Department of Exercise Sciences

Functional potential in chronic stroke patients depends on corticospinal tract integrity

Cathy M. Stinear, P. Alan Barber, Peter R. Smale, James P. Coxon, Melanie K. Fleming and Winston D. Byblow.
Published in Brain (2007) 130, 170-80


To date, no studies have combined Transcranial Magnetic Simulation (TMS) and Magnetic Resonance Imaging (MRI) variables in the evaluation of current upper limb function or potential for functional improvement in chronic stroke patients.


To determine if measures from combined neurophysiological and neuroimaging techniques would allow prediction of an individual’s capacity for functional improvement of upper limb function following stroke.


  • Patients: 21 hemiparetic stroke patients. See Table 1.
  • Single-pulse TMS mapping of wrist extensor muscle was performed using an 18 point grid and map density (MD) and MD asymmetry was calculated between affected and unaffected sides. Absence or presence of affected side motor evoked potentials(MEPs) were also noted.
  • Blood oxygen level-dependent contrast images were acquired to examine lateralization of cortical activity during affected hand use. Ipsilesional activation during affected hand use has been associated with better outcomes. Lateralization of activation was determined across a range of pre-determined cortical areas.
  • Diffusion tensor imaging (DTI) was conducted with a single shot spin echo EPI pulse sequence with 12 uniformly distributed motion-probing gradient orientations. Fractional Anisotropy (FA) was determined within the posterior limb of the internal capsule as an index of corticospinal integrity, and an FA asymmetry index determined the extent of damage due to stroke relative to the intact side.
  • Primary outcome measures were Fugl-Meyer Assessment of Upper limb Motor Function (maximum score 32). We sought to determine which variables could predict current upper limb function and functional potential (change in FM score) for patients who completed 30 days of intensive motor practice involving the affected upper limb.
  • Factors of age, hemisphere affected, time since stroke, clinical stroke at inception, hand grip asymmetry, MD asymmetry, FA asymmetry, presence or absence of motor cortex damage and MEPs in the affected target muscle were included in the stepwise multiple regressions.


  • Presence or absence of MEPs and FA asymmetry predicted FM score at inception (Fig 2). For patients without MEPs, lower FA asymmetry and lateralization of cortical activity toward the ipsilesional M1 (BA 4p) were associated with higher FM scores.
  • The presence or absence of MEPs, FA Asymmetry and time since stroke were predictors of functional potential of these chronic patients. Variables such as current clinical score and lateralization of cortical activity did not predict functional potential. (Fig 1).


In patients with some corticospinal integrity, as assessed by TMS (MEPs), FA, or both, functional improvements in upper limb are possible up to three years post-stroke. This is considerably longer than previously reported. These findings suggest a way of predicting outcomes and targets for neuromodulation based on an individual’s residual corticospinal integrity in the stroke affected hemisphere (Fig 4).



We would like to thank Toby Verryt, Patricia Bennett, Linda Ross, Yvette Baker, Dr Samir Anwar, Dr Timothy Verstynen, Dr Nicole Wenderoth, the Centre for Advanced MRI, and the Stroke Foundation of New Zealand. This research was funded by the Health Research Council of New Zealand.