Invited Commentary

From:David M. Yousem, MD, MBADepartment of Radiology, Johns Hopkins Medical InstitutionsBaltimore, Maryland

In the preceding article, one can find a true tourde force of diffusion-tensor imaging (DTI) fibertractography in patients with congenital centralnervous system anomalies. In the past 2 years,DTI has bridged the chasm between creating as-tounding color-coded images of the normal whitematter tracts in the brain to a technique that cangraphically depict the impact of disease on thewhite matter.

DTI developed as a specialized form of diffu-sion-weighted imaging (DWI) in multiple dimen-sions. The concept of DWI is to assess the localenvironment of the cell to determine the ease ofwater diffusion. As cells swell (as in cytotoxicedema), the ability of water protons to diffuseextracellularly is restricted. This may be on thebasis of a decrease in the extracellular space, achange in membrane permeability, or activetransport restriction. The restriction of diffusabil-ity corresponds to high signal intensity on diffu-sion-weighted images and a reduction in the ap-parent diffusion coefficient (ADC) of local water.The mechanism for the increased intracellularwater is believed to be the failure of the sodium-potassium pump at the cell membrane. DWI in-terrogates the water diffusion at a cellular level.

The concept behind diffusion-tensor tractogra-phy or white matter mapping is a more globalview of the brain in terms of the macroscopic dif-fusivity of water. There is rigid organization to thewhite matter structure in the brain that can becharacterized by a three-dimensional asymmetry,since the water protons cannot diffuse equally inall directions, bounded by the structure of thewhite matter tracts. This is termed diffusion anisot-ropy. DTI is a means to obtain ADC values alongmany orientations and map white matter tractsfrom such measurements (1–8). To perform thisstudy, diffusion gradients are placed in at least sixindependent directions and calculations of thetensor in three-dimensional space are performed.From these studies, one can determine whitematter orientation and, by extension, possible

injuries. The connectivity of white matter tracts, aparameter thought to be disrupted in many dis-eases, can now be assessed with DTI.

Imagine, if you will, the ultimate set of mapsused by the neurosurgeon removing an astrocy-toma. The surgeon has a representation of thesurface features of the scalp and skull that allowsaccurate delineation of the boundaries of the cra-niotomy for tumor removal. Deep to this map, asthe surgeon begins to raise the flap, he or she hasan MR venogram that allows one to avoid inad-vertent puncture of the superficial veins of themeningeal surface. Next, as the surgeon contem-plates the corticotomy, he or she finds overlain inthe surgical microscope not only the three-dimen-sional representation of the tumor but also func-tional MR imaging cortical maps outlining theeloquent areas of the brain to avoid. Avoiding themotor strip displaced by the neoplasm, the sur-geon must determine the approach to resectingthe mass.

The surgeon flips the lens of the microscopeand can visualize the corticospinal tract derivedfrom the DTI pulse sequence and sees that thewhite matter tracts are displaced posteriorly bythe tumor (6,9). The surgeon adjusts the plan topulverize the mass from an anterior approach. Asthe surgeon begins the resection, the margins ofthe enhancing portions of the tumor are repre-sented in the scope at the same time additional“danger zones of functional significance” are pro-jected for the nearby gray and white matter. If thepatient is to wake up with a deficit, it will be aninformed decision that the surgeon makes in aneffort to achieve gross total removal. Finally,probing the depths of the tumor, the surgeonswitches to three-dimensional virtual reality toobtain (a) the location of middle cerebral arterybranches from MR arteriographic data and(b) deep gray matter functional maps from previ-ously performed functional MR imaging analyses.

Some academic centers are close to achievingthis ultimate scenario (6,10–21). Heretofore, itwas the white matter mapping that was the stum-bling block. In the preceding article, Lee et al, indemonstrating quite elegantly the facility of usingDTI in demonstrating abnormal fiber tracts, have

66 January-February 2005 RG f Volume 25 ● Number 1

Radio

Gra

phic

s

provided encouragement to the neuroradiologiccommunity that this hurdle can be overcome. Forpatients with congenital gray matter disorders,Lee et al have shown that the underlying whitematter may also be affected. In cases of resectingseizure foci due to cortical dysplasias, the whitematter fibers can now be visualized for surgicalplanning. This is the first step in the elucidationof how the white matter disease may produce theclinical signs and symptoms that plague the child.Determining the severity of tract disease may al-low more accurate prognostication for families ofchildren with leukopathies (22,23).

As discussed by Lee et al, this technique is verymuch operator dependent. DTI fiber tractogra-phy is not a push button solution yet and still re-quires active participation by physicians familiarwith white matter anatomy. To produce theglobal brain maps gracing many journal coversthese days requires the input of physicians, tech-nologists, and PhDs who donate their variousexpertises to the mission. Congratulations areoffered to the Yonsei University College ofMedicine team for providing a glimpse into thepromising future.

Financial Interest: The author has no financial rela-tionships to disclose.

References1. Ito R, Mori S, Melhem ER. Diffusion tensor brain

imaging and tractography. Neuroimaging Clin NAm 2002; 12:1–19.

2. Golay X, Jiang H, van Zijl PC, Mori S. High-reso-lution isotropic 3D diffusion tensor imaging of thehuman brain. Magn Reson Med 2002; 47:837–843.

3. Jaermann T, Crelier G, Pruessmann KP, et al.SENSE-DTI at 3 T. Magn Reson Med 2004; 51:230–236.

4. Melhem ER, Mori S, Mukundan G, Kraut MA,Pomper MG, van Zijl PC. Diffusion tensor MRimaging of the brain and white matter tractogra-phy. AJR Am J Roentgenol 2002; 178:3–16.

5. Mori S, van Zijl PC. Fiber tracking: principles andstrategies—a technical review. NMR Biomed2002; 15:468–480.

6. Mori S, Frederiksen K, van Zijl PC, et al. Brainwhite matter anatomy of tumor patients evaluatedwith diffusion tensor imaging. Ann Neurol 2002;51:377–380.

7. Mori S, Kaufmann WE, Davatzikos C, et al. Imag-ing cortical association tracts in the human brainusing diffusion-tensor-based axonal tracking.Magn Reson Med 2002; 47:215–223.

8. Xue R, van Zijl PC, Crain BJ, Solaiyappan M,Mori S. In vivo three-dimensional reconstructionof rat brain axonal projections by diffusion tensorimaging. Magn Reson Med 1999; 42:1123–1127.

9. Stieltjes B, Kaufmann WE, van Zijl PC, et al. Dif-fusion tensor imaging and axonal tracking in thehuman brainstem. Neuroimage 2001; 14:723–735.

10. Parmar H, Sitoh YY, Yeo TT. Combined mag-netic resonance tractography and functional mag-netic resonance imaging in evaluation of brain tu-mors involving the motor system. J Comput AssistTomogr 2004; 28:551–556.

11. Lu S, Ahn D, Johnson G, Law M, Zagzag D,Grossman RI. Diffusion-tensor MR imaging ofintracranial neoplasia and associated peritumoraledema: introduction of the tumor infiltration in-dex. Radiology 2004; 232:221–228.

12. Provenzale JM, McGraw P, Mhatre P, Guo AC,Delong D. Peritumoral brain regions in gliomasand meningiomas: investigation with isotropic dif-fusion-weighted MR imaging and diffusion-tensorMR imaging. Radiology 2004; 232:451–460.

13. Jellison BJ, Field AS, Medow J, Lazar M, SalamatMS, Alexander AL. Diffusion tensor imaging ofcerebral white matter: a pictorial review of physics,fiber tract anatomy, and tumor imaging patterns.AJNR Am J Neuroradiol 2004; 25:356–369.

14. Barboriak DP. Imaging of brain tumors with diffu-sion-weighted and diffusion tensor MR imaging.Magn Reson Imaging Clin N Am 2003; 11:379–401.

15. Clark CA, Barrick TR, Murphy MM, Bell BA.White matter fiber tracking in patients with space-occupying lesions of the brain: a new technique forneurosurgical planning? Neuroimage 2003; 20:1601–1608.

16. Beppu T, Inoue T, Shibata Y, et al. Measurementof fractional anisotropy using diffusion tensor MRIin supratentorial astrocytic tumors. J Neurooncol2003; 63:109–116.

17. Price SJ, Burnet NG, Donovan T, et al. Diffusiontensor imaging of brain tumours at 3T: a potentialtool for assessing white matter tract invasion? ClinRadiol 2003; 58:455–462.

18. Lu S, Ahn D, Johnson G, Cha S. Peritumoral dif-fusion tensor imaging of high-grade gliomas andmetastatic brain tumors. AJNR Am J Neuroradiol2003; 24:937–941.

19. Witwer BP, Moftakhar R, Hasan KM, et al. Diffu-sion-tensor imaging of white matter tracts in pa-tients with cerebral neoplasm. J Neurosurg 2002;97:568–575.

20. Sinha S, Bastin ME, Whittle IR, Wardlaw JM.Diffusion tensor MR imaging of high-grade cere-bral gliomas. AJNR Am J Neuroradiol 2002; 23:520–527.

21. Wieshmann UC, Symms MR, Parker GJ, et al.Diffusion tensor imaging demonstrates deviationof fibres in normal appearing white matter adja-cent to a brain tumour. J Neurol Neurosurg Psy-chiatry 2000; 68:501–503.

22. Eichler FS, Itoh R, Barker PB, et al. Proton MRspectroscopic and diffusion tensor brain MR im-aging in X-linked adrenoleukodystrophy: initialexperience. Radiology 2002; 225:245–252.

23. Ito R, Melhem ER, Mori S, Eichler FS, RaymondGV, Moser HW. Diffusion tensor brain MR imag-ing in X-linked cerebral adrenoleukodystrophy.Neurology 2001; 56:544–547.

RG f Volume 25 ● Number 1 Lee et al 67

Radio

Gra

phic

s

Authors’ Response

From:Seung-Koo Lee, MD, Dong Ik Kim, MD, Jinna Kim, MDDepartment of Radiology, Yonsei University College of MedicineSeoul, Korea

We greatly appreciate Dr Yousem’s complimen-tary comments and completely agree with his fu-ture prospect for DTI in the clinical field. As DrYousem mentioned, presurgical planning is oneof the most attractive fields of clinical DTI appli-cation because precise localization of importantwhite matter tracts to avoid inadvertent injury hashigh clinical impact. If DTI is used with a classicmethod such as intraoperative cortical stimula-tion, one can achieve a good postoperative resultand less surgical morbidity (1). Combination withmultimodality approaches such as functional MRimaging, positron emission tomography (PET),and three-dimensionally reconstructed anatomicdata will help in the diagnosis, treatment plan-ning, and surgical management of space-occupy-ing lesions in the brain. Therefore, collaborationwith physicians, technologists, and experts in im-age processing is important to maximize thepower of DTI.

In the preceding article, we showed aberrantfiber pathways in the case of cortical dysplasialocated at the precentral gyrus. Usually, the epi-

leptogenic foci do not exactly match the struc-tural abnormality in malformation of cortical de-velopment and sometimes they are much largerthan the dysplastic cortex. Therefore, the surgicalresection margin can contain normal-appearingcortex and underlying white matter. From thispoint of view, detailed information about the ab-errant white matter connections adjacent to thedysplastic cortex is important before surgery andDTI is the only way to provide such information.

As suggested by Dr Yousem, we have to con-sider the limitations of DTI and be careful in theinterpretation of DTI results. More active col-laborations with neuroscientists and technolo-gists, broadening of the clinical applications ofDTI, and obtaining a better understanding ofwhite matter changes in various pathologic condi-tions will be the cornerstones of DTI in clinicalneuroimaging.

Reference1. Berman JI, Berger MS, Mukherjee P, Henry RG.

Diffusion-tensor imaging-guided tracking of fibersof the pyramidal tract combined with intraoperativecortical stimulation mapping in patients with glio-mas. J Neurosurg 2004; 101:66–72.

68 January-February 2005 RG f Volume 25 ● Number 1

Radio

Gra

phic

s