Author: Bob Kronemyer
Screening tests performed at the bedside to identify patients at high risk for an unanticipated difficult airway may be largely inaccurate. This startling finding comes from an abridged Cochrane Review of these diagnostic tests.
The review consisted of 133 published studies, of which 127 were cohort trials and six were case-control, and totaled 844,206 subjects (Anaesthesia 2019
Seven prespecified tests were evaluated: the Mallampati test (six studies), the modified Mallampati test (105 studies), Wilson risk score (six studies), thyromental distance (52 studies), sternomental distance (18 studies), the mouth-opening test (34 studies), and the upper lip bite test (30 studies). Overall, there were 76 different tests and 32 combinations of tests.
The reference standards for the studies were difficult face mask ventilation (seven studies), difficult laryngoscopy (92 studies), difficult intubation (50 studies), and failed intubation (two studies).
“The use of systematic reviews and meta-analyses as evidence for our decisions is well established. However, in the field of diagnostics, this is a rather new concept and still evolving,” said primary author Dominik Roth, MD, PhD, a consultant emergency physician and docent of emergency medicine at Medical University of Vienna. “We chose this particular topic of difficult airway prediction because it has a huge impact on the daily practice of anesthesiologists and other health care providers, plus there is a ton of literature.”
The review concluded that the sensitivity of the bedside tests ranged from 22% to 67%, whereas specificity fared much better, from 80% to 95%.
“The upper lip bite test showed the most favorable accuracy at 67% sensitivity and 92% specificity for difficult larnygoscopy, but still missed one out of three patients with a difficult airway,” Dr. Roth said.
There were no significant differences in test characteristics for the various domains of difficult airway management.
“The general direction of the results was consistent with what we expected, based on our own daily clinical practice,” Dr. Roth said. “It was surprising, though, to find really low sensitivities. After all, tossing a coin has a sensitivity of 50%. The best test in our analysis achieved a sensitivity of only 67%.”
Direct but Not Video Laryngoscopy
Intuitively, clinicians expect such screening tests to be overcautious, according to Dr. Roth. “Hence, we believe these tests will classify some airways as difficult when in fact they are not, but in exchange they will identify with a good certainty all those airways which are truly difficult. Unfortunately, the opposite is true. Most non-difficult airways will be classified correctly, but a huge portion of difficult ones will be missed.”
Because the review found the bedside tests to be inaccurate, “anesthesiologists should not rely on them,” said Dr. Roth, who spoke last December at the annual Difficult Airway Society meeting in Edinburgh, Scotland, where attendees agreed there should be more focus on preparing for the difficult airway and finding better ways to address the problem, rather than predict it.
The data showed that using several bedside tests in an attempt to improve accuracy will not help either. “Those combinations were not superior to single tests,” Dr. Roth said.
Dr. Roth said some clinicians believe that future technology, like digital facial recognition techniques, might enable better predictions about difficult airways. “While this might be true to some degree, I honestly do not think this will be the magic bullet,” he said. “Instead, we should concentrate on our two core responsibilities: vigilance and preparedness.”
Limitations, but Still Valuable
John Doyle, MD, PhD, a professor of anesthesiology at the Cleveland Clinic in Ohio, and a member of the editorial advisory board of Anesthesiology News, said the review “confirms what many of us in the airway business have believed for quite some time, namely that none of the common bedside screening tests is well suited for detecting unanticipated difficult airways.”
What makes this study so special is its thoroughness, according to Dr. Doyle, with receiver operating characteristic plots used to graphically display the results. “However, a notable limitation of the study is that while it details predictors of difficulty with direct laryngoscopy, many centers now routinely use video laryngoscopy instead of direct laryngoscopy,” he said.
Dr. Doyle was somewhat surprised that the upper lip bite test performed better than the other tests, “and I was even more surprised that video laryngoscopy was not mentioned as one possible means to make intubation easier,” he said.
Despite the limitations of these bedside tests, Dr. Doyle believes they are sufficiently valuable for routine use by anesthesiologists. “If the bedside airway evaluation suggests the going will be problematic, the evaluation allows us to prepare for a difficult intubation, such as having extra help on hand, using video laryngoscopy or performing awake intubation.”
Because a bedside airway evaluation that indicates a straightforward approach is frequently wrong, “it is always wise to be prepared for a Plan B, like having an airway introducer and a video laryngoscope immediately available in the room,” Dr. Doyle said. “In many places, the airway introducer is taped to the anesthesia machine or supply cart for immediate access.”
Dr. Doyle noted that some anesthesiologists have abandoned direct laryngoscopy for video laryngoscopy, due to the favorable intubating conditions that the video technique offers in otherwise difficult settings, such as in patients with an anterior larynx.
In addition, Dr. Doyle often uses several bedside tests to try to improve predictive accuracy, including the comprehensive El-Ganzouri airway assessment tool.