Background

Videofluoroscopy (VFS) is the gold standard in evaluating dysphagia. Water swallow tests (WST) and the Eating Assessment Tool (EAT-10) are commonly used in dysphagia screening. We aimed to determine the feasibility of WST and EAT-10 as screening tools for referral to VFS.

Methods

Patients (n=150, median age: 70.0 years, range: 19–92 years, 58.7% female) referred to VFS completed the WST and EAT-10 before the examination. In the WST, we evaluated both the qualitative parameters (coughing, possible change in voice) and quantitative parameters (average drinking bolus size, swallowing speed). Correlations of EAT-10 total scores and WST parameters to the VFS findings were analyzed both individually and combined.

Results

In the WST, the most specific (89.7%) predictor of normal VFS findings was the absence of coughing, and the most sensitive (79.1%) parameter to predict abnormal findings was a bolus size of ≤20 mL. Using a combination of coughing and a bolus size ≤20 mL (simplified WST), the sensitivity of predicting abnormal findings increased to 83.5%. The most sensitive (84.6%) predictor of penetration/aspiration was failing any parameter in the WST. Lack of coughing indicated an absence of penetration/aspiration with an 82.5% specificity. Swallowing speed or combining the EAT-10 results with the WST results did not enhance the sensitivity or specificity of the WST for predicting the VFS results.

Conclusions

Coughing and average drinking bolus size are the most important parameters in WST when screening for referral to VFS, whereas the swallowing speed does not seem to be useful. The WST is superior to EAT-10 in predicting VFS findings.

Level of Evidence

4 Laryngoscope, 2023