Enhancing Difficult Laryngoscopy Prediction Through A Mixed Scoring System

Overview

This study aims to develop a comprehensive airway assessment score including traditional bedside clinical tests and ultrasound measurements with high accuracy in predicting difficult airway, and to validate it through testing it on a large population sample. The main question it aims to answer is: Is the comprehensive scoring system that includes clinical bedside and ultrasound measurements developed to predict difficult airways valid in a large population sample? Adult patients undergoing general anesthesia with American Society of Anesthesiologists (ASA) classification 1-4 will be recruited. Traditional and ultrasound measurements will be collected.

The ability to predict a difficult airway is the cornerstone of avoiding its dreaded complication. To standardize the task of airway assessment, multiple scores have been developed and subsequently validated, such as the Wilson risk sum score, El-ghanzouri risk index, Arne risk index…Despite utilizing these tools, there are situations in which subjects classified as easy airways surprisingly present with challenging airways. This has raised the need for developing and validating newer airway assessment scores, that include quantitative ultrasound measurements. This study aims to develop a comprehensive airway assessment score including traditional bedside clinical tests and ultrasound measurements with high accuracy in predicting difficult airway, and to validate it through testing it on a large population sample. After obtaining the consent from patients to participate in the study, the day before the surgery in the PAU or patients' rooms, a member of the research team with experience in ultrasound, will collect the required data either in the preoperative unit or in the patient room the day before the surgery (for admitted patients). The traditional metrics collected are: Modified Mallampati score, inter-incisor distance, upper lip bite test, the maximum range of head and neck movement, neck circumference, thyromental distance, receding mandible, buck teeth, edentulousness, presence of beard, and sternomental distance and the ultrasound measurements collected are a distance from skin to epiglottis (DSE), hyomental distance (HMD), the ratio of hyomental distance in maximal head extension to hyomental distance in neutral position (HMDR), tongue thickness (TT), the ratio of pre-epiglottic distance to the epiglottic distance at the midpoint of the VC (Pre-E/E-VC), distance from skin to the hyoid bone (DSHB), and distance from skin to vocal cords (DSVC). After induction, the level of difficulty of laryngoscopy will be collected by a member of the research team. For clinical and sonographic criteria with statistical significance, the investigators will calculate their sensitivity, specificity, and positive and negative predictive values. To determine the quantitative variable cut-off, the investigators will establish the ROC (Receiver Operating Characteristics). After checking that the area under the curve (AUC) is significantly \>0.7, the investigators will choose as the cut-off, the value of the variable that corresponds to the best "sensitivity-specificity" pair. Finally, to identify independent criteria associated with difficult laryngoscopy, the investigators will perform multivariable analysis by logistic regression. These independent predictive criteria are the score parameters defined by values corresponding to their respective odd ratios. Then the investigators establish the cutoff value of the score from its ROC curve. As the patients in this study are almost all Lebanese or Arab, potential bias and influencing factors must be considered when the models are used for patients in other countries. Also, this is a single institution study, a multicenter and multiracial study is needed to develop a score that can be widely used internationally.