Airway‎ > ‎


The anesthesia literature uses three questions to guide their approach to airway: what does the patient need, what is available, what is the provider’s comfort? In my world, elective airway interventions are infrequent. My approach is therefore solely dictated by the patient’s needs. It is my responsibility to assemble all of the equipment I might need for the undifferentiated patient and in a very short amount of time. In response to provider comfort, my mentor jokes: “do what the patient needs; there is always disposable underwear for your own needs.” Emergency medicine is a hard job. And airway is always scary.

Indications for airway interventions is summarized in a previous post. I was inspired to write this because it seems the default answer to airway intervention is intubation - which is the first mistake - and the default answer to intubation is RSI - the second mistake. It is important to recognize that intubation is one of the many means to an end. And RSI makes little sense in urgent, emergent scenarios. One’s approach depends entirely on the indication. The following is in order of the most time-sensitive indications to the least:

1. Airway Emergencies

Aspiration. The patient is actively vomiting and their mental status precludes them spiting, swallowing, or achieving a gravity-friendly position. Do not reach for an ET tube. Turn the patient immediately and reach for suction. If the clinical situation is such that the event occurred while eating solid foods, you may need McGill forceps with a direct laryngoscope to remove foreign bodies. The next step is to protect the airway immediately. I-Gel or LMA may be faster than an ET tube. If there is continued vomiting or aspiration, it might be more prudent to place an OGT, NGT, and Minnesota Tube to stem the reflux. After controlling the airway, you need to provide aggressive suctioning below the cords. Sometimes, an emergent bronch is necessary. The goal here is to provide a route for gas exchange and remove aspirate as immediately as possible. ETT provides the best protection to aspiration but does little to treat aspiration once developed.

Equipment – suction, McGill forceps, LMA/I-Gel, fiberoptic scope, OGT/NGT/Minnesota tube

Obstruction. The patient arrives demonstrating stridor, drooling, inability to phonate, large tongue, neck swelling, or obvious facial burns. This patient is on the precipice of decompensating from a upper airway obstruction. Their genioglossus and tensor palatini are already firing at maximal contraction. Do not lay them flat and push paralytics. The goal is to provide awake, upright, fiberoptic intubation - orally if swelling allows, nasally if not. Prepare for alternative strategies if your initial approach fails. Be mentally ready for a surgical airway. The patient's SpO2 dictates the pace and need for additional interventions.

Plan A: awake fiberoptic intubation
Plan B: RSI then fiberoptic intubation
Plan C: indirect intubation
Plan D: direct intubation
Plan E: direct intubation w/ bougie
Plan F: supraglottic airway
Plan G: supraglottic airway with exchange
Plan H: Melker cric
Plan I: knife cric

2. Respiratory Emergencies

Hypoxic Respiratory Insufficiency. The end-game is SpO2. An ET tube offers nothing to the physiology of hypoxia and may worsen it by introducing a period of apnea. FiO2 and PEEP are the solutions for hypoxia. Start with supplemental O2 by nasal cannula, and advance to non-rebreather or high flow nasal cannula (60lpm, 100% FiO2). This temporizes almost every etiology of hypoxia except shunt and severe hypoventilation. This is important to remember because if hypoxia is refractory to these measures (i.e. no improvement) think about intracardiac shunt. A normal CXR or a positive bedside TTE with agitated saline suggests intracardiac shunt. Positive pressure ventilation in these patients will only make things worse. 

Refractory hypoxia from intrapulmonary shunts, however, may benefit from noninvasive positive pressure ventilation. BiPAP if there is time but BVM with PEEP valve is generally the fastest to set up. Remember to hook up supplemental oxygen and to use adjuncts like NPA/OGA. Try to synchronize with the patient's breaths and squeeze very slowly to minimize gastric insufflation - hyperventilation does little to improve oxygenation.

Think creatively to address a patient's hypoxia. Unilateral pneumonias, for example, might benefit from having the affected lung up. Hemoptysis and airway bleeds might benefit from having the affected lung down. In pulmonary edema, shunt is worsened in the supine position. These patients should be allowed to sit upright with their legs dangling down. Nitroglycerin drip and loop diuretics may be helpful to decrease excessive RV preload. Alternatively nitroprusside drip and loop diuretics may be helpful when there is excessive LV afterload on a diseased heart.

If you are committed to intubating this patient anyways, it might be helpful to hook up the ventilator directly to the facemask from a BVM. Respiratory rate and flow are better detected by a machine than your tactile faculties under stress. Plus, the ventilator can guarantee 100% FiO2 whereas the BVM cannot. When studied, mechanical ventilation via face-mask is superior to BVM by lowering peak airway pressures and improving oxygenation. Utilize volume AC with low flow rate (30lpm), tidal volumes of 550cc, respiratory rate 12, and FiO2 100%, PEEP 5-15. Or just use PSV with PEEP 5-15, IPAP 0, and FiO2 100%. The patient may need ketamine or dexmedotomidine as procedural sedation to tolerate this mode of oxygenation. You will need to provide a good two-handed seal.

Weingart’s papers on Preoxygenation, Reoxygenation, and Delayed Sequence Intubation and Preoxygenation and Prevention of Desaturation During Emergency Airway Management fully address this topic.

Hypercarbic Respiratory Insufficiency. The end-game is pH, which is not as easy to monitor as SpO2. The surrogate is mental status; hypercarbia can produce delirium and somnolence. Note that oxygenation can worsen hypercarbia in some COPD patients through the Haldane effect. Treat the underlying etiology: reverse drugs that cause respiratory depression, and use bronchodilators for obstructive patients.

Akin to hypoxia, ET tube provides nothing to address hypercarbic physiology and could potentially worsen it by introducing a period of apnea. Positive pressure ventilation aids by improving ventilation but only in patients who are synchronous with BVM or BiPAP. Simply cranking up the IPAP may actually be harmful if this leads to dysynchrony from patient discomfort. This can be provided by BiPAP, which does not require invasive airways. Intubation is reserved for when the mental status becomes so depressed that airway protection is questionable or the patient appears to be tiring out.

Mechanical Respiratory Insufficiency. This is a clinical judgement and probably under-recognized. Paradoxical abdominal movement, retractions, and worsening tachypnea with shallow breathing are all clues that the respiratory muscles are fatigued and breathing will soon cease. Oftentimes, pCO2 and SpO2 are completely normal and provide false reassurance.

BiPAP can help decrease the work of breathing. Benzos can decrease air hunger and quell the brain into tolerating a slight amount of hypercarbia. The true solution in severe cases is complete rest of the respiratory muscles with the work supplanted by mechanical ventilation, in which case intubation is appropriate for airway protection during deep sedation.

Summary. Respiratory emergencies requiring airway intervention should probably utilize delayed sequence intubation because:

1) BiPAP treats all of the above disorders, which can potentially obviate the need for intubation and invasive ventilation

2) Patients with lung disease generally have poor pulmonary reserve and do not tolerate the apnea of intubation.

3. Cardiovascular Emergencies

Cardiac arrest. AHA makes no recommendation for intubation in cardiac arrest. In fact, many experts regard intubation as detrimental to the vital act of compressions and superfluous on its own. The more prudent maneuver is to immediately place an LMA to provide uninterrupted CPR. If you are adamant on intubating, do not RSI. All RSI drugs are myocardial depressants and antithetical to your efforts of stimulating the heart.

Acidosis. Severe acidosis produces bradycardia and hypotension through myocardial depression and inducing a refractory state to pressors. Bicarb can temporize metabolic acidosis but is detrimental respiratory acidosis. Many patients in severe acidosis may need assistance with ventilation to produce respiratory alkalosis. Intubation for the procedure of mechanical ventilation is tricky in these circumstances because traditional approaches all induce a period of apnea. It is tempting to argue that RSI produces the least amount of apnea and therefore should be the default. But Airway Gods are fickle and if the airway was unexpectedly difficult, this patient will die from refractory hypotension. Unlike oxygen desaturation which can be prolonged by denitrogenation, carbon dioxide levels rise immediately.

The more cautious approach would be to use zofran and gastric decompression to decrease the likelihood of massive aspiration. Sedate and immediately place a supraglottic airway connected to a swivel adapter and continue ventilation.  Vent settings should be PRVC: RR 30-35; and tidal volumes of 550cc; PEEP 0 unless hypoxemia. PRVC minimizes PIP and minimizes gastric insufflation. Then introduce a fiberoptic scope preloaded with an Aintree catheter through the swivel adapter to pass the cords. One does not need to pray for expediency because ventilation is maintained during the entire procedure. Once the path through the cords is secured, the act of intubation is rapid because it is an exchange over a modified bougie. One can even hyperventilate prior to this railway technique to provide additional alkaltoic buffer. There is almost no apnea time. To pay homage to Levitan: NA DESAT!

This approach is limited by the impracticality (you can’t sterile process a scope for each septic patient you intubate). It is also overshadowed by RSI, which works most of the time. Just take heed of ventilation. Weingart advocates for using the vent and mask up until laryngoscopy. But strongly consider this option as plan B.

Shock. Positive pressure ventilation will worsen most forms of shock (except left sided heart failure). If you do ventilate the patient, try to use shock friendly settings. Start with low tidal volumes (4-6cc//kg) and PEEP (5) and titrate up slowly.

Furthermore, all sedatives used for intubation are myocardial depressants, including ketamine (contrary to popular belief). The myodepressant effects of ketamine are counteracted by the drug’s propensity to exocytose adrenergic vesicles. Patients chronically ill or in prolonged shock deplete these vesicles and can also become hypotensive.

Dr. Dutton advocates for decreasing sedative dosing to 10-20% of original for patients in shock because the disease state is already a CNS depressant. Additionally, the risk of cardiovascular collapse with sedation/intubation is high in the already unstable patient.

Summary. RSI can be performed in shock, though with extreme caution. It’s probably better to preform hemodynamic DSI.

4. Airway Protection

Patient is not in trouble yet but might be because you plan on sedating them or there is an expected decline in clinical status.

Anticipated easy intubation. RSI is ideal.

Anticipated difficult intubation. RSI sacrifices consciousness which is the primary defense against aspiration. Zofran, NGT, awake fiberoptic. If that fails, march down algorithm (see Plan A-I above)

5. Surgical Airway