Predicting Language Recovery in Acute Stroke Patients in the Neurovascular Intensive Care Unit: An Exploratory Study With the Core Assessment of Language Processing.

Overview

Introduction: Stroke affects one person every 4 minutes in France (i.e. more than 140,000 new cases per year) resulting in cognitive and motor disorders. Aphasia is one of the most devastating cognitive disorders that persist in the late phase. However, early treatment of aphasia can improve the effects of rehabilitation. Identifying, as early as possible, the patients most at risk of presenting persistent language disorders in the late phase would make it possible to improve their management and increase the effects of cognitive rehabilitation on their language abilities. The aim of this project is to evaluate whether the Core Assessment of Language Processing (CALAP) assessed in the acute phase of stroke can predict language abilities in the late phase. Hypothesis/Objective: The primary objective is to determine whether the language abilities of patients in the acute phase of stroke can be used to predict language abilities in the late phase. Secondary objectives are to determine whether prediction can be improved with (1) brain MRI data and (2) neuropsychological assessment data. The (3) secondary objective is to determine whether cognitive abilities at the chronic phase can be predicted by language performance in the acute phase. The (4) secondary objective is to assess whether language rehabilitation modifies the predictive power of the language abilities assessed with the CALAP. Method: Patients will be included during their hospitalization after a brain vascular injury (acute phase, up to 21 days of hospitalization). After discharge, they will return for a post-stroke assessment between 3 and 18 months after the acute phase. During these two visits, a clinical and neurological examination, a neuropsychological assessment and an MRI will be performed. A prediction model (development and validation) will be used for all objectives using a linear regression model with cross validation. The entire sample consists of stroke patients. The study is single-center and will have a total duration of 6 years with an estimated 570 patients included. Conclusion: Predicting the language abilities of a post-stroke patient will improve clinical management and direct patients requiring language rehabilitation to appropriate care.

Conditions