ABSTRACT
Subtle blood-brain barrier (BBB) disruption is involved in numerous neurological conditions. This disruption is found diffusely in the brain and requires quantitative methods for assessment. We propose a statistical method to identify individual voxels where the BBB is disrupted using T1-weighted MRI. We used models of severe and focal vs. mild and generalized disruption of the BBB to show proof of principle with the cold injury model, hypoxia, and a model of inflammation using low- and high-dose lipopolysaccharide (LPS) treatment. Using voxel-based analysis, we found that mild hypoxia resulted in diffuse disruption of the BBB, whereas more severe hypoxia and high-dose LPS treatment resulted in prominent leakage, particularly in the periventricular area, suggestive of blood-cerebrospinal fluid (CSF) barrier disruption. Our data suggest that the periventricular area may be compromised first in conditions of inflammation and hypoxia. Voxel-based analysis could be used in future studies assessing subtle blood-CSF or BBB disruption.