Elsevier

NeuroImage

Volume 30, Issue 4, 1 May 2006, Pages 1100-1111
NeuroImage

Characterization of displaced white matter by brain tumors using combined DTI and fMRI

https://doi.org/10.1016/j.neuroimage.2005.11.015Get rights and content

Abstract

In vivo white matter tractography by diffusion tensor imaging (DTI) has become a popular tool for investigation of white matter architecture in the normal brain. Despite some unresolved issues regarding the accuracy of DTI, recent studies applied DTI for delineating white matter organization in the vicinity of brain lesions and especially brain tumors. Apart from the intrinsic limitations of DTI, the tracking of fibers in the vicinity or within lesions is further complicated due to changes in diseased tissue such as elevated water content (edema), tissue compression and degeneration. These changes deform the architecture of the white matter and in some cases prevent definite selection of the seed region of interest (ROI) from which fiber tracking begins. We show here that for displaced fiber systems, the use of anatomical approach for seed ROI selection yields insufficient results. Alternatively, we propose to select the seed points based on functional MRI activations which constrain the subjective seed ROI selection. The results are demonstrated on two major fiber systems: the pyramidal tract and the superior longitudinal fasciculus that connect critical motor and language areas, respectively. The fMRI based seed ROI selection approach enabled a more comprehensive mapping of these fiber systems. Furthermore, this procedure enabled the characterization of displaced white matter using the eigenvalue decomposition of DTI. We show that along the compressed fiber system, the diffusivity parallel to the fiber increases, while that perpendicular to the fibers decreases, leading to an overall increase in the fractional anisotropy index reflecting the compression of the fiber bundle. We conclude that definition of the functional network of a subject with deformed white matter should be done carefully. With fMRI, one can more accurately define the seed ROI for DTI based tractography and to provide a more comprehensive, functionally related, white matter mapping, a very important tool used in pre-surgical mapping.

Introduction

The white matter possesses about 50% of the adult brain volume (Filley, 2001) and consists of a complex array of neuronal fiber networks. This network can be divided into specific tracts built of thousands of aligned neuronal fibers (axons) connecting different brain regions and target cells. Within the tracts, each fiber is usually coated with a multiple cell membrane layers (myelin) providing efficient and fast electrical transmission. Many brain diseases affect the white matter fibers either by disruption, degeneration or deviation of the fibers. Disruption of fiber continuity or disintegration of the myelin membrane surrounding the axons (demyelination) as what happens in multiple sclerosis is the most common form of white matter pathology. One of the less studied pathologies of white matter is displacement of neuronal fibers. Pressure applied on the white matter may lead to significant loss of neuronal transmission, demyelination and axonal loss leading to disability (Bergstrom et al., 1986, O'Brien et al., 1987, Siegal et al., 1987). Indeed, permanent damage to white matter integrity is known to occur in cases of severe and chronic exposure to mechanical intra-cranial pressure as happens in hydrocephalus (Del Bigio et al., 2003, Ding et al., 2001, Hanlo et al., 1997).

Diffusion tensor imaging (DTI) is an MRI based methodology that maps white matter (Basser and Pierpaoli, 1998, Pajevic and Pierpaoli, 1999, Pierpaoli and Basser, 1996, Pierpaoli et al., 1996) which has recently become extensively used for clinical purposes. DTI is based on the anisotropic nature of water motion in white matter fibers (Basser and Pierpaoli, 1998, Pierpaoli and Basser, 1996, Pierpaoli et al., 1996). Along the fibers, the motion of water molecules is relatively free, while perpendicular to them, it is more hindered. Based on this observation, DTI enables to extract the orientation of the fibers on a pixel by pixel basis and to quantify the motional anisotropy, with measures such as fractional anisotropy (FA) (Basser and Pierpaoli, 1998). FA was used over the last decade to study several white matter related diseases (Wieshmann et al., 1999, Tievsky et al., 1999, Pierpaoli et al., 2001, Takahashi et al., 2002, Sotak, 2002, Lim and Helpern, 2002, Horsfield and Jones, 2002, Holodny and Ollenschlager, 2002, Mori et al., 2002, Sinha et al., 2002, Witwer et al., 2002). Typically, disruption to organization of the fibers whether in the form of demyelination, axonal loss or other processes of cellular degeneration will cause a decrease of FA in white matter. One of the applications of DTI is to distinguish between distinct effects of SOLs (space occupying lesions) on white matter such as displacement of the tracts, destruction of the fibers or infiltration of the white matter tracts (Witwer et al., 2002, Mori et al., 1999). In contrast to the large number of studies on DTI in brain diseases, there are only a handful of reports on the investigation of white matter deviation by this technique. Barker's group (Eriksson et al., 2001, Wieshmann et al., 2000) was the first to show that DTI can help in delineation of displaced white matter fibers and suggested that the increased anisotropy might be related to white matter compression. Another work of mapping displaced white matter was done by Witwer et al. (2002), who used directionally color-coded maps (also termed: directionally encoded color (DEC)) mapping of DTI to mark displaced white matter as opposed to destructed or infiltrated white matter. Recently, Assaf et al. (2003) have shown that DTI measures could help in characterizing displaced white matter in 2D DEC-FA maps.

The introduction of 3D fiber tracking based on DTI extended the clinical use of this method for pre-surgical mapping (Mori et al., 1999, Mori et al., 2002). Fiber tractography enables three-dimensional visualization of specific fiber bundles that are either in proximity to a lesion or that are influenced by it. DTI based fiber tractography necessitates definition of a seed region of interest (ROI) that is located at the path of the investigated fiber network system in order to initiate the fiber tracking procedure (Basser et al., 2000, Conturo et al., 1997, Ito et al., 2002, Mori et al., 1999). Selecting seed ROIs based on known anatomical landmarks has led to the identification of a large number of fiber bundle systems in healthy subjects (Catani et al., 2002, Wakana et al., 2004). Fiber tractography was also used to investigate large white matter bundles in vicinity to brain lesions, particularly the pyramidal tract of the motor system (Akai et al., 2005, Berman et al., 2004, Clark et al., 2003, Hendler et al., 2003, Yamada et al., 2003, Stieltjes et al., 2001). However, brain lesions, especially SOLs, often affect the white matter and may alter the known anatomical path in which the fibers pass. Nonetheless, in many of these cases, only partial or no functional deficits are observed leading to the assumption that the fibers, though deviated, are still partially functionally intact. In such cases, white matter mapping using seed ROI based on known normal anatomical locations might be misleading, since the white matter is deviated from its normal location. This becomes even more complicated when edema masks the path of the white matter tracts. Recently, it has been shown that DTI based tractography can be combined with functional magnetic resonance imaging (fMRI) to mark both gray and white matter integrity and functionality in relation to brain lesions (Hendler et al., 2003, Parmar et al., 2004). In this work, we used fMRI driven seed ROI selection procedure in patients with SOLs where probable deviation of white matter tracts was observed.

Guye et al. (2003) have also proposed an approach where fMRI might help in defining a seed region of interest for functional connectivity mapping. While they showed that fMRI might help in identifying the seed ROI in the presence of SOL, they did not explore tracking or connectivity measures of different white matter pathologies in the presence of SOL. In this work, we specifically distinguish between cases of lesion infiltration into white matter and white matter displacement by a lesion. While tracking procedure might be successful in cases of displaced white matter, given the appropriate seed ROI, in infiltrated white matter, reduced fractional anisotropy might significantly complicate the tracking ability. Our aim is to show that the combination and co-registration of routinely used, clinically feasible, fMRI and DTI data to the same anatomical volume constrains the seed ROI selection procedure for fiber tracking. This procedure helps the tractographer to define seed ROI which are relevant to desired fiber system. The fMRI activated areas constrain the ROI selection to regions where desired, functionally related, fiber bundle pass. By using the fMRI activations as landmarks for the seed ROI selection, we were able to track the complete path of the fiber system when the fibers were displaced. Following the identification of the functional fiber system, we could then specifically characterize the effect of the SOL on the white matter. To that end, we used analysis of the principal diffusivities to characterize the displaced white matter.

Section snippets

Patients

Nine patients and five healthy subjects participated in this research. All the patients were referred to the fMRI clinic for pre-surgical functional mapping. The mapping included both regular anatomical MRI, functional MRI, which was designed to detect brain activity relevant to their lesion, and DTI based fiber tractography, in order to determine the involvement of white matter structures either in the lesioned area, or in the vicinity of the lesion. The DTI was used to estimate the

Results

In this work, we focused on the pyramidal tract (part of the corona radiata) and the superior longitudinal fasciculus (SLF, arcuate fasciculus), two large bundles of white matter that connect critical areas of motor and language function, respectively. Fig. 1 shows tracking of the pyramidal tract based on known anatomical landmarks for 3 healthy subjects. The anatomically based ROIs are all marked on the directionally encoded color FA maps (internal capsule) and FA maps (pre-central gyrus) of

Discussion

This work indicated that incorporation of fMRI information could enhance the ability to identify tracts of interest in brains with deformed anatomy. DTI based fiber tractography attracts the attention of radiologists, neurologists and neurosurgeons as a possible way to visualize white matter paths in the diseased brain. Yet, in some of these cases due to brain pathology such as SOLs, the white matter fibers are displaced, thus exposing the tractography result to artifacts caused by the common

Acknowledgments

The authors wish to thank the following funds for financial support: The ministry of science and technology (The Eshkol program, to YA), The Israel Science Foundation (To YA), The Adams Super Center for Brain Research of Tel Aviv University (To TH and YA) and The Levie-Gitter-Edershiem institute for functional brain imaging (to TS). We also wish to thank the neurosurgeons and the patients who participated in this study.

References (56)

  • H. Akai et al.

    Diffusion tensor tractography of gliomatosis cerebri

    J. Comput. Assist. Tomogr.

    (2005)
  • D.C. Alexander et al.

    Spatial transformations of diffusion tensor magnetic resonance images

    IEEE Trans. Med. Imaging

    (2001)
  • Y. Assaf et al.

    Deviation of fiber tracts in the vicinity of brain lesions: evaluation by diffusion tensor imaging

    Isr. J. Chem.

    (2003)
  • P.J. Basser et al.

    A simplified method to measure the diffusion tensor from seven MR images

    Magn. Reson. Med.

    (1998)
  • P.J. Basser et al.

    A. In vivo fiber tractography using DT-MRI data

    Magn. Reson. Med.

    (2000)
  • K. Bergstrom et al.

    Magnetic resonance imaging of brain tissue displacement and brain water contents during progressive brain compression. An experimental study in dogs

    Acta Radiol., Suppl.

    (1986)
  • J.I. Berman et al.

    Diffusion tensor imaging guided tracking of fibers of the pyramidal tract combined with intraoperative cortical stimulation mapping in patients with gliomas

    J. Neurosurg.

    (2004)
  • T.E. Conturo et al.

    Tracking neuronal fiber pathways in the living human brain

    Proc. Natl. Acad. Sci. U. S. A.

    (1997)
  • M.R. Del Bigio et al.

    Chronic hydrocephalus in rats and humans: white matter loss and behavior changes

    Ann. Neurol.

    (2003)
  • Y. Ding et al.

    Axonal damage associated with enlargement of ventricles during hydrocephalus: a silver impregnation study

    Neurol. Res.

    (2001)
  • S.H. Eriksson et al.

    Diffusion tensor imaging in patients with epilepsy and malformations of cortical development

    Brain

    (2001)
  • K.J. Friston et al.

    Spatial registration and normalization of images

    Hum. Brain Mapp.

    (1995)
  • O.A. Glenn et al.

    DTI-based three-dimensional tractography detects differences in the pyramidal tracts of infants and children with congenital hemiparesis

    J Magn. Reson. Imaging

    (2003)
  • P.W. Hanlo et al.

    The effect of intracranial pressure on myelination and the relationship with neurodevelopment in infantile hydrocephalus

    Dev. Med. Child Neurol.

    (1997)
  • T. Hendler et al.

    Delineating gray and white matter involvement in brain lesions: 3D alignment of fMRI and DTI

    J. Neurosurg.

    (2003)
  • M.A. Horsfield et al.

    Applications of diffusion-weighted and diffusion tensor MRI to white matter diseases—A review

    NMR Biomed.

    (2002)
  • D.K. Jones

    Determining and visualizing uncertainty in estimates of fiber orientation from diffusion tensor MRI

    Magn. Reson. Med.

    (2003)
  • Cited by (0)

    View full text