Background: The German expert recommendations on the management of dysphagia in patients after acute stroke suggest an algorithm for clinical and technical investigations to identify patients at risk for aspiration and thus reduce the rate of aspiration pneumonia. The effectiveness of this algorithm has, however, not yet been prospectively validated .
Methods: In this study 144 consecutive stroke patients were assessed by a full bedside swallowing assessment including the screening procedures of standardized swallowing assessment (SSA) and 2 out of 6. Flexible endoscopic evaluation of swallowing (FEES) was performed in all patients.
Results: Aspiration was diagnosed in 25 patients (17.4%) by FEES. The SSA predicted aspiration with a sensitivity of 76% and a specificity of 55.5% and the 2 out of 6 screening with a sensitivity of 68.0% and a specificity of 61.0%. Of the patients 7 with negative screening for 2 out of 6 and 6 patients with negative SSA showed silent aspiration with the penetration aspiration scale (PAS 8) during FEES (28% of all patients with aspiration). Significant predictors for aspiration were dysarthria, dysphonia, abnormal volitional cough and cough after swallowing water; however, in multivariable analysis only dysarthria and cough after swallowing water were identified as independent predictors for aspiration. The rate of aspiration pneumonia was 2.8%.
Conclusion: Clinical screening alone is not sufficient to identify patients at risk for aspiration pneumonia. The FEES should be used at a low threshold in cases of severe stroke and minor clinical abnormalities, especially concerning isolated dysarthria and cough after swallowing water; therefore, current recommendations should be correspondingly modified.
Keywords: Acute stroke; Aspiration; Flexible endoscopic evaluation of swallowing (FEES); Management of dysphagia; Screening.