Elsevier

NeuroImage

Volume 19, Issue 4, August 2003, Pages 1349-1360
NeuroImage

Regular article
Combined functional MRI and tractography to demonstrate the connectivity of the human primary motor cortex in vivo

https://doi.org/10.1016/S1053-8119(03)00165-4Get rights and content

Abstract

In this study, we combined advanced MR techniques to explore primary motor cortex (M1) connectivity in the human brain. We matched functional and anatomical information using motor functional MRI (fMRI) and white matter tractography inferred from diffusion tensor imaging (DTI). We performed coregistered DTI and motor task fMRI in 8 right-handed healthy subjects and in 1 right-handed patient presenting with a left precentral tumour. We used the fast-marching tractography (FMT) algorithm to define 3D connectivity maps within the whole brain, from seed points selected in the white matter adjacent to the location of the maximum of fMRI activation. Connectivity maps were then anatomically normalised and analysed using statistical parametric mapping software (SPM99) allowing group comparisons (left versus right hemisphere in control subjects and patient versus control subjects). The results demonstrated, in all control subjects, strong connections from M1 to the pyramidal tracts, premotor areas, parietal cortices, thalamus, and cerebellum. M1 connectivity was asymmetric, being more extensive in the dominant hemisphere. The patient had differences in M1 connectivity from the control group. Thus, fMRI-correlated DTI-FMT is a promising tool to study the structural basis of functional networks in the human brain in vivo.

Introduction

Functional anatomy and connectivity of motor cerebral cortex have been widely studied in nonhuman primates. It has been demonstrated that primary motor cortex (M1) is connected not only with the spinal cord but also with other cortical areas and subcortical structures Geyer et al 2000, Passingham 1997, Holsapple et al 1991, Matelli et al 1989, Orioli and Strick 1989, Strick 1985, Luppino et al 1993, Strick and Preston 1983, Barbas and Pandya 1987, Petrides and Pandya 1984. Numerous human studies, based on positron emission tomography (PET) Honda et al 1998, Grafton et al 1993, Fink et al 1997, functional MRI (fMRI) (Rao et al., 1997), transcranial magnetic stimulation (TMS) (Munchau et al., 2002), or scalp or depth electrode electroencephalography (EEG) Hallett and Toro 1996, Lim et al 1996, Chauvel et al 1996, have suggested various similar cortico-cortical and subcortical M1 functional connections in man. To date, however, no anatomical motor network has been demonstrated in man in vivo. The most reliable technique to study axonal pathways that make up these connections used invasive procedures which are only applicable in animals or in human postmortem brain (Rye, 1999). Furthermore the exact location and variability of long fibre tracts in human adult postmortem brain have only recently been extensively studied Burgel et al 1997, Rademacher et al 2001.

Diffusion tensor imaging (DTI) is a noninvasive technique using quantitative three-dimensional measurements of passive tissue water diffusion to infer orientation of white matter axonal fibres in vivo Le Bihan et al 2001, Melhem et al 2002. In the brain, motion of water molecules is constrained by the structure of axons and myelin sheaths Basser and Pierpaoli 1996, Pierpaoli et al 1996. In consequence, the directionality, or anisotropy, of the diffusion of water molecules depends on the orientation of brain fibre pathways. The dominant orientation and magnitude of diffusion within an imaging voxel can be quantified as a principal eigenvector and eigenvalue of diffusion obtained from the measured diffusion tensor. This information can be interrogated using colour-coding (Douek et al., 1991) or vector maps (Makris et al., 1997). The remaining information in the tensor—three eigenvectors are required to describe the anisotropy of diffusion—may be included using a display of the tensor ellipsoid (Pierpaoli and Basser, 1996). Visualisation and isolation of specific white matter pathways require an evaluation of connectivity between voxels, which can be inferred using a range of mathematical algorithms Mori et al 1999, Mori et al 2000, Conturo et al 1999, Jones et al 1999, Basser et al 2000, Poupon et al 2000, Parker et al 2002a, Parker et al 2002b. In this present study, we used fast marching tractography (FMT) (Parker et al., 2002a). This technique allows branching of tracts from a single seed point and provides 3D connectivity maps representing a metric (or informal probability) of connectivity between a seed point and each voxel in the brain. These connectivity maps can be displayed and analysed in an anatomically normalised framework using statistical parametric mapping (SPM99).

Our aim was to study in vivo, using this technique, normal and pathological human primary motor cortex connectivity. The anatomical correlates of functional areas can vary, even in primary cortices such as M1 (Rademacher et al., 2001), in particular in the presence of structural lesion. FMRI reliably localises motor functions Lotze et al 2000, Beisteiner et al 2001. Thus we selected FMT seed points from the location of the signal activation during a simple fMRI motor task.

Section snippets

Subjects

Eight right-handed healthy volunteers, 7 males, aged 29 to 46 years (mean 34), without any history of neurological disorders, were included in this study. One right-handed patient, aged 30 years, presenting with a left frontal lesion, responsible for frontal lobe epilepsy, was also included.

MR data acquisition

All studies were performed on a 1.5 T General Electric Signa Horizon imager. Standard imaging gradients with a maximum strength of 22 mT m−1 and slew rate 120 Tm−1 s−1 were used. All data were acquired using

Controls

The unthresholded averaged connectivity maps showed broad connections from the activation maxima in the hand area in all subjects (Fig. 3, Fig. 4). Evaluation based on the connectivity values of tracts (values of voxels involved, colour-coded in figures) permitted the identification of common features in the left and right hemispheres in the control group.

Different groups of tracts could be described according to the range of their connectivity values, suggesting a hierarchy in the

Validity and limitations of the FMT method

In this study, we used the FMT method, to demonstrate for the first time, the connectivity of the primary motor cortex in man. Validity and reliability of FMT as a tractography method has been assessed in a previous study demonstrating accurate connectivity maps of corticospinal tracts and optic radiations in macaque and human brain (Parker et al., 2002b). The two advantages of FMT over other tractography algorithms are its capability to estimate a metric of connections which rank the different

Conclusion

The tractography methods, despite technical limitations at this stage in comparison with invasive anatomical techniques, have the advantage of being feasible in vivo. This study opens several perspectives for further investigations. The combination of fMRI activation tasks and DTI-FMT may be helpful in defining the structural basis of functional connectivity in the normal brain, and the derangement of these networks in disease states.

Acknowledgements

This work was funded by National Society for Epilepsy. M. Guye was supported by the French League Against Epilepsy (Pfizer grant).

References (64)

  • C. Pierpaoli et al.

    Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture

    NeuroImage

    (2001)
  • C. Poupon et al.

    Regularization of diffusion-based direction maps for the tracking of brain white matter fascicles

    NeuroImage

    (2000)
  • G. Rizzolatti et al.

    The organization of the cortical motor systemnew concepts. Electroencephalogr

    Clin. Neurophysiol.

    (1998)
  • D.B. Rye

    Tracking neural pathways with MRI

    Trends Neurosci.

    (1999)
  • J.N. Sanes

    Motor cortex rules for learning and memory

    Curr. Biol.

    (2000)
  • B. Stieltjes et al.

    Diffusion tensor imaging and axonal tracking in the human brainstem

    NeuroImage

    (2001)
  • P.L. Strick

    How do the basal ganglia and cerebellum gain access to the cortical motor areas?

    Behav. Brain Res.

    (1985)
  • P.L. Strick et al.

    Input to primate motor cortex from posterior parietal cortex (area 5). I. Demonstration by retrograde transport

    Brain Res.

    (1978)
  • I. Toni et al.

    Movement preparation and motor intention

    NeuroImage

    (2001)
  • D.J. Werring et al.

    A direct demonstration of both structure and function in the visual systemcombining diffusion tensor imaging with functional magnetic resonance imaging

    NeuroImage

    (1999)
  • H. Barbas et al.

    Architecture and frontal cortical connections of the premotor cortex (area 6) in the rhesus monkey 2

    J. Comp. Neurol.

    (1987)
  • P.J. Basser et al.

    In vivo fiber tractography using DT-MRI data

    Magn. Reson. Med.

    (2000)
  • U. Burgel et al.

    Histological visualization of long fiber tracts in the white matter of adult human brains

    J. Hirnforsch.

    (1997)
  • Ciccarelli, O., Parker, G.J.M., Toosy A.T., Wheeler-Kingshott, C.A.M., Barker, G.J., Boulby, P.A., Miller D.H.,...
  • P.Y. Chauvel et al.

    What stimulation of the supplementary motor area in humans tells about its functional organization

    Adv. Neurol.

    (1996)
  • T.E. Conturo et al.

    Tracking neuronal fiber pathways in the living human brain

    Proc. Natl. Acad. Sci. USA

    (1999)
  • P. Douek et al.

    MR color mapping of myelin fiber orientation

    J. Comput. Assist. Tomogr.

    (1991)
  • G.R. Fink et al.

    Multiple nonprimary motor areas in the human cortex

    J. Neurophysiol.

    (1997)
  • K.J. Friston et al.

    Statistical parametric maps in functional imaginga general linear approach

    Hum. Brain Mapp.

    (1995)
  • S. Geyer et al.

    Functional neuroanatomy of the primate isocortical motor system

    Anat. Embryol. (Berl)

    (2000)
  • S.T. Grafton et al.

    Within-arm somatotopy in human motor areas determined by positron emission tomography imaging of cerebral blood flow 4

    Exp. Brain. Res.

    (1993)
  • M. Hallett et al.

    Generation of movement-related potentials in the supplementary sensorimotor area

    Adv. Neurol.

    (1996)
  • Cited by (0)

    1

    Current address: Laboratoire de Neurophysiologie et Neuropsychologie, Faculté de Médecine, Université de la Méditerranée, Marseille, France.

    View full text