Tissue signature characterisation of diffusion tensor abnormalities in cerebral gliomas

Eur Radiol. 2004 Oct;14(10):1909-17. doi: 10.1007/s00330-004-2381-6. Epub 2004 Jun 25.


The inherent invasiveness of malignant cells is a major determinant of the poor prognosis of cerebral gliomas. Diffusion tensor MRI (DTI) can identify white matter abnormalities in gliomas that are not seen on conventional imaging. By breaking down DTI into its isotropic (p) and anisotropic (q) components, we can determine tissue diffusion "signatures". In this study we have characterised these abnormalities in peritumoural white matter tracts. Thirty-five patients with cerebral gliomas and seven normal volunteers were imaged with DTI and T2-weighted sequences at 3 T. Displaced, infiltrated and disrupted white matter tracts were identified using fractional anisotropy (FA) maps and directionally encoded colour maps and characterised using tissue signatures. The diffusion tissue signatures were normal in ROIs where the white matter was displaced. Infiltrated white matter was characterised by an increase in the isotropic component of the tensor (p) and a less marked reduction of the anisotropic component (q). In disrupted white matter tracts, there was a marked reduction in q and increase in p. The direction of water diffusion was grossly abnormal in these cases. Diffusion tissue signatures may be a useful method of assessing occult white matter infiltration.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Algorithms
  • Body Water / metabolism
  • Brain / metabolism
  • Brain / pathology
  • Brain Neoplasms / diagnosis*
  • Brain Neoplasms / pathology
  • Corpus Callosum / pathology
  • Diffusion Magnetic Resonance Imaging / methods*
  • Echo-Planar Imaging / methods
  • Female
  • Frontal Lobe / pathology
  • Glioblastoma / diagnosis
  • Glioblastoma / pathology
  • Glioma / diagnosis*
  • Glioma / pathology
  • Humans
  • Image Enhancement / methods*
  • Image Processing, Computer-Assisted / methods*
  • Male
  • Middle Aged
  • Neoplasm Invasiveness
  • Occipital Lobe / pathology
  • Prospective Studies