Accuracy of the Brazilian version of the DYMUS questionnaire for the screening of oropharyngeal dysphagia in multiple sclerosis

Abstract

Background: Oropharyngeal dysphagia is a common symptom of many neurological diseases, including Multiple Sclerosis (MS). Early identification of the risk of dysphagia in neurological patients is very important for early referral for specialized evaluations of oropharyngeal swallowing and treatments. The Dysphagia in Multiple Sclerosis (DYMUS) questionnaire has been translated and validated in different countries over the last 10 years. We aimed to analyze the accuracy of the Brazilian Portuguese version of the DYMUS (DYMUS-BR) questionnaire in identifying dysphagia in patients with MS. Methods: The DYMUS questionnaire and a videofluorographic swallowing study (VFSS) were conducted in 30 patients with MS. Dysphagia was identified by at least one abnormal response and was considered alarming when the DYMUS scores were equal to or higher than 3. Patients were considered to have dysphagia in the VFSS when one or more signs of impairment in the efficiency and/or safety of swallowing were detected. Results: According to the initial self-assessment, 37% (N = 11) of patients with MS self-reported with dysphagia. According to the DYMUS-BR scores, 53% (N = 16) of the patients with MS were classified as having dysphagia. The sensitivity, specificity, and positive and negative predictive values of the DYMUS-BR questionnaire for the detection of dysphagia as measured by the VFSS were 50% [95% confidence interval (CI) 29–71], 78% (95% CI 61–90), 60% (95% CI 42–76), and 70% (95% CI 60–78), respectively. The area under the receiver-operating characteristic curve for detecting dysphagia was 64% (95% CI 49–79). Conclusion: The accuracy of the DYMUS-BR questionnaire is poor to detect mild swallowing impairment in patients with MS. However, we suggest longitudinal follow-up in patients with low DYMUS-BR scores for early detection of oropharyngeal dysphagia.