A Model to Predict Difficult Airway Alerts After Videolaryngoscopy in Adults with Anticipated Difficult Airways – the VIDIAC Score

K. Kohse, H. K. Siebert, P. B. Sasu et al.

Anaesthesia 2022; 77, 1089-1096 https://doi.org/10.1111/anae.15841

Aim of Study

• Develop a model to classify the difficulty of videolaryngoscopic tracheal intubation

Design and Location

• Hamburg; Observational study

Methodology

• Patient selection: Adults over 18 years old, undergoing ENT or OMFS between 1^st April 2019 – 3^rd April 2020 who were expected difficult airway management and underwent tracheal intubation aided by videolaryngoscopy (C-MAC)
• 374 Anaesthetics in 320 participants (400 Anaesthetics eligible)
• Anaesthetist and two observers independently assessed intubation-related characteristics during/directly following intubation

Primary outcome

• Whether Anaesthetist expected future videolaryngoscopic tracheal intubations to be difficult and issued ‘difficult airway alert’ after videolaryngoscopy

Secondary outcomes

• Compare the discrimination of VIDIAC and Cormack-Lehane classifications for difficult airway alerts
• Assess the discrimination of the model for transition to hyperangulated blade

Statistics

• Fleiss’ to calculate agreement between anaesthetist + 2 observers ratings of each factor before model developed
• Developed multivariable prediction model
• 100,000 random forest analysis decision trees to select variables for the model
• Calculated Gini impurity to quantify importance of intubation related variables to correctly classify patients with an alert
• Lasso regression for variable selection, model development & internal validation

Results

• Anaesthetists issued airway alerts after 183/374 (49%) tracheal intubations
• Produced a model with variables independently associated with issuing an alert:
• VIDIAC score discriminated participants more or less likely to have secondary laryngoscopy with hyperangulated blades
• Cormack-Lehane classification indicated less discrimination than the VIDIAC score

Limitations from the Study

• Reproducibility of results – single-center, specific patient group & surgeries
• Small sample size – 320 patients
• Anaesthetic experience ranged from 1month – 14 years
• Observers assessments truly ‘independent’
• Score validated for use with C-MAC only

Discussion from journal club

• Cormack-Lehane classification not validated for use with videolaryngoscopy
• Need to document findings on videolaryngoscopy and ease of intubation – including type of videolaryngoscope and blade used
• Mixed opinions on whether a specific scoring system is necessary
• Most felt VIDIAC scoring system not especially ‘simple’

Summary by Dr K Edgerton. Journal Club Meeting 3 Nov 2022.