Summary
There is no consensus as to the ideal approach for the anaesthetic management of the adult obstructed airway and there are advocates of awake fibreoptic intubation, inhalational induction and intravenous induction techniques. This review considers the different options available for obstruction at different anatomical levels. Decisions must also be made on the urgency of the required intervention. Particular controversies revolve around the role of inhalational vs intravenous induction of anaesthesia, the use or avoidance of neuromuscular blockade and the employment of cannula cricothyroidotomy vs surgical tracheostomy.
Optimal management of the obstructed adult airway remains controversial with no consensus as to the ideal approach. Opinions differ as some experts propose techniques that other experts find unacceptable and explicitly criticise [1–3]. There are advocates of inducing general anaesthesia by an inhalational route and avoiding neuromuscular blockade [4]; of inducing general anaesthesia by the intravenous route and using neuromuscular blockade [5]; of avoiding general anaesthesia altogether and securing an airway by an awake fibreoptic intubation technique [6, 7]; of tracheostomy under local anaesthesia [8]; or of insertion of a transtracheal catheter under local anaesthesia [9]. The advocates of these various techniques claim that each respectively provides the best management and the range of choices can appear very confusing for the anaesthetist who is actually faced with a patient who has an obstructed airway. Will they be later criticised for using any particular technique? What should they do if the technique they have chosen fails?
It is important to remember that the quality of evidence for the management of the obstructed airway is of a relatively low level. At best, reports are non-analytical case control or cohort studies with a high risk of confounding or bias [10]. So a more pertinent question may be: what is the best technique or techniques for the management of the obstructed airway, considering the experience of the anaesthetist, the team available, the location, the surgical expertise, the equipment available, the urgency, the patient’s comorbidity and the site and extent of airway obstruction?
It is also useful to remember that all airway plans can fail. Back-up plans (‘Plans B, C, D’, etc.) are probably just as, if not more, important than any primary plan (‘Plan A’) [11]. These back-up plans require the same level of thought as the primary airway plan. There needs to be early recognition by the team that the original plan is not working, and good communication and execution of the predetermined back-up plan. Whilst this may seem self-evident, unless this has been thought through, the clinical situation can quickly deteriorate.
For example, unless there is prior communication, the absence of a scrubbed surgeon for an emergency surgical airway when the airway becomes unmanageable, or the absence of vital equipment such as high-pressure source ‘jet’ ventilation and a rigid bronchoscope, may all contribute to failure. While there may be little consensus on the initial plan for the management of the obstructed airway, the concept of a team approach with surgeon and theatre team is essential [11].
The Fourth National Audit Project (NAP4) of the Royal College of Anaesthetists and the Difficult Airway Society is the largest and most comprehensive airway project undertaken in the UK, and important lessons can be learnt for general airway management and more specifically, airway management of patients with head and neck pathology [12, 13]. Seventy-two patients with head and neck pathology developed severe complications of airway management including death, brain damage, emergency surgical airway and an unanticipated intensive care unit (ICU) admission or prolongation of ICU stay. Seventy percent of these reported cases were associated with an obstructed airway, and overall these 72 patients accounted for 40% of all the cases reported to NAP4 [11].
The obstructed airway describes any obstruction within the airway from the nasopharynx, pharynx and larynx through to the trachea and lower airways. The classic symptoms and signs of obstruction at the various sites vary and include stertor at a nasopharyngeal level, gurgling at an oropharyngeal level, inspiratory stridor at a supraglottic level, inspiratory or biphasic stridor at a glottic or subglottic level, and expiratory wheeze at a tracheobronchial level [14]. The site of obstruction will have a great influence on the efficacy and suitability of any given technique.
Patients can present acutely with signs of airway obstruction including altered breathing position, an inability to lie flat, stridor, tachypnoea, accessory muscle use, sternal retraction, tracheal tug, an inability to maintain alveolar ventilation, hypoxia and ultimately exhaustion leading to quiet breathing that may not be appreciated as serious. Patients may also present with a more chronic obstruction in which they can appear comfortable with no obvious signs or symptoms of airway obstruction despite airway diameters of only a few millimetres [11]. The difference in symptoms and signs between acute (minutes/hours) and chronic (weeks/months) airway obstruction is due largely to respiratory muscle conditioning. In acute obstruction the untrained respiratory muscles tire early when trying to generate the intrapleural pressure changes required during the respiratory cycle. The oxygen requirement of respiratory muscles at maximal effort may be such a substantial proportion of available oxygen delivery that the supply/demand situation is unstable with a rapid onset of failure to maintain alveolar ventilation. Fully trained respiratory muscles can easily sustain an adequate alveolar minute ventilation at rest through a 3-mm orifice but cannot meet the requirement in exercise.
Primary medical management is directed at maintaining oxygenation and involves careful observation, humidified oxygen, nebulised adrenaline, intravenous corticosteroids (both to reduce airway oedema) and in some instance helium/oxygen mixtures (to improve gas flow into the lungs). The urgency of any subsequent surgical intervention will be determined by the clinical presentation and usually, for a chronically obstructed airway, there is time to undertake investigations that may include computed tomography (CT) or magnetic resonance imaging (MRI) scans and careful nasal endoscopy to determine the extent and site of the obstruction. Ideally, for every patient presenting with airway obstruction the site, size, level, extension and nature of the lesion should be established, but when urgent intervention is needed in acute obstruction, this is not always possible.
There are three key questions that determine the management of the obstructed airway: (1) What and where is the lesion? (2) How urgent is the surgical intervention? (3) Is the obstruction so significant that we should abandon attempts for general anaesthesia and use an awake technique?
Site and cause of obstruction
Causes of airway obstruction vary from obstructing oropharyngeal lesions, supraglottic lesions such as epiglottitis, tongue base lesions, vocal cord paralysis, laryngeal lesions, intratracheal masses and extrinsic tracheal compression from mediastinal masses. Airway management after head and neck trauma has been discussed elsewhere, but that approach also emphasised the importance of site and urgency of the requirement for intervention [15]. Thus, also in non-traumatic causes, the level at which obstruction exists makes a significant difference to the suitability of different techniques. Optimising the ability to oxygenate is fundamental to any management strategy for the obstructed airway and recent work has challenged traditionally held views around neuromuscular blockade and routine facemask ventilation [5, 16–19].
For obstructing oral cavity and oropharyngeal lesions the problems are first the ability to ventilate by facemask following induction of anaesthesia and second, difficult or impossible direct laryngoscopy around obstructing masses. The challenge in this group is one of bypassing a large obstructing mass without traumatising it, while maintaining a patent airway. If the glottis and lower airway are normal, an awake technique such as awake fibreoptic intubation allows (spontaneous) ventilation to be preserved until the airway is secured. Alternatively, the airway can be secured by an awake transtracheal catheter or tracheostomy.
Obstructing lesions at the tongue base and supraglottis compromise airflow significantly and as they push the epiglottis downwards, further worsen the obstruction. There is a danger of total airway obstruction following intravenous induction in this group as a result of loss of supporting tone from the soft tissues. Oral or nasal airways may be ineffective in relieving the obstruction as they do not extend this low into the airway, but experienced forceful jaw thrust manoeuvres may displace the mandible and tongue base forwards, creating an airspace. Laryngoscopy can cause trauma, resulting in bleeding, swelling and complete airway obstruction. An awake fibreoptic technique, awake transtracheal catheter or awake local anaesthetic tracheostomy should certainly be considered.
At the glottis the most common causes of obstructing lesions are tumours. As these enlarge, patients initially are able to compensate by changes in breathing patterns or position, but acute deterioration results at a point when a critical narrowing is reached. The key decision is to identify if it will be possible to pass a tracheal tube through the narrowing and if this is felt to be unlikely, an awake transtracheal catheter or awake local anaesthetic tracheostomy should be considered.
Tracheal compression can occur as a result of lesions within the trachea or compression by thyroid and mediastinal masses. The upper airway may be normal at laryngoscopy and it may be possible to pass the tracheal tube beyond the glottis but not beyond the obstruction (or even place a surgical airway beyond the obstruction).
Fibreoptic intubation
The role of awake fibreoptic intubation for the management of patients with a difficult airway is universally recognised as useful, but its specific role in the management of an obstructed airway is dependent on the site of the obstruction. Successfully used for mass lesions within the oral cavity or tongue base, where passing the fibrescope around the lesion may be possible in skilled hands [6, 7], its role for advanced obstructing lesions within the glottis is more controversial, as the fibrescope has to pass through the small orifice of the obstructive mass with the risk of total airway occlusion [4].
Technically, administering topical anaesthesia for distorted, large, vascular, friable, necrotic tumours is challenging and application of local anaesthetic or passage of the fibrescope can cause coughing and total airway obstruction [20, 21]. Even if the fibrescope can help to show a way through the obstruction, complete airway obstruction can still occur and it may not be possible to pass the tracheal tube [22]. The more significant the airway obstruction, the more likely this is to occur. Initial nasendoscopy to inspect the larynx has been advocated by skilled users of fibreoptic techniques, to assess if it is safe to continue with awake fibreoptic intubation or defer to an awake local anaesthetic tracheostomy [7].
Awake ‘fibrecapnic’ intubation [23] has been described as a technique in which a very narrow catheter is passed through the suction channel of a bronchoscope and advanced into the airway for carbon dioxide measurement. When capnography has confirmed the catheter position, the fibrescope is railroaded over the catheter and then the tracheal tube passed in turn over this. This technique may still cause total airway obstruction with advanced glottic disease, and with severe tumours bleeding and acute airway obstruction can occur [23].
The Fourth National Audit Project provides important information on the use of fibreoptic techniques and describes several challenging cases where the initial decision to perform awake fibreoptic intubation was suitably changed to an awake surgical approach after problems obtaining a view, contamination of the airway, or a very narrow airway [24]. In patients with head and neck pathology and airway obstruction, 14 of 23 attempted flexible fibreoptic intubations failed. Of these, four were attempted in awake patients and failed because the glottic inlet could not be identified or it was not possible to pass the fibrescope or tracheal tube. In 10 patients, fibreoptic intubation was attempted after induction of anaesthesia and despite repeated attempts, there was an inability to identify the glottic inlet, pass the fibrescope, or pass the tracheal tube (with bleeding and airway obstruction common) [11]. In those patients where fibreoptic techniques were unsuccessful, a surgical airway was usually required.
Successful use of awake fibreoptic techniques for an obstructed airway requires skill, an understanding of the nature and level or site of obstruction, and recognition that with advanced glottic obstruction, fibreoptic techniques fail [4, 11, 24].
Inhalational induction
Inhalational induction and maintenance of spontaneous ventilation was the traditional method of managing a patient with a difficult airway or airway obstruction [4]. In a recent debate about managing a patient with stridor and a retrosternal thyroid mass compressing the trachea, only two of the eight experts chose an inhalational induction technique. Two experts chose an intravenous induction technique (including neuromuscular blockade), while three experts selected an awake fibreoptic intubation technique [1]. Clearly, even amongst experts, anaesthetic practice is varied.
The theoretical advantage of an inhalational induction is the maintenance of spontaneous ventilation and therefore oxygenation. Yet, airway collapse is common as the collapsible pharyngeal tissues are affected by the dynamic effects of negative intraluminal pressures during inspiration, which favours collapse [25]. The traditional view is that the technique is safe because as the patient looses consciousness, this airway collapse and obstruction prevents further uptake of inhaled volatile agent, allowing the patient to waken. Such arousal is purported to relieve the airway obstruction. Although modern practice suggests sevoflurane is now the drug of choice for inhalational induction, studying the influence of blood solubility during simulated airway occlusion in healthy volunteers suggests otherwise. The end-tidal concentration of halothane may fall more quickly than that of sevoflurane [26], and halothane compared with sevoflurane may thus lead to a quicker awakening [26]. The application of continuous positive airway pressure (CPAP) and positive end-expiratory pressure (PEEP) when assisting ventilation with bag and mask may help relieve any obstruction by acting as a pneumatic splint.
In clinical practice, when an inhalational induction is commenced in a patient with a severely obstructed airway the induction is slow, despite the application of CPAP and PEEP, there are apnoeic periods, and the patient often becomes more hypoxic and hypercarbic. It is sometimes an unstable scenario with arrythmias and episodes of total airway obstruction, following which the patient in fact often does not awaken, but instead apnoea continues and the hypoxia worsens. Something has to be done and in practice this often involves manual lung ventilation with bag and mask to relieve severe hypoxia. At this stage the technique is no longer a spontaneous ventilation technique at all, and the theoretical advantages of the method are lost. The administration of neuromuscular blockers may provide optimum ventilation and intubating conditions, and should be considered (notwithstanding current debate as to the role of neuromuscular blockade in the ability to mask ventilate) [27, 28].
Recent work [5] during induction and surgery has shown that positive-pressure ventilation following intravenous induction of anaesthesia and neuromuscular blockade may be physiologically superior to spontaneous respiration in adult stridulous patients with airway compromise due to laryngotracheal stenosis.
During inhalational induction where spontaneous ventilation is maintained, there is a reduction in airflow and respiratory drive, and increased collapsibility of the airway, leading to an increased work of breathing and critical instability at points of airway narrowing. With a reduction in functional residual capacity, the only mechanism to try to counteract these changes whilst holding a facemask is to provide PEEP and CPAP (logically in that order) manually with a bag. For initially non-obstructed airways in adults and children this is often enough and an inhalational induction technique is successful. However, for a critically obstructed adult airway, the application of PEEP and CPAP is often not enough to counteract all these changes. Active positive-pressure ventilation, by contrast, produces positive pressures during both phases of ventilation and potentially better maintains the airway.
The deterioration in the airway following inhalational induction and subsequent inability to maintain spontaneous ventilation is described in NAP4 [11]. In 12 patients the airway was compromised and spontaneous ventilation became more difficult with oxygen desaturation. In 11 of these patients spontaneous ventilation became impossible, either because the airway deteriorated further or after direct laryngoscopy attempts were made. In situations when the airway was lost there was reluctance to use (or even an active decision to avoid) neuromuscular blockade and controlled ventilation, despite respiratory distress, airway obstruction, hypoxia and a peri-arrest state. It was clear that in these cases patients do not ‘rapidly awaken’ [11].
Intravenous induction and administration of neuromuscular blockade
For routine airway management neuromuscular blockade can facilitate mask ventilation [28] (occasionally even if initial mask ventilation is impossible [29]), abolish laryngeal reflexes and so make tracheal intubation easier and less traumatic [30, 31]. Intravenous induction and administration of neuromuscular blockade is more often ‘the answer rather than the problem’ [19].
The reluctance in administering neuromuscular blockade to a critically obstructed adult airway is based on the fear of an inability to mask ventilate and the impossibility of a prompt return to spontaneous ventilation. Pandit’s recent algorithm describes the thought process commonly employed but this may not apply in full to the case of an obstructed airway [32]. This is because an obstructed airway (as described in this algorithm) can equally happen with inhalational induction techniques that attempt to maintain spontaneous ventilation from the outset [11].
No plan is always successful and both inhalational and intravenous induction techniques can fail. Perhaps what is more important is to judge, for any given patient, which technique is more likely to be successful in terms of the ability to maintain oxygenation, and second, which technique leads to fewer ‘can’t intubate, can’t ventilate’ (CICV) scenarios when used in a critically obstructed airway. There is some evidence in adult stridulous patients with laryngotracheal stenosis that ventilation after intravenous induction and neuromuscular blockade is physiologically superior to seeking to maintain spontaneous respiration [5] The answer to the second question remains unknown, since evidence from NAP4 has highlighted failures of both inhalational and intravenous induction techniques [11]. No study has yet compared alternate modes of managing CICV. Perhaps this is why experts have such different approaches when dealing with an adult obstructed critical airway and advocate techniques with which they are most comfortable [1, 33, 34]. With the introduction of sugammadex it may be theoretically possible to reverse deep neuromuscular block and re-establish spontaneous ventilation within minutes [35], making it possible that newer methods will appear based around use of this drug. However, time will tell whether this theoretical property of the drug is always translated into practice. Whatever technique is chosen, either inhalational induction or intravenous induction and administration of neuromuscular blockade, there will be failures and the importance of well thought through back-up plans become important.
Awake tracheal access
Awake direct tracheal access under local anaesthetic involves the placement of an airway into the trachea either using a narrow or wide bore cannula or through a surgical cricothyroidotomy or tracheostomy. Generally, an awake surgical tracheostomy is placed by a surgeon and an awake cannula device by an anaesthetist. One of the most important decisions a team looking after a patient with an obstructed airway has to make is to decide if it is safe to proceed with some form of general anaesthesia or whether awake tracheal access is more suitable. This decision will be guided by a knowledge that, with repeated attempts to secure the airway, the risk of total airway obstruction may increase considerably [13]. If a CICV situation does arise, an immediate surgical airway is required in a patient with a totally obstructed airway and no other means of oxygenation. Yet, these are not ideal conditions to perform a surgical tracheostomy.
Therefore, when these high-risk patients have been identified in advance, the anaesthetist and surgeon should jointly decide which type of awake tracheal access to make. An awake surgical tracheostomy is generally undertaken in patients with advanced tumours in or around the airway, where postoperatively, the calibre of the airway is worse and a definitive airway is required. This decision is not easy and depends on urgency, experience, location, the patient’s pathology and the knowledge and skills to perform the back-up plans.
On the other hand, in an emergency a junior anaesthetist, with no immediate senior help available and with limited experience of back-up plans or equipment, may well consider an awake surgical tracheostomy the safest option. The same patient presenting when experienced anaesthetists and surgeons are immediately available might be managed in other ways. Technically undertaking an awake surgical tracheostomy can be challenging, particularly in a patient who has airway obstruction and cannot lie flat, extend their head, or tolerate surgical manipulation in the neck. With fewer routine critical care tracheostomies being undertaken, current junior surgical staff inevitably have less experience than did their predecessors [36].
The use of narrow bore cannula as an elective technique inserted awake either through the cricothyroid membrane or upper trachea has been described [37–39] and in experienced hands allows ventilation throughout surgery in patients with significant airway obstruction and stridor [9]. If there are concerns about the calibre of the patient’s airway at the end of surgery, the narrow bore cannula can be left in place, or a definitive surgical tracheostomy performed. Narrow bore cannulae are not a substitute for a definitive surgical tracheostomy but in some patients where airway obstruction is improved as a result of the surgery, the technique may avoid an unnecessary surgical tracheostomy [9]. The NAP4 data for patients with head and neck pathology in CICV described 27 uses of cannula cricothyroidotomy of which 12 were successful and 15 failed. These failures were due to misplacement, inability to place, fracture, kinking, blockage, dislodgement and barotrauma. This failure rate is of concern and as an accompanying editorial observed [12], more research is needed on the relative merits of cannula cricothyrotomy vs surgical tracheostomy [11].
Central airway obstruction and mediastinal masses
Central airway obstruction can be particularly challenging because of the risk of fatal airway obstruction in a site distal to rescue by a surgical airway, and because of adverse effects of large anterior mediastinal masses on cardiovascular stability. Tracheobronchial masses, inflammatory stenoses or extrinsic compression from mediastinal masses can all result in airway obstruction. Chest radiograph and CT scans identify the extent, site and severity of mediastinal masses. Awake CT-guided needle biopsy of large mediastinal masses avoids the risks of a general anaesthetic.
Similar to the issues surrounding airway obstruction at supraglottic and glottic sites, controversy exists as to the best management of central airway obstruction. There are advocates of inhalational induction and maintenance of spontaneous ventilation [34] and other groups who believe ‘gaseous inductions are usually contraindicated’ and advocate instead intravenous induction incorporating neuromuscular blockade [33]. Following induction of anaesthesia, either spontaneous ventilation is maintained [34] or a rigid bronchoscope placed and jet ventilation commenced with neuromuscular blockade [33]. If ventilation becomes impossible, repositioning or more usually, rigid bronchoscopy to beyond the obstruction, is needed [33, 34].
Cardiopulmonary bypass commenced before induction under local anaesthesia [40–42] has been described in extreme circumstances (e.g. where complete airway obstruction will result from general anaesthesia). However, the concept of cardiopulmonary bypass as a ‘standby’ technique in case of problems is a myth [34]: even when a team is prepared and ‘pump-primed’ (i.e. the cardiac bypass machine ready to start) it takes at least 5–10 min to cannulate and establish adequate circulation and oxygenation [34]. If cardiopulmonary bypass is thought to be needed it needs to be established early.
Conclusions
In summary, all airway techniques need back-up plans. Adequate assessment to identify the site and level of obstruction, in particular with nasendoscopy, CT and MRI imaging, is useful in guiding the initial airway plan. Awake fibreoptic intubation, inhalational induction, direct tracheal access techniques, and even cardiopulmonary bypass (all sometimes argued to be panaceas) in fact have their limitations. No plan is always successful but several back-up plans, good communication within the team, experience, training, adaptability and skill will all help determine the outcome.
Competing interests
No external funding or competing interests declared.
