Kristina Shkirkova, BSc

Wyss A, Dawson J, Arba F, Wardlaw JM, Dickie DA, on behalf of the VISTA-Prevention Collaborators. “Combining Neurovascular and Neurodegenerative Magnetic Resonance Imaging Measures in Stroke.” Stroke. 2019; 50:1136-1139.

To characterize age and stroke-related tissue damage, the total small vessel disease score and the brain health index have recently been developed for clinical use. The total small vessel disease score combines presence of lacunes, microbleeds, and moderate to severe white matter hyperintensities (WMH) by visual scoring based on clinical imaging. The brain health index uses automatic processing of MRI scans to quantify visible injury from small vessel disease and brain atrophy. However, the total small vessel disease score is prone to granularity and measurement limitations, whereas the brain health index requires high-resolution T1, T2, T2 gradient echo, and fluid attenuated inversion recovery scans, which are not often available in routine clinical imaging.

The study by Wyss et al. argues that individual markers of cerebral small vessel disease and brain atrophy have limited potential to explain high proportion of variance in neurovascular and neurodegenerative disease. The authors propose to combine markers of white matter hyperintensity and cerebral atrophy, represented by cerebrospinal fluid (CSF) volume, into a single measure capable of more accurate predictions of cognitive impairment.

To test the new measure, authors used the Virtual International Stroke Trials Archive to retrospectively extract data for 317 patients with a diagnosis of ischemic stroke or transient ischemic attack. WMH and CSF volumes were assessed on T1 and fluid attenuated inversion recovery MRI images at baseline and 2-year follow-up scans. Additionally, 6-month and 2-year Mini Mental State Examination (MMSE) scores as well as reported vascular risk factors were extracted. A linear regression analysis was performed to establish a relationship between brain volumes, MMSE scores, and hypertension, hypercholesterolemia, diabetes mellitus, smoking, atrial fibrillation and measured systolic and diastolic blood pressure.

The results of this study show that the combined measure of WMH and CSF volumes increases variance accounted for by vascular risk factors and improves prediction of cognitive impairment poststroke. On a linear regression analysis, the association between 2-year follow-up MMSE score, adjusted by 6 months MMSE score, with combined measure of WMH and CSF was 31% greater than with WMH alone and 11% greater than with CSF alone. Furthermore, the combined WMH and CSF measure reduced sample size needed to detect potential treatment effects by 20% compared to WMH measure alone. However, authors note that the final sample size estimation will depend on the required effect size. Although the combined measure reduces sample size, to show a moderate treatment effect of 20% in a clinical trial 3,000 patients per arm are required to detect the effect. The authors conclude that the use of combined white matter hyperintensity and brain atrophy measure may increase predictive accuracy of cognitive impairment and further reduce required sample size in early phase clinical studies.