Related Experiment Videos
A quantitative MRI study of vascular dementia
C K Liu1, B L Miller, J L Cummings
1Department of Neurology, Harbor-UCLA Medical Center.
Abstract:
We studied the MRI and clinical factors associated with dementia following stroke by quantifying ventricle-to-brain ratio (VBR), anatomic region of infarction, and cortical, subcortical, and white matter areas of infarction in 24 stroke patients with dementia and 29 nondemented stroke patients. The factors that most strongly correlated with dementia were total white matter lesion (WML) area, left WML, VBR, right WML, age, left cortical infarction area, left parietal infarction area, and total infarction area. Using discriminant analysis, these factors correctly classified 28 of 29 nondemented patients and 18 of 24 demented patients. Both cortical and white matter total infarction area measurements were strongly associated with dementia in stroke patients, suggesting that these factors strongly influenced the development of dementia following stroke. There was a strong association between dementia and left- but not right-hemisphere infarction area. The only demographic factor that strongly associated with dementia was age.
Insights
White matter lesions and ventricle-to-brain ratio on MRI scans are key indicators of dementia after stroke. These factors, along with infarction size, help predict cognitive decline in stroke survivors.
Area of Science:
- Neurology
- Radiology
- Neuroimaging
Background:
- Stroke is a leading cause of disability, and post-stroke dementia is a significant complication.
- Identifying factors predicting dementia after stroke is crucial for patient management and prognosis.
Purpose of the Study:
- To investigate MRI and clinical factors associated with dementia in stroke patients.
- To determine the predictive power of these factors for classifying demented and non-demented stroke patients.
Main Methods:
- Retrospective analysis of MRI scans and clinical data from 53 stroke patients (24 with dementia, 29 without).
- Quantification of ventricle-to-brain ratio (VBR), white matter lesion (WML) area, and infarction areas (cortical, subcortical, and regional).
- Discriminant analysis was used to assess the classification accuracy of identified factors.
Main Results:
- Total WML area, left WML, VBR, right WML, age, left cortical infarction area, left parietal infarction area, and total infarction area were strongly correlated with dementia.
- These factors achieved a classification accuracy of 96% for non-demented and 75% for demented patients.
- Both cortical and white matter infarction areas were significantly associated with dementia, with a stronger link to left-hemisphere lesions.
Conclusions:
- MRI-derived measures, particularly white matter lesion burden and ventricle enlargement, are significant predictors of dementia post-stroke.
- Infarction extent and location, especially in the left hemisphere, also contribute to dementia development.
- Age is the primary demographic factor associated with post-stroke dementia.