Asymmetry of Subinsular Anisotropy by In Vivo Diffusion Tensor Imaging

 

Y. Cao, J. Huang, K. L. Berger, M. DeLano

Department of Radiology, Michigan State University, East Lansing, MI, USA

 

Abstract

The cortical regions specialized in speech-language exhibit a left-right asymmetry, e.g., a larger cortical size in the left auditory cortex and Wernicke’s area.   The possibility of developmental asymmetry in axonal fibers interconnecting speech-language cortical areas can be investigated by in vivo diffusion tensor imaging.   Seven right-handed native English speakers showed a markedly significant asymmetry (p<0.0005) in the relative anisotropy of diffusion in the subinsular white matter, greater on the left.  These results suggest a possible association between the hemispheric functional specialization and the axonal fiber development.

 

Introduction

A cortical asymmetry has been described in human auditory cortex (the planum temporal) and Wernicke’s area (1,2).  The asymmetry of the Wernicke’s area mainly concerns the size of the cortex, usually larger in the left side that specializes in speech and language, implying an association of the cortical asymmetry with the hemispheric functional lateralization.  Similarly, axonal fiber bundles that interconnect the cortical areas might be developed asymmetrically, either in size, degree of myelination, and/or organization.  Applying in vivo diffusion tensor imaging (3-5),  we investigated anisotropy in the external/extreme capsule beneath the insular cortex, which may provide connectivity between the inferior frontal cortex and the temporal/parietal cortex.   

 

Methods and Materials

Seven normal right-handed native English speakers (1 female, 6 male, ages 23 to 70 years) participated in the study.  The handedness of the subjects was assessed by the Edinburgh Handedness Inventory (6).  Axial diffusion tensor images of a whole head were acquired on a GE 1.5 T clinical scanner using a spin echo Echo-Planar-Imaging pulse sequence (TE/TR = 69.3/10000 ms, in-plane resolution 2.3 mm, slice thickness 3 mm, gap 0 mm, 46 sections, and NEX 6).  The diffusion encoding was accomplished in six non-colinear directions with b value 1000 s/mm2. 

Data Analysis

Images were assessed and corrected for possible in-plane translation and rotation.  After six diffusion tensor elements were calculated by singular value decomposition (4), the trace of the diffusion tensor (Dtr) and the relative anisotropy (RA) were computed (7).  The analysis in the region of interest (ROI) was applied to the RA, Dtr, and T2-weighted (T2W) images.  The ROI was chosen in the space that is lateral to putamen and medial to insular cortex corresponding to the external capsule, extreme capsule and claustrum, which are not resolved.  The ROIs were selected from successive voxels having the largest RA on the left or right (see Figure).  

 

Results

The relative anisotropy (RA) of diffusion was markedly asymmetric between the left and right external/extreme capsule.  The average difference of the RA between the left and right ROIs was 0.155 + 0.014 (p<0.0005), see Table.  This asymmetry was also demonstrated in T2-weighted signal intensities when the analysis was done in the same ROIs, (Table).  The average difference of the T2W signal intensities was –36.1 + 5.9 (p<0.001).  However, the trace of the diffusion tensor did not show any left and right asymmetry (p=0.12).  All subjects were strongly right-handed, the handedness index (6) varied between 100% and 82%.  There was no significant correlation between age and any asymmetry of the measures.

 
Discussion

The relative contribution to the anisotropic measurements of the three structures of interest– external capsule, extreme capsule and claustrum--cannot be resolved given the spatial resolution of diffusion tensor imaging.  Thus, the MRI measurements in the interested space may arise from any or all these structures.   The relative size differences of the subinsular white matter may contribute to partial volume effects, which could explain the T2W signal intensity asymmetry.  The marked asymmetry of the RA suggests that the left and right external/extreme capsule may develop differently in size and/or organization, associated with speech-language hemispheric lateralization and right-handedness. 

 

 

Figure illustrates the relative anisotropic and T2W images of a right-handed subject in the respective left and middle panels.  Voxels in yellow and dark pink showed the selected ROIs.  The right panel shows an overlay of the RA on the T2W image and greater anisotropy within the left subinsular white matter. 

 

Table: The relative anisotropy and T2W signal intensity

 

RA(L)

RA(R)

Diff.

T2WI(L)

T2WI(R)

Diff.

1

0.49

0.30

0.19

721

787

-66

2

0.42

0.26

0.16

568

590

-22

3

0.51

0.33

0.18

650

680

-30

4

0.41

0.26

0.15

603

645

-41

5

0.34

0.23

0.11

679

698

-19

6

0.45

0.29

0.16

582

621

-39

7

0.37

0.23

0.14

754

791

-36

mean

 

 

0.155

 

 

-36.1

sem

 

 

0.014

 

 

5.87

p

 

 

0.0005

 

 

0.001

 

 

References

1.        Geschwind N, et al. Science 1968, 161:186-187.

  1. Galaburda, AM, et al. Arch Neurol 1978, 35:812-817.
  2. Le Bihan D, et al, Radiology 1986, 161:401-407.
  3. Basser PJ, et al,  J Magn Reson B 1994, 103:247-254.
  4. Nakada T, et al, Neurosci Res 1995, 22:389-398.
  5. Olfield RC, Neuropsychologia 1971;9:97-113.
  6. Ulug Am, et al, J Magn Reson Imag 1999, 9:804-813.