Background: Current individual and combined bedside airway tests can predict difficult laryngoscopy with low sensitivity.

Objective: To develop a composite scoring model to predict difficult laryngoscopy with higher sensitivity.

Materials and Methods: The retrospectively recruited samples were randomly divided into two groups, 80% for the Training group, and 20% for the Validation group. In the Training group, the extracted data included 1) the body mass index (BMI), 2) pre-operative bedside airway tests, i.e., inter-incisor gap (IIG), thyromental distance (TMD), sternomental distance (SMD), and modified Mallampati test (MMT), and 3) laryngoscopic view according to Cormack-Lehane (C-L) classification. The authors classified grade 3 and 4, according to the C-L classification, as difficult laryngoscopy. A predictive scoring model was developed using multivariate logistic regression analyses. The model was then validated in the Validation group.

Results: Seven thousand eight hundred five patients with a prevalence of difficult laryngoscopy of 3.6%, were divided as 6,251 and 1,554 patients in the Training and Validation group, respectively. In the Training group, the developed thyromental-modified Mallampati-sternomental (TMS) score comprised of three tests, the TMD, MMT, and SMD. The TMS had a total score of 6 and a score of 2 or more can predict difficult laryngoscopy with a respective sensitivity and specificity of 66.1% (95% CI 59.4 to 72.3) and 92.5% (95% CI 91.8 to 93.2). The performance was comparable when validated in the Validation group.

Conclusion: The TMS score can predict difficult laryngoscopy with a higher AUC and sensitivity albeit slightly decreased specificity than the TMD, SMD, MMT, or IIG.

Keywords: Airway test, Predictive model, Validation study, Difficult laryngoscopy

Received 23 Sep 2019 | Revised 19 Nov 2019 | Accepted 26 Nov 2019