Dynamic corticospinal white matter connectivity changes during stroke recovery: a diffusion tensor probabilistic tractography study

J Magn Reson Imaging. 2009 Mar;29(3):529-36. doi: 10.1002/jmri.21627.

Abstract

Purpose: To investigate corticospinal tract connectivity changes at the cortical surface using diffusion tensor imaging (DTI) tractography during recovery from stroke.

Materials and methods: Using data from 10 stroke patients (four subcortical) and six elderly controls, we developed an automated method to quantify altered motor connectivity that involves the use of a simplified cortical surface model as a seed mask with target regions defined within the corticospinal tracts to initiate a probabilistic tractography algorithm.

Results: We found no change in volume overlap of the generated corticospinal tracts in the stroke patients compared to controls, but significant connectivity changes at the boundary of the simplified cortical surface mask, especially within the ipsilesional hemisphere of stroke patients over time. Using the cortical regions with significantly enhanced connectivity as a seed mask on the patient data, tracts that are directly associated with stroke recovery can be delineated. Measures of uncertainty in fiber orientation within these fiber tracts significantly correlated with functional outcome.

Conclusion: The novel findings from this study highlight the usefulness of this methodology to study white matter repair/reorganization during stroke recovery.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Algorithms*
  • Brain Mapping / methods
  • Cerebral Cortex / pathology*
  • Diffusion Magnetic Resonance Imaging / methods*
  • Feasibility Studies
  • Female
  • Humans
  • Image Processing, Computer-Assisted / methods
  • Imaging, Three-Dimensional / methods
  • Male
  • Middle Aged
  • Models, Statistical
  • Monte Carlo Method
  • Neural Conduction*
  • Pyramidal Tracts / pathology
  • Stroke / pathology*
  • Stroke / physiopathology*