A prospective randomized crossover manikin study comparing GlideScope® video laryngoscopy and direct laryngoscopy for impacted fish bone removal in a simulated difficult airway

Abstract Background Fish bone ingestion is a common emergency presentation. Direct laryngoscopy (DL) is the conventional bedside technique, but it can be limited in anatomically challenging airways. Video laryngoscopy (VL) improves glottic visualization for intubation, yet its role in foreign body retrieval is unclear. Because difficult airway features (e.g., restricted mouth opening, cervical spine immobility) occur in the general population and may complicate retrieval, we compared VL and DL in a simulated difficult airway model. Methods In this prospective, multicenter, randomized crossover manikin study, 56 emergency physicians (24 specialists, 32 non‐specialists) performed two timed retrieval attempts (maximum 120 s each) on a manikin fitted with a cervical collar. Each attempt required the removal of three simulated fish bones (vallecula, left pyriform sinus, post‐cricoid) using Magill forceps. Device order was block‐randomized and stratified by specialist status. The primary outcome was complete retrieval success. Secondary outcomes included retrieval times, Cormack‐Lehane grade, ease of use, and device preference. Results VL achieved higher success than DL (83.9% vs. 66.1%; p = 0.013). Adjusted analysis favored VL (adjusted odds ratio 3.10, 95% CI 0.90–10.71; p = 0.073). VL yielded superior glottic views (Grade 1: 55.4% vs. 12.5%; p < 0.001), was rated easier ( p < 0.001), and was preferred by most operators (57.1%). Total retrieval times were similar, but VL was slower to first retrieval (17.5 vs. 13.0 s; p = 0.001). Conclusions In a simulated difficult airway, VL improved success, visualization, and operator preference; slower initial retrieval suggests a need for instrument optimization and targeted training.