Predictors of Postextubation Dysphagia in Nonstroke Patients

Abstract

Purpose: Postextubation dysphagia (PED) is an ICU-acquired condition encompassing swallowing disorders after extubation. PED is associated with complications such as aspiration and increased mortality. For non-stroke patients, there is no bedside screening evaluation by the nursing staff to predict PED. The primary objective of this study is to find predictors of PED from clinical factors and to validate the 3-oz water test (3OWT) as a screening test in non-stroke patients. Methods: This is an observational, prospective cohort study. Patients with more than 48 hours of intubation, without acute stroke or neuromuscular dysphagia, were included. A nurse completed a questionnaire of clinical factors. PED was diagnosed by a Speech and Swallow Therapist (SST) through the Bedside Swallowing Evaluation (BSE). The Fiberoptic Evaluation of Swallowing (FEES) was offered to all patients and was used as a diagnostic test of PED. The 3OWT was performed after extubation. Patients were evaluated for aspiration and diet tolerance after 48 hours. Results: A total of 44 patients were enrolled. All underwent a BSE by SST and 3OWT, 28 patients had FEES. PED, diagnosed by BSE, was found in 13 patients (29.5%). Abnormal cough was the only clinical factor correlated with PED (OR: 4.86, CI-95% 1.19-19.87; p<0.05). The duration of intubation more than 120 hours (OR: 1.53, CI-95% 0.39-5.95) and reintubation (OR: 0.95, CI-95% 0.165.63) were not found to be risk factors for PED. The 3OWT was found to be related to PED (p<0.05), with a sensitivity of 69.23% (CI 95%, 40.30 to 98.17), specificity of 93.55% (CI 95%, 83.29 to 100), positive predictive value of 81.82% (CI 95%, 54.48 to 100) and negative predictive value 87.88% (CI 95%, 75.23 to 100). The FEES was performed on 28 patients; 8 had PED, 12 patients refused the procedure, and 2 stopped the procedure due to discomfort. FEES and BSE were 100% in agreement for diagnosing PED. After 48 hours there were no aspiration events in 42 patients, 2 died due to their critical conditions. Conclusions: Abnormal cough reported by an ICU nurse was the only clinical factor found to be a predictor of PED. Prolonged intubation was not found to be a risk factor for PED. The 3OWT is a valuable screening test. It can be utilized as an aid for identifying patients that would benefit from SST evaluation. Further studies involving a larger sample should be conducted to confirm these findings. Clinical Implications: Abnormal cough was found to be a predictor of PED. 3OWT could be used as a screening test for PED.