patients with significant pulmonary shunt physiology. In light of this study and PreVent³, one may consider the use of positive-pressure ventilation of some sort (NIPPV or BVM with PEEP valve) for preoxygenation and after induction to prevent hypoxemia in patients being intubated with pulmonary shunt physiology.

Caveat: Another well-designed study, with thought-provoking results. There was no difference in new infiltrates on CXR or ventilator-associated pneumonias after intubation between groups. Again, however, these patients were not ED patients, and there remains clinical equipoise regarding both the efficacy and safety of positive pressure ventilation prior to and after induction of ED patients undergoing urgent/emergent intubations. 

IRIS: cricoid pressure vs sham procedure during RSI⁵
In emergency scenarios, patients requiring intubation are at higher risk of aspiration given their high prevalence of a non-fasting state, abdominal pathology, severe illness, and more. Traditionally, cricoid pressure has been used in attempt to reduce the risk of aspiration, though opposing data has suggested it is not effective and may even increase the risk of aspiration while making intubation more difficult^6,7. IRIS hypothesized that the absence of application of cricoid pressure would not lead to increased aspiration rates.

• P: Adults intubated in the operating room (OR) for surgery requiring RSI due to their non-fasting state, or at least one risk factor for aspiration (emergency conditions, body mass index>30, previous gastric surgery, ileus, within 48hrs postpartum, diabetic gastroparesis, gastroesophageal reflux, hiatal hernia, pre-operative nausea/vomiting, and/or pain)
• I: Cricoid pressure group: via planned 30N force applied using the first three fingers on the cricoid cartilage (blinded by use of opaque cover over the neck). All investigators providing this intervention were trained specifically on this maneuver
• C: Sham group: opaque cover applied over neck (in both groups, to facilitate blinding) and investigator simulated (placed their hand behind the cover near the cricoid so that the intubator was unable to tell if cricoid pressure was applied or not) but did not apply cricoid pressure
• O: Primary outcome: pulmonary aspiration, either reported as witnessed by the intubator as gastric fluid in the vocal cords during laryngoscopy, or evidence of gastric fluid on endotracheal aspiration performed just after intubation. Secondary outcomes assessed intubation difficulty and Cormack-Lehane grade

Study design: Multicenter (10 centers), double-blinded, noninferiority RCT. The sham procedure was considered noninferior if the incidence of aspiration was not more than 50% higher (RR of 1.5) as compared to the cricoid group (i.e., if the upper limit of the 95% CI<1.5, then noninferiority is accepted).

Results: 3,471 patients were analyzed; 1,736 randomized to the sham group and 1,735 to the cricoid pressure group. The groups were well-balanced in regard to baseline characteristics and difficult intubation risks/features. The majority of patients were intubated for abdominal surgery, and the majority of the patients’ risk factor for aspiration was emergency condition (65% and 66%, respectively), with the second most common risk factor being ileus (31% and 32%, respectively). The primary outcome of pulmonary aspiration occurred in 0.5% of patients in the sham group and 0.6% in the cricoid group (RR: 0.90 [95% CI: 0.33-2.38]). There was a trend towards more difficult intubation in the cricoid group, and requested interruption by the intubator of the cricoid/sham procedure occurred more frequently in the cricoid group (14% of patients) as compared to the sham group (5% of patients), which also resulted more often in an improvement in Cormack-Lehane grade (62% of the time), as compared to the sham group (33% of the time). 

Takeaway: Sham procedure, as compared to cricoid pressure, for adults at higher risk for aspiration undergoing RSI for surgery, was not shown to be noninferior. Limitations of this study include the non-ED based population, whose rates of aspiration are low and likely lower than patients being intubated emergently in the ED. The pre-specified noninferiority margin was not met, but the overall incidence of aspiration was lower in the sham group, and intubation trended more difficult in the cricoid group with significant improvement in laryngoscopy views after removal of the cricoid pressure. While noninferiority was not met, there is little evidence to suggest cricoid pressure is effective at preventing aspiration and may actually have downsides such as making intubation more difficult. It is important to remember, however, that cricoid pressure is a different technique with different intended purposes than external laryngeal manipulation. 

Use of the intubating laryngeal mask airway⁸
This was a 10-year retrospective review at one institution of the use of the intubating laryngeal mask airway (ILMA) in the ED. The ILMA is used often in the OR and can be used as a primary extraglottic airway device to oxygenate/ventilate, but it can also be used as a rescue device and to help facilitate blind intubation with an endotracheal tube through its lumen. In this study, the ILMA was used on 218 patients over the ten-year period; 118 patients as a primary airway device and in 100 patients as a rescue device after failed intubation. The ILMA provided effective oxygenation/ventilation in 98% of patients (99% of patients when used as primary airway device and 96% as a rescue device). Blind intubation through the ILMA was successful in 81% of attempts. The ILMA was used to facilitate intubation in seven patients who could not be intubated with video-assisted laryngoscopy nor endoscopic techniques, who likely would have undergone surgical airway otherwise.

Takeaway: Consider the ILMA as a primary airway device, especially in cardiac arrest to avoid compression interruptions and help facilitate blind intubation for a secure airway later, or as an effective rescue device for patients who fail endotracheal intubation with other conventional techniques.

End tidal oxygen to assess preoxygenation status⁹
End tidal oxygen (ETO₂) has been used frequently in the OR, but not often in the ED, to assess preoxygenation effectiveness prior to intubation attempts. The goal is an ETO₂>85% which suggests adequate preoxygenation in critically ill patients being intubated per anesthesia guidelines¹⁰. This study was a prospective observational cohort study in two EDs where patients being intubated were preoxygenated for at least three minutes with either BVM (with the use of assisted breaths and PEEP as deemed necessary by the provider), or non-rebreather (15 L/min, 19 L/min or flush rate at the discretion of the provider). ETO₂ was recorded by a separate independent observer (data blind to the treating clinicians). 100 adult patients were included. The median ETO₂ at the start of preoxygenation was 53% (interquartile range [IQR]: 43-65), and 78% (IQR: 64-86) at the start of induction. Only 26% of patients were able to achieve an ETO₂>85%. Median ETO₂ levels were similar between NRB (80%, IQR: 59-87) and BVM (77%, IQR: 65-86) groups. NRB at flush rate demonstrated the highest median ETO₂ (86%, IQR: 80-90), and NRB at 15 L/min demonstrated the lowest median ETO₂ (57%, IQR: 53-60). SpO₂ levels were higher during intubation in patients with higher ETO₂ at induction.

Takeaway: ETO₂, while frequently used in the OR to assess preoxygenation, has not been adopted in the ED, and may be helpful in assessing adequacy of preoxygenation prior to intubation, though further research is needed in the ED setting. While a separate gas analyzer would be needed in the ED to measure ETO₂, the setup is similar to continuous capnography monitoring and could be adopted into ED practices if found to be beneficial.

Post-intubation sedation¹¹
This was a retrospective analysis of prospectively collected data from the National Emergency Airway Registry (NEAR)—a multicenter ED airway monitoring database—to evaluate post-intubation sedation in the ED. 11,748 patients intubated in the ED were included over a one-year period. 77.5% of patients received post-intubation sedation within 15 minutes after intubation. Of patients who received post-intubation sedation, the majority (93%) underwent RSI, whereas of those patients who did not receive post-intubation sedation, only 47% underwent RSI. Of patients that did not receive post-intubation sedation, the majority were intubated for cardiac arrest (57%, as compared to 7% of patients that received post-intubation sedation), whereas patients that received post-intubation sedation predominantly were intubated for trauma or altered mental status/overdose. 37% of patients that did not receive post-intubation sedation were hypotensive, whereas only 14% of patients that received post-intubation sedation were hypotensive. In regression analyses, pre- and post-intubation hypotension and cardiac arrest were associated with significantly less odds of receiving post-intubation sedation, while medical indication for intubation (as compared to trauma), RSI (as compared to non-RSI), and succinylcholine use (as compared to long-acting neuromuscular blockade) were associated with significantly increased odds of receiving post-intubation sedation.

Takeaway: While the study is at risk for recall bias, this large study indicates post-intubation sedation in the ED is provided more often than historically thought, and that patients that are hypotensive, in cardiac arrest, or do not undergo RSI are less likely to receive post-intubation sedation. Those who receive short-acting neuromuscular blockade are more likely to receive post-intubation sedation, likely due to their subsequent movements prompting clinicians to provide sedation, as compared to their counterparts that received long-acting neuromuscular blockade and are still paralyzed but not sedated. This is important information for the emergency physician to be cognizant of when intubating patients, to prompt them to provide post-intubation sedation as medically appropriate in their patients.

Other high-yield articles to review:

• Driver BE, Klein LR, Prekker ME, et al. Drug Order in Rapid Sequence Intubation. Acad Emerg Med. 2019;26(9):1014-1021.
• Driver BE, Scharber SK, Horton GB, Braude DA, Simpson NS, Reardon RF. Emergency Department Management of Out-of-Hospital Laryngeal Tubes. Ann Emerg Med. 2019;74(3):403-409.
• Carlson JN, Zocchi M, Marsh K, et al. Procedural Experience With Intubation: Results From a National Emergency Medicine Group. Ann Emerg Med. 2019;74(6):786-794.
• Vourch M, Baud G, Feuillet F, et al. High-flow Nasal Cannulae Versus Non-invasive Ventilation for Preoxygenation of Obese Patients: The PREOPTIPOP Randomized Trial. EClinicalMedicine. 2019;13:112-119.
• Sakles J, Augustinovich C, Patanwala A, Pacheco G, Mosier J. Improvement in the Safety of Rapid Sequence Intubation in the Emergency Department with the Use of an Airway Continuous Quality Improvement Program. Western J Emerg Med. 2019;20(4):610-618.
• Guihard B, Chollet-Xémard C, Lakhnati P, et al. Effect of Rocuronium vs Succinylcholine on Endotracheal Intubation Success Rate Among Patients Undergoing Out-of-Hospital Rapid Sequence Intubation. JAMA. 2019;322(23):2303.
• Guitton C, Ehrmann S, Volteau C, et al. Nasal high-flow preoxygenation for endotracheal intubation in the critically ill patient: a randomized clinical trial. Intensive Care Med. 2019;45(4):447-458.

References

1. Janz DR, Casey JD, Semler MW, et al. Effect of fluid bolus on cardiovascular collapse among critically ill adults undergoing tracheal intubation (PrePARE): a randomized controlled trial. Lancet Respiratory Medicine. 2019;7(12):1039-1047.
2. Green RS, Turgeon AF, Mcintyre LA, et al. Postintubation hypotension in intensive care unit patients: A multicenter cohort study. Journal of Critical Care. 2015;30(5):1055-1060. doi:10.1016/j.jcrc.2015.06.007.
3. Casey JD, Janz DR, Russell DW, et al. Bag-Mask Ventilation during Tracheal Intubation of Critically Ill Adults. New England Journal of Medicine. 2019;380(9):811-821. doi:10.1056/nejmoa1812405.
4. Frat JP, Ricard JD, Quenot JP, et al. Non-invasive ventilation versus high-flow nasal cannula oxygen therapy with apnoeic oxygenation for preoxygenation before intubation of patients with acute hypoxaemic respiratory failure: a randomised, multicentre, open-label trial. The Lancet Respiratory Medicine. 2019;7(4):303-312.
5. Birenbaum A, Hajage D, Roche S, et al. Effect of Cricoid Pressure Compared With a Sham Procedure in the Rapid Sequence Induction of Anesthesia. JAMA Surgery. 2019;154(1):9. doi:10.1001/jamasurg.2018.3577.
6. Smith KJ, Dobranowski J, Yip G, Dauphin A, Choi PT-L. Cricoid Pressure Displaces the Esophagus: An Observational Study Using Magnetic Resonance Imaging. Anesthesiology. 2003;99(1):60-64. doi:10.1097/00000542-200307000-00013.
7. Ellis DY, Harris T, Zideman D. Cricoid Pressure in Emergency Department Rapid Sequence Tracheal Intubations: A Risk-Benefit Analysis. Annals of Emergency Medicine. 2007;50(6):653-665. doi:10.1016/j.annemergmed.2007.05.006.
8. Driver BE, Martel M, Lai T, Marko TA, Reardon RF. Use of the intubating laryngeal mask airway in the emergency department: A ten-year retrospective review. The American Journal of Emergency Medicine. 2019. doi:10.1016/j.ajem.2019.11.017.
9. Caputo ND, Oliver M, West JR, Hackett R, Sakles JC. Use of End Tidal Oxygen Monitoring to Assess Preoxygenation During Rapid Sequence Intubation in the Emergency Department. Annals of Emergency Medicine. 2019;74(3):410-415. doi:10.1016/j.annemergmed.2019.01.038.
10. Higgs A, Mcgrath B, Goddard C, et al. Guidelines for the management of tracheal intubation in critically ill adults. British Journal of Anaesthesia. 2018;120(2):323-352. doi:10.1016/j.bja.2017.10.021.
11. Lembersky O, Golz D, Kramer C, et al. Factors associated with post-intubation sedation after emergency department intubation: A Report from The National Emergency Airway Registry. The American Journal of Emergency Medicine. 2019. doi:10.1016/j.ajem.2019.05.010.