The Decision to Intubate

In theory, the decision to intubate is straight forward. Intubate if the patient is not oxygenating, ventilating, or protecting their airway. In practice, it can be a LOT harder to decide to pull the trigger on intubation. This is compounded when incomplete knowledge of patient history and data are available, such as in the emergency department or for a crashing floor patient. Maybe the decision would be easy if you had all the information - but you often just don't.

This decision is important. Intubating a patient that doesn't need it puts them at risk of peri-intubation complications (such as hypotension or arrest), failed airway, and complications of ventilation/sedation. At the same time, delaying intubation in a patient that needs it puts them at an even higher risk of peri-intubation complications, aspiration, or worsening of underlying disease process.

80% of the time it's easy

• The patient has intubation in their near future, so just do it now. A prime example is a major intracranial hemorrhage with borderline mental status or a patient with severe trauma.
• The patient is markedly hypoxic and in respiratory distress
• The patient is completely unconscious and not responding at all
• The patient has obvious airway compromise
• The patient has a DO NOT INTUBATE order/status/MOLST (don't intubate)

15% of the time a little data gathering or response to treatments makes it easy

• COPD or CHF patients with borderline mental and clinical status
• Narcan for opioid overdose
• Dextrose for hypoglycemia
• Benzodiazepines for seizure

5% of the time it's hard

• Questionable stridor/airway compromise
• Semi-obtunded hypercarbic patients
• Very borderline or waxing/waning mental status

Real world considerations when it's not easy (aka the 5%)

The less I know about the patient, the more I will lean towards intubation in a borderline case.

If a semi-crashing and altered patient presents to the ED and I don't have any history, imaging, or lab data - then I'd be more aggressive if it's borderline.

In contrast, when I know the details about the patient and have a trajectory and understanding of their disease process  (such as in the ICU), then I may be more likely to give them some treatment, lasix?, antibiotics?, HFNC, or NIPPV and see if they can turn around.

There is a special case for patients with suspected acute right heart failure, severe metabolic acidosis, or severe obstructive lung disease, such as patients with likely pulmonary embolism, diabetic ketoacidosis, or asthma. I really try to do everything possible to not intubate in these cases since doing so can REALLY be dangerous. If intubation is clearly the only option - then special precautions MUST BE MADE to make sure the patient can handle the transition to positive pressure ventilation.

Clinical pearls

• EtCO2 and Pulse oximetry are INVALUABLE - make sure to get the patient hooked up to these as soon as possible.
• Protecting the airway means keeping oral fluid and gastric fluid OUT OF THE TRACHEA. This results from airway TONE. Talking also requires airway tone.
• If the patient can wake up enough to answer simple questions and make a sentence, then they are probably protecting their airway.
• Inability to swallow secretions can be a good reason to pull the trigger on intubation. Pooled secretions in the posterior oropharynx is BAD.
• Gag reflex is NOT a sensitive or specific indictor of airway protection. It's absent in a large portion of NORMAL PEOPLE.
• If the patient is stable enough to get an ABG, then they can probably have a trial of NIPPV so long as the disease process is amenable to this.
• The more frail a patient, the less likely a trial of NIPPV will work.
• DON'T BE A COWBOY if you don't have to be. High risk airways are EXTREMELY dangerous. If you have resources in your shop USE THEM. Don't hesitate to call for help from anesthesia, ENT, or a surgeon if you have the time.
• Always be prepared for a surgical airway.
• Be sure to examine the mouth, posterior pharynx, and neck