PRediction Of Cognitive Recovery After Stroke(PROCRAS study): a role for new MRI modalities in daily clinical practice?

RATIONALE: Every day, in the Netherlands, about 100 persons are struck by an ischemic stroke, with huge physical, psychological and financial impact on patients and society. Cognitive dysfunction affects up to 70% of patients. About half of them will manifest (partial) recovery, but the others are left with sequelae or even deteriorate. This is a key aspect of long-term outcome because of its negative impact on independent daily life functioning, social reintegration and quality of life. The spectrum of ischemic strokes impacting cognition is wide:it ranges from extensive cortical to small lacunar infarcts. Particularly in the elderly a stroke may add to the burden of pre-existing vascular and degenerative pathologies, thus increasing the chance of cognitive deficits (1) Cognitive deficits after stroke can easily be missed in the acute stage, especially in patients with minimal motor impairments and no language problems (‘the walking and talking patient’). Therefore, cognitive screening is performed in all patients at our stroke-unit. It improves guidance of patients and the planning of rehabilitation. However, the predictive value of this early screening for long-term cognitive outcome is limited. A better prediction of the cognitive status later on is needed to improve prognostic information for patients and a more tailored allocation of rehabilitation facilities. Large cohort studies have identified predictors of cognitive function in the long term after stroke, e.g. age, location and size of the infarct, premorbid status, cognitive function in the acute stage, but individualized prediction of cognitive outcome using these clinical parameters is imprecise. New imaging techniques have come forward that not only allow a precise localization of the lesion but also provide detailed individualized information about the structural cerebral networks involved in various cognitive functions. From a conceptual viewpoint it is likely that the extent to which a lesion disturbs the networks that are responsible for the main flow of information through the brain would be a key determinant of functional deficits and recovery,with an important potential to improve prediction on top of existing clinical parameters.Up till now these techniques have been explored in experimental designs but have not been evaluated in daily clinical practice. In the PRESENT PROJECT we will establish the feasibility and value of integration of these new imaging techniques in daily clinical practice in a dedicated stroke clinic in order to predict cognitive outcome one year post stroke. AIMS OF THE STUDY: 1: assess feasibility of a standardized work–up including cognitive evaluation and advanced brain MRI in daily clinical practice in a dedicated stroke clinic. 2: determine the additional predictive value of MRI measures of brain connectivity on top of existing clinical parameters for cognitive recovery or deterioration in the first year after stroke. APPROACH: All ischemic stroke patients admitted to the stroke unit of the St.Elisabeth Hospital Tilburg who are impaired on regular cognitive screening (MoCA:score < 26)),but did not manifest dementia prior to stroke, are eligible for inclusion in the study. After informed consent, these patients will undergo formal neuropsychological tests in the second week post stroke, according to a standardized 6o min protocol for vascular cognitive impairment(2). Patients will be graded as non–impaired or having a mild or major vascular cognitive disorder according to the VasCog criteria (3). Patients with a mild or major vascular cognitive disorder are the target group for our prognostic study. In these patients a 3Tesla MRI scan will be acquired in the second or third week after stroke. Infarct volume and location,as well as brain volumes will be assessed. Networks will be reconstructed with state of the art fiber tractography and processing techniques. At one year after stroke again patients will undergo the neuropsychological test battery. Besides, participation, functional status and Quality of Life will be assessed. Clinical variables, cognitive function in the acute stage and imaging parameters will be used to predict cognitive status at one year. If the above mentioned will prove to be feasible, we will implement the results at the stroke unit of the St.Elisabeth hospital and validity of the prediction algorithm will be tested in a new prospective cohort. EXPECTED RESULTS We expect that this approach will support a more accurate diagnosis, prognostication and management of patients with post-stroke cognitive deficits. It will establish the feasibility and value of high–end image analysis techniques in daily practice in an unselected population in a dedicated stroke center. It will lead to a more precise identification of patients who are at risk of long term problems due to the presence of cognitive deficits. This will help to offer a better organized, more tailored after care possible.