Non-invasive mapping of corticofugal fibres from multiple motor areas--relevance to stroke recovery

Abstract

Recovery of motor function after subcortical stroke appears to be related to the integrity of descending connections from the ipsilesional cortical motor system, a view supported by the observation of greater than normal movement-related activation in ipsilesional motor regions in chronic subcortical stroke patients. This suggests that damage to the descending output fibres from one region of the cortical motor system may be compensated by activity in areas that retain corticofugal outputs. Though the trajectories of corticofugal fibres from each major component of the motor system through the corona radiata and internal capsule are well described in non-human primates, they have not been described fully in humans. Our study set out to map the trajectories of these connections in a group of healthy volunteers (8 male, 4 female; age range = 31-68 years, median = 48.5 years) and establish whether this knowledge can be used to assess stroke-induced disconnection of the cortical motor system and better interpret functional reorganization of the cortical motor system. We describe the trajectories of the connections from each major component of the motor system to the cerebral peduncle using diffusion-weighted imaging and probabilistic tractography in normal subjects. We observed good reproducibility of these connections over subjects. The comparative topography of these connections revealed many similarities between humans and other primates. We then inferred damage to corticofugal pathways in stroke patients (n = 3) by comparing the overlap between regions of subcortical white matter damage with the trajectories of the connections to each motor area. In a small series of case studies, we found that inferred disconnections could explain enhanced hand-grip-related responses, as assessed with functional MRI, in the ipsilesional motor system. These results confirm that selective disruption of motor corticofugal fibres influences functional reorganization and outcome in individual patients.

Fig. 1

Definition of seed points and cortical motor areas for tractography. (A) Seed points for tractography shown on the cross-section of the right cerebral peduncle on the z = −16 mm plane of the normalized FA map for an individual. (B) Cortical motor areas of the right hemisphere, as defined on the same individual, shown on the individual’s volume-rendered T₁-weighted image.

Fig. 2

Trajectory variability maps for the peduncle/M1 connections shown on representative sections of the average of the T₁-weighted structural. Colour bar shows number of subjects with connections in each voxel: range = 2–12. The top panel shows sagittal slices x = −24 and x = +24 either side of coronal slice y = −18. The lower two panels show transverse slices. The right side of the brain is depicted on the right side of the transverse and coronal images.

Fig. 3

Trajectory variability maps for the peduncle/PMd connections shown on representative sections of the average of the T₁-weighted structural. Colour bar shows number of subjects with connections in each voxel: range = 2–12. The top panel shows sagittal slices x = −24 and x = +24 either side of coronal slice y = −16. The lower two panels show transverse slices. The right side of the brain is depicted on the right side of the transverse and coronal images.

Fig. 4

Trajectory variability maps for the peduncle/PMv connections shown on representative sections of the average of the T₁-weighted structural. Colour bar shows number of subjects with connections in each voxel: range = 2–12. The top panel shows sagittal slices x = −24 and x = +24 either side of coronal slice y = −12. The lower two panels show transverse slices. The right side of the brain is depicted on the right side of the transverse and coronal images.

Fig. 5

Trajectory variability maps for the peduncle/SMA connections shown on representative sections of the average of the T₁-weighted structural. Colour bar shows number of subjects with connections in each voxel: range = 2–12. The top panel shows sagittal slices x = −24 and x = +24 either side of coronal slice y = −12. The lower two panels show transverse slices. The right side of the brain is depicted on the right side of the transverse and coronal images.

Fig. 6

Sections showing the overlap between the thresholded trajectory variability maps (eight or more subjects) for M1, PMd and SMA on axial and sagittal sections. Coordinates for x and z correspond to MNI space.

Fig. 7

Sections showing the regions of reduced anisotropy in each patient (black outline) overlaid on the maps of the overlap between the thresholded trajectory variability maps (eight or more subjects) for M1, PMd and SMA shown on the T₁-weighted image for the respective patient. Colour key as for Fig. 6.

Fig. 8

fMRI results for Patients A and B shown on individual patient’s volume-rendered T₁-weighted images. (A) Regions of significantly greater left (affected) hand-grip activation for Patient A compared with the control group. Cut-out reveals cluster in right PMd in the depth of the precentral sulcus (x = 33, y = −3, z = +42). (B) Region of significantly greater right (affected) hand-grip activation for Patient B compared with the control group. Cut-out reveals cluster in left precentral gyrus (x = −48, y = −3, z = +51). CS = central sulcus, PCS = precentral sulcus, HK = hand knob region of M1.