To the Editor:
We read with great interest and some concern the revised Practice Guidelines for Management of the Difficult Airway.1 The Difficult Airway Algorithm, presented in figure 1, has some confusing recommendations.
Before entering the flow chart in figure 1, the reader is asked to assess the difficulty of supraglottic airway (SGA) placement. Multiple factors are known to increase the probability of difficult mask ventilation and difficult laryngoscopy and/or intubation.2,3 However, similar evidence is absent for the assessment of difficult SGA placement. Excluding an inadequate mouth opening for insertion, difficulty with an SGA is usually unanticipated.
Next, in the “AWAKE INTUBATION” pathway, if awake intubation fails, the authors suggest inducing general anesthesia assuming that “mask ventilation will not be problematic.” It seems that if an awake intubation was attempted, difficult mask ventilation and/or laryngoscopy is anticipated, and induction of general anesthesia is not in the best interest of patient safety. This is supported by the 2005 closed claims analysis that found that two thirds of difficult airway claims were associated with an induction of anesthesia.4 Given recent discussion suggesting the decreased skill of providers with bronchoscopy in the presence of newer airway equipment (i.e., video laryngoscopes),5,6 perhaps part of this pathway should include consulting a colleague or considering an alternative awake airway approach including, as mentioned by the authors, invasive airway access.
Finally, if the “NONEMERGENCY PATHWAY” (“FACE MASK VENTILATION ADEQUATE”) shifts to the “EMERGENCY PATHWAY” (“BOTH FACE MASK AND SGA VENTILATION BECOME INADEQUATE”), the algorithm details to use “Emergency noninvasive airway ventilation.” Similarly, the use of “Emergency noninvasive airway ventilation” is also recommended when “SGA NOT ADEQUATE OR NOT FEASIBLE.” However, as defined in figure 1E, “Emergency noninvasive airway ventilation consists of an SGA”. If SGA ventilation was already proven to be inadequate, how does this recommendation improve the situation?
We recognize and appreciate the authors’ efforts to provide a clear algorithm for this complicated topic. However, we feel that the above issues may create confusion.