Clinical predictors for perioperative anticipated and unanticipated difficult intubation: a matched case-control study

We aimed to determine the clinical predictors of perioperative anticipated and unanticipated difficult intubation using a matched case-control study. We recruited patients undergoing surgery with endotracheal intubation from 2015 to 2020. Difficult intubation was defined as at least 3 attempts to perform intubation with conventional or video laryngoscopy before surgery. Controls were randomly selected in a ratio of 3:1 matching on year of surgery, site of operation and age within 5 years. Clinical predictors were evaluated. A multivariate conditional logistic regression analysis was performed and presented with adjusted odds ratios (OR) and 95% confidence intervals (CI). We selected 168 cases and 504 controls out of 62,111 intubated patients. The predictors for anticipated difficult intubation were previous history of difficult airway (OR [95% CI]: 6.4 [1.3,32.5]) and abnormal facial appearance/syndrome (OR [95% CI]: 6.1 [1.3,28.0]). The predictors of unanticipated difficult intubation were BMI < 15 kg/m² (OR [95% CI]: 4.6 [1.5,14.4]), ASA physical status of 3 (OR [95% CI]: 3.6 [1.1,11.3]), airway/neck/oral deformity (OR [95% CI]: 2.1 [1.03,4.3]) and tumors at intraoral, airway or thyroid (OR [95% CI]: 2.4 [1.1,4.9]). Undiagnosed airway/neck/oral deformity and tumors at intraoral, airway or thyroid sites might be encountered with unanticipated difficult intubation, especially in patients who have a normal general appearance.