Spatial Distribution of Perivascular Spaces on MRI Is Associated with Disease Severity in Idiopathic Intracranial Hypertension.

Background

AND

Purpose

Glymphatic dysfunction, affecting cerebrospinal fluid transport and clearance, is increasingly implicated in idiopathic intracranial hypertension (IIH). Because perivascular spaces (PVS) reflect glymphatic activity, we evaluated whether their volume and spatial distribution could serve as imaging markers of IIH severity.

Materials And Methods

In this case-control study, 47 patients with IIH (mean age at MRI 32.5 ± 9.7 years; 44 women) and 47 epilepsy controls underwent non-contrast MRI with 3D FLAIR. PVS were manually segmented, volumetrically quantified, and normalized to standard space. Group-level frequency maps identified common PVS locations. Outcome-weighted maps (continuous and binary) assessed associations with ophthalmological severity, defined by visual field mean deviation (VFMD), peripapillary retinal nerve fiber layer thickness (pRNFL), and visual loss (best-corrected visual acuity <20/20). Odds ratio maps compared spatial distributions between groups. All maps were tested by permutation (one-tailed, p < 0.05). PVS volumes were compared using the Mann-Whitney U test, and multivariable linear regression assessed clinical associations.

Results

Whole-brain PVS volume was higher in IIH than in controls (median 146.9 mm³ [IQR 95.4-206.2] vs. 105.3 mm³ [IQR 79.5-149.1]; p = 0.015), but was not associated with ophthalmological severity in adjusted models (all p > 0.05). On voxel-wise mapping, patients with more severe disease showed PVS preferentially in the amygdala, posterior internal capsules, and superior basal ganglia, with voxels significant on permutation testing within these regions for all three severity measures (p < 0.05). These regions were at least three times more likely to show PVS in IIH than in controls.

Conclusions

While total PVS volume is increased in IIH, it was not associated with disease severity. Descriptive mapping suggests a distinct regional distribution of PVS in more severe disease particularly in the amygdala, posterior internal capsules, and basal ganglia. Validation in larger cohorts is required before this distribution can be adopted as a severity marker.