Airway Assessment Before Intervention: What We Know and What We Do

One of our colleagues failed to intubate the trachea of a patient whom he had easily intubated under direct laryngoscopy (DL) several weeks earlier for a craniotomy.1 On this occasion, the patient’s mouth opening was severely limited by contracture of the temporalis muscle after the previous temporal fossa craniotomy. The patient was confused and uncooperative, precluding a complete airway examination. This led to the failure to detect the limited mouth opening before the induction of general anesthesia. In other words, failure to assess the airway was the reason for the failed tracheal intubation. One year later, another colleague gave an anesthetic to the same patient and again, without consulting old anesthetic records or fully assessing the patient’s airway before induction, he also encountered a failed tracheal intubation under general anesthesia. That was in the 1980s. We have learned a lot since then. Are we doing better today? The American Society of Anesthesiologists closed claims project in the United States2,3 and the fourth National Audit Project in the United Kingdom4,5 have reported that difficulty with and failure to successfully manage the airway are associated with serious adverse respiratory events. The single most important factor likely to lead to an inability to intubate and oxygenate is the failure to predict difficulties before managing the airway.6 In the 1980s, Mallampati et al.,7 Samsoon and Young,8 Wilson et al.,9 and other investigators10 proposed several methods of assessing the airway in an effort to predict a difficult DL intubation, the most common method of tracheal intubation at the time. Despite some success and wide acceptance, all these methods have limitations. Because of the low incidence of a failed DL, we now know that almost all univariate predictors have low positive predictive value11 and that the use of multivariate predictors appears to improve the positive predictive value of these airway assessment tools for DL.12–14 A report from the Danish Anaesthesia Database,15 which included 188,064 patients, showed that the diagnostic accuracy of the anesthesiologists’ predictions of difficult DL intubation and difficult mask ventilation was poor. Of 3391 difficult intubations, 3154 (93%) were unanticipated. When difficult DL intubation was anticipated, only 25% (229 of 929) turned out to actually be difficult to intubate. One of the limitations of the study is that only 2 questions were asked in the preoperative airway assessment: (1) Is difficult DL intubation anticipated? and (2) Is difficult bag-mask-ventilation (BMV) anticipated? Unfortunately, no airway assessment tools were disclosed in the preinduction assessment. It is difficult to draw any conclusion regarding the value of preanesthesia airway assessment if many of the accepted airway evaluation tools for DL were not used. Around the turn of the 21st century and with advances in technology, the focus of contemporary airway management practice has changed from tracheal intubation to the broader issue of oxygenation and ventilation.16–18 This has involved emphasizing the utility of all of BMV, use of extraglottic devices (EGDs), tracheal intubation using DL and indirect laryngoscopes as well as other alternative techniques (visual and nonvisual), and the invasive surgical airway in managing the airway. Accordingly, in addition to assessing the predictors of a difficult DL, emphasis has been placed on expanding the preinduction airway examination to include evaluations of predictors of difficult BMV, difficult use of an EGD, difficult use of DL and indirect laryngoscopic intubation, and a difficult surgical airway.18,19 It is encouraging to see that during the past decade, a considerable number of studies have emerged to investigate the predictors of difficult use of BMV, EGD, DL, and cricothyrotomy.19 During the same period, a large number of rigid fiber-optic and videolaryngoscopic devices have been introduced into clinical practice. Despite their proven efficacy and increased use in patients with unanticipated and anticipated difficult DL intubations,20 there has been little critical evaluation of the difficulties and challenges associated with using these devices, and little practical advice has been provided in the specifics of their utilization. Few investigations have reported factors associated with difficult use of these devices so far.15,21–23 It is welcome news that, in this issue of the Journal, Nowakowski et al.24 report findings regarding predictors of difficulty when using the Bonfils Rigid Fiberscope in a study involving 400 patients. Limited mouth opening, high body mass index, and high Cormack and Lehane grade are shown to be associated with longer intubation times with the Bonfils Rigid Fiberscope. Interestingly, these also are some of the predictors of difficult DL. It is encouraging to see that more investigators are beginning to show an interest in analyzing the specific strengths and weaknesses of devices that are promoted for airway management, and these findings will be reflected in a requirement for careful airway assessment. Clearly, it is not a simple task to assess the difficulties associated with using these newer devices while their designs are constantly changing with new modifications. In addition, an airway evaluation is not complete without an assessment of old anesthetic records and seeking any information present in databases. For the third anesthetic given to the patient presented earlier, a failed DL would have been anticipated had the anesthesia practitioner checked the patient’s previous anesthetic records. In hopes of minimizing the risk of complications, anesthesia organizations around the globe advocate for the identification of potential problems with oxygenation and ventilation before induction of general anesthesia, so that strategies and plans can be made beforehand.4,5,19,25 Recognizing that there is no reliable way of predicting difficult airway management,14,15 airway practitioners are being advised to identify and prepare alternative airway techniques such as an EGD and to call for assistance should oxygenation and ventilation become compromised. More importantly, if airway practitioners anticipate difficulties in using most of the airway techniques (e.g., BMV, EGD, and DL), it would be prudent to manage the airway awake with the patient breathing spontaneously. Furthermore, it is also important to discuss such an airway management plan before the induction of general anesthesia during the team briefing and the preinduction phase of the World Health Organization Surgical Safety Checklist.26,27 To understand the methods of preanesthetic airway assessment currently in use worldwide, we solicited preanesthetic airway assessment forms from colleagues in 10 countries. Several forms were electronic versions (mostly from North America) and the remainder was hard copies of standard preanesthetic assessment forms. Not surprisingly, the airway assessment section on these forms varied substantially, ranging from free text (descriptive) to a more detailed list of predictors of difficult DL, such as the Mallampati score, thyromental distance, jaw protrusion, mouth opening, and cervical spine movement. Interestingly, with the exception of the free text forms, all asked for a Mallampati score. Less frequently mentioned were the thyromental distance (60%), cervical spine movement (50%), dentition (50%), and mouth opening (40%). None of the forms listed other predictors of difficult DL, such as the sternomental distance, Wilson risk scores, neck circumference, or the upper lip bite test. Only one center asked for predicted difficulty in BMV and surgical airway (cricothyrotomy) and not one mentioned an assessment for predicted difficulty in using the EGD. Recognizing that some of the predictors of difficult DL are also predictors of difficulty in the use of other airway techniques such as BMV,19,28 it appears that assessment for a difficult DL remains the main focus of airway assessment at most centers around the globe. More importantly, the broader approach in airway assessment, including forewarnings of difficulties with BMV, EGD use, videolaryngoscopy, and cricothyrotomy are largely ignored. It is disturbing to realize that, although we have made great strides in technology to improve airway management, efforts to assess risks, in this small sample, vary greatly. In addition, it is alarming to contemplate that little attention is being paid to assessing the overall difficulties that may arise in providing oxygenation and ventilation. There appears to be a disconnect between what we know and what we actually do in clinical practice. What is the missing link? Perhaps human factors may be blamed for most of the discrepancies between recommendations and actual clinical practice.29 In fourth National Audit Project,4,5 human factors were considered to have contributed to almost all serious adverse outcomes. In that report, widespread deficiencies in judgment, communication, planning, equipment, and training were identified. How can these findings be applied to our clinical practice to improve outcomes? Perhaps we can borrow the approach from the Process Safety Management Guidelines and Compliance of the US Department of Labor.30 Using this approach, “…the process design, process technology, process changes, operational and maintenance activities and procedures, non-routine activities and procedures, emergency preparedness plans and procedures, training programs, and other elements that affect the process are all considered in the evaluation….” This may be the right approach for us, but it would be costly and painful. Perhaps education through reading, seminars, workshops, and simulations would be a good start. Leadership from all organizations must embrace changes that would improve outcomes, implement programs with proven successes, engage collaborative efforts with other departments, and promote continuing medical education and simulation-based training addressing the issues around human factors, including planning and decision making. In summary, the case we presented earlier1 and other outcome studies2–5 serve to remind us that preanesthetic assessment of the airway is essential to avoid catastrophe. Unfortunately, during the past 30 years, although tremendous advances in airway management devices have taken place, failure to identify difficulties before airway intervention continues with disturbing frequency. Recommendations have been made, and yet our practice does not appear to reflect these needed changes. In the Danish Anaesthesia Database study,15 the investigators concluded that “…prediction of difficulties remains a challenging task. There may be ample room for improvement, based on a rigorous, evidence based and systematic approach…” This is a call to action. We must find solutions to improve and correct problems related to incomplete airway assessment so that appropriate airway management strategies can be planned before induction of anesthesia. We must do it ourselves and do it now. DISCLOSURES Name: Orlando Hung, MD, FRCPC. Contribution: This author helped prepare the manuscript. Attestation: Orlando Hung approved the final manuscript. Name: J. Adam Law, MD, FRCPC. Contribution: This author helped prepare the manuscript. Attestation: J. Adam Law approved the final manuscript. Name: Ian Morris, MD, FRCPC, DABA, FACEP. Contribution: This author helped prepare the manuscript. Attestation: Ian Morris approved the final manuscript. Name: Michael Murphy, MD, FRCPC. Contribution: This author helped prepare the manuscript. Attestation: Michael Murphy approved the final manuscript. This manuscript was handled: Steven L. Shafer, MD.