Diagnostic accuracy of anaesthesiologists’ prediction of difficult airway management in daily clinical practice: a cohort study of 188 064 patients registered in the Danish Anaesthesia Database. A.K. Norskov et al. Anaesthesia (2015) Mar;70(3):272-81. doi: 10.1111/anae.12955.
Accurate prediction of “difficult airways” has for many years continued to be an elusive goal for Anaesthetists. With probably more than a million different ways to oxygenate a patient, the challenge is a daunting one, with low levels of correct prediction achieved and no consistency or standardisation in the approach taken.
No single predictor of difficult intubation is sufficiently reliable.
It remains of vital importance to ensure that appropriate skills, equipment and facilities are available to ensure a good clinical outcome for the patient.
A different approach may be needed if this situation is to be improved and reduce the reliance on the Anaesthetist to always “expect the unexpected”.
NAP 4 highlighted poor airway assessment contributed to poor airway outcomes but doesn’t specify pre-operative airway assessment tool.
ASA recommends a pre-op assessment of a patient’s airway based on 11 anatomical variables, but doesn’t elaborate on which factors are mandatory or how they should be weighted.
Pre-operative airway assessment carried out subjectively by the Anaesthesiologist reflecting pragmatic daily clinical practice and differed from previous studies based on the predictive value of single or several combined risk factors.
Mandatory yes/no answers to questions on predicting difficult airway management completed pre-operatively as required by DAD.
Immediately post airways management, an intubation score was registered as 1 unproblematic or ≥2 difficult.
An analogue score for mask ventilation was registered.
Approved by the National Board of Health and Data Protection Agency.
Inclusion criteria: All patients in whom mask ventilation was attempted & all patients undergoing attempted tracheal intubation.
In the primary analysis, 93% of difficult intubations were not predicted pre-operatively. Only 25% of anticipated difficult intubations proved correct; and 94% of difficult mask ventilations were not predicted. When anticipated, difficult mask ventilation occurred in 22% of cases.
The low positive predictive values make this unsatisfactory as a diagnostic test. There are several issues that would lead to concerns in the interpretation of the findings e.g. the pre-operative subjective assessment prediction whilst reflective of pragmatic daily practice lacks the rigor of a controlled study:
- widespread population of patients
- seniority of the anaesthetist not defined
- speciality of intubating doctor not documented
- preparations prior to intubation not known
- level of difficulty anticipated not known (questions required just yes or no response)
- registration errors (data not externally verified)
- data registration process biased towards low intubation scores
- default setting on the questions set to “No” for anticipating difficulties
Large database, representative of population.
Aimed to see if Anaesthetists are accurately predicting difficult airways. Anaesthetists and patients randomised as retrospective. All patients were accounted for and those with known prior difficult intubations were discounted.
Results show a very poor outcome for predicting airway management difficulties by pre-operative assessment.
Several factors in the study may have influenced this poor outcome however the underlying message remains that with so many variables, it is unlikely that the current method and approach can achieve acceptable levels of prediction.
WOULD IT CHANGE MY PRACTICE?
I will continue to do routine airway assessments as part of the pre-op assessment. The conclusion is that we are bad at predicting difficult airways with current techniques but there is no single proven different technique that will aid us in our assessment. Therefore, on a day to day basis it will probably not change my practice until evidence of other factors/ airway assessments is proven.
Summary by Dr Orlanda Allen