Tongue Thickness and Skin-to-Hyoid Distance as Ultrasonographic Markers of Difficult Airway in Patients Without Anticipated Difficulty: A Prospective Diagnostic Accuracy Study of 713 Adults.

Background/

Objectives

In patients without clinically anticipated difficulty, an unexpected difficult airway carries the greatest potential for harm, yet routine bedside tests exclude difficulty poorly, and evidence for point-of-care ultrasonography remains heterogeneous, with difficult mask ventilation rarely studied as a distinct outcome. We evaluated three rapid anterior-neck ultrasonographic parameters-tongue thickness (TT), skin-to-hyoid-bone distance (SHBD), and cricothyroid membrane length (CTML)-for their diagnostic performance and independent association with a difficult airway in adults without anticipated difficulty.

Methods

In this prospective, single-center, diagnostic-accuracy study reported per STARD 2015 and STROBE, 713 adults (ASA I-III) underwent neutral-position measurement of TT, SHBD, and CTML by a sonographer blinded to the clinical assessment and airway-management outcomes. A composite difficult airway (difficult mask ventilation, laryngoscopy, or intubation) was assessed independently. Discrimination was evaluated by ROC analysis with DeLong comparisons, multivariable logistic regression, and bootstrap internal validation; p-values used Benjamini-Hochberg correction.

Results

A difficult airway occurred in 197/713 patients (27.6%). TT and SHBD were associated with the composite outcome and all components (AUC 0.68-0.73) and remained independent after adjustment for age, sex, BMI, and Mallampati class; CTML discriminated poorly (AUC ≈ 0.55), with only marginal significance after correction. Negative predictive values were high (78-99%) but prevalence-dependent, with modest likelihood ratios. Adding TT and SHBD to clinical variables improved discrimination (AUC 0.69 → 0.78; DeLong p < 0.001).

Conclusions

TT and SHBD show a limited but independent diagnostic signal that may complement, not replace, clinical screening; CTML showed no clinically meaningful discrimination in this cohort. Proposed cut-offs are exploratory and require standardized measurement, reproducibility testing, and external validation before clinical adoption.