Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.

x

CRITICAL REVIEW AND INVITED COMMENTARY

Diffusion-based magnetic resonance imaging and tractography in epilepsy
Mahinda Yogarajah and John S. Duncan
Department of Clinical and Experimental Epilepsy and National Society for Epilepsy, Institute of Neurology, University College London, Queen Square, London, United Kingdom

SUMMARY
Diffusion-based imaging is an advanced MRI technique that is sensitive to the movement of water molecules, providing additional information on the micro-structural arrangement of tissue. Qualitative and quantitative analysis of peri, post and interictal diffusion images can aid the localization of seizure foci. Diffusion tensor tractography is an extension of diffusion-based imaging, and can provide additional information about white matter pathways. Both techniques are able to increase under-

standing of the effects of epilepsy on the structural organization of the brain, and can be used to optimize presurgical planning of patients with epilepsy. This review focuses on the basis, applications, limitations, and future directions of diffusion imaging in epilepsy. Literature search strategy: We searched Pubmed using the terms “diffusion MRI or diffusion tensor MRI or tractography and epilepsy.” KEY WORDS: Diffusion, Diffusion tensor, MRI, Tractography, Ictal, Postictal.

Magnetic resonance imaging (MRI) is central to the assessment of individuals with refractory epilepsy, enabling the identification of the underlying epileptogenic substrate, and if surgical treatment is considered, depicting the relationship of the epileptogenic lesion and zone to eloquent areas of the brain such as the motor, language, or memory areas. Diffusion-based MRI and tractography can provide valuable information in the evaluation of an individual with epilepsy. Diffusion-based MRI has the potential to identify potentially epileptogenic abnormalities, including those that appear normal on standard MRI sequences. Tractography may be used to map white matter tracts, and their relationship to epileptogenic tissue and eloquent cortex. This information may be used to improve surgical planning in order to minimize postoperative deficits including memory, language, and visual field loss. Furthermore, it also has the potential to aid understanding of the acute

and chronic pathophysiological effects of seizures on the brain.

T HE B IOLOGICAL AND P HYSICAL BASIS OF D IFFUSION I MAGING
In a free medium the molecular diffusion of water refers to the random translational motion (Brownian motion) of molecules resulting from the thermal energy carried by these molecules. In the brain, diffusion is restricted by intra- and extracellular boundaries, and represents the effects of several variable, independent factors. These include the presence of impermeable or semipermeable membranes (Hansen, 1971), macromolecules that hinder the diffusion of small molecules, and intra- and extracellular microcirculatory effects (Le Bihan et al., 1992; Le Bihan & Turner, 1992). The measurement of water diffusion therefore provides a means of probing cellular integrity and pathology (Le Bihan, 2003). The principles of diffusion MRI were first developed in vivo in the mid 1980s (see Le Bihan, 1995 for review). In diffusion-weighted imaging (DWI), images are sensitized to the diffusional properties of water by the incorporation of pulsed magnetic field gradients into a standard spin echo sequence (Merboldt et al., 1985; Taylor & Bushell, 1985). By taking measurements in at least three directions,

Accepted August 31, 2007; Online Early publication October 18, 2007. Address correspondence to Prof J. S. Duncan, Department of Experimental and Clinical Epilepsy, Institute of Neurology, University College London, Queen Square, London WC1N 3BG, United Kingdom E-mail: j.duncan@ion.ucl.ac.uk Blackwell Publishing, Inc. C 2008 International League Against Epilepsy

189

190 M. Yogarajah and J. S. Duncan it is possible to characterize the mean diffusion properties within a voxel in the image by way of a single scalar apparent diffusion coefficient (ADC). Early diffusion studies discovered that ADC measurements depended on a subject’s orientation relative to the magnet and gradient coils (Hajnal et al., 1991). White matter tracts parallel to an applied gradient had the greatest ADC whereas those lying oblique or transverse to a gradient had smaller ADC values. This gave rise to the concept of asymmetry of diffusion of molecules in three directions, or “anisotropy” (Basser, 1995). Diffusion tensor imaging (DTI) enables not only the quantification of water molecule diffusion, but also the characterization of the degree and direction of anisotropy (see Le Bihan et al., 2001 for review). The diffusion tensor is a mathematical construct that can be calculated from a non–diffusion-weighted image plus six or more diffusionweighted measurements along noncollinear directions. The tensor can be diagonalized to give three eigenvectors, ε 1, ε 2 , and ε 3 representing the principal directions of diffusion, and three eigenvalues λ 1 , λ 2 , and λ 3 representing the magnitude of diffusion (or the corresponding ADC values) along these directions. Furthermore, a number of diffusion parameters can be derived in each voxel, which are insensitive to subject positioning and fiber tract alignment within the diffusion gradients of the MRI scanner (Basser et al., 1994; Pierpaoli et al., 1996). Mean diffusivity (MD) is a summary measure of the average diffusion properties of a voxel and is equivalent to the estimated ADC over three orthogonal directions. Fractional anisotropy (FA) on the other hand is an estimate of what proportion of the magnitude of the diffusion tensor is due to anisotropic diffusion. Quantitative maps of these parameters can also be constructed, and used to make comparisons between individuals or populations (Fig. 1). Diffusion anisotropy in cerebral tissue is highly heterogeneous due to several factors including, the concentration of macromolecules and intracellular organelles, regional differences in the density of nerve fibers, the degree of myelination, fiber diameter and the density of neuroglial cells (Beaulieu, 2001). Anisotropy in white matter results from the organization of tissue as bundles of axons and myelin sheaths run in parallel, and the diffusion of water is freer and quicker in the long axis of the fibers, than in the perpendicular direction (Beaulieu, 2001). Malformations or acquired insults cause disruption to the microstructural environment, and more often than not, a subsequent reduction in anisotropy. Such abnormalities may also lead to a reduction in cell density and/or expansion of the extra cellular space, resulting in an increase in MD/ADC.

P ERI - AND P OSTICTAL C HANGES IN D IFFUSION
Seizure-associated changes in diffusion parameters are not static, but have a dynamic profile. These changes are observed in both animal and human studies, and generally show a pattern of early postictal depression, followed by

Figure 1. (A) Axial diffusion-weighted image. The dark end of the gray scale represents areas of increased diffusion, and the bright end areas of restricted diffusion. Diffusion is greatest in the CSF, which therefore appears dark. Diffusionweighted images are T2 sensitive, such that bright regions of high T2 signal that are not diffusion restricted persist in the diffusion-weighted images (“T2 shine through”). For this reason the calculation of an ADC map that is independent of this effect is useful. Bright T2 signal and decreased ADC drive DWI signal intensity up whereas low T2 signal and high ADC drive DWI signal intensity down. (B) Axial ADC map describing the ADC value in each voxel. The bright end of the gray scale represents increased diffusion, and the dark end areas of decreased diffusion. Diffusion is greatest in the CSF, which therefore appears bright. (C) Fractional anisotropy map describes the degree of diffusion anisotropy in each voxel. In white matter where anisotropy is high the bright end of the gray scale is assigned. In gray matter where anisotropy is low, the dark end of the gray scale is applied. Epilepsia C ILAE
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

191 Diffusion-based MRI and Tractography in Epilepsy normalization, and then transient or chronic elevation of the ADC/MD (Righini et al., 1994). Animal studies A considerable body of animal data has shown that diffusion-weighted MRI can visualize the histopathological changes that result from seizures in animal models. The first reported study, by Zhong et al., demonstrated a fall of 15% in the ADC in bicuculline-induced status epilepticus in rats (Zhong et al., 1993). Other models have shown similar reductions in ADC values that are in proportion to the severity of seizure activity (Prichard et al., 1995; Zhong et al., 1995, 1997). The ictal and postictal changes seen in the ADC are similar to those seen in cerebral ischaemia, and both share a common biological basis, namely the loss of membrane function and ion homeostasis. Cerebral ischaemia leads to a failure of energy metabolism, membrane dysfunction, and cell death. Sustained seizures on the other hand lead to an increased metabolic rate. This is coupled to an increase in cerebral blood flow (Szabo et al., 2005), so that cellular energy values are close to normal, though in prolonged ictal activity, the increased metabolic activity may not be matched by enhanced blood flow (Bruehl et al., 1998). The early ADC decline seen in prolonged seizures is thought to reflect cytotoxic oedema (Wang et al., 1996), and a decrease in the extra cellular space volume fraction of up to 30% at the area of maximum neuronal activity in the cortex (Lux et al., 1986). This in turn leads to increased extracellular tortuosity and decreased diffusivity. Seizures cause increased membrane ion permeability (McNamara, 1994) leading to an influx of sodium, calcium, and water along the osmotic gradient (Wang et al., 1996), which cannot be compensated for by an energy deficient sodium–potassium ion ATP pump. Intracellular cytoskeletal fragmentation that increases intracellular tortuosity and viscosity, may also contribute to restricted diffusion (van der Toorn et al., 1996). While cytotoxic oedema is the most common pathophysiological effect of seizures found in cortical gray matter, vasogenic oedema has also been reported less commonly in subcortical white matter (Tanaka et al., 1992). Animal studies have demonstrated that seizures can also trigger acidosis and the breakdown of the blood brain barrier (Nitsch & Klatzo, 1983). This, together with local vasodilatory effects, can give rise to vasogenic oedema and an increase in intercellular space and diffusivity (Nedelcu et al., 1999). Though cytotoxic oedmatous changes are not necessarily irreversible, with prolonged seizures, diffusivity and the ADC can change permanently (Nedelcu et al., 1999). Excitotoxic mechanisms mediated by excitatory amino acids, calcium influx, ATP depletion, and lactate accumulation eventually lead to cell atrophy and death (Wasterlain et al., 1993). This cell lysis results in an increase in extracellu-

Figure 2. (Adapted with permission from Wieshmann et al., 1997) DWI in a patient with complex partial status epilepticus affecting the right leg. Decreased diffusion is visible in the motor cortex, and increased diffusion is visible in the subcortical white matter. This corresponded to a relative decrease in ADC of 27% and increase in ADC of 31% in the cortical and subcortical tissue, respectively. Epilepsia C ILAE

lar space and an increase in diffusion above normal values, which correlates with histopathological changes in both the seizure focus and secondarily affected areas (Pitkanen et al., 2002; Hasegawa et al., 2003). Clinical studies Status epilepticus Early clinical studies assessed diffusion-weighted imaging in patients with status epilepticus. In a patient with focal motor status epilepticus consisting of clonic jerking of the right leg, a 27% relative decrease in the ADC was demonstrated in the motor cortex of the right leg (Wieshmann et al., 1997). There was also a 31% relative increase in the ADC of the subcortical white matter (Fig. 2). This finding was thought to represent a shift of water into cortical cells at the seizure focus, and a shift of water into extracellular space in remote white matter due to vasogenic oedema (Lux et al., 1986). Similar findings in other case reports hint at the complex osmotic relationship between epileptogenic and surrounding areas, and cytotoxic and vasogenic oedema (Kim et al., 2001; Hong et al., 2004). Other studies have broadly corroborated these results including a small case series, where cortical ADC reductions of up to 36% were found during partial status epilepticus (Lansberg et al., 1999). Early clinical reports also suggested that there were significant correlations between the areas of diffusion abnormalities, and increased perfusion and electrocorticographic abnormalities (Diehl et al., 1999; Flacke et al.,
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

192 M. Yogarajah and J. S. Duncan 2000; Calistri et al., 2003). In a study of 10 patients with complex partial status epilepticus there was correlation between focal swelling and hyperintensity on T 2 -weighted images and increased signal on DWI images (Szabo et al., 2005). ADC values were reduced by 11 to 37%, and there was a close spatial correlation of diffusion weighted and perfusion imaging (PI) changes, hyperperfusion on SPECT, and localization of EEG focus. These abnormalities normalized in most patients by day 14. In many cases however, DWI revealed abnormalities in several different regions, and it was problematic to differentiate changes in areas of seizure focus, and changes in the epileptic cortical and subcortical networks that underlie seizure spread (Lansberg et al., 1999; El Koussy et al., 2002). The authors therefore concluded that it would be difficult to locate the epileptogenic focus using DWI and PI alone. Single seizures There have also been several studies of diffusion imaging following single seizures (Table 1). The interpretation of these studies is limited by a number of factors. These include, small numbers of heterogeneous patients, varying methods of analysis (including a priori region of interest (ROI) and whole brain voxel-based methods), lack of control groups or follow-up scanning, and wide variability in the duration of both seizure, and interval from seizure to scan. Salmenpera et al. (2006b) used DWI to study changes in diffusivity after single seizures. In 21 patients with intractable focal epilepsy, postictal decreases were found in 52% seizures, but in 17% of seizures there were increases in MD. The analysis used voxel-based methods to include data from the whole brain, and the resulting spatial distribution of diffusion changes was complex, with postictal changes in MD often being found distant to the putative seizure focus. This implied involvement of a widespread epileptic network, and not a single focus (Fig. 3). The increases in MD, which were detected together with the decreases in postictal scans that were acquired soon after seizures, were thought to be due to vasogenic oedema. Concordance with the presumed epileptogenic focus was seen in only four patients, all of whom had postictal scans within 45 min of seizure onset. Repeated postictal scans showed a gradual return to baseline for both the increases and decreases in MD. In an effort to minimize the delay between seizure and scan, Konermann et al. administered intravenous flumazenil, during scanning. They consistently demonstrated significantly reduced ADC in hippocampi and parahippocampal gyri, ipsilateral to the seizure onset, in a series of 10 patients with refractory TLE (Konermann et al., 2003). Further work by the same group without the use of flumazenil, identified diffusion changes postictally in only two out of nine patients, in whom complex partial seizures were of duration greater than 60 s, and seizure to scan time was less than 15 min. Generalized seizures were associated with global ADC change (Hufnagel et al., 2003). These studies suggest that the diffusion changes visualized with MRI after single seizures are more transient than those after status epilepticus, and are complex in terms of their distribution and evolution of change. The inherent difficulties in scanning patients directly after seizures,

Table 1. Postictal diffusion studies
Authors Diehl et al., 2001 Findings • 1/7 had significant ↓ADC compared with contralateral side after a single seizure • Areas of diffusion change maximal adjacent to hippocampus (unclear if seizure onset or seizure spread zone) Hufnagel et al., 2003 • 2/9 patients had significant ↓ADC postictally compared to interictally • Changes colocalized with postulated seizure focus Konermann et al., 2003 • 10 patients with TLE scanned before and after injection with flumazenil • Significant ↓ADC in all patients postictally compared with interictally • Changes colocalized with postulated seizure focus Oh et al., 2004 • 9/14 patients had significant ↓ADC postictally compared to interictally • Changes colocalized with postulated seizure focus • Significant difference seen only in patients with neocortical ictal onset zones or in neocortical portion of temporal lobe—authors hypothesize this is due to interictal chronic ↑ADC in hippocampus of mTLE patients masking any postictal decrease Diehl et al., 2005 • 8/18 patients had significant ↓MD postictally compared with interictally which were focal in seven patients (including one with ↑MD) • In 3 patients presumed epileptogenic zone colocalized with the area of MD decrease • No changes in FA seen suggesting that single short seizures cause changes in cell hydration but not the directionality of diffusion Salmenpera et al., 2006b • 21 patients scanned after 23 seizures • Focal diffusion changes (significant ↑ or ↓ in MD) seen in 52% of seizures postictally compared with interictally • Changes colocalized with postulated seizure focus in 4 patients

Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

193 Diffusion-based MRI and Tractography in Epilepsy

Figure 3. (Adapted with permission from Salmenpera et all. 2006b) Difference analysis (areas of decreased diffusivity postictally are compared with interictal values) of a patient with left temporal lobe epilepsy scanned 40 min after a complex partial seizure. The areas of change are overlaid in color on the patient’s normalized b0 image, and this shows decreased mean diffusivity postictally in the bilateral cingulate cortex compared to the interictal image. Red arrows refer to the mean diffusivity values measured from the areas of change at different time points (II = interictal, PI = postictal). Epilepsia C ILAE

the evident involvement of a cerebral network and not of a single focus, and the physical limitations of spatial resolution limit the sensitivity of the technique in the localization of seizure foci. Technological advancements such as, real time motion correction, open access scanners, and fast acquisitions may overcome these limitations and result in postictal diffusion MRI becoming a useful clinical tool. Interictal studies Early interictal diffusion imaging studies of patients with epilepsy concentrated on temporal lobe epilepsy (TLE) and hippocampal Sclerosis (HS) and found increased average ADC values in sclerotic hippocampi, compared with the contralateral side and control subjects. This suggested structural disorganization and an expansion of extra cellular space, and was thought to reflect neuronal loss, reduction of dendritic branching, and microstructural changes associated with epileptogenesis (Hugg et al., 1999; Wieshmann et al., 1999a; Kantarci et al., 2002; Yoo et al., 2002; Assaf et al., 2003; Hakyemez et al., 2005). Furthermore, in those patients who undergo surgery, ADC measures may be a useful postoperative prognostic indicator (Goncalves Pereira et al., 2006). Studies using highresolution DTI have also found abnormal anisotropy values in the hippocampus compared to healthy control subjects, though to a lesser magnitude than mean diffusivity changes (Salmenpera et al., 2006a). Abnormalities in the diffusion parameters of hippocampi ipsilateral to seizure onset, which are normal on conventional MRI have also

been found. This suggests that diffusion MRI may be more sensitive in identifying abnormal cerebral tissue than standard MRI sequences (Assaf et al., 2003; Londono et al., 2003). Wehner et al. assessed 22 patients with TLE, and found that in the 14 patients with MRI defined hippocampal sclerosis, the ADC was significantly greater in the ipsilateral HC compared with the contralateral side, and could be used to lateralize the seizure focus (Wehner et al., 2007). In the remaining patients without HS, the ADC of the hippocampi were not significantly different to the contralateral side, but were significantly less than in controls. Analysis of the resected specimens confirmed hippocampal sclerosis in those MR positive patients, but revealed gliosis only without any apparent neuron loss or hippocampal sclerosis in the MR negative group. The authors postulated that bilateral temporal lobe abnormalities in some patients with TLE might explain why diffusivity did not provide lateralizing information in patients with nonlesional MRI, and this appears to be have been borne out by other studies (Lee et al., 2004). Diffusion-based studies that have specifically assessed normal looking tissue, beyond ipsilateral mesial temporal lobe structures in TLE patients, have demonstrated bilateral changes together with extratemporal abnormalities (Arfanakis et al., 2002; Thivard et al., 2005b; Gross et al., 2006). This suggests that structural or functional abnormalities (metabolic changes, subtle structural lesions) may extend beyond the seizure onset zone in unilateral mesial TLE associated with HS. Diffusion imaging is also sensitive to patients with epilepsy and nonprogressive acquired lesions such as
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

194 M. Yogarajah and J. S. Duncan

Figure 4. (Adapted with permission from Dumas et al., 2005) (A–C) Frontal lobe dysplasia in a 31-year-old patient. (A) FLAIR sequences showing subtle hyperintensity of the left frontal lobe (arrow). (B) Fractional anisotropy (FA) map. Area of decreased FA of the left frontal lobe, more extensive than the area of FLAIR signal abnormality. (C) Superposition of FA (using a color scale) and FLAIR images. Epilepsia C ILAE cerebral ischaemic lesions and perinatal hypoxia (Wieshmann et al., 1999b, 1999c, Rugg-Gunn et al., 2001). Areas of increased MD and decreased FA correlate with abnormalities identified on visual inspection of conventional MR imaging, and are concordant with neuronal loss, gliosis, and structural disorganization. Moreover, diffusion imaging can often pick up areas of pathology beyond the conventional margins of acquired lesions seen on standard MRI, again suggesting additional sensitivity from DTI (Rugg-Gunn et al., 2001). Patients with epilepsy and malformations of cortical development (MCD) have also been studied with diffusionbased MRI (Wieshmann et al., 1999b). Eriksson et al. used a voxel-based method to assess the whole brains of 22 patients with several types of MCD (Eriksson et al., 2001). Fifteen and eight patients had reduced anisotropy and increased diffusivity within the MCD respectively, which suggests a loss of directional organization and relatively preserved cell density. Moreover, diffusion abnormalities were also found beyond the margins of the evident MCD in areas that appeared normal on conventional MRI. Consistent with these findings, Dumas et al. used a ROI-based method to assess both areas of MR visible abnormality, and normal appearing cerebral tissue in 15 patients (Dumas et al., 2005). They identified significantly reduced anisotropy in normal appearing white matter adjacent to, and 2–3 cm distant from several types of cerebral lesion, including MCDs (Fig. 4). Histological examination of resected normal looking tissue revealed the presence of occult abnormalities such as gliosis, infiltrative tumor cells, and axonal loss. Together these findings suggest that diffusion imaging can often pick up areas of pathological abnormality beyond the conventional margins seen on standard MR images, which has implications for the surgical resection margins of these areas.
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

Interictal DTI is also able to identify focal abnormalities in patients with focal epilepsy, and unremarkable conventional MRI (Fig. 5). In one of the earliest studies, increased diffusivity was found in eight patients and reduced anisotropy was found in two patients out of a total of 30 patients with refractory focal epilepsy, and unremarkable conventional MRI (Rugg-Gunn et al., 2001). In seven, the areas of abnormal diffusion corresponded with the localization of EEG focus. A group analysis of the nine patients with electroclinical seizure onset localizing to the left temporal region revealed a significant increase in diffusivity, and reduction in anisotropy within the white matter of the left temporal lobe. The areas of abnormal diffusion were postulated to be caused by disruption in the microstructural environment due to etiological factors such as occult dysgenesis, or acquired damage, or as a result of repeated seizures leading to neuron loss, gliosis, and expansion of the extra cellular space. This study suggested that diffusivity is a more sensitive diffusion index than anisotropy for identifying occult abnormalities in patients with normal, conventional MRI. This may represent expansion of the extra cellular space but retention of the overall structural organization of the white matter tracts. A patient from this study, with cryptogenic frontal lobe epilepsy, had focally increased MD in the right frontal lobe. Subsequent intracranial EEG concluded that this was the area of seizure onset, and led to a resection. Histopathologic examination of the resected specimen showed marked white matter gliosis, associated with structural disorganization, and expansion of the extra cellular space (Rugg-Gunn et al., 2002). Six years following surgery, seizures have been reduced by more than 50% the preoperative rate. Subsequent studies have corroborated these findings, and investigated the correlation between DTI measurements and stereo-electroencephalographic (SEEG)

195 Diffusion-based MRI and Tractography in Epilepsy

Figure 5. (Adapted with permission from Thivard et al., 2006) View of a 3D representation of the brain of a patient with occipital lobe epilepsy and reported normal “conventional” MRI. Diffusion imaging revealed a region of increased ADC (red) in the left temporo-occipital junction. This corresponded to the onset zone (blue) and irritative zone (green) as delineated by intracranial EEG. Epilepsia C ILAE recordings in patients with cryptogenic focal epilepsy (Thivard et al., 2006; Guye et al., 2007). These studies have found good spatial concordance between epileptiform activity on EEG and diffusion abnormalities in nearly 50% (6/13 and 4/9, respectively) patients. They also found that diffusivity, rather than anisotropy measures, correlated better with electroclinical data. Furthermore, in those patients who have undergone surgery for their epilepsy, this has often translated into a good postoperative outcome, suggesting that DTI can provide additional information over conventional MRI in the identification of occult abnormalities. Despite the encouraging nature of these results, it is important to note that in several of the aforementioned studies, details of the “conventional” MRI sequences used were not available. Tertiary referral centers can increase their diagnostic yield in patients with refractory epilepsy, with the use of epilepsy-specific, high-resolution volumetric imaging (Von Oertzen et al., 2002). In those cases that remain “MR negative” after such imaging, interictal diffusion imaging has a role to play. The derivation of quantitative ADC/MD/FA maps and their analysis either by ROI or VBM methods provides a useful tool in the localization of subtle structural abnormalities, as part of a multimodality evaluation that should include interictal and ictal EEG recordings, neuro psychiatric and psychological evaluations and other imaging modalities such as PET, SPECT, or magnetoencephalography (MEG).

T RACTOGRAPHY AND E PILEPSY
Knowledge of the anatomy of white matter connections is crucial to the understanding of normal and abnormal brain function (Ffytche & Catani, 2005). With conventional MRI variations in white matter signal are subtle, and white matter tracts cannot be accurately parcellated. In most studies, DTI quantitative measures have been assessed using either region of interest or voxel-based analysis. The former has limitations in that it is user dependent, and has a possibility of error that other fiber tracts, gray matter and CSF or other white matter structures may be included. The latter, though observer independent, has problems associated with the need for spatial normalization and smoothing due to anatomical variations in ventricular size, gyral patterns, etc. Both methods have limited ability to quantify specific white matter pathways along their entire trajectories. Tractography is an extension of DTI, whereby the directional information obtained in each voxel is used to generate virtual, three-dimensional white matter maps. These maps are based on similarities between the diffusion properties of neighboring voxels in terms of shape (quantitative diffusion anisotropy measures) and orientation (principal eigenvector map), and several mathematical algorithms have been devised to generate white matter tracts (Mori & van Zijl, 2002).
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

196 M. Yogarajah and J. S. Duncan Tractography does not therefore trace fibers in the sense that injected tracers do; rather it demonstrates the path of least resistance to water diffusion. The size of typical imaging voxels is a 2–3 mm3 so a single voxel could contain thousands of axons. In addition most methods assume that fibers at each voxel are well described by a single orientation estimate. This can lead to tracking difficulties in areas of fiber kissing or crossing. As methodological developments occur in orientational (Tuch et al., 2002) and spatial resolution (Nunes et al., 2005), and in diffusion modeling (Tournier et al., 2004; Tuch, 2004; Alexander, 2005; Perrin et al., 2005) and tractography algorithms (Parker & Alexander, 2005), these limitations should prove less of a problem. Despite these limitations tractography is the only technique available for tracing the white matter pathways in the living brain. By isolating specific pathways from adjacent gray and white matter and CSF, tract-specific qualitative and quantitative information such as volume, anisotropy, and connectivity indices can also be derived (Ciccarelli et al., 2003a). Tracts can also be normalized and combined to generate group maps that indicate how reproducible a given tract or connection is across a group of subjects (Ciccarelli et al., 2003b). This information can be used to locate and assess the pathophysiological effects of chronic epilepsy on the white matter anatomy, including the structural reorganization of higher cortical functions such as language and memory. The technique can also be used to investigate white matter anatomy (Catani et al., 2002), which can aid preoperative planning, and prevent damage to eloquent cortical functions, particularly when combined with functional activation studies (Guye et al., 2003). Reorganization of language and memory networks Refractory TLE due to HS has a good outcome following anterior temporal lobe resection (ATLR). TLE may be associated with disrupted lateralization of language and material specific memory, and these functions may be further impaired by ATLR. Significant, selective language deficits have been reported in up to 40% of patients following dominant ATLR (Davies et al., 1998). Patients undergoing unilateral ATLR for refractory TLE also typically show a decline in verbal memory following surgery involving the language-dominant hemisphere (Ivnik et al., 1987) and deficits in topographical memory following nondominant temporal lobe resection (Spiers et al., 2001). Functional MRI studies have demonstrated the reorganization of both memory (Powell et al., 2007b) and language functions in TLE patients (Adcock et al., 2003; Thivard et al., 2005a). DTI tractography has the potential to demonstrate the structural reorganization of networks involved in memory and language, which mirror changes in cerebral function. Powell et al. (Powell et al., 2007a) combined fMRI and tractography in patients with unilateral TLE, and
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

in controls. Verb generation and reading comprehension paradigms were used to define functional regions that were used to generate starting regions for tractography. Tractography was carried out using diffusion images acquired with a high angular resolution technique, and a probabilistic algorithm. This technique is thought to cope better with crossing or kissing fibers (Parker & Alexander, 2003). Controls and right TLE patients had a left-lateralized pattern of both language-related activations, and associated white matter organization. Left TLE patients showed more symmetrical language activations, along with reduced left hemisphere and increased right hemisphere white matter pathways, in comparison with both controls and right TLE patients (see Fig. 6). Correlations between measures of structure and function in both groups were found, with subjects with more lateralized functional activation having more lateralized white matter pathways. Other tractography studies have assessed memoryrelated structures within the limbic system. Concha et al. found that patients with unilateral TLE have bilateral changes in the fornix and cingulum bundle, characterized by impaired tracking of these pathways, and increased mean diffusivity and reduced FA along them. This was thought to be consistent with the degeneration of pathways connecting to the hippocampus (Concha et al., 2005). Other studies have assessed the progression of Wallerian degeneration in the limbic structures in patients with refractory epilepsy who have undergone surgical procedures such as corpus callostomy (Concha et al., 2006) and temporal lobe resections (Concha et al., 2007). Together, these results suggest that the use of tractography-derived quantitative measures may have a significant role to play in the longitudinal evaluation of the effects of epilepsy on the brain, and on cognitive functions such as memory and language, particularly when correlated with neuropsychological measures (Lui et al., 2005). Visual pathways and preoperative planning ATLR can also cause visual field defects (VFD) in up to 10% of patients. Indeed, in 5% it can be severe enough to render the patient ineligible for a driving license, despite being seizure-free (Manji & Plant, 2000). Typically, VFDs after ATLR occur in the superior homonymous field contralateral to the resection and are due to disruption of fibers of Meyer’s loop. The anterior extent of the Meyer loop is not visualized on conventional imaging and varies from person to person (Ebeling & Reulen, 1988). As a consequence the occurrence and extent of a postoperative VFD cannot be accurately predicted by conventional MRI, or from the extent of resection performed. Tractography has been used to demonstrate the optic radiation in normal subjects (Yamamoto et al., 2005), and has been applied to preand postoperative surgical patients with AV malformations and tumors in and around the visual pathways. Kikuta et al. (Kikuta et al., 2006) carried out pre- and postoperative

197 Diffusion-based MRI and Tractography in Epilepsy

Figure 7. (Adapted with permission from Powell et al., 2005) Tractography of the optic radiation in two patients who underwent anterior temporal lobe resections, superimposed on each subject’s sagittal non diffusion weighted (b = 0) MR image. Preoperative images on the left and postoperative images on the right. Patient A suffered a quadrantic field deficit postoperatively due to surgical interruption of the optic radiation (arrow). The visual fields of patient B remained intact. Epilepsia C ILAE Figure 6. (Adapted with permission from Powell et al., 2007a) Group variability maps of the connecting paths tracked from left and right functionally defined frontal ROIs for each of the three groups controls, left TLE and right TLE patients. Each image shows the maximum intensity of the commonality maps in each plane of view as a brain surface rendering. The color scale indicates the degree of overlap among subjects (expressed as commonality value); for example, a value of 1 (pure red) represents 100% subject overlap (i.e., every subject’s identified tract contains the voxel in question). Controls and right TLE patients show a similar pattern of connections with greater SLF connections to the temporal lobe on the left (arrowed) than the right. In the left TLE group the opposite pattern is seen with greater temporal lobe connections on the right. Epilepsia C ILAE

ity of tractography in the resection of neoplasms that are in close proximity to eloquent subcortical white matter tracts (Yu et al., 2005). There are, however, technical challenges to be overcome to enable the coregistration of preoperative tractography with the T 1 -weighted MR images used to guide neurosurgical interventions. When these are surmounted, preoperative tractography of the optic radiation and other vital white matter connections, will be able to be displayed when planning and undertaking surgical procedures (Kamada et al., 2005). Further, the advent of peroperative MRI will allow the correction of the movement of tracts caused by craniotomy, and will improve the accuracy of the data, to aid surgical planning and result in a lower risk of postoperative deficits.

C ONCLUSION
The advent of diffusion-based MRI and tractography heralds an exciting period in the neuroimaging of epilepsy patients. The interpretation of peri- and postictal diffusion changes remains complex, but has the potential to improve understanding of seizure physiology. Interictal diffusion MRI studies have some localizing value in patients with focal epilepsy, but normal conventional MRI scans. The use of quantitative maps derived from diffusion imaging, especially within the context of a multimodality assessment, is a powerful tool to search for subtle lesions. Furthermore this technique may also have advantages for delineating the
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

tractography in 10 such patients, and were able to predict the magnitude of pre- and postoperative visual field loss from the geometrical relationship between the optic radiation and AV malformation. A recent study demonstrated application to temporal lobe surgery for epilepsy (Fig. 7). The optic radiation was visualized before and after ATLR, and disruption of Meyer’s loop was demonstrated in a patient who developed a quadrantanopia (Powell et al., 2005). In a similar vein, other studies have demonstrated the util-

198 M. Yogarajah and J. S. Duncan extent of a structural abnormality, and have a particularly important role in the planning of intracranial EEG and tissue resection. The place of tractography in the imaging armory available to epileptologists remains to be determined. Its findings need to be interpreted with a degree of caution due to the limitations described above. Despite this, it has already demonstrated its potential in increasing our understanding of the structural and functional plasticity that occurs in chronic TLE. Further studies are needed to evaluate the role of tractography in presurgical planning, particularly studies incorporating postoperative findings. We anticipate that as data acquisition and tracking algorithms improve, and tractography data are combined with EEG and fMRI data, these improvements will be forthcoming. Ultimately, it may be possible to visualize white matter organization with tractography, such that potentially novel approaches to functionally disconnect the seizure focus from the surrounding brain can be developed.
Ciccarelli O, Parker GJ, Toosy AT, Wheeler-Kingshott CA, Barker GJ, Boulby PA, Miller DH, Thompson AJ. (2003a) From diffusion tractography to quantitative white matter tract measures: a reproducibility study. Neuroimage 18:348–359. Ciccarelli O, Toosy AT, Parker GJ, Wheeler-Kingshott CA, Barker GJ, Miller DH, Thompson AJ. (2003b) Diffusion tractography based group mapping of major white-matter pathways in the human brain. Neuroimage 19:1545–1555. Concha L, Beaulieu C, Gross DW. (2005) Bilateral limbic diffusion abnormalities in unilateral temporal lobe epilepsy. Ann Neurol 57:188– 196. Concha L, Gross DW, Wheatley BM, Beaulieu C. (2006) Diffusion tensor imaging of time-dependent axonal and myelin degradation after corpus callosotomy in epilepsy patients. Neuroimage 32:1090– 1099. Concha L, Beaulieu C, Wheatley BM, Gross DW. (2007) Bilateral white matter diffusion changes persist after epilepsy surgery. Epilepsia 48:931–940. Davies KG, Bell BD, Bush AJ, Hermann BP, Dohan FC, Jaap AS. (1998) Naming decline after left anterior temporal lobectomy correlates with pathological status of resected hippocampus. Epilepsia 39:407–419. Diehl B, Najm I, Ruggieri P, Foldvary N, Mohamed A, Tkach J, Morris H, Barnett G, Fisher E, Duda J, Luders HO. (1999) Periictal diffusionweighted imaging in a case of lesional epilepsy. Epilepsia 40:1667– 1671. Diehl B, Najm I, Ruggieri P, Tkach J, Mohamed A, Morris H, Wyllie E, Fisher E, Duda J, Lieber M, Bingaman W, Luders HO. (2001) Postictal diffusion-weighted imaging for the localization of focal epileptic areas in temporal lobe epilepsy. Epilepsia 42:21–28. Diehl B, Symms MR, Boulby PA, Salmenpera T, Wheeler-Kingshott CA, Barker GJ, Duncan JS. (2005) Postictal diffusion tensor imaging. Epilepsy Res 65:137–146. Dumas dlR, Oppenheim C, Chassoux F, Rodrigo S, Beuvon F, DaumasDuport C, Devaux B, Meder JF. (2005) Diffusion tensor imaging of partial intractable epilepsy. Eur Radiol 15:279–285. Ebeling U, Reulen HJ. (1988) Neurosurgical topography of the optic radiation in the temporal lobe. Acta Neurochir (Wien) 92:29–36. El Koussy M, Mathis J, Lovblad KO, Stepper F, Kiefer C, Schroth G. (2002) Focal status epilepticus: follow-up by perfusion- and diffusion MRI. Eur Radiol 12:568–574. Eriksson SH, Rugg-Gunn FJ, Symms MR, Barker GJ, Duncan JS. (2001) Diffusion tensor imaging in patients with epilepsy and malformations of cortical development. Brain 124:617–626. Ffytche DH, Catani M. (2005) Beyond localization: from hodology to function. Philos Trans R Soc Lond B Biol Sci 360:767–779. Flacke S, Wullner U, Keller E, Hamzei F, Urbach H. (2000) Reversible changes in echo planar perfusion- and diffusion-weighted MRI in status epilepticus. Neuroradiology 42:92–95. Goncalves Pereira PM, Oliveira E, Rosado P. (2006) Apparent diffusion coefficient mapping of the hippocampus and the amygdala in pharmaco-resistant temporal lobe epilepsy. AJNR Am J Neuroradiol 27:671–683. Gross DW, Concha L, Beaulieu C. (2006) Extratemporal white matter abnormalities in mesial temporal lobe epilepsy demonstrated with diffusion tensor imaging. Epilepsia 47:1360–1363. Guye M, Parker GJ, Symms M, Boulby P, Wheeler-Kingshott CA, SalekHaddadi A, Barker GJ, Duncan JS. (2003) Combined functional MRI and tractography to demonstrate the connectivity of the human primary motor cortex in vivo. Neuroimage 19:1349–1360. Guye M, Ranjeva JP, Bartolomei F, Confort-Gouny S, McGonigal A, Regis J, Chauvel P, Cozzone PJ. (2007) What is the significance of interictal water diffusion changes in frontal lobe epilepsies? Neuroimage 35:28–37. Hajnal JV, Doran M, Hall AS, Collins AG, Oatridge A, Pennock JM, Young IR, Bydder GM. (1991) MR imaging of anisotropically restricted diffusion of water in the nervous system: technical, anatomic and pathologic considerations. J Comput Assist Tomogr 15(1):1–18. Hakyemez B, Erdogan C, Yildiz H, Ercan I, Parlak M. (2005) Apparent diffusion coefficient measurements in the hippocampus and amygdala of patients with temporal lobe seizures and in healthy volunteers. Epilepsy Behav 6:250–256. Hansen JR. (1971) Pulsed NMR study of water mobility in muscle and brain tissue. Biochem Biophys Acta 230:482–486.

ACKNOWLEDGEMENTS
We are grateful to the Welcome Trust for supporting our work (Programme Grant No 067176) and The Big lottery Fund, Wolfson Trust, and the National Society for Epilepsy for supporting The NSE MRI scanner. Conflict of interest: We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. M. Yogarajah – nil J. Duncan – nil

R EFERENCES
Adcock JE, Wise RG, Oxbury JM, Oxbury SM, Matthews PM. (2003) Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy. Neuroimage 18:423–438. Alexander DC. (2005) Multiple-fiber reconstruction algorithms for diffusion MRI. Ann NY Acad Sci 1064:113–133. Arfanakis K, Hermann BP, Rogers BP, Carew JD, Seidenberg M, Meyerand ME. (2002) Diffusion tensor MRI in temporal lobe epilepsy. Magn Reson Imaging 20:511–519. Assaf BA, Mohamed FB, Abou-Khaled KJ, Williams JM, Yazeji MS, Haselgrove J, Faro SH. (2003) Diffusion tensor imaging of the hippocampal formation in temporal lobe epilepsy. AJNR Am J Neuroradiol 24:1857–1862. Basser PJ. (1995) Inferring microstructural features and the physiological state of tissues from diffusion-weighted images. NMR Biomed 8(7– 8):333–344. Basser PJ, Mattiello J, Le Bihan D. (1994) Estimation of the effective self-diffusion tensor from the NMR spin echo. J Magn Reson B 103(3):247–254. Beaulieu C. (2001) The basis of anisotropic water diffusion in the nervous system—a technical review. NMR Biomed 15:435–455. Bruehl C, Hagemann G, Witte OW. (1998) Uncoupling of blood flow and metabolism in focal epilepsy. Epilepsia 39:1235–1242. Calistri V, Caramia F, Bianco F, Fattapposta F, Pauri F, Bozzao L. (2003) Visualization of evolving status epilepticus with diffusion and perfusion MR imaging. AJNR Am J Neuroradiol 24:671– 673. Catani M, Howard RJ, Pajevic S, Jones DK. (2002) Virtual in vivo interactive dissection of white matter fasciculi in the human brain. Neuroimage 17:77–94.
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

199 Diffusion-based MRI and Tractography in Epilepsy
Hasegawa D, Orima H, Fujita M, Nakamura S, Takahashi K, Ohkubo S, Igarashi H, Hashizume K. (2003) Diffusion-weighted imaging in kainic acid-induced complex partial status epilepticus in dogs. Brain Res 983:115–127. Hong KS, Cho YJ, Lee SK, Jeong SW, Kim WK, Oh EJ. (2004) Diffusion changes suggesting predominant vasogenic oedema during partial status epilepticus. Seizure 13:317–321. Hufnagel A, Weber J, Marks S, Ludwig T, de Greiff A, Leonhardt G, Widmann G, Stolke D, Forsting M. (2003) Brain diffusion after single seizures. Epilepsia 44:54–63. Hugg JW, Butterworth EJ, Kuzniecky RI. (1999) Diffusion mapping applied to mesial temporal lobe epilepsy: preliminary observations. Neurology 53:173–176. Ivnik RJ, Sharbrough FW, Laws ER Jr. (1987) Effects of anterior temporal lobectomy on cognitive function. J Clin Psychol.43:128–137. Kamada K, Todo T, Morita A, Masutani Y, Aoki S, Ino K, Kawai K, Kirino T. (2005) Functional monitoring for visual pathway using realtime visual evoked potentials and optic-radiation tractography. Neurosurgery 57:121–127. Kantarci K, Shin C, Britton JW, So EL, Cascino GD, Jack CR Jr. (2002) Comparative diagnostic utility of 1H MRS and DWI in evaluation of temporal lobe epilepsy. Neurology 58:1745–1753. Kikuta K, Takagi Y, Nozaki K, Hanakawa T, Okada T, Miki Y, Fushimi Y, Fukuyama H, Hashimoto N. (2006) Early experience with 3-T magnetic resonance tractography in the surgery of cerebral arteriovenous malformations in and around the visual pathway. Neurosurgery 58:331–337. Kim JA, Chung JI, Yoon PH, Kim DI, Chung TS, Kim EJ, Jeong EK. (2001) Transient MR signal changes in patients with generalized tonicoclonic seizure or status epilepticus: periictal diffusion-weighted imaging. AJNR Am J Neuroradiol 22:1149–1160. Konermann S, Marks S, Ludwig T, Weber J, de Greiff A, Dorfler A, Leonhardt G, Wiedemayer H, Diener HC, Hufnagel A. (2003) Presurgical evaluation of epilepsy by brain diffusion: MR-detected effects of flumazenil on the epileptogenic focus. Epilepsia 44:399–407. Lansberg MG, O’Brien MW, Norbash AM, Moseley ME, Morrell M, Albers GW. (1999) MRI abnormalities associated with partial status epilepticus. Neurology 52:1021–1027. Le Bihan D, Turner R. (1992) The capillary network: a link between IVIM and classical perfusion. Magn Reson Med 27:171–178. Le Bihan D, Breton E, Lallemand D, Aubin M, Vignaud J, Laval-Jenatet M. (1992) Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 168:497–505. Le Bihan D. (1995) Molecular diffusion, tissue microdynamics and microstructure. NMR Biomed 8:375–386. Le Bihan D, Mangin JF, Poupon C, Clark CA, Pappata S, Molko N, Chabriat H. (2001) Diffusion tensor imaging: concepts and Applications. J Magn Reson Imaging 13:534–546. Le Bihan D. (2003) Looking into the functional architecture of the brain with diffusion MRI. Nat Neurosci 4:469–480. Lee JH, Chung CK, Song IC, Chang KH, Kim HJ. (2004) Limited utility of interictal apparent diffusion coefficient in the evaluation of hippocampal sclerosis. Acta Neurol Scand 110:53–58. Londono A, Castillo M, Lee YZ, Smith JK. (2003) Apparent diffusion coefficient measurements in the hippocampi in patients with temporal lobe seizures. AJNR Am J Neuroradiol 24:1582–1586. Lui YW, Nusbaum AO, Barr WB, Johnson G, Babb JS, Orbach D, Kim A, Laliotis G, Devinsky O. (2005) Correlation of apparent diffusion coefficient with neuropsychological testing in temporal lobe epilepsy. AJNR Am J Neuroradiol 26:1832–1839. Lux HD, Heinemann U, Dietzel I. (1986) Ionic changes and alterations in the size of the extracellular space during epileptic activity. Adv Neurol 44:619–639. Manji H, Plant GT. (2000) Epilepsy surgery, visual fields, and driving: a study of the visual field criteria for driving in patients after temporal lobe epilepsy surgery with a comparison of Goldmann and Esterman perimetry. J Neurol Neurosurg Psychiatry 68:80–82. McNamara JO. (1994) Cellular and molecular basis of epilepsy. J Neurosci 14:3413–3425. Merboldt KD, Hanicke W, Frahm J. (1985) Self-diffusion NMR imaging using stimulated echoes. J Magn Reson 64:479–486. Mori S, van Zijl PC. (2002) Fiber tracking: principles and strategies – a technical review. NMR Biomed 15:468–480. Nedelcu J, Klein MA, Aguzzi A, Boesiger P, Martin E. (1999) Biphasic edema after hypoxic-ischemic brain injury in neonatal rats reflects early neuronal and late glial damage. Pediatr Res 46:297–304. Nitsch C, Klatzo I. (1983) Regional patterns of blood-brain barrier breakdown during epileptiform seizures induced by various convulsive agents. J Neurol Sci 59:305–322. Nunes RG, Jezzard P, Behrens TE, Clare S. (2005) Self-navigated multishot echo-planar pulse sequence for high-resolution diffusionweighted imaging. Magn Reson Med 53:1474–1478. Oh JB, Lee SK, Kim KK, Song IC, Chang KH. (2004) Role of immediate postictal diffusion-weighted MRI in localizing epileptogenic foci of mesial temporal lobe epilepsy and non-lesional neocortical epilepsy. Seizure 13:509–516. Parker GJ, Alexander DC. (2003) Probabilistic Monte Carlo based mapping of cerebral connections utilising whole-brain crossing fibre information. Inf Process Med Imaging 18:684–695. Parker GJ, Alexander DC. (2005) Probabilistic anatomical connectivity derived from the microscopic persistent angular structure of cerebral tissue. Philos Trans R Soc Lond B Biol Sci 360:893–902. Perrin M, Poupon C, Rieul B, Leroux P, Constantinesco A, Mangin JF, Lebihan D. (2005) Validation of q-ball imaging with a diffusion fibrecrossing phantom on a clinical scanner. Philos Trans R Soc Lond B Biol Sci 360:881–891. Pierpaoli C, Jezzard P, Basser PJ, Barnett A, Di Chiro G. (1996) Diffusion tensor MR imaging of the human brain. Radiology 201(3):637– 648. Pitkanen A, Nissinen J, Nairismagi J, Lukasiuk K, Grohn OH, Miettinen R, Kauppinen R. (2002) Progression of neuronal damage after status epilepticus and during spontaneous seizures in a rat model of temporal lobe epilepsy. Prog Brain Res 135:67–83. Powell HW, Parker GJ, Alexander DC, Symms MR, Boulby PA, WheelerKingshott CA, Barker GJ, Koepp MJ, Duncan JS. (2005) MR tractography predicts visual field defects following temporal lobe resection. Neurology 65:596–599. Powell HW, Parker GJ, Alexander DC, Symms MR, Boulby PA, WheelerKingshott CA, Barker GJ, Koepp MJ, Duncan JS. (2007a) Abnormalities of language networks in temporal lobe epilepsy. Neuroimage 15;36(1):209–221. Powell HW, Richardson MP, Symms MR, Boulby PA, Thompson PJ, Duncan JS, Koepp MJ. (2007b) Reorganization of verbal and nonverbal memory in Temporal Lobe Epilepsy dut to unilateral hippocampal sclerosis. Epilepsia [Epub ahead of print]: Prichard JW, Zhong J, Petroff OA, Gore JC. (1995) Diffusion-weighted NMR imaging changes caused by electrical activation of the brain. NMR Biomed 8:359–364. Righini A, Pierpaoli C, Alger JR, Di Chiro G. (1994) Brain parenchyma apparent diffusion coefficient alterations associated with experimental complex partial status epilepticus. Magn Reson Imaging 12:865– 871. Rugg-Gunn FJ, Eriksson SH, Symms MR, Barker GJ, Duncan JS. (2001) Diffusion tensor imaging of cryptogenic and acquired partial epilepsies. Brain 124:627–636. Rugg-Gunn FJ, Eriksson SH, Symms MR, Barker GJ, Thom M, Harkness W, Duncan JS. (2002) Diffusion tensor imaging in refractory epilepsy. Lancet 359:1748–1751. Salmenpera TM, Simister RJ, Bartlett P, Symms MR, Boulby PA, Free SL, Barker GJ, Duncan JS. (2006a) High-resolution diffusion tensor imaging of the hippocampus in temporal lobe epilepsy. Epilepsy Res 71:102–106. Salmenpera TM, Symms MR, Boulby PA, Barker GJ, Duncan JS. (2006b) Postictal diffusion weighted imaging. Epilepsy Res 70:133–143. Spiers HJ, Burgess N, Maguire EA, Baxendale SA, Hartley T, Thompson PJ, O’Keefe J. (2001) Unilateral temporal lobectomy patients show lateralized topographical and episodic memory deficits in a virtual town. Brain 124:2476–2489. Szabo K, Poepel A, Pohlmann-Eden B, Hirsch J, Back T, Sedlaczek O, Hennerici M, Gass A. (2005) Diffusion-weighted and perfusion MRI demonstrates parenchymal changes in complex partial status epilepticus. Brain 128:1369–1376. Tanaka T, Tanaka S, Fujita T, Takano K, Fukuda H, Sako K, Yonemasu Y. (1992) Experimental complex partial seizures induced by a microinjection of kainic acid into limbic structures. Prog Neurobiol 38:317– 334.
Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x

200 M. Yogarajah and J. S. Duncan
Taylor DG, Bushell MC. (1985) The spatial mapping of translational diffusion coefficients by the NMR imaging technique. Phys Med Biol 30:345–349. Thivard L, Adam C, Hasboun D, Clemenceau S, Dezamis E, Lehericy S, Dormont D, Chiras J, Baulac M, Dupont S. (2006) Interictal diffusion MRI in partial epilepsies explored with intracerebral electrodes. Brain 129:375–385. Thivard L, Hombrouck J, du Montcel ST, Delmaire C, Cohen L, Samson S, Dupont S, Chiras J, Baulac M, Lehericy S. (2005a) Productive and perceptive language reorganization in temporal lobe epilepsy. Neuroimage 24:841–851. Thivard L, Lehericy S, Krainik A, Adam C, Dormont D, Chiras J, Baulac M, Dupont S. (2005b) Diffusion tensor imaging in medial temporal lobe epilepsy with hippocampal sclerosis. Neuroimage 28:682–690. Tournier JD, Calamante F, Gadian DG, Connelly A. (2004) Direct estimation of the fiber orientation density function from diffusion-weighted MRI data using spherical deconvolution. Neuroimage 23:1176–1185. Tuch DS. (2004) Q-ball imaging. Magn Reson Med 52:1358–1372. Tuch DS, Reese TG, Wiegell MR, Makris N, Belliveau JW, Wedeen VJ. (2002) High angular resolution diffusion imaging reveals intravoxel white matter fiber heterogeneity. Magn Reson Med 48:577–582. van der Toorn A, Sykova E, Dijkhuizen RM, Vorisek I, Vargova L, Skobisova E, van Lookeren CM, Reese T, Nicolay K. (1996) Dynamic changes in water ADC, energy metabolism, extracellular space volume, and tortuosity in neonatal rat brain during global ischemia. Magn Reson Med 36:52–60. Von Oertzen J, Urbach H, Jungbluth S, Kurthen M, Reuber M, Fernandez G, Elger CE. (2002) Standard magnetic resonance imaging is inadequate for patients with refractory focal epilepsy. J Neurol Neurosurg Psychiatry 73:643–647. Wang Y, Majors A, Najm I, Xue M, Comair Y, Modic M, Ng TC. (1996) Postictal alteration of sodium content and apparent diffusion coefficient in epileptic rat brain induced by kainic acid. Epilepsia 37:1000– 1006. Wasterlain CG, Fujikawa DG, Penix L, Sankar R. (1993) Pathophysiological mechanisms of brain damage from status epilepticus. Epilepsia 34 (Suppl 1):S37–S53. Wehner T, Lapresto E, Tkach J, Liu P, Bingaman W, Prayson RA, Ruggieri P, Diehl B. (2007) The value of interictal diffusion-weighted imaging in lateralizing temporal lobe epilepsy. Neurology 68:122– 127. Wieshmann UC, Symms MR, Shorvon SD. (1997) Diffusion changes in status epilepticus. Lancet 350:493–494. Wieshmann UC, Clark CA, Symms MR, Barker GJ, Birnie KD, Shorvon SD. (1999a) Water diffusion in the human hippocampus in epilepsy. Magn Reson Imaging 17:29–36. Wieshmann UC, Clark CA, Symms MR, Franconi F, Barker GJ, Shorvon SD. (1999b) Reduced anisotropy of water diffusion in structural cerebral abnormalities demonstrated with diffusion tensor imaging. Magn Reson Imaging 17:1269–1274. Wieshmann UC, Symms MR, Clark CA, Lemieux L, Parker GJ, Barker GJ, Shorvon SD. (1999c) Blunt-head trauma associated with widespread water-diffusion changes. Lancet 353:1242–1243. Yamamoto T, Yamada K, Nishimura T, Kinoshita S. (2005) Tractography to depict three layers of visual field trajectories to the calcarine gyri. Am J Ophthalmol 140:781–785. Yoo SY, Chang KH, Song IC, Han MH, Kwon BJ, Lee SH, Yu IK, Chun CK. (2002) Apparent diffusion coefficient value of the hippocampus in patients with hippocampal sclerosis and in healthy volunteers. AJNR Am J Neuroradiol 23:809–812. Yu CS, Li KC, Xuan Y, Ji XM, Qin W. (2005) Diffusion tensor tractography in patients with cerebral tumors: a helpful technique for neurosurgical planning and postoperative assessment. Eur J Radiol 56:197– 204. Zhong J, Petroff OA, Prichard JW, Gore JC. (1993) Changes in water diffusion and relaxation properties of rat cerebrum during status epilepticus. Magn Reson Med 30:241–246. Zhong J, Petroff OA, Prichard JW, Gore JC. (1995) Barbiturate-reversible reduction of water diffusion coefficient in flurothyl-induced status epilepticus in rats. Magn Reson Med 33:253–256. Zhong J, Petroff OA, Pleban LA, Gore JC, Prichard JW. (1997) Reversible, reproducible reduction of brain water apparent diffusion coefficient by cortical electroshocks. Magn Reson Med 37:1– 6.

Epilepsia, 49(2):189–200, 2008 doi: 10.1111/j.1528-1167.2007.01378.x