INTRODUCTION — Determining the presence of an anatomically difficult airway is a critical step in deciding the best approach to tracheal intubation. This topic review will discuss the incidence, assessment, and management of the anatomically difficult airway in adults outside of the operating theater. Other aspects of airway management, including pediatric airway management, are discussed separately.
●For pediatric airway management: (see "The difficult pediatric airway" and "Emergency airway management in children: Unique pediatric considerations" and "Rapid sequence intubation (RSI) outside the operating room in children: Approach" and "Basic airway management in children" and "Supraglottic airway devices in children with difficult airways")
●For basic and advanced adult airway management: (see "Basic airway management in adults" and "Direct laryngoscopy and endotracheal intubation in adults" and "Extraglottic devices for emergency airway management in adults" and "Video laryngoscopes and optical stylets for airway management for anesthesia in adults")
DETERMINING THE PROPER APPROACH TO AIRWAY MANAGEMENT — Most emergency department (ED) endotracheal intubations are performed on an emergency basis (ie, intubation cannot be delayed or avoided). The universal emergency airway management algorithm provides the recommended approach to emergency intubation (algorithm 1 and algorithm 2) [1]. This approach is based on two key assessments of the patient prior to intubation.
●Crash airway? – The first assessment is to determine if the patient has a "crash" airway (ie, presenting in extremis with little or no cardiovascular or respiratory activity and unlikely to respond to insertion of a laryngoscope). If so, the crash airway algorithm is used (algorithm 3) [1].
●Anatomic difficulty? – If the patient does not have a crash airway, the next step is to determine if the patient represents an anatomically difficult airway. This requires assessment of specific patient attributes to predict the likelihood of difficulty in performing any of the major procedures in airway management: direct laryngoscopy and intubation, bag-mask ventilation (BMV), surgical airway management, and ventilation using an extraglottic airway.
If the patient is felt to have neither a crash nor an anatomically difficult airway, then rapid sequence intubation (RSI) is the recommended method for managing the airway, provided the administration of induction medications and neuromuscular blocking agents (NMBAs) is not contraindicated by the presence of impaired physiology. Such impaired physiology would include severe metabolic acidosis or refractory hypoxemia, making the apneic phase of RSI intolerable; and hemodynamic instability, creating a risk for circulatory collapse following the administration of RSI medications. (See "Rapid sequence intubation for adults outside the operating room".)
INCIDENCE
Incidence of difficult intubation and first-pass success — A difficult emergency department (ED) intubation can be defined as one that requires multiple attempts, multiple operators, multiple devices, excessive lifting force, external laryngeal manipulation, or performance with an inadequate glottic view. The precise incidence of difficult intubation in the ED is unknown.
The third phase of the multicenter National Emergency Airway Registry (NEAR III) project analyzed over 17,500 ED intubations and found that, in approximately 3 percent of cases, the airway was ultimately secured by a means other than the first method chosen [2]. Approximately 0.5 percent of cases required a surgical airway, with 75 percent of these performed as a rescue after failure of another method. Using these data and incorporating the success rate on first pass, use of a tracheal tube introducer ("bougie") or flexible fiberoptics, and cricothyrotomy rate, one would estimate the incidence of a difficult-to-manage airway in the ED to be approximately 10 to 15 percent. The incidence of failed airway, defined as an intubation requiring more than three attempts or dropping oxygen saturation (regardless of the numbers of attempts), was in the 3 to 5 percent range. Overall failure, meaning that no airway was secured, was reported rarely, and data regarding whether hypoxic injury or other severe airway-related injury occurred were incomplete. In a single-center study, the incidence of anticipated airway difficulty based on bedside assessment and clinician gestalt was much higher, with nearly 50 percent of ED intubations falling into that category, although assessments varied by clinician expertise and experience [3]. A retrospective study of emergency intubations performed by anesthesiologists outside the operating room reported that 351 of 3423 intubations (10.3 percent) were difficult (Cormack-Lehane grade 3 or 4) [4].
Analysis of the NEAR data found that PGY-3 and PGY-4 residents in emergency medicine performed most intubations at academic hospitals, with first-attempt success rates of 80 percent when using the GlideScope video laryngoscope, 83 percent for direct laryngoscopy, and 90 percent with the C-MAC video laryngoscope [5]. Fewer than 2 percent of intubations required more than three attempts. Ninety-seven percent of intubations by residents beyond the level of intern were successfully performed by the first operator. (See "Devices for difficult emergency airway management in adults outside the operating room".)
Historically, intubation difficulty has been defined based on the relative ease or difficulty of obtaining an adequate view of the glottis using a direct laryngoscope. The Cormack-Lehane scale is the standard gauge for such airway difficulty (figure 1) [6]. However, the increasing use of video laryngoscopes and other advanced airway tools is changing traditional conceptions of airway difficulty. In the NEAR III study, video laryngoscopy use increased from less than 2 percent in 2002 to 2004 to almost 30 percent after 2008 [5], and NEAR data from 2016 through 2018 show that video laryngoscopes are now used as the first device in nearly two-thirds of all ED intubations [5]. Many of the anatomic challenges that confront clinicians using conventional laryngoscopes are mitigated or eliminated with video-enhanced devices. Consequently, the approach to anatomic airway difficulty is evolving as video laryngoscopy becomes ever more prevalent. As an example, in a single-center study evaluating methods of difficult airway management, nearly 80 percent of all predicted difficult airways were still managed with rapid sequence intubation (RSI) when a C-MAC video laryngoscope was used, with a 90 percent first-attempt success rate [3].
When a difficult airway is anticipated, video laryngoscopy should be used. Although conventional laryngoscopes are still favored by some clinicians, and challenging direct laryngoscopy can be improved by laryngeal manipulation, proper patient positioning, and a tracheal tube introducer, analysis of the NEAR VII dataset, looking at more than 11,000 adult intubations performed during 2016 and 2017, found video laryngoscopy to be superior [5]. Multivariable regression analysis, controlling for confounders of first-attempt success, found that intubators were two to three times more likely to achieve first-attempt success using a video laryngoscope compared with "optimized" direct laryngoscopy. In a meta-analysis of 222 trials (only seven in the emergency department), video laryngoscopy improved first-attempt success and reduced rates of failed intubation, although the effect was more pronounced in the elective operating room setting as compared with the emergency department [7].
Direct laryngoscopy continues to play an important role in hospitals without access to video equipment and possibly in particular clinical situations. In such settings, it is important to maintain direct laryngoscopy skills.
Incidence of difficult bag-mask ventilation — Difficult bag-mask ventilation (BMV) has been studied extensively in the operating room but is challenging to assess in other settings [8-14]. Overall, although definitions vary among studies, the incidence of difficult BMV is universally low, and true failed BMV is rare.
Researchers in one prospective observational study of 1502 operating room patients defined difficult BMV as the inability of a single clinician to maintain oxygen saturation above 92 percent or provide adequate ventilation. They found BMV to be difficult in about 5 percent of patients and impossible in only a single patient [8]. Subsequent, larger studies found difficult BMV to occur in approximately 3 percent of patients, with the lower incidence possibly due to exclusion of some patients who would have been difficult to ventilate [9,15,16]. In a cluster randomized trial of pre-assessment for difficult BMV involving 94,006 patients in 26 Danish anesthesia departments, researchers reported an incidence of unpredicted difficult BMV below 1 percent in both the intervention group, in which potential BMV difficulty was assessed using 11 explicit criteria (0.91 percent); and the control group, in which assessment was left to the anesthesiologist's discretion (0.88 percent) [14].
Among the selected patients studied in the operating room, the combined inability to intubate and inability to perform BMV occurs in only 1 in 5000 to 1 in 10,000 patients [17]. This likely represents a reasonable estimate of the incidence of unexpected difficulty in the ED (ie, patients without identifiable characteristics associated with difficulty but for whom intubation and BMV are difficult). This observation underscores the safety of an approach to airway management that includes a preintubation assessment for difficult intubation and difficult BMV. In the absence of markers of difficult intubation or difficult BMV, the clinician will rarely encounter a patient who is both impossible to intubate and impossible to ventilate with a bag-mask. (See "Basic airway management in adults", section on 'Bag-mask ventilation'.)
Incidence of difficult cricothyrotomy — The incidence of difficult surgical cricothyrotomy in the ED, where the procedure is most often performed, or in the operating room is not known. Cricothyrotomy is not a technically difficult procedure in most patients, but infrequency of performance, time pressure during the procedure, lack of regular training, and the anxiety accompanying airway failure undoubtedly contribute to any difficulty encountered.
Several case series of emergency surgical cricothyrotomies have reported success rates approaching 100 percent with complication rates of 5 to 14 percent. Difficulty related to the procedure, including cases requiring excessive time to secure the airway, is not properly captured in these retrospective series [18,19]. (See "Emergency cricothyrotomy (cricothyroidotomy)".)
Incidence of difficult extraglottic airways — Extraglottic airways are most often used as rescue devices in the ED. These include the supraglottic airways (ie, laryngeal mask airways), designed to sit just above the glottis; and the retroglottic airways, designed to be inserted into the esophagus, and which provide proximal (oropharyngeal) and distal (esophageal) seals to permit "side-stream" ventilation (eg, Combitube, King LT Airway, EasyTube, and others).
The incidence of difficult insertion is not known for most of these devices. Disposable and reusable laryngeal mask airway devices have been shown to permit effective ventilation in 92 and 97 percent, respectively, of anesthetized patients [20]. Emergency medical technicians (EMTs) have used the Combitube in the prehospital setting, obtaining ventilation rates ranging from 79 to 95 percent [21-23]. One retrospective review of 162 patients managed by EMTs reported a success rate of only 70 percent with a 27 percent complication rate using the Combitube [24]. In a single-center registry of 14,480 elective operating room cases managed with either an iGel or LMA, the rate of initial difficult ventilation was 0.5 percent; however, 60 percent were corrected with either repositioning or re-insertion. Overall, successful placement and ventilation occurred in 99.8 percent of all cases [25].
Success rates for retroglottic airways used in the ED are unknown. Analysis of data from the NEAR registry showed extraglottic airways were seldom used as rescue devices and never used as a primary airway in 7712 ED patients [26]. Patients with disrupted upper airway anatomy are poor candidates for a supraglottic or retroglottic airway device, as both rely on creating a seal between an inflatable cuff and the patient's mucosa.
IDENTIFYING THE DIFFICULT AIRWAY — Failure to identify, in advance, characteristics associated with difficult intubation or ventilation is one of the leading causes of failed airways in the operating room [27,28]. This concept likely extends to emergency airway management outside the operating room. Difficult airway assessment requires determination of the potential for difficulty with bag-mask ventilation (BMV), laryngoscopy and intubation, cricothyrotomy, and extraglottic airway insertion.
The LEMON assessment for difficult intubation — The LEMON assessment uses a series of physical evaluations to determine whether difficult laryngoscopy and intubation is anticipated. Each step is described:
L: Look externally — This refers to the clinician's general impression that the airway will be difficult. Does the patient have abnormal facies or body habitus, unusual anatomy, or facial trauma [29-31], any of which can be expected to create difficulty? A general impression of airway difficulty is reasonably specific but not particularly sensitive. If an operator observes the patient and anticipates difficulty, that assessment likely is correct. Conversely, absence of obvious external markers of airway difficulty does not ensure success.
E: Evaluate (3-3-2 rule) — The size of the mandible, the distance between the mentum and the hyoid bone, and the extent of mouth opening are all important geometric determinants of the success of direct laryngoscopy [32,33]. These relationships are represented by the 3-3-2 rule (picture 1) [34,35]. The rule describes three measurements found in normal patients (ie, patients in whom difficult laryngoscopy is not expected).
●3: This assessment indicates the ease of access to the airway. A normal patient can open his mouth sufficiently to permit three of his own fingers to be placed between the incisors. Adequate mouth opening facilitates both insertion of the laryngoscope and obtaining a direct view of the glottis.
●3: This assessment provides an estimate of the volume of the submandibular space. A normal patient is able to place three of his fingers along the floor of the mandible between the mentum and the neck/mandible junction (near the hyoid bone).
●2: This assessment identifies the location of the larynx relative to the base of the tongue. A normal patient is able to place two fingers in the superior laryngeal notch (ie, the space between the superior notch of the thyroid cartilage and the neck/mandible junction, near the hyoid bone). If the larynx is too high in the neck, direct laryngoscopy is difficult or impossible because of the angles that have to be negotiated to permit visualization.
Variations in patient size are accommodated by using the patient's fingers as the standard for measurement. Since the emergency department (ED) patient is often uncooperative or unable to perform the required steps, size is estimated by comparing the examiner's fingers with the patient's. The examiner's fingers are then used to estimate the proportions involved. Failure to achieve these three dimensions predicts difficulty visualizing the glottis during direct laryngoscopy.
M: Mallampati score — The Mallampati classification is a simple scoring system to help predict difficult intubation (figure 2) [36,37]. It has been prospectively validated in several studies, although not as a solitary (ie, sufficient) predictor of difficult intubation [38,39]. The Mallampati score is best used as one part of a global airway assessment [40,41].
The Mallampati class, ranging from I to IV, relates the amount of mouth opening to the size of the tongue and provides an estimate of space for oral intubation by direct laryngoscopy. In general, Mallampati class I or II predicts easy laryngoscopy, class III predicts difficulty, and class IV predicts extreme difficulty.
Many ED patients are unable to cooperate with a Mallampati assessment. In such cases, the examiner should gently open the mouth, if possible, and use a direct laryngoscope blade in the manner of a conventional tongue blade to assess the size of the tongue compared with that of the oropharynx. If this assessment reveals a large tongue-to-oropharynx ratio or it cannot be done, the clinician should assume that direct laryngoscopy will be difficult.
O: Obstruction/obesity — The presence of upper airway obstruction interferes with both laryngoscopy and intubation. A supraglottic mass or infection, trauma with hematoma, injury with disruption of the upper airway, and vocal cord masses (eg, tumor), among other conditions, can obstruct the view of the glottis, block access for tube insertion by narrowing the airway, or both. The redundant tissues in the upper airway of the obese patient make visualization of the glottis by direct laryngoscopy more difficult, and an oversize laryngoscope blade may be required.
N: Neck mobility — Ideally, the patient is placed in the sniffing position for intubation. The sniffing position is achieved by flexing the neck forward on the body (thoracic spine) and elevating the head. Thus, decreased cervical spine mobility compromises the direct laryngoscopic view [42,43]. Proper positioning for direct laryngoscopy is discussed in detail separately. (See "Direct laryngoscopy and endotracheal intubation in adults", section on 'Positioning the patient'.)
Medical conditions such as psoriatic or rheumatoid arthritis, ankylosing spondylitis, or simply the degenerative joint disease that accompanies aging can greatly reduce neck mobility. In uncooperative, nontrauma patients, neck mobility can be assessed by passively extending the neck.
Blunt trauma patients require in-line stabilization of the cervical spine during intubation, which also limits glottic view. Most trauma patients, although identified as difficult airways because they require in-line cervical spine stabilization, can nonetheless be intubated successfully orally, unless other difficult airway markers are present.
Multiple studies have attempted to identify patient characteristics predictive of difficult laryngoscopy and intubation [33,40]. Not all markers of difficulty are applicable to ED patients, and many are too complex or require evaluation (eg, magnetic resonance imaging [MRI]) that is not feasible in the emergency situation. The LEMON mnemonic described above (table 1) was developed by researchers in emergency airway management [34,35]. A prospective observational study of 156 patients undergoing intubation in the ED found that the LEMON evaluation accurately stratified patients according to the risk of difficult intubation [44]. A subsequent large registry study performed in Japanese EDs evaluated the benefit of a "modified" LEMON assessment (ie, LEMON without Mallampati classification or measurement of thyromental distance) and discovered its most impactful feature was its ability to rule out difficulty [45]. They found that when patients were completely LEMON negative, difficulty was rarely encountered (negative predictive value of 98 percent). Subsequently, the LEMON mnemonic was adopted by the American College of Surgeons' Advanced Trauma Life Support course.
Difficult video laryngoscopy — Video laryngoscopes are a relatively recent innovation, and difficult intubation using these tools is less common. Thus, assessment tools for predicting difficult video laryngoscopy are incomplete. The LEMON mnemonic applies to direct laryngoscopy, not to video laryngoscopy, although some of the attributes (eg, obesity) are likely to have some validity for video laryngoscopy as well.
Although newer mnemonics encompassing both direct laryngoscopy and video laryngoscopy have been developed, the criteria are broad and do not provide the clinician with specific guidance for accurate identification [46]. Research to date has identified the upper lip bite test (ULBT) as one predictor of difficult video laryngoscopy. In a systematic review of 27 studies involving over 18,000 patients, the ULBT showed high specificity and high negative predictive value [47]. The ULBT is performed by having the patient extend their jaw and cover the upper lip with their lower incisors. If the patient can fully cover their upper lip with their lower incisors, difficult laryngoscopy is unlikely; if the patient cannot reach their upper lip, difficulty is significantly more likely. (See "Airway management for induction of general anesthesia", section on 'Airway examination'.)
It will continue to be challenging to develop a reliable set of predictors of difficult intubation using a video laryngoscope as these instruments almost universally provide grade 1 or 2 views (400/400 grade 1 or 2 in one study of the GlideScope [48] and 60/60 in a study using the Storz C-MAC [49]). In addition, differences in blade shape (hyper-angulated versus standard geometry) mean that any set of patient characteristics might have different effects on intubation success based solely on the design of the device.
Difficult bag-mask ventilation — Assessment for difficult BMV is performed immediately after the LEMON evaluation for difficult intubation. The LEMON evaluation and BMV technique are reviewed separately. (See 'The LEMON assessment for difficult intubation' above and "Basic airway management in adults", section on 'Bag-mask ventilation'.)
Difficult BMV has been studied in anesthesia populations, and the incidence appears to be low. (See 'Incidence of difficult bag-mask ventilation' above.)
The predictors of difficult BMV are summarized by the mnemonic ROMAN (table 2) and discussed immediately below [8,9,14,15,35,50].
There is no clear correlation between each individual attribute and the degree of difficulty, but assessment of all the attributes helps to determine whether difficulty is likely [14]. If no markers are present, BMV is unlikely to be difficult. However, while uncommon, difficult BMV does occur, and clinicians who perform airway management must be prepared.
●R: Radiation/restriction – Head and neck radiation is strongly associated with difficult rescue mask ventilation, likely due to reduced pliability of upper airway soft tissue. Restriction is used to connote restriction of forward gas flow (ie, from the bag and mask apparatus into the patient's lungs) and resistance to ventilation that occurs in conditions that increase the required inspiratory pressure to ventilate the lungs, and includes asthma, chronic obstructive pulmonary disease (COPD), pulmonary edema, widespread infiltrates, and any other condition that decreases pulmonary compliance.
●O: Obstruction/obesity/obstructive sleep apnea – Obstruction of the upper airway, although not widely studied, will make BMV more difficult, as increased pressures will be required to ensure that gas flows past the obstruction in both directions. Obesity (body mass index [BMI]>26) is an independent marker of difficult BMV. Redundant upper airway tissue and the combination of chest wall weight and resistance from abdominal contents all impede airflow. Late third trimester pregnancy is a surrogate for obesity with respect to BMV as it creates many of the same problems. Placing the bed at an angle with the head higher than the feet (ie, reverse Trendelenburg) may reduce impedance to airflow from abdominal weight. (See "Emergency airway management in the morbidly obese patient".)
●M: Mask seal/Mallampati/male – Mask seal requires reasonably normal anatomy, absence of facial hair, lack of interfering substances (such as excessive vomitus or bleeding), and the ability to apply pressure to the face with the mask. Poor Mallampati classification and male sex are associated with challenging mask ventilation as well.
●A: Age – In one study, age >55 years was a marker of difficult BMV [8]. The general loss of elasticity of tissues and the increased incidence of restrictive or obstructive pulmonary disease most likely make ventilation more difficult. Fifty-five years is not a distinct cutoff, but as patients age or appear to be physiologically aged, it is reasonable to assume that BMV difficulty will increase.
●N: No teeth – Edentulousness creates difficulty with BMV. Teeth provide a framework against which the mask sits and support the cheeks, enhancing mask seal. If a patient has dentures, they should be left in situ during BMV, where they are of benefit, then removed for direct laryngoscopy, where they are detrimental [51].
Difficult cricothyrotomy — Assessment for difficult cricothyrotomy is performed after the LEMON evaluation for difficult intubation and the ROMAN evaluation for difficult BMV.
Difficult cricothyrotomy is caused by difficult access to the anterior neck, inability to identify landmarks, distortion of the anatomy, or abnormalities of the tissues; and can be assessed using the mnemonic SMART (table 3) [35]. Evaluation for difficult cricothyrotomy requires palpating the structures overlying the larynx, identifying the cricothyroid membrane, and identifying potential problems with surgical access. Identification of the cricothyroid membrane is more difficult in the obese and in women [52]. Compared with bedside ultrasound, many commonly used palpation techniques for identifying the cricothyroid membrane are only marginally accurate (46 to 62 percent) [53].
Difficult extraglottic airway placement — Placement of a rescue device (eg, laryngeal mask airway) can be difficult if mouth opening is limited, if the airway is disrupted or distorted (eg, by swelling), or if debris such as teeth or bone fragments are present (table 4). Increased airway resistance can prevent effective ventilation via an extraglottic airway.
THE DIFFICULT AIRWAY ALGORITHM — When a difficult airway is identified using the LEMON, ROMAN, or other bedside criteria, the difficult airway algorithm is used (algorithm 4). Predictors of airway difficulty when using a video laryngoscope remain to be fully defined; pending that, it is reasonable to apply the same principles of airway assessment and decision-making, regardless of whether a traditional direct or a video laryngoscope will be used.
Key questions guiding the approach — Although the difficult airway algorithm appears complex, it is really a series of simple questions:
●Is oxygenation adequate? – If oxygen saturation is falling despite oxygenation maneuvers, this is a failed airway, even before any intubation attempt, and the failed airway algorithm should be used (algorithm 5).
●Is a "forced-to-act" scenario present? – If there is dynamic airway deterioration (eg, anaphylaxis), rapid sequence intubation (RSI) is permitted despite anatomic challenges.
●Is RSI a reasonable approach? – RSI may be reasonable even though the airway has been identified as either anatomically or physiologically difficult.
●If RSI is not reasonable, what can be learned from an awake look at the glottis?
●Is there still sufficient time (ie, oxygenation is adequate and airway is not deteriorating)? – If so, a number of alternatives remain. If not, the situation represents a failed airway.
Applying the algorithm — Before initiating the steps in the difficult airway algorithm, the first action is to obtain any necessary assistance (personnel, equipment, airway devices) at the earliest opportunity (algorithm 4). As this is being done, the stepwise management of the difficult airway can proceed.
Is the operator forced to act? — In the face of precipitous airway deterioration, especially with a patient who is agitated and unable to cooperate or a patient with rapidly progressing airway swelling, immediate administration of an induction agent and neuromuscular blocking agent (NMBA) may be indicated, even though the airway is identified to be difficult. In such a situation, patient conditions force the operator to act immediately to forestall deterioration to respiratory arrest or complete airway obstruction.
As an example, consider a morbidly obese patient with severe status asthmaticus, who is combative and fatigued, with oxygen saturations falling into the upper 80s despite maximal therapy, including high-flow oxygen (if it can be kept in place). In such circumstances, a prompt decision to give RSI drugs and create the best possible situation for a single best attempt at tracheal intubation, whether by laryngoscopy or surgical airway, often is preferable to considering other (likely impossible) approaches as the patient progresses toward respiratory arrest and death. If this single best attempt is not successful, a failed airway is present, and the operator proceeds to the failed airway algorithm.
As another example, consider a patient with rapidly worsening airway swelling due to anaphylaxis. In such a patient, the rate of decline precludes the standard approach to partial airway obstruction (ie, awake intubation with sedation, topical anesthesia, and flexible nasal or oral laryngoscopy) as the time required for the procedure would result in complete airway obstruction.
An alternative approach, termed "delayed sequence intubation" (DSI), has been advocated for use in the "forced to act" category of patients who are agitated or uncooperative as described above. Proponents of DSI recommend administration of intravenous ketamine sufficient to gain control of the patient to permit pre-oxygenation, followed by administration of an intubating dose of an NMBA. Although DSI has been recommended on podcasts and in educational presentations, formal assessment of the technique is limited to case series and retrospective studies [54,55]. In a retrospective, single-center, before–and-after study, the use of ketamine prior to RSI as part of a multi-interventional "no desaturation" bundle for out-of-hospital tracheal intubation led to lower rates of peri-intubation hypoxia [55]. No increase in adverse events occurred in the ketamine bundle group. Pending further study, we do not feel there is sufficient evidence at present to support routine use of DSI in the emergency department (ED).
Is there time? — If the operator is not forced to act, the next determination is whether the patient's oxygenation is adequate. Adequate oxygenation is defined as the ability to maintain the patient's oxyhemoglobin saturation (SpO2), as measured by a pulse oximeter, at 90 percent or higher. A patient with chronically low, but stable, oxygenation might be considered to have adequate oxygenation at a lower threshold value, but the risk of rapid desaturation is much higher [56]. Adequate SpO2 allows the clinician time to plan and implement a stepwise approach to airway management. Techniques for maximizing oxygenation are discussed separately. (See "Rapid sequence intubation for adults outside the operating room", section on 'Preoxygenation'.)
If oxygenation is inadequate, attempts are made to improve it with supplemental oxygen, bag-mask ventilation (BMV), or assisted ventilation. If there is insufficient time to plan a methodical approach to the airway because adequate SpO2 cannot be maintained, the airway is considered to be a failed airway (algorithm 5). The combination of an anticipated difficult intubation and inability to maintain SpO2 at adequate levels is a surrogate for the "can't intubate, can't ventilate" failed airway. (See "Approach to the failed airway in adults outside the operating room".)
Is RSI reasonable? — Many difficult airways are managed using RSI. The presence of a difficult airway does not preclude use of RSI but requires the operator to ask three key questions:
●Am I confident I can oxygenate the patient using a bag and mask or an extraglottic device?
●Am I confident I can intubate?
●Is the patient at risk of severe hypoxia or circulatory collapse following administration of RSI medications?
RSI is not a prudent option if the operator is not confident the patient can be ventilated adequately with a bag and mask or an extraglottic device, in which case an awake technique is advised. Even if BMV is deemed possible, RSI is not advisable unless the clinician believes oral intubation will likely be successful.
In patients with deranged physiology (eg, profound hypoxemia, metabolic acidosis, hypotension) who are thus at risk for rapid oxygen desaturation, the presence of a challenging airway or even a longer-than-normal laryngoscopy may place the patient in danger, as a contest emerges between the clinician's efficiency placing the tracheal tube and initiating ventilation and the patient's descent along their oxyhemoglobin desaturation curve [57]. If the clinician loses that race, the patient is at risk for death or anoxic brain injury. RSI can be performed (following the universal emergency airway algorithm) if the clinician answers "yes" to the first two questions above and severely compromised physiology does not preclude the use of RSI medications (algorithm 1 and algorithm 2). Methods for reducing the risks associated with deranged physiology are discussed separately (table 5 and table 6). (See "Rapid sequence intubation for adults outside the operating room", section on 'Preintubation optimization'.)
Often, a double setup is used when RSI is planned in the setting of a potentially difficult airway. Double setup means the alternative (ie, backup) airway device and approach (often a surgical airway) are identified and readied before beginning the intubation sequence.
Awake technique — "Awake" is an imprecise term that connotes direct examination of the upper airway, facilitated by light to moderate sedation and topical anesthesia, using a direct laryngoscope, flexible endoscope, or rigid fiberoptic or video laryngoscope. The term is imprecise because it refers to a number of different approaches, having only in common that each is done while the patient is breathing on their own and, although technically awake, is sedated as for a painful procedure to permit laryngoscopy.
The patient receives procedural sedation and topical anesthesia (depending upon their clinical condition) to permit laryngoscopy and intubation if indicated. The laryngoscopy may identify barriers that preclude oral intubation and mandate cricothyrotomy or, conversely, may identify a straightforward upper airway amenable to RSI. The patient may simply be inspected and a decision made to proceed with an intubation plan, or may be intubated during the awake look depending upon the circumstances, likelihood of deterioration, and the judgment of the operator.
Alternative approaches — If the awake examination identifies that RSI is not advisable, and the patient cannot be intubated during laryngoscopy (eg, glottis cannot be visualized during awake look), a number of airway management options remain, provided an adequate SpO2 is maintained. The goal with the difficult airway, as with any emergency intubation, is to place a cuffed endotracheal tube in the trachea. A number of devices can be used to accomplish this end.
An intubating laryngeal mask airway (ILMA) (picture 2) may be placed, then an endotracheal tube passed through it. A common approach is to load an endotracheal tube over top of a flexible fiberoptic device. The flexible scope is navigated into the airway using the ILMA device as a conduit to the laryngeal inlet, where it acts as an intubation guide over which an endotracheal tube can be advanced into place. A video laryngoscope likely will achieve a glottic view superior to that of a standard laryngoscope, enabling intubation. A fiberoptic or video intubating stylet may also be used. (See "Devices for difficult emergency airway management in adults outside the operating room".)
Primary cricothyrotomy (ie, cricothyrotomy as a planned airway intervention, rather than a rescue) may be indicated if oral access or the supraglottic region of the airway is anatomically disrupted. Blind nasotracheal intubation may rarely have a role but is best reserved for situations in which no other device is available or felt to be appropriate (eg, massive hemorrhage obscuring vision).
Failed airway — If, at any time during the evaluation or management of the difficult airway, an SpO2 cannot be maintained at 90 percent, or at least held stable in a viable range, the difficult airway becomes a failed airway, and the failed airway algorithm is followed (algorithm 5). (See "Approach to the failed airway in adults outside the operating room".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Airway management in adults".)
SUMMARY AND RECOMMENDATIONS
●Determine proper airway approach – Whenever possible, the clinician preparing to perform rapid sequence intubation (RSI) should perform an airway assessment to determine the difficulty of intubation, bag-mask ventilation (BMV), extraglottic device placement, and cricothyrotomy. Impaired physiology (eg, hypoxia, metabolic acidosis, hemodynamic instability) should also be considered in airway management decision-making (table 5 and table 6). (See 'Determining the proper approach to airway management' above.)
●Importance of planning – Failure to recognize and plan for a difficult intubation is a leading factor contributing to a failed airway and poor patient outcomes. (See 'Identifying the difficult airway' above.)
●Predicting the difficult airway – Attributes predictive of difficult direct laryngoscopy and intubation can be identified using the LEMON mnemonic: Look, Evaluate (3-3-2), Mallampati, Obstruction/obesity, Neck mobility (table 1). (See 'The LEMON assessment for difficult intubation' above.)
●Predicting difficult bag-mask ventilation – Difficult BMV can be predicted using the ROMAN mnemonic: Radiation and restriction, Obstruction/obesity/obstructive sleep apnea, Mask seal/Mallampati/male, Age over 55, No teeth (table 2). (See 'Difficult bag-mask ventilation' above.)
●Difficult airway algorithm – The difficult airway is managed according to the difficult airway algorithm (algorithm 4). The key determinations are whether the operator is "forced to act" and, if not, whether the patient's oxygenation is adequate (ie, oxyhemoglobin saturation [SpO2] >90 percent). (See 'The difficult airway algorithm' above.)
•Obtain assistance and equipment – The first action is to obtain all necessary assistance (personnel, equipment, airway devices).
•If "forced to act" situation (eg, imminent airway obstruction) – Administration of RSI drugs is permitted even if intubation is anticipated to be difficult. (See 'Applying the algorithm' above.)
•Improve oxygenation – If the operator is not forced to act, but oxygenation is inadequate, attempts are made to improve it with supplemental oxygen or BMV.
•If adequate SpO2 cannot be maintained – In the case of inadequate oxygen saturation, the airway is considered to be a failed airway and is managed accordingly (algorithm 5). (See "Approach to the failed airway in adults outside the operating room".)
●Awake intubation – When the clinician anticipates that intubation will be difficult and doubts whether the patient can be successfully ventilated using a bag-mask, we suggest neuromuscular blockade be avoided and an awake technique used to manage the airway (Grade 2C). An awake approach is also preferred in patients with hypoxemia or metabolic acidoses who are at risk of profound deterioration in response to RSI medications. (See 'Awake technique' above.)
ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge Michael Murphy, MD, who contributed to an earlier version of this topic review.
