Distinct spatiotemporal patterns of white matter hyperintensity progression.

Area of Science:

• Neuroimaging
• Neurology
• Cardiovascular Disease

Background:

• White matter hyperintensity (WMH) is a crucial neuroimaging biomarker reflecting brain health and predicting stroke outcomes.
• Understanding WMH progression patterns is essential for refining stroke risk stratification and management strategies.

Purpose of the Study:

• To identify and characterize distinct subtypes of white matter hyperintensity progression using advanced modeling.
• To investigate the associations between WMH subtypes, demographic factors, vascular risks, and stroke outcomes.
• To evaluate the utility of WMH subtyping for improving post-stroke outcome prediction and potentially for stroke prediction.

Main Methods:

• Utilized a large, multicenter MRI dataset comprising 9179 stroke patients and 36,210 controls from the UK Biobank.
• Employed Subtype and Stage Inference (SuStaIn) modeling to classify WMH progression into distinct subtypes.
• Performed longitudinal validation to confirm the stability and reliability of the identified WMH subtypes.

Main Results:

• Identified three distinct WMH progression subtypes: fronto-parietal, radial, and temporo-occipital.
• Fronto-parietal subtype associated with delayed onset, hypertension, small vessel stroke, and higher ischemic stroke recurrence.
• Temporo-occipital subtype linked to atrial fibrillation, coronary heart disease, cardioembolism, early neurological deterioration, and worse outcomes.

Conclusions:

• The developed subtyping-staging model effectively captures WMH progression, associated demographics, and vascular risks.
• WMH subtyping significantly enhances the prediction of post-stroke outcomes, including recurrence and early deterioration.
• This novel subtyping approach holds promise for improving both post-stroke outcome prediction and future stroke risk assessment.