Can We Make Intubation a Safer Procedure for Patients?

Abstract & Commentary

By Andrew M. Luks, MD, Pulmonary and Critical Care Medicine, University of Washington, Seattle. Dr. Luks reports no financial relationship to this field of study. This article originally appeared in the May 2010 issue of Critical Care Alert. It was edited by David J. Pierson, MD, and peer reviewed by William Thompson, MD. Dr. Pierson is Professor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle, and Dr. Thompson is Staff Pulmonologist, VA Medical Center; Associate Professor of Medicine, University of Washington; they both report no financial relationships relevant to this field of study.

Synopsis: This two-phase, prospective, multicenter study demonstrated that implementation of an intubation management protocol reduced the incidence of severe hypoxemia and cardiovascular collapse during endotracheal intubation when compared to standard practice, but did not improve other patient outcomes such as ICU mortality or duration of mechanical ventilation.

Source: Jaber S, et al. An intervention to decrease complications related to endotracheal intubation in the intensive care unit: A prospective, multiple-center study. Intensive Care Med 2010;36:248-255.

Endotracheal intubation is a procedure fraught with multiple, potentially life-threatening complications. Given that "care bundles" have been associated with improved management of various critical care problems including severe sepsis and ventilator sedation and weaning, Jaber and colleagues sought to determine whether implementation of an intubation management protocol would decrease the incidence of intubation-related complications.

To investigate this question, they conducted a two-phase intervention study involving all ICU endotracheal intubation procedures, except those performed for cardiac arrest, at three different hospitals. During a six-month control phase, intubation was performed by the clinician caring for the patient without use of a protocol. This was followed by a four-month lead-in period during which an intubation management bundle was developed and all ICU staff received training in the bundle practices and a six-month intervention period during which all intubations were performed using the management bundle. The bundle comprised a total of 10 interventions including the presence of two operators, fluid administration prior to intubation (500 mL normal saline or 250 mL of hydroxyethyl starch), preparation of long-term sedation, preoxygenation for three minutes using non-invasive positive pressure ventilation (NIPPV), rapid sequence intubation using etomidate or ketamine with succinylcholine, cricoid pressure, confirmation of tube placement by capnography, norepinephrine for low diastolic pressure post-intubation, initiation of long-term sedation and selection of initial ventilator settings (tidal volume 6-8 mL/kg IBW; FIO2 1.0; PEEP < 5 cm H2O; and respiratory rate 10-20 breaths/min). During each study period, they recorded the number of severe life-threatening complications (death, cardiac arrest, hypotension that persisted > 30 minute or required vasopressor support, severe hypoxemia) and mild-to-moderate complications (intubation requiring > 3 attempts, > 10 minutes, or need for another operator; esophageal intubation, gastric aspiration, arrhythmia requiring intervention, severe agitation, or dental injury). Other outcome measures included the duration of mechanical ventilation, the number of ventilator-free ICU days, ICU length of stay, and vital status upon ICU discharge.

There were 121 intubations during the control period and 123 intubations in the intervention period, with the two groups being well matched in terms of reasons for intubation and other clinical factors. Among those patients intubated during the intervention period, 75% of the total recommended number of procedures were followed by practitioners. Intubation during the intervention phase was associated with a lower incidence of life-threatening (21% vs. 34%; p = 0.03) and mild-to-moderate (9% vs. 21%; p = 0.01) complications. This result appears to be driven by a large decrease (~ 50%) in the incidence of severe hypoxemia (SpO2 < 80% during intubation attempts) and cardiovascular collapse during the intervention phase, as there were no statistically significant differences in the incidence of other complications, including cardiac arrest or death, esophageal intubation, aspiration, agitation, or dental injury. With regard to the other outcome measures, there were no differences in the duration of mechanical ventilation or ICU stay, the number of ventilator-free days, or ICU mortality between the two patient groups.


Care bundles are becoming an increasingly prevalent part of ICU practice as we now have bundles for a large number of processes including ventilator management, ventilator weaning, sedation, and treatment of sepsis, pneumonia, and myocardial infarction. Given the risks associated with endotracheal intubation and the potential for life-threatening complications, any practice that decreased such risk would be a worthwhile intervention, and the study by Jaber et al suggests there may be opportunities for improvement on this front. Although the incidence of many types of complications (e.g., esophageal intubation, gastric aspiration) was not decreased, they did show a significant decrease in the incidence of severe hypoxemia and hemodynamic instability, two important outcomes.

One of the interesting phenomena of evidence-based practice in critical care and other aspects of medicine is that when a study demonstrates a positive result, the practices used in that study are often adopted wholesale with little modification. An excellent example of this is the positive end-expiratory pressure-inspired oxygen fraction "ladder" that is frequently used in management of hypoxemia in patients with the acute respiratory distress syndrome. Although there is no physiologic basis for the ladder, it has been widely adopted at many institutions because it was the procedure used in the original ARDS Network study. While the study by Jaber and colleagues supports the notion of having an intubation bundle, there are aspects of their procedures that may not be ideal and should be re-evaluated before adoption of such a bundle at other institutions. Their protocol, for example, called for starting individuals on a tidal volume of 6-8 mL/kg predicted body weight, even though there is no evidence to support these tidal volumes in all individuals and such low tidal volumes at the initiation of mechanical ventilation may lead to atelectasis and worsening oxygenation beyond the short time frame analyzed in this study. They also call for a respiratory rate of 10-20 breaths/min, a number that is likely adequate for many patients but would lead to severe hypoventilation in patients with severe metabolic acidosis, particularly in light of the low tidal volumes in the protocol. In addition, there were items that were omitted from the protocol that might be of benefit. One could imagine, for example, adding an element to the protocol about airway assessment and mandating the availability of Eschmann stylets or ultrasound-guided laryngoscopy for any patients with unfavorable airway characteristics.

Given these concerns about the particular protocol elements, it is best to view the study by Jaber and colleagues as establishing "proof of concept" rather than providing a detailed roadmap we should all follow. What is needed prior to widespread adoption of these protocols is to refine the procedures contained within the protocol to ensure we are delivering optimal care.