Background
Dysphagia with subsequent aspiration occurs in up to 60% of acute respiratory failure (ARF) survivors. Accurate bedside tests for aspiration can reduce aspiration-related complications while minimizing delay of oral nutrition. In a cohort of ARF survivors, we determined the accuracy of the bedside swallowing evaluation (BSE) and its components for detecting aspiration.
Methods
Patients who were extubated after ≥24 hours of mechanical ventilation were eligible for enrollment. Within three days of extubation, patients underwent comprehensive BSE including 3-ounce water swallowing test (3-WST), followed by a gold standard test for aspiration, flexible endoscopic evaluation of swallowing (FEES).
Results
45 patients were included in the analysis. Median patient age was 55 years (IQR=47- 65). Median duration of mechanical ventilation was 3.3 days (IQR 1.8–6.0). 14 patients (31%) aspirated on FEES. Physical exam findings on BSE and abnormal swallowing during trials of different consistencies were variably associated with aspiration. Compared to FEES, the 3-WST yielded a sensitivity of 77% (95%CI=50–92%), specificity of 65% (95%CI=47–79%), and AUC=0.71; an SLP’s recommendation for altered diet yielded a sensitivity of 86% (95%CI=60– 96%), specificity of 52% (95%CI=35–68%), and AUC=0.69; an SLP’s recommendation for NPO yielded a sensitivity of 50% (95%CI=27–73%), specificity of 94% (95%CI=79–98%), and AUC=0.72.
Conclusions
The BSE and its components, including the 3-WST, demonstrated variable accuracy for aspiration in survivors of ARF. Investigation to determine the optimal non-invasive test for aspiration in ARF survivors is warranted.
Clinical Trials Registration
ClinicalTrials.gov Identifier: NCT02363686, Aspiration in Acute Respiratory Failure Survivors