Toward accurate diagnosis of white matter pathology using diffusion tensor imaging

Abstract

Diffusion tensor imaging (DTI) has been widely applied to investigate injuries in the central nervous system (CNS) white matter (WM). However, the underlying pathological correlates of diffusion changes have not been adequately determined. In this study the coregistration of histological sections to MR images and a pixel-based receiver operating characteristic (ROC) analysis were used to compare the axial (lambda( parallel)) and radial (lambda( perpendicular)) diffusivities derived from DTI and histological markers of axon (phosphorylated neurofilament, SMI-31) and myelin (Luxol fast blue (LFB)) integrity, respectively, in two different patterns of injury to mouse spinal cord (SC) WM. In contusion SC injury (SCI), a decrease in lambda( parallel) matched the pattern of axonal damage with high accuracy, but lambda( perpendicular) did not match the pattern of demyelination detected by LFB. In a mouse model of multiple sclerosis (MS), lambda( perpendicular) and lambda( parallel) did not match the patterns of demyelination or axonal damage, respectively. However, a region of interest (ROI) analysis suggested that lambda( perpendicular)-detected demyelination paralleled that observed with LFB, and lambda( parallel) decreased in both regions of axonal damage and normal-appearing WM (NAWM) as visualized by SMI-31. The results suggest that directional diffusivities may reveal abnormalities that are not obvious with SMI-31 and LFB staining, depending on the type of injury.