Redefinition of multiple sclerosis plaque size using diffusion tensor MRI.

Published

Journal Article

OBJECTIVE: We used diffusion tensor MRI to redefine the size of multiple sclerosis (MS) plaques on fractional anisotropy (FA) maps. MATERIALS AND METHODS: Thirty-six white matter (WM) plaques were identified in 20 patients with MS. Plaque FA was measured by placing regions of interest (ROIs) on plaques on diffusion tensor images. We compared FA values in identical mirror-image ROIs placed on normal-appearing WM in the contralateral hemisphere. This comparison showed a mean decrease in FA of 41% in plaques, serving as the threshold for outlining abnormal regions in normal-appearing WM surrounding plaques. ROIs were placed around each plaque and FA values were compared with those in the mirror-image ROIs. Combined areas of perilesional normal-appearing WM with 40% or more FA reduction plus plaque were compared with the areas of abnormality on T2-weighted images using a paired Student's t test. A p value of 0.05 or less was considered significant. RESULTS: Mean plaque area was 60 mm(2) (range, 15-103 mm(2)), mean plaque FA was 0.251 (range, 0.133-0.436), and mean FA of contralateral normal-appearing WM was 0.429 (range, 0.204-0.712). Applying a threshold of 40% FA reduction, mean combined area of abnormal WM (including plaque seen on T2-weighted sequences) was 87 mm(2) (range, 30-251 mm(2)) or 145% of the mean plaque area that was seen on T2-weighted images (p < 0.001). CONCLUSION: Using an operator-defined threshold of abnormal FA values based on plaque anisotropy characteristics, we saw a statistically significant increase in plaque size.

Full Text

Duke Authors

Cited Authors

  • Kealey, SM; Kim, Y; Provenzale, JM

Published Date

  • August 2004

Published In

Volume / Issue

  • 183 / 2

Start / End Page

  • 497 - 503

PubMed ID

  • 15269047

Pubmed Central ID

  • 15269047

International Standard Serial Number (ISSN)

  • 0361-803X

Digital Object Identifier (DOI)

  • 10.2214/ajr.183.2.1830497

Language

  • eng

Conference Location

  • United States