Upper airway angulation: a prospective cohort to evaluate the predictive values of a new feature for difficult laryngoscopies

Abstract

Background: There is currently some discussion over the actual role of airway prediction as we don't have an accurate predictor available so far. In this regard, some angles of the upper airways have demonstrated reasonable predictive performance and from that we aimed to assess the predictive valeus of a new angle (Figure 1) - between upper lip, mandibular angle, and thyroid cartilage prominence - for airway prediction as well as to compare these values with those from modified Mallampati test. Methods: An accuracy study was performed with 125 patients undertaking general anesthesia for surgical procedures. We collected preoperatively data on age, sex, weight, height, body mass index, and the aforementioned angle (upper airway angulation). The main outcome was difficult laryngoscopy defined as Cormack and Lehane's grades 3 or 4. Bivariate analyses were performed and two predictive models built with their ROC curves and AUC. Results: Difficult laryngoscopies were present in 9 patients (7.2%). Sex (p=0.009); MMT (p=0.020), and upper airway angulation (p=0.000) were associated with difficult laryngoscopies. The threshold of MMT - by Youden index - for difficulty was grade II and the predictive values for this test were as follows: AUC of 68.9% (95% CI: 59-78.7%); sensitivity of 100%; specificity of 45.7%; positive predictive value of 12.5%; negative predictive value of 100%; overall accuracy of 49.6%; balanced accuracy of 72.8%. For upper airway angulation the predictive values were AUC of 88.7% (95% CI: 80-97.4%); sensitivity of 88.9%; specificity of 88.8%; positive predictive value of 38.1%; negative predictive value of 99%; overall accuracy of 88.8%; and balanced accuracy of 88.8%. Conclusion: Upper airway angulation is a new feature associated with difficult laryngoscopies and has better predictive performance as compared to the widely used modified Mallampati test.