Association between Malnutrition Severity and Swallowing Function in Convalescent Rehabilitation Wards: A Multi-Center Cohort Study in Malnourished Patients with Sarcopenic Dysphagia

Objectives To evaluate the characteristics of sarcopenic dysphagia (SD) and the prognosis of swallowing function in convalescent rehabilitation hospital patients; and to investigate the association between malnutrition severity and SD. Design A prospective, multi-center, cohort study. Setting We extracted registry data from the Japanese Sarcopenic Dysphagia Database, focusing on patients admitted to convalescent rehabilitation hospitals. Participants A total of 207 participants were recruited and stratified according to the presence or absence of SD. Next, the participants were divided into groups based on nutrition status using the Global Leadership Initiative on Malnutrition criteria normal nutrition, moderate malnutrition, and severe malnutrition. We also compared the outcomes between patients with SD (SD group) and those without SD (no-SD group) according to malnutrition status. Measurements The Food Intake LEVEL Scale (FILS) score was the outcome measure. Higher scores on the FILS indicate better swallowing function. We compared the patient characteristics between the SD and non-SD groups among all patients and then according to the severity of malnutrition. Results A total of 207 patients were recruited. 11 were diagnosed with normal nutrition, 72 with moderate malnutrition, and 124 with severe malnutrition. There were 128 participants with SD and 79 participants without SD; the prevalence of SD was 61.8%. Hip fracture was the most common disease among the SD patients (34.4%). The median time to the end of follow-up was 73.5 days for the SD group and 84.0 days for the no-SD group. There was no significant difference in the FILS score between the SD and no-SD groups, but the increase in the FILS score was significantly lower in the SD group than the no-SD group among patients with severe malnutrition after adjusting for confounding factors (age, sex, FILS at admission, BMI, cognitive functional independence measure, and care level before onset) (beta = -0.206, p = 0.011, 95% confidence interval = -0.723, -0.098). Conclusion Orthopedic diseases are the most common type of disease among SD patients in convalescent rehabilitation hospitals. Swallowing dysfunction was particularly severe in malnourished patients with SD. This result suggests the importance of the definition of SD for malnourished patients. We should practice nutritional management as soon as possible in severely malnourished patients diagnosed with SD.