MRI-Based Small Vessel Disease Score Predicts Risk for Dementia

A simple SVD score using MRI scans, may aid in identifying patients with SVD who are likely to develop cognitive decline.

A simple small vessel disease (SVD) score based on routine clinical magnetic resonance imaging (MRI) scans, may aid in identifying patients with SVD who are likely to develop cognitive decline and dementia, according to study results published in Neurology.

Previous studies have reported that cerebral SVD features on MRI are associated with increased risk for dementia and the goal of the current study was to assess if a simple MRI score can predict cognitive decline and dementia in patients with SVD.

The score was investigated via 3 prospective longitudinal cohorts that included patients with SVD ranging from mild and asymptomatic to severe and symptomatic: St. George’s Cognition and Neuroimaging in Stroke (SCANS, n=121), Radboud University Nijmegen Diffusion Imaging and Magnetic Resonance Imaging Cohort (RUN DMC, n=503), and the Austrian Stroke Prevention Study (ASPS, n=1218. In all studies, MRI was performed at baseline, cognitive tests repeated during follow-up, and progression to dementia recorded prospectively.

The addition of the simple SVD score to a model that included age, sex, and education years improved the prediction of dementia as area under the curve [AUC] from 0.76 (95% CI, 0.72-0.80) to 0.81 (95% CI, 0.77-0.85; P =.098) and to 0.83 (95% CI, 0.80-0.87) with an amended score in which there were a number of categories for both the number of lacunar infarcts and the severity of white matter hyperintensities (P =.011). Using the amended score the prediction of vascular dementia was stronger, compared with that from clinical risk factors alone (AUC, 0.85; 95% CI, 0.81-0.89 vs AUC 0.76; 95% CI, 0.71- 0.81, respectively).

SCANS had participants with the most severe cases of SVD; ASPS participants had the least severe cases of SVD. Further, because the amended SVD score had a higher predictive performance in patients with more severe SVD, the score improved prediction in SCANS slightly more than in the other cohorts. In contrast, ASPS and RUN DMC, with patients with milder SVD, had little prediction improvement.

Related Articles

The researchers also aimed to determine whether using the SVD score to select a group of patients with more severe disease would reduce sample sizes for clinical intervention trials. Results found that with dementia onset as the outcome measure, preselection of patients with higher SVD score reduced sample size by 57%.

The researchers noted that the study was limited due to pooling data from a number of different populations with different follow-up periods and methods of assessing dementia incidence.

“A simple SVD score, easily obtainable from clinical MRI scans and therefore applicable in routine clinical practice, aided prediction of future dementia risk”, concluded the researchers.


Amin Al Olama A, Wason JMS, Tuladhar AM, et al. Simple MRI score aids prediction of dementia in cerebral small vessel disease [published online ahead of print, 2020 Mar 2]. Neurology. doi:10.1212/WNL.0000000000009141