Résumé : Background: Clinical evaluation of swallowing disorders postextubation is often neglected. Videofluoroscopy is the gold standard with fiber-optic endoscopic evaluation of swallowing (FEES) having a high sensitivity. The aim of this study was to analyze the correlations between clinical, FEES, and videofluoroscopic evaluations in the intensive care unit. Methods: Twenty-one patients extubated after prolonged intubation were subjected to a clinical evaluation of swallowing and FEES within 24 hours. This was repeated at 48 hours with a videofluoroscopic evaluation with identical swallowing-namely, boluses of liquid and thickened water. The patients were scored from 0 (normal) to 3 (worst). Results: There was no correlation between the oral phase (bedside evaluation) and FEES. The correlation between pharyngeal phase (palatal and laryngeal elevation, pharyngeal rales, and gag reflex) before and after swallowing at 24 and 48 hours was statistically significant (liquid water P =.025 [24 hours] vs P <.001 [48 hours]; thickened water P <.001 [24 and 48 hours]). Clinical assessment, although not statistically significant, failed to detect silent aspiration (P =.58). There was a good correlation between FEES and videofluoroscopy as opposed to clinical assessment and videofluoroscopy (P <.001 vs P =.762). Conclusion: Cough is a reliable sign of swallowing disorder but does not exclude silent aspiration and contraindicates oral feeding. Cough induced by liquid water should lead to modification of diet in terms of consistency and viscosity with cough reassessment. © 2011 American Society for Parenteral and Enteral Nutrition.