Special Feature

Management of the Difficult Airway

Michael Gibbs, MD, FACEP Dr. Gibbs is Chief, Department of Emergency Medicine, Maine Medical Center, Portland, ME. Dr. Gibbs has reported no relationships with companies having ties to the field of study covered by this CME program.

Effective and timely management of the airway is our paramount responsibility. Once the decision to intubate has been made, the emergency physician faces a number of daunting questions — often with little information, no time, and zero tolerance for failure. Does the patient have a difficult airway? Can this be predicted using physical examination? What is the best approach if standard techniques are unsuccessful? To avoid disaster, these questions must be considered before the succinylcholine is pushed. This article provides a rational framework for doing that.

Step 1. Defining the Problem

In 2003 the American Society of Anesthesiology (ASA) Difficult Airway Task Force provided specific definitions for both the difficult and the failed airway that since have become widely accepted.1 A difficult airway is broken down into three unique attributes that may exist alone or in combination: 1) difficult bag-valve-mask ventilation; 2) difficult laryngoscopy and intubation; and 3) difficult surgical airway. The failed airway is present when either of the following exists: 1) inability to ventilate or intubate the paralyzed patient; and 2) three failed intubation attempts by an experienced operator. Strict adherence to this level of definition specificity helps guide effective clinical decision-making; it raises the bar for performance, and it allows clinicians to communicate accurately.

Step 2. Understanding the Scope of the Problem

A recent review of the National Emergency Airway Registry (NEAR) described data from more than 6,000 airway cases at 31 university-affiliated EDs in the United States and Canada.2 In this large, diverse cohort of adults and children, the first intubator was successful 90% of the time (95% CI 89-91%), including 83% (95% CI 82-84%) on the first attempt. The overall success rate for the first intubator was highest with rapid sequence intubation (RSI) (91%). Intubation success rates were directly proportional to years of training. Further analysis of this group demonstrates that the first technique failed in about 3% of cases.3 Two-thirds of these first technique failures occurred when methods other than rapid sequence intubation were used (e.g., nasotracheal intubation and intubation with sedation alone). This fact provides additional compelling evidence that RSI should be the technique of choice in the majority of our patients. Rescue techniques included RSI (49%), cricothyroidotomy (21%), and a number of alternative airway techniques (30%).

These two studies illuminate several important principles. First, we are good at what we do. Second, we will fail. Third, because we don't fail very often, becoming an airway rescue expert is a challenge.

Step 3. Predicting the Problem

A number of clinical features have been proposed as potential markers of difficult intubation (Table 1). The accuracy of any single feature or combination of features for predicting the presence or absence of a difficult airway is not clear. An evidence-based literature review performed by the ASA Difficult Airway Taskforce lacked sufficient evidence to definitively evaluate the accuracy of the physical examination at predicting the presence of a difficult airway, although data suggested that "abnormal findings obtained during an airway exam may be associated with a difficult airway."1 Shiga and colleagues performed a meta-analysis of 35 studies (n = 50, 760) relating to the accuracy of predictive tests for difficult intubation. Study selection criteria included: 1) prospective design, 2) at least one bedside diagnostic test performed, 3) data reported for true positives, true negatives, false positives, false negatives, 4) adequate blinding and consecutive enrollment. Once again, the features examined (Mallampati score of 3 or more, thyromental distance < 2 fingerbreadths, sternomental distance < 3 fingerbreadths, and mouth opening < 3 finger breadths) were associated with difficult airways, but had unacceptably low specificity as single markers. Adding several markers did increase specificity.4

Table 1. Difficult Airway Predictors

More recently, a prospective single-center study evaluated the predictive value of the 10-point LEMON criteria (Table 2) in 156 consecutive ED patients.5 For each patient, the score was compared with the widely accepted Cormack laryngoscopic view score (1 = full view, 4 = glottis not visualized). At intubation, 114 patients were classified as Cormack grade 1 (easy intubation), and 42 were classified as grade 2 or higher (difficult intubation). Patients in the difficult intubation group had significantly higher LEMON scores. Of the criteria used to calculate the score, only large incisors, inter-incisor distance less than 3 fingerbreadths, and thyromental distance less than 2 fingerbreadths were associated with difficult intubation. These data underscored the critical function of a focused and sophisticated airway evaluation whenever intubation is planned. However, it must be acknowledged that there is no perfect formula. I like the LEMON law, and use it routinely.

Table 2. The LEMON Law

Step 4. Managing the Problem

Dealing with a perilously difficult or failed airway is perhaps the most treacherous set of circumstances facing the emergency physician. While a number of difficult airway algorithms have been published in the anesthesia literature,1,6,7 these are not relevant to the ED. In addition, evidence-based, prospectively validated guidelines are lacking in our specialty. That being said, a number of fundamental principles can help guide decision-making. There are multiple alternative airwaymanagement devices available today, with more on the way (Table 3). Cricothyroidotomy is no longer the only option. Airway rescue devices can be classified logically as supraglottic (inserted above the glottis) and infraglottic (inserted below the glottis). With the exception of fiberoptic-guided devices, supraglottic rescue devices are blind—they are inserted without direct vision of the airway. As such, these generally are not recommended in patients with significant disruption of airway anatomy (e.g., hemorrhage, edema, tumor, or foreign body). Obviously this recommendation is not absolute; clinical judgment and experience are key. The element of time is critical. In patients who can be ventilated effectively, our menu of choice is potentially wide. Conversely, it should be recognized that the solution to a can-not-intubate-can-not-ventilate scenario must be an immediate definitive airway. Using the device you know will be the most effective method in your hands. It is our responsibility to become proficient with a number of devices that will allow each of us to manage the spectrum of airway disasters landing on our doorstep.

Table 3. Airway Rescue Devices


1. ASA Difficult Airway Taskforce. Practice guidelines for management of the difficult airway. Anesthesiology 2003; 98:1269-1277.

2. Sagarin MJ, et al. Airway management by US and Canadian emergency medicine residents: A multicenter analysis of more than 6,000 endotracheal intubation attempts. Ann Emerg Med 2005; in press.

3. Bair AE, et al. The failed intubation attempt in the emergency department: Analysis of prevalence, rescue technique and personnel. J Emerg Med 2002; 23:131-140.

4. Shiga T, et al. Predicting difficult intubation in apparently normal patients. Anesthesiology 2005;103:429-437.

5. Reed MJ, et al. Can an airway assessment score predict difficulty at intubation in the emergency department? Emerg Med J 2005; 2:99-109.

6. Henderson JJ, et al. Society guidelines for management of the unanticipated difficult airway. Anesthesia 2004;59:675-694.

7. Crosby ET, et al. The unanticipated difficult airway with recommendations for management. Canad J Anaesthesiol 1998, 45:757-776.