BACKGROUND: Dysphagia prevalence increases with age and a significant contribution is given by stroke survivors; its treatment is mainly based on rehabilitation, but outcome cannot be easily predicted. OBJECTIVE:The aim of this study is to detect possible predictors of the outcome of dysphagia in patients beginning rehabilitation after a major Central Nervous System injury. METHODS: Dysphagia severity was measured in 95 consecutive patients (71 with ischemic or hemorrhagic stroke) upon admission to our neurorehabilitation unit and at discharge, during the year 2017. The initial evaluation included also demographic data, functional and geriatric multidimensional assessment, laboratory test results and comorbidities. Their possible predictive value on the degree of recovery of the swallowing process at discharge has been analyzed. RESULTS: Poor functional conditions and the presence of cognitive impairment on admission appear to be associated with a worse outcome of dysphagia at discharge. A significant correlation exists between scores at functional scales at the beginning of rehabilitation and dysphagia score at discharge. Patients with cognitive impairment at the beginning (n = 60) showed a significantly lower degree of recovery of dysphagia at discharge. CONCLUSION: Other factors, beside the degree of dysphagia itself, are important to predict its outcome. Their knowledge not only allows an initial prognostic assessment; it can also be useful to decide which aspects should receive greater attention when treating patients with dysphagia.

We assessed the outcome of dysphagia rehabilitation in all the 139 patients with post-stroke dysphagia admitted to our Neurorehabilitation Unit during 2 years (2017 and 2018), divided into two groups: old (aged 65–84 years) and oldest-old (aged 85 or above). We studied which factors predicted dysphagia improvement in the two groups. The potential association of improvement with type of discharge was also evaluated. On admission, ‘old’ patients had more frequently aphasia (P = 0.02) and less frequently dysarthria (P = 0.03); ‘oldest old’ had more severe pressure ulcers (P = 0.008), higher levels of c reacting protein (P = 0.01) and more heart problems (P = 0.004). None of these factors was associated with the outcome of dysphagia. We found no difference between the two groups in the severity of dysphagia, as measured with Dysphagia Outcome and Severity Scale on admission and discharge, but due to minor differences, the degree of improvement was higher in the ‘old’ group (P = 0.02). The number of patients discharged with improved swallowing was also comparable. Cognitive impairment had a negative predicting role in the ‘old’ group (odds ratio 0.270, 95% confidence interval 0.101–0.725, P = 0.007). Norton Scale score predicted dysphagia improvement in the ‘oldest old’ group (odds ratio 1.611, 95% confidence interval 1.102–2.355, P = 0.007). Dysphagia improvement was associated with discharge home in general (P = 0.011) and in the ‘old’ group (P = 0.04). Our data, though preliminary, could give a contribution to implement patient-specific rehabilitation strategies; these could increase swallowing improvement in post-stroke dysphagia.