What is meant by ‘difficult intubation’? If we consider the definition of the American Society of Anesthesiologists (ASA), a difficult intubation is ‘an intubation during which the insertion of the endotracheal tube takes more than 10 min, and/or requires more than three attempts by an experienced anesthesiologist’. This definition is not entirely satisfactory, because a difficult intubation can often be diagnosed within seconds, if laryngoscopy is carried out under optimal conditions, defined in Table 2 . This allows a more logical definition, which is ‘an intubation is difficult when insertion of an endotracheal tube is impossible despite optimal laryngoscopy’.
Recently, a scale allowing a posteriori quantification of the true difficulty of an intubation has been proposed . The IDS (intubation difficulty scale) consists of seven criteria, which are summated (Table 3); its use permits more objective comparison of the real difficulties encountered during intubation. The IDS takes into consideration the use of different techniques, as well as the skill and experience of the operator. This scale includes both qualitative and quantitative dimensions of difficult intubation. A recent study has confirmed the adequacy of this scale  but a prospective, large-scale validation is desirable.
Review of airway anatomy
The upper airways include all structures allowing the passage of air from the nostrils or the mouth to the lungs (Figure 2). The nasal and oral cavities open into the nasopharynx and oropharynx, respectively. They are separated anteriorly by the bony palate, then by the soft palate, while they are communicate posteriorly. The tongue, which occupies a large part of the oral cavity, is attached principally to the hyoid bone, and the epiglottis is attached at its base. This structure provides the functional separation between the glottis and larynx (which open into the trachea) and the hypopharynx, which opens into the oesophagus. The plane of the vocal cords is approximately at the same level as the thyroid cartilage (‘Adam’s apple').
Airway patency is guaranteed by bones, cartilages, ligaments, and by the tone of the surrounding muscles. The nose, mouth, larynx, and trachea, which are primarily made up of bone and semirigid cartilage, are naturally patent. However, the airway patency of the oropharynx depends on the tone in the muscles of the tongue and pharynx. When the central nervous system is depressed, as after the induction of general anaesthesia, the tongue easily falls backwards into the hypopharynx, and (especially in the supine position) comes into contact with the posterior pharyngeal wall. This degree of airway obstruction can be overcome using manoeuvres such as chin-lift, head-tilt and jaw-thrust. Sometimes, patency can be regained by insertion of oropharyngeal or nasopharyngeal cannulae. The glottic orifice and the base of the tongue are very close together. Under certain circumstances, following loss of muscle tone, the base of the tongue can prolapse into the glottic opening. This type of obstruction is not easily reversed by the above mentioned manoeuvres. In such a case, endotracheal intubation is necessary to restore the airway.
The difficult airway
Brain damage occurs when upper airway patency is lost, at times for surprisingly short intervals. Gas exchange can be maintained using mask ventilation after re-establishing the patency of the upper airway – or by use of a tube that entirely bypasses the upper airway, passing through the glottis directly into the trachea. The difficulty of mask ventilation can vary on a scale ranging from 0 (where no manipulation is necessary to maintain airway patency) to infinity, where despite external manoeuvres and use of appropriate cannulae, ventilation of the lungs remains impossible. Difficulty with intubation can also vary on a scale from 0 to infinity . Cerebral ischaemia only occurs if both mask ventilation and endotracheal intubation are impossible (Figure 3).
Diagnosis of potential airway problems before an anaesthetic is given is essential . When such problems are not anticipated, several factors intervene to make the situation worse: operator stress, unavailability of necessary equipment, improvised responses to a difficult situation, late calls for help and multiple attempts, which usually make subsequent ventilation and/or intubation more difficult. It is therefore necessary to improve the reliability of the diagnosis of potential airway problems in order to guide the initial anaesthetic approach to the patient, and to choose appropriate alternative strategems .
It is therefore important to consider two aspects of the problem during the preoperative visit:
- • will it be possible to maintain upper airway patency using external manoeuvres or cannulae, if necessary?
- • if the patency of the airway is compromised, will it be possible to visualize the laryngeal orifice and to insert an endotracheal tube?
Translaryngeal insertion of an endotracheal tube requires: recognition of pharyngeal and laryngeal structures, anterior displacement of the base of the tongue (using the laryngoscope blade) away from the glottic opening and optimal alignment of three axes (pharyngeal, laryngeal and oral), in order to allow the vocal cords to be seen (Figure 4).
While numerous causes of difficult intubation can be identified (congenital malformations, anatomical variations, and acquired pathology such as trauma, radiotherapy, tumours, etc.) , all of them can be assigned to one of four categories (Table 4).
However, in clinical practice, only two types of difficult intubation need be considered. In the first, the laryngeal structures are easily visualized, but present such a degree of distortion of the anatomy as to be unrecognizable or impossible to cross. The second type is encountered more often. Here, although the laryngeal structures are normal they cannot be seen. This is because either the base of the tongue cannot be mobilized adequately, or else the axes cannot be aligned properly. When both conditions exist, intubation is even more difficult. Usually, an appropriately thorough history suggests to the clinician that there may be problems with the airway or larynx, and subsequent physical examination allows some causes to be identified. Further tests are often of not much help in the diagnosis of difficult intubation. These are considered briefly below.
Diagnosis of the difficult airway
The preanaesthetic interview should include questions concerning previous difficult intubation. The problems encountered, and the solutions found should be detailed. A history of maxillofacial trauma or surgery of the face, neck, larynx, or pharynx should also be noted. A history of radiotherapy of the neck is also sought. Local problems of the mouth, the teeth, the thyroid, or the neck are also considered. Signs of dyspnoea, stridor, changes in the voice, dysphagia, and/or snoring should alert the clinician to the possibility of pharyngolaryngeal disproportion or an endobuccal tumour.
General patient morphology.
The anaesthetist's experience often leads to the recognition of certain conditions that may be associated with difficult intubation. Careful observation of the patient, both from the front and side aspects, should be carried out systematically. Examination of the face, neck, and mouth should allow the anaesthetist to anticipate difficulty with mask ventilation and/or intubation. It should be remembered that the plane of the vocal cords is located at the mid-level of the thyroid cartilage. When observing the patient from the side, the anaesthetist knows the positions of the mouth and larynx and can therefore estimate the approximate anatomical relations between the glottis, the mandible, and the base of the tongue. He or she can visualize the trajectory that the endotracheal tube will follow, and the structures that may prove to be obstacles.
Certain conditions are classically associated with difficult intubation. The patient with a short neck is one instance where visualization of the larynx is often difficult. Similarly, it is often difficult to see the larynx in obese patients. This is frequently aggravated by an increased amount of perilaryngeal fat in these patients. The pregnant patient is at increased risk of difficult intubation, especially because the pregnancy-related salt and water retention may lead to oedema of the periglottic structures.
This general evaluation is often rather intuitive, and must be followed by a more objective search for signs of potential difficulty.
Crowding of the pharynx by the tongue.
The tongue is the principle structure occupying the oral cavity. The posterior part of the tongue (its base) is intimately related to the glottic opening. The tongue has to be displaced by the laryngoscope blade during intubation in order to see the larynx.
If the tongue is disproportionately large with respect to the oral cavity, laryngoscopy and intubation are likely to be difficult. Because it is difficult to quantify the size of the tongue in relation to the oropharynx, Mallampati has suggested an evaluation technique to assess this factor . He argued that if the tongue was so large that the oropharyngeal structures could not be seen when the mouth was opened, it was likely that the glottis would not be seen either and therefore intubation would be difficult. This classification was modified by Samsoon and Young  to define four categories (Figure 1a). There is a good correlation between the Mallampati score and the different grades of visualization of the larynx described by Cormack . This seems reasonable, because the larger the tongue, the more space it occupies within the pharynx, proportionately reducing the view of the palate. Although this classification can point to the possibility of difficult intubation, it has certain limits. The results of the evaluation vary according to patient co-operation, level of consciousness, position, and strength, and may therefore present considerable interobserver variability. Moreover, other factors can make intubation difficult despite a favourable Mallampati class.
The mandibular space, bounded by the arch of the mandible, contains the tongue and the larynx. Depending on the size of this compartment and on the relative position of the various structures within, the adequacy of visualization of the glottis using a laryngoscope will vary. If the mandibular space is small, the structures will be constrained and difficult to mobilize. In this case, the base of the tongue will not be easy to displace with the laryngoscope blade, and a view of the glottis will be difficult to obtain.
Furthermore, the tongue is inevitably pushed back towards the larynx when it has insufficient room within the mandibular space. In this case, the angle between the pharyngeal and laryngeal axes is more acute. The alignment of these axes, by positioning the head, will be more difficult to obtain.
The mandibular space can be estimated using three measurements:
- •The thyro-mental distance. This is the distance that separates the point of the chin from the larynx. It represents the space into which the tongue will be displaced by the laryngoscope. This is the easiest measurement to make because the Adam's apple is easy to locate. In general, difficult intubation is may occur when this distance is less than 6 cm [
- •The hyo-mental distance is the distance between the point of the chin and the hyoid bone, to which the tongue is attached. This measurement also gives an idea of the available space for the tongue during laryngoscopy. However, it is more difficult to estimate, because in some patients (especially those with thick necks), the hyoid bone is difficult to palpate.
- •The mandibular length. This is a third measurement that allows estimation of the mandibular space. It does not take into account the position of the tongue with respect to the larynx.
To see the glottis, the anaesthetist must optimally align the oral, pharyngeal, and laryngeal axes (Figure 4). Three articulations are involved in this process:
- •The temporo-mandibular joint is responsible for mouth opening. This must be sufficient to permit introduction of the laryngoscope blade. It can be measured objectively.
- •The articulations of the cervical spine must allow moderate flexion in order to align the pharyngeal and laryngeal axes.
- •The atlanto-occipital joint, between the cervical spine and the base of the skull, must allow extension. This movement may be limited by ankylosis (e.g. rheumatoid arthritis), osteoarthritis or by hypertrophy of the cervical spinous processes.
The anaesthetist must quantify mouth opening, and the capacity to flex and extend the cervical spine and the head. Measurement of the sterno-mental distance with the head in hyperextension allows objective evaluation of head extension . This measurement also takes into account the length of the cervical spine.
- • Measurement of the sterno-mental distance is an objective evaluation of the anaesthetist's instinctive impression of a ‘short neck’.
- • Some patients with longstanding, poorly controlled insulin-dependent diabetes mellitus present with a generalized ankylosis. The ‘prayer sign’ (which is assessed by asking the patient to place his hands together to determine whether the interphalangeal joints are touching) raises suspicion of this problem. The sign is considered positive when the distal interphalangeal joints remain separated. The atlanto-occipital, or the temporo-mandibular joints, or both, may also be affected.
- • If the atlanto-occipital joints are fixed, or have reduced mobility, attempts to position the head correctly transform the desired flexion of the cervical spine into extension, and displaces the laryngeal axis forwards, instead of into a correct alignment (
Status of the teeth.
Prominent upper incisors (‘buck teeth’) can hinder satisfactory alignment of the axes and make introduction of the laryngoscope blade difficult. A careful preoperative dental examination is essential, not only to diagnose possible difficulty with intubation, but also to reveal diseased, fragile, or loose teeth that could be damaged. Prominent teeth alone are not sufficient to raise the possibility of a difficult intubation, but when associated with other factors increases the likelihood. On the contrary in the edentulous, a Mallampati grade of III or IV which normally would indicate difficult intubation, could be of less concern. In this case, the absence of teeth could partially compensate for the large volume of the tongue, so allowing alignment of the necessary axes.
Additional testing – medical imaging
Further investigations are not usually required for diagnosis of difficult intubation. It may be necessary, in certain situations, to define specific anatomical anomalies, but such examinations need not be carried out routinely. The information gained by such investigation rarely justifies the additional cost. However, obviously when such investigations (e.g. head and neck computed tomographic scans) are available for any other reasons, the anaesthetist should make use of them.
Detection of difficult intubation: which tests should be used?
Ideally, a screening test for difficult intubation should allow diagnosis of all problem cases (sensitivity = 100%), without false positives in patients who are easy to intubate (specificity = 100%). Such a test does not exist, and it seems unlikely that one will be developed in the future. Mallampati originally found a sensitivity close to 100%, with a specificity of 80% . Unfortunately, subsequent studies did not show such promising results. In general, the same test gives variable values of sensitivity and specificity in different studies [19–21]. Some have tried to improve the quality of the screening for difficult intubation by associating several variables [1,17,21–24]. Because difficult intubation is multifactorial, adding variables to such an index can increase the sensitivity, but at the risk of reducing the specificity. However, it is vital to alert the clinician to the possibility of difficult intubation, in order that all precautions are taken; false positives are clearly, in such instances, less dangerous than false negatives.
We have established an intubation score based on the various criteria discussed above. This score, which we call the ADS (airway difficulty score), by analogy to the IDS, represents the sum of the points for five criteria of difficult intubation (Table 5). For each evaluation we obtain a score which varies from 5 to 15. A potentially difficult intubation is defined as a score ≥ 8. This score also allows the clinician to distinguish difficulty in maintaining the patency of the upper airway (preferentially influenced by the thyro-mental distance and by the Mallampati score), and the difficulty with alignment of the axes and in visualizing the larynx (influenced by mouth opening, cervical spine mobility, and the presence of upper incisors). This is why this index is termed the ‘Airway Difficulty Score’ as opposed to the ‘Difficult Intubation Score’. Studies are in progress to validate this new concept .
Approach to difficult intubation
Various algorithms have been developed by professional organizations, such as the American Society of Anesthesiologists  and the French Society of Reanimation (SFAR) . Essentially, three situations must be considered: (a) anticipated difficult intubation, (b) difficult intubation in the anaesthetized patient but ventilation of the lungs is possible, (c) both intubation and ventilation are impossible.
The anaesthetist must define the management strategy for securing the airway. This must take into account the history and physical examination, the material at hand, and the preferences of the clinician responsible for the patient. The strategy must consider the difficulties foreseen in intubation or lung ventilation, or both, and the potential for the patient to co-operate. The advantages and disadvantages of the following options must be weighed: (a) surgical or nonsurgical control of the airway, (b) spontaneous or controlled ventilation of the lungs, (c) an awake intubation, or intubation after the induction of anaesthesia.
Based on the above, the anaesthetist must plan an initial approach, with several alternatives available in case of failure.
Anticipated difficult intubation
In this situation, spontaneous ventilation must be maintained until the airway is secured. A surgical airway using local anaesthesia can be obtained, but in most cases an awake intubation is chosen. Blind nasotracheal intubation has been the technique of choice for many years. The frequent occurrence of nontrivial complications of this technique (laryngospasm, nasopharyngeal haemorrhage, etc.) and the increased availability of fibreoptic bronchoscopes in the operating room have tended to relegate the method to the alternative list. Awake fibreoptic intubation has become the ‘gold standard’ in cases of anticipated difficult intubation. Meticulous preparation of both patient and equipment are necessary in order to maximize the success of the technique. Patient preparation includes an adequate explanation of the procedures involved, often the administration of an antisialogogue (glycopyrrolate or scopolamine are often more effective than atropine), and use of a topical nasal vasoconstrictor. Anaesthesia can be provided with nebulized 4% lidocaine, or with topical nasal and oropharyngeal application combined with superior laryngeal and transtracheal nerve blocks. All equipment and materials must be verified and that the operator must be experienced in the technique. The services of a closely available surgeon capable of quickly obtaining a surgical airway is useful. Confirmation of correct placement of the endotracheal tube (by fibreoptic bronchoscopy or end-tidal CO2 detection) must be obtained before abolishing spontaneous ventilation. In case of failure, several options are available: (a) establishment of a surgical airway, (b) postponing the intervention, with a new attempt at awake intubation under better conditions, (c) general anaesthesia is induced and maintained by facemask, (d) tracheal intubation is attempted after the induction of general anaesthesia, or (e) use of local or regional anaesthesia.
Difficult intubation in the patient whose lungs can be ventilated
When a difficult intubation is encountered in a patient in whom spontaneous ventilation has ceased, whether anticipated or not, it is essential that mask ventilation is adequate. If doubt exists, the possibility of rapid resumption of spontaneous ventilation, and/or awakening the patient should be considered. If ventilation by facemask is possible without a problem, the situation is less urgent and the anaesthetist can choose techniques to optimize subsequent attempts at laryngoscopy (see above). Such options include: alteration of the position of the patient, use of optimal external laryngeal mobilization, ensure satisfactory muscular relaxation, change laryngoscopes, or use an intubating stylet. In all cases, three ‘golden rules’ must always be respected. First, oxygenation of the patient must be scrupulously ensured between each attempt. Secondly, the anaesthetist must change at least one element at each attempt. Thirdly, a limit of three or four attempts at classical intubation is imposed. This is necessary because repeated attempts at laryngoscopy and intubation can, and probably will, cause laryngeal oedema and bleeding, rendering lung ventilation more and more difficult. In certain cases, use of the laryngeal mask airway can restore ventilation that has become marginal.
A large number of techniques exist for the difficult intubation scenario. A detailed description of these is beyond the scope of this paper. Several variations of the classical laryngoscope are commercially available [PVC – a French laryngoscope blade (Piquet Crinquette Vilette, Lille, France)], McCoy, Bullard, etc.), as are lighted tracheal stylets (TrachLight®). Retrograde intubation is another useful alternative; as is fibreoptic intubation during controlled ventilation. The choice of technique should depend more on the experience of the attending anaesthetist than on arbitrary rules.
If these techniques fail, the airway should be secured surgically or the patient should be awakened, and an awake intubation carried out subsequently.
Intubation and ventilation impossible
When the first attempt at intubation fails and ventilation is found to be impossible, the anaesthetist is confronted with a life-threatening emergency. Help should be sought immediately; ideally, a more experienced colleague should be summoned. In any event, the presence of another person, even if less experienced, is indispensable.
If the anaesthetist feels that the initial attempt at laryngoscopy was not optimal, a second (and final) attempt may be made. If this fails, and in the absence of a surgeon capable of surgically securing the airway, three solutions are available: (a) insertion of a laryngeal mask airway, (b) insertion of a Combitube®, or (c) institution of transtracheal ventilation.
The first two alternatives involve supraglottic ventilation, which in case of laryngeal oedema and/or laryngospasm may be inadequate. Transtracheal ventilation, on the other hand, allows satisfactory lung ventilation if performed correctly even with upper airway obstruction. However, this comes at the ever-present risk of pulmonary barotrauma.
The Fastrach® laryngeal mask airway will probably replace the standard laryngeal mask airway at this point in the difficult airway algorithm. Insertion of an endotracheal tube is possible in most patients using this device and several studies are currently underway in various situations. The three techniques mentioned above are salvage pathways, which allow the clinician to avoid the death of the patient while attempting to obtain a definitive airway. The Fastrach® laryngeal mask airway not only allows control of the emergency situation (‘can’t intubate, can't ventilate' scenario), but also definitively secures the airway with an endotracheal tube.
Appropriate management of the difficult intubation situation is one of the main challenges facing anaesthetists. Use of the Intubation Difficulty Scale allows retrospective quantification of this problem, and should facilitate objective comparisons of various predictors of difficult intubation proposed in the literature. Ideally, the difficult intubation situation requires the anaesthetist to anticipate the problem, to plan the initial management, and to define the alternative stratagems to be used if unforeseen difficulties are encountered.
Various criteria and scores of difficult intubation can be found in the literature. None are perfect: it is unlikely that we will soon see one that is both easy to perform and 100% successful. In this case, an overestimation of the predicted difficulty is preferable to underestimation. Major clinical problems are more frequent, and more severe, when difficulty is unanticipated.
The anaesthetist must always approach control of the airway with great respect. He must never choose a strategy which, in case of failure, leaves few or no alternatives, and which therefore leaves the patient at risk of irreversible hypoxic lesions. Various professional societies have therefore proposed a series of airway management algorithms. These include numerous options that can be selected depending upon circumstances. It should be remembered that while an ever-increasing number of techniques and devices are introduced, none are perfect in every situation. At any point in a management algorithm, the anaesthetist must always choose the technique with which he is most comfortable, and has had the most experience.
1 Bellhouse CP, Doré C. Criteria for estimating likelihood of difficulty of endotracheal intubation
with the Macintosh laryngoscope. Anaesth Intens Care
1988; 16: 329–337.
2 Caplan RA, Posner KL, Ward RJ, Cheney FW. Adverse respiratory events in anesthesia: a closed claims analysis. Anesthesiology
1990; 72: 828–833.
3 Benumof JL. Definition and incidence of the difficult airway. In: Benumof JL, ed. Airway Management: Principles and Practice.
Philadelphia, USA: Mosby, 1995: 121–125.
4 Rose DK, Cohen MM. The airway: problems and predictions in 18,500 patients. Can J Anaesth
1994: 41: 372–383.
5 Schwartz DE, Matthay MA, Cohen NH. Death and other complications of emergency airway management in critically ill adults. A prospective investigation of 297 tracheal intubations. Anesthesiology
1995; 82: 367–376.
6 Cormack RS, Lehane J. Difficult tracheal intubation
in obstetrics. Anaesthesia
1984; 39: 1105–1111.
7 Benumof JL. The ASA difficult airway algorithm: new thoughts/considerations. Annual Refresher Course Lectures.
Park Ridge, IL, USA: Am Soc Anesthesiologists 1997; 241: 1–7.
8 Benumof JL, Cooper SD. Quantitative improvement in laryngoscopic view by optimal external laryngeal manipulation. J Clin Anesth
1996; 8: 136–140.
9 Takahata O, Kubota M, Mamiya K et al.
The efficacy of the ‘BURP’ maneuver during a difficult laryngoscopy. Anesth Analg
1997; 84: 419–421.
10 Adnet F, Borron SW, Racine SX et al.
Difficulty Scale (IDS): Proposal and evaluation of a new score characterizing the complexity of endotracheal intubation
1997; 87: 1290–1297.
11 Janssens M, Lamy M. Quantification of intubation
difficulty: comparison of subjective assessment of the anesthesiologist and the Intubation
Difficulty Scale (IDS). Eur J Anaesthesiol
2000; 17 (Suppl.): 35, A112.
12 Benumof JL. Recognition of the difficult airway. In: Benumof JL, ed. Airway Management: Principles and Practice.
Philadelphia, USA: Mosby, 1995: 126–142.
13 Practice guidelines for management of the difficult airway. A Report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology
1993; 78: 597–602.
14 Schwartz DE, Wiener-Kronish JP. Management of the difficult airway. Clin Chest Med
1991; 12: 483–495.
15 Mallampati SR, Gatt SP, Gugino LD et al.
A clinical sign to predict difficult tracheal intubation
: a prospective study. Can Anaesth Soc J
1985; 32: 429–434.
16 Samsoon GLT, Young JRB. Difficult tracheal intubation
: a retrospective study. Anaesthesia
1987; 42: 487–490.
17 Frerk CM, Till CBW, Bradley AJ. Difficult intubation
: thyromental distance and the atlanto-occipital gap. Anaesthesia
1996; 51: 738–740.
18 Ramadhani SAL, Mohamed LA, Rocke DA, Gouws E. Sternomental distance as the sole predictor of difficult laryngoscopy in obstetric anaesthesia. Br J Anaesth
1996; 77: 312–316.
19 Oates JDL, Macleod AD, Oates PD, Pearsall FJ, Howie JC, Murray GD. Comparison of two methods for predicting difficult intubation
. Br J Anaesth
1991; 66: 305–309.
20 Randell T. Prediction of difficult intubation
. Acta Anaesthesiol Scand
1996; 40: 1016–1023.
21 Savva D. Prediction of difficult tracheal intubation
. Br J Anaesth
1994; 73: 149–153.
22 Nath G, Sekar M. Predicting difficult intubation
– a comprehensive scoring system. Anaesth Intens Care
1997; 25: 482–486.
23 Tse JC, Rimm EB, Hussain A. Predicting difficult endotracheal intubation
in surgical patients scheduled for general anesthesia: a prospective blind study. Anesth Analg
1995; 81: 254–258.
24 Wilson ME, Spiegelhalter D, Robertson JA, Lesser P. Predicting difficult intubation
. Br J Anaesth
1988; 61: 211–216.
25 Janssens M, Lamy M. Airway difficult score (ADS): a new score to predict difficulty in airway management. Eur J Anaesthesiol
2000; 17 (Suppl.): 35, A113.
26 Boisson-Bertrand D, Bourgain JL, Camboulives J et al. Intubation
difficile. Société Française d'Anesthésie-Réanimation. Expertise collective. Ann Fr Anesth Reanim
1996; 15: 207–214.
Keywords:© 2001 European Academy of Anaesthesiology
AIRWAY OBSTRUCTION; INTUBATION; INTRATRACHEAL.