Objective To investigate the clinical factors associated with successful gastrostomy tube weaning in patients with prolonged dysphagia after stroke. Methods This study involved a retrospective medical chart review of patients diagnosed with prolonged dysphagia after stroke who underwent gastrostomy tube insertion between May 2013 and January 2020. Forty-seven patients were enrolled and consequently divided into gastrostomy tube sustaining and weaning groups. The numbers of patients in the sustaining and weaning groups were 31 and 16, respectively. The patients' demographic data, Korean version of Mini-Mental State Examination (K-MMSE) score, Korean version of the Modified Barthel Index (K-MBI), Functional Dysphagia Scale (FDS) score, and Penetration-Aspiration Scale (PAS) score were compared between the two groups. A videofluoroscopic swallowing study was performed before making the decision of gastrostomy tube weaning. The clinical factors associated with gastrostomy tube weaning were then investigated. Results There were significant differences in age; history of aspiration pneumonia; K-MMSE, FDS, and PAS scores; and K-MBI between the groups. In the multiple logistic regression analysis, the FDS (odds ratio [OR]=0.791; 95% confidence interval [CI], 0.634-0.987) and PAS scores (OR=0.205; 95% CI, 0.059-0.718) were associated with successful gastrostomy tube weaning. In the receiver operating characteristic curve analysis, the FDS and PAS were useful screening tools for successful weaning, with areas under the curve of 0.911 and 0.918, respectively. Conclusion In patients with prolonged dysphagia, the FDS and PAS scores are the only factors associated with successful gastrostomy tube weaning. An evaluation of the swallowing function is necessary before deciding to initiate gastrostomy tube weaning.

Objective: We investigated the usefulness of maximal expiratory pressure (MEP) in evaluating dysphagia subsequent to ischemic stroke. Methods: This study included patients with ischemic stroke who underwent MEP testing and videofluoroscopic swallowing study (VFSS), from October 2016 to February 2020. The VFSS findings were interpreted using the penetrationaspiration scale (PAS) and functional dysphagia scale (FDS). Patients were stratified into the non-aspiration (n=59) and aspiration (n=47) groups. Partial correlation analysis among MEP, PAS, and FDS was performed after adjusting for age. Binary logistic regression using PAS was conducted to investigate the risk factors predisposing patients to inclusion in the aspiration group. Multiple linear regression using FDS was conducted to investigate the risk factors according to dysphagia severity. Receiver operating characteristic (ROC) curve analysis was applied to investigate factors which could be useful for detecting aspiration. Results: Student's t-test revealed a significant difference in MEP between the non-aspiration and aspiration groups. MEP showed a positive correlation with PAS and FDS. MEP was also determined to be a risk factor for inclusion into the aspiration group, and a risk factor according to the severity of dysphagia. In the ROC curve analysis, MEP showed good diagnostic properties to help classify patients with aspiration. Conclusion: Our results indicate that swallowing assessment can predict and help prevent aspiration pneumonia in patients with ischemic stroke. In the present study, MEP showed significant association with aspiration and the severity of dysphagia. Thus, determining the MEP during swallowing assessment in patients with ischemic stroke is potentially a useful parameter to predict dysphagia.