A retrospective review of swallow dysfunction in patients with severe traumatic brain injury.

In the acute-care setting, it is difficult for clinicians to determine which patients with severe traumatic brain injury will have long-term oropharyngeal dysphagia (>6 weeks) and which patients will begin oral nutrition quickly. Patients frequently remain in the acute-care setting while physicians determine whether to place a percutaneous endoscopic gastrostomy (PEG) tube. To improve the acute-care clinician's ability to predict long-term oropharyngeal dysphagia and subsequent need for PEG tube placement in patients with severe traumatic brain injury [Glascow Coma Scale (GCS) ≤ 8), a novel prediction model was created utilizing clinical information and acute-care swallowing evaluation findings. Five years of retrospective data were obtained from trauma patients at a Level 1 trauma hospital. Of the 375 patients who survived their hospitalization with a GCS ≤ 8, a total of 269 patients received Ranchos Los Amigos (RLA) scores. Of those patients who were scored for RLA, 219 patients underwent swallowing evaluation. Ninety-six of the 219 patients were discharged from the hospital with a feeding tube, and 123 patients were discharged without one. Logistic regression models examined the association between clinical and patient characteristics and whether a patient with severe traumatic brain injury exhibited long-term oropharyngeal dysphagia. Multivariable logistic regression analysis revealed that increased age, low RLA score, tracheostomy tube placement, and aphonia observed on the initial swallowing evaluation significantly increased the odds of being discharged from the acute-care hospital with a feeding tube. The resultant model could be used clinically to guide decision making and to counsel patients and families.