Sensitivity and Specificity of Bedside Screening Tests for Detection of Aspiration in Patients Admitted to a Public Rehabilitation Hospital

Abstract

Early detection of dysphagia and specifically aspiration is essential to prevent and reduce complications of hospitalized patients in rehabilitation centers. Bedside screening test are often used to evaluate swallowing disorders, but their results may be questionable due to insufficient and inconsistent sensitivity and specificity. To compare the sensitivity and specificity of various bedside screening tests for detecting aspiration in hospitalized rehabilitation patients. A prospective observational study was performed in 150 consecutive patients of a tertiary rehabilitation hospital. Patients were evaluated regarding clinical predictors for aspiration, maximum phonation time (MPT), Eating Assessment Tool 10 (EAT-10) questionnaire, tongue strength and endurance (Iowa Oral Performance Instrument [IOPI]) and a swallowing test (Volume-Viscosity Swallow Test [V-VST]). Flexible Endoscopic Evaluation of Swallowing (FEES) was the reference test. Of the 144 patients included, 22% aspirated on FEES. Previous history of pneumonia, dysarthria, wet voice, and abnormal cough reflex were significantly associated with aspiration. The sensitivity, specificity and accuracy for V-VST (83.3%, 72.6%, 74.8%, respectively) and EAT-10 (82.8%, 57.7%, 62.8%, respectively) to detect aspiration were superior than those of other methods. Maximum tongue strength on IOPI and MPT presented high sensitivity but low specificity to detect aspiration. Clinical predictors of aspiration (previous history of pneumonia, dysarthria, wet voice, and abnormal cough reflex) associated with either V-VST or EAT-10 may be good screening methods to detect aspiration in patients hospitalized in a rehabilitation center.