Reliability and main findings of the flexible endoscopic evaluation of swallowing-Tensilon test in patients with myasthenia gravis and dysphagia

Background and purpose: Diagnosis of pharyngeal dysphagia caused by myasthenia gravis (MG) based on clinical examination alone is often challenging. Flexible endoscopic evaluation of swallowing (FEES) combined with Tensilon (edrophonium) application, referred to as the FEES-Tensilon test, was developed to improve diagnostic accuracy and to detect the main symptoms of pharyngeal dysphagia in MG. Here we investigated inter- and intra-rater reliability of the FEES-Tensilon test and analyzed the main endoscopic findings. Methods: Four experienced raters reviewed a total of 20 FEES-Tensilon test videos in randomized order. Residue severity was graded at four different pharyngeal spaces before and after Tensilon administration. All interpretations were performed twice per rater, 4 weeks apart (a total of 160 scorings). Intra-rater test–retest reliability and inter-rater reliability levels were calculated. Results: The most frequent FEES findings in patients with MG before Tensilon application were prominent residues of semi-solids spread all over the hypopharynx in varying locations. The reliability level of the interpretation of the FEES-Tensilon test was excellent regardless of the rater's profession or years of experience with FEES. All four raters showed high inter- and intra-reliability levels in interpreting the FEES-Tensilon test based on residue clearance (kappa = 0.922, 0.981). The degree of residue normalization in the vallecular space after Tensilon application showed the highest inter- and intra-rater reliability level (kappa = 0.863, 0.957) followed by the epiglottis (kappa = 0.813, 0.946) and pyriform sinuses (kappa = 0.836, 0.929). Conclusion: Interpretation of the FEES-Tensilon test based on residue severity and degree of Tensilon clearance, especially in the vallecular space, is consistent and reliable.