Postextubation swallowing disorders (SD) are common in the intensive care unit (ICU) and are associated with severe complications, including aspiration pneumonia, a three-fold increase in reintubation risk and higher mortality. While fibreoptic endoscopic evaluation of swallowing (FEES) and videofluoroscopy are gold standards for diagnosis, they are often impractical or impossible to perform on intubated patients. The use of ultrasound offers a non-invasive, bedside alternative to evaluate the musculoskeletal structures involved in swallowing. The Echographie Identifier les troubles de Déglutition Acquis en Réanimation (EIDAR) study aims to evaluate the diagnostic performance of pre-extubation ultrasound in identifying patients at risk of SD following mechanical ventilation.

This prospective, monocentric diagnostic study conducted at the Dijon University Hospital ICU will include 100 adult patients ventilated for ≥48 hours. The primary outcome is the presence of SD, defined as a Penetration-Aspiration Scale score >2 during a FEES procedure performed 3 to 24 hours postextubation and independently assessed by an otolaryngologist blinded to index test results. Pre-extubation cervical ultrasound (Index Test) will be performed within 3 hours prior to extubation and measure hyoid bone ascension (primary variable of interest), geniohyoid muscle surface area and digastric muscle cross-sectional area. The diagnostic performance of cervical ultrasonographic parameters will be assessed using their discriminative capacity via a receiver operating characteristic curve. The feasibility of the ultrasound procedure in a critical care setting will also be assessed.

The study received a favourable opinion from the independent ethics committee CPP Ouest III and is registered with the French health authority ANSM (national agency on safety in medicine and health products). It is conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Participants or their proxies provide free and informed oral consent. Results will be submitted for publication in peer-reviewed medical journals and presented at international conferences.

RCB 2023-A00461-44 and NCT05922085