Perivascular spaces in MS patients at 7 Tesla MRI: a marker of neurodegeneration?

Virchow-Robin spaces (VRS) are associated with vascular and neurodegenerative disease. In multiple sclerosis (MS), VRS have been associated with neuroinflammation.

Previous studies that investigated VRS in MS have been performed at 1.5 Tesla (T) MRI. Because of the small size of VRS, visibility may be hampered at lower field strengths. Increased spatial resolution at 7T may result in improved detection of VRS.

In this study, the authors investigated VRS in MS patients using high-resolution 7T MRI. They compared their frequency and size to those of healthy controls. The aim was to analyze frequency and size of VRS in MS patients, at 7T. Additionally, the radiological link between VRS and inflammation was revisited: the authors aimed to relate VRS to neuroinflammatory (MS lesion counts) and to neurodegenerative aspects of the disease (brain atrophy).

The study included 34 MS patients and 11 age- and gender-matched healthy controls without vascular or neurological comorbidity. They consisted of 22 relapsing–remitting (RR), five secondary progressive (SP) and seven primary progressive (PP) MS patients. Clinical disability of the patients was measured using the Extended Disability Status Scale (EDSS). Imaging was performed on a whole body 7T MRI system.

VRS were manually delineated on the axially reformatted T1-weighted images by one of the authors under the supervision of an experienced neuroradiologist, using MIPAV software. Per subject, VRS were counted and the area and largest cross-section were measured for each VRS at five levels in the brain. These five levels correspond to regions where VRS are common, and were selected using predetermined anatomical landmarks that were easily reproducible across subjects. Total MS lesion counts (white matter (WM) and GM) were assessed on the 7T FLAIR images.

Both in the MS patients and the controls, VRS were detected at all five levels of the brain, although not at each level in every subject. The differences in presence of VRS between MS patients and healthy controls varied between the levels. The total number of VRS per subject was significantly larger in MS patients than in healthy controls. The size of VRS, as measured by the area and the largest cross-section did not differ between MS patients and controls.

These results confirmed the presence of VRS both in MS patients and healthy controls, using 7T MRI. Significantly more VRS were detected in MS patients than in healthy controls, with a median VRS count of 11 per MS patient and four per healthy control, across five commonly affected levels in the brain. This difference in VRS count was particularly true for VRS in superior areas of the brain (levels a and b in our rating system).

Although both groups displayed the highest number of VRS at the level of the anterior commissure, VRS count at this level did not differ between MS patients and healthy controls. In the MS group, a few very large VRS were found. However, no systematic difference could be detected in the size of the VRS between MS patients and healthy controls. To explore the relationship of VRS with other radiological hallmarks of MS, the authors correlated the number of VRS with measures of lesion load and brain atrophy. As expected, lower supratentorial brain volume (as measured by sBVF) was observed in MS patients compared to healthy controls. These results furthermore show that the number of VRS in MS patients is related to sBVF, age and disease duration, but not to lesion load.

Authors: Kilsdonk I, Steenwijk M
Source: Mult Scler. 2014 Jul 10. pii: 1352458514540358. [Epub ahead of print]
Read the abstract


Latest MS research news

Main MS research areas

Challenges of MS research

Atacicept in multiple sclerosis: a randomised, placebo-controlled, double-blind, phase 2 trial