Critically ill patients commonly require tracheal intubation for a variety of indications from airway obstruction to shock. With an aging and increasing population, and the seemingly increasing infectious threats from respiratory viruses such as influenza and Sars-CoV-2, tracheal intubation is increasingly required and more complex. We are often asked: “Why do we need to focus on safe airway management practices when intubation is such a brief moment in the patient’s entire hospital course”? Our answer is simple: To be competent in “life support,” you must be skilled in putting your patient on life support. If you are not skilled at airway management, it does not matter how good you are with a ventilator, how smart you are, or what your expertise is in any other procedure. If a patient’s airway is not secured, and oxygenation/ventilation is not restored, it does not matter what else you can do. Airway management in critically ill patients, regardless of location (prehospital, emergency department [ED], intensive care unit, hospital ward, operating room [OR]), requires procedural and cognitive expertise, and that expertise needs to be durable in the presence of high stress and difficulty. This book will elaborate this philosophy in all its nuance.

Intubation in the ICU is not just common, it is quite dangerous. “Severe complications” occur in one out of every four intubations and are most defined as those that are immediately life threatening such as severe desaturation, cardiovascular collapse, and cardiac arrest. Every study to date that looked at the peri-intubation adverse events with intubation in critically ill patients has reported cardiac arrest rates between 2% and 4%; higher than any other procedure we commonly perform. When cardiac arrests do happen, only about half of those patients achieve return of spontaneous circulation.

The major precipitants to cardiac arrest are desaturation (20% to 40% incidence) and hypotension (10% to 40% incidence), which carry adjusted odds ratios of 3.99 and 3.41, respectively, for cardiac arrest.¹ The INTUBE study recently prospectively evaluated nearly 3,000 intubations across 29 countries and reported some remarkable findings.² While the rate of severe desaturation <80% was relatively low (9.3%) in INTUBE, the rate of cardiovascular instability was high (43%), and overall, 45% of all intubations suffered a major adverse event including a 3% cardiac arrest rate. INTUBE also uncovered many modifiable risks such as the common use of mask ventilation for preoxygenation, low uptake of video laryngoscopy as the first device, and the overuse and overdose of vasodilating induction medications such as propofol for rapid sequence intubation.

Complications short of cardiac arrest are even more common. “Moderate complications,” which are reported in one out of every three intubations include those that are either functionally injurious such as dental injury, procedurally dangerous such as difficult intubation, and those that are potentially life threatening such as moderate desaturation, aspiration, or esophageal intubation.³

Difficult intubations, those that take more than two attempts at laryngoscopy or a prolonged time with laryngoscopy (generally >10 minutes), increase these risks dramatically. Difficult
intubations inside the operating room (OR) have a reported incidence of 0.5% to 1%. Outside of the OR, 8% to 10% of intubations are considered difficult.⁴ Failed intubations are equally disparate, occurring in one in 2,000 cases in the OR, 1 in 250 in obstetrics, 1 in every 100 in the ED, and 1 in 50 in the ICU.⁵ Thus, you are far more likely to encounter a difficult intubation in the ED or the ICU than you are in the OR.

Our tribalism instincts in medicine surface when trying to explain the high incidence of difficulty, complications, cardiac arrest, and death; especially compared to the lower incidence in the OR. It is tempting to explain these findings as those intubating in the ICU or ED are not as skilled at intubation as those intubating inside the OR. On the surface, this is illogical but an elegant study in 2018 exposed this heuristic. Taboada and colleagues compared intubating conditions in the OR and in the ICU by observing all nonpregnant adult patients intubated using direct laryngoscopy in the ICU within a month of an elective intubation in the OR by the same group of anesthesiologists in both locations.⁶

The mean patient age was 70 ± 12 years, 30% had a body mass index >30, the mean BMI was 28 ± 5, and two-thirds of the patients had a Mallampati score ≤2. Patients were most commonly in the ICU postoperatively (66%), but still one-third had acute respiratory failure (25%) or shock (5%), and overwhelmingly patients were intubated for acute respiratory failure (83%), shock (16%) or an acute neurologic event (18%); and nearly two-thirds of patients were on noninvasive ventilation before intubation (63%). These demographics are quite like the INTUBE demographics. Taboada and colleagues found that hemodynamically neutral induction agents (etomidate 67%) and rapid-onset paralytic agents (succinylcholine 90%, rocuronium 5%), were used more often in the ICU compared to the OR (etomidate 31%, succinylcholine 15%, rocuronium 29%, cisatracurium 55%); and there was far more bougie use in the ICU (19% vs. 10%). However, despite using more hemodynamically neutral induction drugs, there were more patients with hypotension (28% vs. 4%) in the ICU. Overall, there were worse Cormack-Lehane grades of view of the airway, more attempts, and more hypoxemia (14% vs. 2%) in the ICU. Finally, difficult intubations in the ICU occurred more than twice as often compared to the OR despite the same patients and the same operators for each location within a month of each other.

Why are critically ill patients so much more difficult and dangerous to intubate? First, critically ill patients are not as tolerant of laryngoscopy and tube placement. Anatomic consequences of critical illnesses and their treatment on the upper airway, the physiologic threats to apnea, hemodynamic changes with positive pressure ventilation, and induction drugs all render patients less physiologically tolerant of the transition to mechanical ventilation. These factors combine to make critically ill patients less forgiving of human factor errors and process flaws.

Additionally, noninvasive respiratory support (NIRS) strategies like noninvasive positive pressure ventilation and high-flow nasal oxygen have changed the landscape of how acute respiratory failure is commonly managed. NIRS modalities are being utilized with increasing frequency and expanding indications. While NIRS modalities can be amazing at reducing the work of breathing, improving dyspnea and restoring gas exchange to avoid the perils of invasive mechanical ventilation, they are not a guaranteed success and NIRS failure comes with excess consequences. NIRS failure for diseases like acute decompensated heart failure or decompensated obstructive lung diseases is uncommon, but if it does fail, it will usually occur early (hours) because of fatigued respiratory muscles. Acute decompensated heart failure, COPD, and asthma also tend to improve quickly, within hours, with aggressive medical therapies and respiratory support so a trial of NIRS is not only prudent but is the standard of care. Acute hypoxemic respiratory failure due to pneumonia, acute lung injury, or acute respiratory distress syndrome (ARDS), on the other hand, is far more complicated. NIRS failure in this situation is reported up to 40% to 50% of the time, and failure usually occurs late (days) due to worsening lung injury that is often potentially amplified by patient self-inflicted lung injury. When NIRS does fail in these patients, they are now more physiologically deranged, harder to preoxygenate, often have a more threatened right ventricle, and are at higher risk of critical desaturation and cardiovascular collapse.

In critically ill or injured patients, peri-intubation adverse events involve far more than one’s skill with a laryngoscope. As Fig. 1.1 shows, there are many factors that contribute to harm, or safety, independently and in combination. Inadequacies or errors in one domain can cause harm but the opposite is not true. Excellence in a single variable cannot overcome deficiencies in other
variables and lead to safety. For example, excellence with a laryngoscope cannot overcome the harm associated with intubating a physiologically deranged patient that is not adequately preoxygenated, induced using hemodynamically embarrassing drugs, and the absence of rescue or backup equipment in the event of difficulty, or without a skilled and prepared team. Each of these domains will be explored in this book with their own chapters, but the guiding overarching message is that to achieve excellence with airway management, the entire portfolio of variables need attention, practice, and exploration.