Brain, Vol. 124, No. 2, 427-436,
February 2001
© 2001 Oxford University Press
Quantitative short echo time proton magnetic resonance spectroscopic imaging study of malformations of cortical development causing epilepsy
1 The MRI Unit, National Society of Epilepsy and Epilepsy Research Group, University Department of Clinical Neurology and 2 NMR Research Unit, Institute of Neurology, University College London, London, UK
Correspondence to:
Professor J. S. Duncan, National Society for Epilepsy, Chalfont St Peter, Gerrards Cross, Bucks SL9 0RJ, UK E-mail: j.duncan{at}ion.ucl.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In patients with malformations of cortical development (MCD), widespread structural abnormalities of the brain have been demonstrated using volumetric MRI, and associated with poor post-surgical outcome in patients with localization-related epilepsy. Proton magnetic resonance spectroscopic imaging (1H-MRSI) studies permit the non-invasive measurement of concentrations of a variety of cerebral metabolites implicated in cerebral structure and function. There is a dearth of quantitative 1H-MRSI studies of MCD. Ten controls and 10 patients with localization-related epilepsy who were found to have MCD on high resolution MRI underwent 1H-MRSI on a 1.5 T GE Signa scanner [TE (echo time) = 30 ms, TR (repetition time) = 3 s]. In all patients, the axial area studied contained lesional and perilesional tissue. In seven unilaterally affected patients, the area studied contained also apparently normal contralateral grey and white matter; in three patients with bilateral but asymmetrical MCD, it contained visually normal and abnormal tissue from both hemispheres. N-acetyl aspartate + N-acetyl aspartyl glutamate (NAA), creatine + phosphocreatine (Cr), choline-containing compounds (Cho), glutamate + glutamine (Glx) and myo-inositol (Ins) were automatically quantified in voxels covering these different regions. Metabolite concentrations were corrected for CSF content and correlated with the grey and white matter of the MRSI voxels. In control subjects, there were significant positive correlations between grey matter content and concentrations of NAA, Glx, Ins and Cr. Compared with a normal range that took grey matter content into account, defined as the control mean ± 2 SD, all lesions but one showed metabolic abnormalities. The most common abnormality was a decrease in NAA, but findings were heterogeneous and there was increased NAA in one lesion. Perilesional tissue was abnormal in eight patients, with increased NAA in three. Tissue contralateral to the main MCD was abnormal in all three patients with bilateral but asymmetrical MCD, and in six of the seven apparently unilaterally affected patients. Spectroscopic grey and white matter abnormalities in patients with MCD exceeded the apparently focal abnormality shown by MRI, indicating widespread abnormalities of cerebral function.
epilepsy; 1H-MRSI; human brain; malformations of cortical development
Cho = choline-containing compounds; Cr = creatine + phosphocreatine; FOV = field of view; Glx = glutamate + glutamine; Ins = myo-inositol; MCD = malformations of cortical development; MRSI = magnetic resonance spectroscopic imaging; NAA = N-acetyl aspartate + N-acetyl aspartyl glutamate; PRESS = point-resolved spectroscopy; TE = echo time; TI = inversion time; TR = repetition time
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Malformations of cortical development (MCD) are increasingly recognized as a cause of epilepsy; they are identified in up to 60% of children and 20% of adults undergoing epilepsy surgery (Palmini et al., 1991a
; Raymond et al., 1995). Results of surgical treatment depend on the extent of the MCD (Palmini et al., 1991b). Poorer surgical outcome is likely to be the result of MCD being more widespread and extending beyond the visualized lesion (Sisodiya et al., 1995; Sisodiya, 2000).
Few proton magnetic resonance spectroscopy studies have been performed in patients with MCD. These studies have been performed on magnetic resonance scanners with high field strengths not commonly used clinically (Kuzniecky et al., 1997), in single voxels restricting the region investigated (Castillo et al., 1993; Hanefeld et al., 1995; Simone et al., 1999), with long echo times (TE) restricting the number of measured metabolites (Marsh et al., 1996; Li et al., 1998; Simone et al., 1999), or using pattern recognition or intensity ratios to discriminate diseased from normal grey and white matter, without absolute metabolite quantitation (Castillo et al., 1993; Kuzniecky et al., 1997; Marsh et al., 1996; Li et al., 1998).
Metabolite quantitation is particularly important because the common assumptions involved in the use of ratios may not hold true in this patient group. Ratios to creatine can be difficult to interpret even in normals, due to the strong correlation between creatine concentration and grey matter content; in the visibly abnormal tissues of MCD, it is not known whether the creatine content is similar to normal grey or white matter, or neither. Additionally, in the analysis of semi-quantitative data from focal lesions, it is often assumed that the contralateral hemisphere is normal. However, in MCD this is not a safe assumption (Sisodiya et al., 1995). The extent of metabolic abnormality beyond the limits of the apparent focal lesion is a particular focus of this study.
The use of short echo times not only improves quantitation by reducing the effects of differential T2-weighting of metabolite signals, it also allows the investigation of the distribution of two important metabolic markers: glutamate + glutamine (Glx) and myo-inositol (Ins). Since MCD is often associated with epilepsy, the concentrations of the neurotransmitter glutamate and related compounds are of particular interest, and this is the first study to attempt to estimate the Glx concentration over a wide area in this patient group. Myo-inositol is thought to be largely glial in origin, so an increase in its concentration might reflect glial proliferation (Brand et al., 1993). Recently, we have reported an elevation in Ins in hippocampal sclerosis (Woermann et al., 1996).
The aim of this study was to quantitate metabolites in visually abnormal grey matter and adjacent as well as remote apparently normal white and grey matter in patients with partial seizures and MCD, using proton magnetic resonance spectropic imaging (1H-MRSI) with quantitation of metabolites on a 1.5 T MR scanner.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Subjects
Ten control subjects (six males, four females; median age 34.5 years, range 20–45 years) and 10 patients with chronic, medically intractable localization-related epilepsy and frontoparietal malformations of cortical development on MRI (six men, four women; median age 43.5 years, range 22–57 years) were investigated using a 1.5 T GE Signa scanner and a standard quadrature head coil (GE, Milwaukee, Wis., USA). Informed consent was obtained from subjects prior to scanning, and the project had the approval of the combined medical ethics committees of the National Hospital for Neurology and Neurosurgery and the Institute of Neurology, London. The age at scanning did not differ significantly between the two groups. Clinical features are summarized in Table 1
.
|
Structural image acquisition
Studies were performed on a 1.5 T GE Signa scanner using the standard birdcage head coil. Axial T1-weighted images were acquired using an IRP-FSPGR (inversion-recovery-prepared fast spoiled gradient echo sequence) [repetition time (TR) = 15.9 ms, TE = 4.2 ms, inversion time (TI) = 450 ms, flip angle 20°, matrix size 256 x 128, field of view (FOV) 28 x 28 cm, 124 slices, slice thickness 1.5 mm, scan time 5 min 47 s].
Spectroscopic image acquisition
Volume-selective, two-dimensional 1H-MRSI was performed, using the above T1-weighted images as scouts, on a 15 mm thick slab superior to the ventricles [point-resolved spectroscopy (PRESS) localization; TE = 30 ms, TR = 3000 ms]. The nominal voxel size was 2.04 cm3 (24 x 24 phase encode steps over a 28 cm FOV). Scan time was 29 min.
Post-processing
We analysed the data using a locally developed program, which allowed the user to define regions of interest by drawing with a cursor on an axial reference image (McLean et al., 1999; Fig. 1). In the patients, drawn regions were centred over lesional, perilesional and contralateral apparently normal grey and white matter areas separately, as identified visually on axial T1-weighted images. In control subjects, selected spectroscopic imaging voxels were centred over areas with predominately grey or white matter on visual inspection and over areas containing a mixture of both tissues. The program then extracted voxels that were shifted to be aligned with the drawn region for subsequent quantitative analysis. For each voxel, metabolites were quantitated using the frequency-domain fitting routine LCModel calibrated to an external standard (Provencher, 1993). No T1 and T2 corrections were applied as these would be small given the short TE and long TR used.
|
The three-dimensional T1-weighted axial IRP-FSPGR data set was automatically segmented using SPM96 (Wellcome Department of Cognitive Neurology, Institute of Neurology, London, UK) to estimate grey matter, white matter and CSF content (Fig. 2
) (Woermann et al., 1999a). The grey matter, white matter and CSF distribution image slices corresponding to the MRSI slab were extracted, combined and reduced to 1H-MRSI resolution to estimate the CSF, grey matter and white matter content of each voxel. Each voxel's metabolite concentrations were normalized to 100% brain tissue by multiplying measured metabolites by [1/(1–CSF)] (McLean et al., 2000). This approach assumed no metabolites in CSF detectable by clinical proton spectroscopy (Lynch et al., 1993). The individual's metabolite information from voxels overlying the same areas was averaged. Since we have found previously a marked dependence of metabolite concentration on grey matter composition of voxels (McLean et al., 2000), the metabolites were analysed with grey matter content as a covariant, using ANCOVA (analysis of covariance) in SPSS version 7.
|
We also produced a chemical–shift–artefact-corrected estimate of the efficiency of the PRESS excitation at each location (McLean et al., 2000
). If the efficiency for any of the metabolites studied was <95%, the voxel was not included in the analysis. |
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Control subjects
In individual control subjects, the number of voxels averaged for each region (overlying predominantly grey matter, white matter or areas containing a mixture of both) ranged from two to six voxels. The total number of voxels in the group was 115, of which five had to be rejected on the basis of the PRESS excitation profiles. The mean grey matter content, expressed as the fraction grey/(grey + white), for the grey, white and mixed tissue regions was 0.82 ± 0.09, 0.06 ± 0.03 and 0.31 ± 0.10, respectively (Table 2
).
|
Displaying metabolite concentration as a function of the grey matter content of the MRSI voxels revealed a positive correlation of all measured metabolites with the grey matter content apart from choline-containing compounds (Cho) [Fig. 3A
, N-acetyl aspartate + N-acetyl aspartyl glutamate (NAA)]. Correlation coefficients of metabolite concentration to grey matter content were NAA r2 = 0.8 (P < 0.0001), Glx r2 = 0.84 (P < 0.0001), Ins r2 = 0.74 (P < 0.0001), Cho r2 = 0.01 (P > 0.05) and creatine + phosphocreatine (Cr) r2 = 0.86 (P < 0.0001). There was a negative correlation between grey matter content and the ratio of NAA to Cr (r2 = 0.55, P < 0.0001).
|
A normal range consisting of the mean concentration ±2 SD of each metabolite was constructed using the correlation curve representing the slope of the above correlations over the whole range of grey/white matter content. The standard error of the intercept [at gm (grey matter) = 0] and the number of controls was used to calculate 2 SD, as indicated by the dotted lines in Fig. 3
.
Patients
Six patients had gyral abnormalities (two of which were bilateral) and four patients had heterotopias, with one patient being affected bilaterally. None have been resected at the present time.
In individual patients, the number of voxels overlying each region (lesional, perilesional, contralateral less affected or apparently normal tissue) ranged from one to four voxels. The total number of voxels in the group was 105, of which six had to be rejected on the basis of the PRESS excitation profiles. The mean grey matter content, expressed as the fraction grey/(grey + white), for the lesional, perilesional, contralateral grey and contralateral white regions was 0.55 ± 0.2, 0.12 ± 0.09, 0.58 ± 0.10 and 0.10 ± 0.07, respectively.
ANCOVA tests with grey matter fraction as a covariant were needed to compare control and patient data: ANOVA tests without grey matter as a covariant found P < 0.001 for all metabolites studied except choline. With grey matter as a covariant, only the ratio NAA/Cr remained highly significant, although NAA and inositol concentrations were also significant at P < 0.05 (Table 2).
Spectroscopic abnormalities were identified by comparing metabolite concentrations (expressed as a function of grey/white matter content) in patients with the normal range. Abnormal results of individual metabolites and the ratio NAA/Cr are displayed in Table 3. Abnormal results of single metabolites in individual patients were found in all but one MCD (including all bilateral cases), in all but one of the perilesional regions and in six of seven contralateral apparently normal areas overlying grey or white matter. Comparing the number of abnormal metabolites, lesional and perilesional tissues were most frequently abnormal, while apparently normal contralateral white matter showed abnormalities to a lesser degree compared with apparently normal grey matter (Fig. 3B and C).
|
Decreased NAA was the abnormality most frequently found, but increases in NAA were also seen (Fig. 3B and C
). There was no single combination of individual abnormal metabolites that characterized lesional, perilesional or remote tissue.
All three patients with bilateral MCD detected visually had bilateral metabolic abnormalities (Table 3). NAA was decreased in two patients on the visually more markedly affected side and increased in one case. On the visually less affected side, there were two increases and one decrease of NAA. Four of the seven patients with apparently unilateral MCD had reduced lesional NAA concentrations; two unilaterally affected patients had increased NAA and one had decreased NAA in the perilesional regions.
There was no consistent pattern of results in different forms of MCD. Of the three patients with polymicrogyria, the lesions in two had low NAA and one was unremarkable. The perilesional area was normal in one and two had reduced NAA or a reduced NAA/Cr ratio. Distant grey matter was normal in one, one had increased NAA, Glx and Cr and one had reduced NAA and Cr. Distant white matter had increased Cr in one and reduced NAA in two.
One of the three macrogyrias showed reduced NAA, one increased Glx and one increased NAA and Glx. Perilesional tissue was normal in one, one had increased Glx and one had increased NAA, Glx and Cr. Distant grey matter was normal in one, NAA was increased in one and NAA and Cho were reduced in one. Distant white matter was normal in one, had reduced NAA in one and increased NAA, Glx and Cr in one.
A subependymal nodule had reduced NAA in the lesion, increased NAA in perilesional tissue and distant grey matter, and increased Cr in distant white matter.
Of the three schizencephalic lesions, one had reduced NAA, one had reduced NAA, Glx and Cr and one had increased Cr. Perilesional tissue had increased NAA in one and reduced NAA in two. One of the latter had raised Cr and one reduced Cr. Distant grey matter had elevated NAA and Cr in one, reduced NAA and Cho in one and reduced Glx in Cho in the third. Distant white matter was normal in one, had reduced NAA and Glx in one and reduced Glx and Cho in one.
All in all, a reduction of NAA characterized main lesions in six out of 10 MCD patients. One patient had no lesional metabolic abnormality. Perilesional NAA was significantly increased in three patients but was significantly decreased in three more patients. In seven unilaterally affected patients, 1H-MRSI voxels covering apparently normal grey matter detected reduced NAA in two patients and increased NAA in another two patients. Apparently normal contralateral white matter was characterized by abnormally low NAA in three of these patients.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Using absolute quantitation and short echo time 1H-MRSI, we demonstrated abnormal absolute metabolite concentrations in MCD, perilesional tissue and brain tissue remote from MCD in patients with localization-related epilepsy. These findings support the concept of widespread abnormalities in patients with apparently focal MCD. Spectroscopic abnormalities do not necessarily represent widespread structural changes as described earlier (Sisodiya et al., 1995
), but might demonstrate dysfunction. Measurements of individual metabolites were abnormal in some malformations and normal in others, suggesting metabolic heterogeneity in MCD.
Methodological considerations
Compared with single voxel spectroscopy, 1H-MRSI is a method for collecting spectroscopic data from adjacent voxels covering a large area of brain tissue in a single study. Multidimensional Fourier transformation yields localized spectral data that can be examined in different ways: as single spectra related to individual voxels, as spectral maps or as metabolite images. We chose the absolute quantitation of spectral data from single voxels to avoid the ambiguity of metabolite ratios or of visual assessment of metabolite maps used in other studies (Castillo et al., 1993; Kuzniecky et al., 1997; Marsh et al., 1996; Li et al., 1998).
Spatial resolution was limited to 2.04 cm3 to obtain reasonable signal-to-noise in a total scan duration of ~60 min (including imaging). Volume-selective two-dimensional 1H-MRSI was used to avoid unwanted signals from the subcutaneous fat of the scalp. Enclosing a rectangular MRSI slab inside the brain limited the examination of peripheral grey matter. Normal grey matter was usually examined in the medial interhemispheric area. All lateral cortical MCD in this study had components that extended medially and were covered by the spectroscopic slab. As a consequence of these limitations, however, we did not examine `pure' cerebral cortex adjacent to MCD (Palmini et al., 1995a, b). It is possible, therefore, that our approach missed the parts of the MCD that were metabolically most abnormal, as these may have been situated more peripherally than the 1H-MRSI slab. However, 1H-MRSI with alternative means of lipid suppression was used in patients with MCD to cover peripheral grey matter, but showed similarly heterogeneous metabolite concentrations in MCD compared with our study (Kuzniecky et al., 1997).
1H-MRSI offers the unique advantage of allowing adjustment in voxel positioning during post-processing. This is done by phase multiplication prior to the spatial Fourier transform. It is also possible to combine adjacent voxels to replicate, as closely as possible, the shape of a structure before adding the corresponding spectra. We recently demonstrated the feasibility of this approach to clinical 1H-MRSI (McLean et al., 1999) and used it in this study to maximize content of one tissue type or another in each region analysed.
Even with high spatial resolution spectroscopy, most spectra will arise from 1H-MRSI voxels that contain a mixture of grey and white matter and CSF, adding ambiguity to the interpretation of spectral data. To address this problem, automated segmentation of cerebral T1-weighted MRI indicated the amounts of CSF, grey matter and white matter in every voxel (Woermann et al., 1999a; McLean et al., 2000). Measured metabolite concentrations were corrected for CSF content of the voxel, on the basis that CSF contains no MRSI-detectable metabolites (Lynch et al., 1993; Hetherington et al., 1996). These concentrations were plotted as a function of tissue composition and used to construct a normal range over the whole range of grey and white matter concentrations (Fig. 3). We used this approach to allow assessment of abnormality in voxels encompassing the whole range of the neocortex, including heterotopic grey matter and cortical malformations.
Results in controls
We found positive correlations between grey matter content of the MRSI voxel and NAA, Cr, Glx and Ins concentrations in controls. There was no significant difference in Cho between voxels containing more grey or white matter. Similar studies combining quantitative proton magnetic resonance spectroscopy and image segmentation have been conducted mainly in control subjects (Doyle et al., 1995; Hetherington et al., 1996; Pan et al., 1998; Wang and Li, 1998; Noworolski et al., 1999; Pfefferbaum et al., 1999). Although quantitative 1H-MRSI studies display their results as absolute concentrations, they represent to some degree `institutional units', due to differences in data acquisition (localization, TE), in standards and in correction for partial volume or for inhomogeneity. This makes the direct comparison of values between studies difficult. However, it is possible to compare grey : white matter ratios of metabolites (Noworolski et al., 1999), such as in studies that identified spectroscopy voxels overlying grey and white matter visually, i.e. without automatic segmentation and CSF correction (Hetherington et al., 1994; Tedeschi et al., 1995; Soher et al., 1996). In most control studies, NAA and Cr were found to have higher spectral peaks or concentrations in grey than in white matter. These findings corroborated results from extract studies (Peeling et al., 1993; Petroff et al., 1995). In some of the MRSI studies that found lower NAA or lower Cr in grey matter, no correction for CSF contamination was made, which most probably explains the low measures in the grey matter (Hetherington et al., 1994; Tedeschi et al., 1995; Soher et al., 1996). One group, however, using MRSI and correction for CSF demonstrated lower NAA and higher Cr concentration in control grey matter compared with metabolites of white matter (Pan et al., 1998). Similarly, in some magnetic resonance spectroscopy studies without correction for CSF, Cho was lower in grey than in white matter. In our study, using absolute quantitation and correction for CSF contamination, we found no significant difference between grey and white Cho, which was consistent with other MRSI studies (Doyle et al., 1995; Hetherington et al., 1996) and with extract studies (Petroff et al., 1995).
Biological and clinical implications
The group analysis of metabolite concentrations with grey matter content as a covariant confirmed findings from earlier studies of decreased NAA/Cr in the region of the malformation (Kuzniecky et al., 1997), though it suggests that these abnormalities may be more widespread. Quantification allowed us to investigate whether the low NAA/Cr could be due to decreased NAA, increased Cr, or both. Increases in Cr were seen in several patients, but were usually accompanied by increases in NAA. Decreased NAA was the more common finding, and group analysis suggests that NAA showed more significant variation than Cr when the grey matter content of voxels was taken into account (Table 2). Our results are in agreement with Li et al. (1998), who found abnormal ratios of NAA/Cr in cerebral areas distant from the visible MCD, but we did not find any clear pattern of metabolic abnormalities in patients with different forms of MCD.
The group analysis also suggested a significant decrease in inositol in the patient groups compared with controls. This appears to be a global effect and the cause is not clear.
In individual MCD in our study, a decrease in NAA was the most frequent finding. This is in keeping with group findings in earlier magnetic resonance spectroscopic findings in vivo and with results in individual MCD patients from these studies (Kuzniecky et al., 1997; Li et al., 1998; Simone et al., 1999). A recent in vitro magnetic resonance spectroscopy study showed a reduction in NAA in biopsies from MCD (Aasly et al., 1999). As NAA is believed to be located primarily within neurones (Moffett et al., 1991), reduced NAA suggests either neuronal loss or dysfunction (Petroff et al., 1995). The finding of low NAA in an area of increased grey matter distinguishes MCD from hippocampal sclerosis or cortical gliosis, which is characterized by a low NAA and atrophy (Woermann et al., 1999). Recent longitudinal MRSI studies before and after temporal lobectomy for hippocampal sclerosis have shown postoperative increases of NAA in the unoperated temporal lobe, suggesting that NAA might represent neuronal dysfunction rather than a fixed neuronal loss (Hugg et al., 1996; Cendes et al., 1997). Low NAA in areas depicted and segmented as grey matter, but considerably disordered on structural imaging, might therefore represent neurones that were abnormal in structure and function. There was, however, no correlation between low NAA and neurological deficit or other clinical features. The abnormally high lesional NAA in two patients with bilateral heterotopic grey matter may represent an increase of neurones in disordered and densely packed grey matter: neuronal density has been described as both increased or normal in MCD (Battaglia et al., 1996). Further correlative histological studies are needed to evaluate this.
Metabolites other than NAA were more often normal than abnormal in MCD. Lesional measurements of Glx were heterogeneous (two abnormal increases, one abnormal decrease). Glx concentration was positively correlated with grey matter content in control subjects. Normal grey matter has a greater density of synapses than white matter, so its larger metabolic pool perhaps may reflect more active turnover, related to neurotransmission. A recent study has found the rate of incorporation of 13C-labelled glucose into Glx in grey matter to be almost five times higher than in white matter (Mason et al., 1999). If the abnormal grey matter in MCD contains less densely packed active synapses than normal grey matter, a lower Glx concentration than in normal grey matter would not be surprising. In a case report, decreased glutamate in MCD was thought to indicate damage, but was hard to interpret in the absence of CSF correction (Hanefeld et al., 1995). The finding in two subjects of Glx increased above what would be predicted from the grey matter content of the voxels is intriguing. Perhaps the synapses are even more densely packed than in normal grey matter. Another possibility is that the Glx turnover in MCD is even higher than in normal grey matter, perhaps related to epileptogenesis; it is of note that we have shown an increased ipsilateral Glx in patients with temporal lobe epilepsy with no structural abnormality (Woermann et al., 1996).
In the current study, Cr was increased in three cases of MCD and was decreased in two cases of MCD. This heterogeneous finding supports the decision to use absolute concentrations rather than ratios to screen for abnormalities.
MCD are found to be heterogeneous clinically (Raymond et al., 1994), electrophysiologically (Rosenow et al., 1998), pathologically (Battaglia et al., 1996) and on structural, quantitative and functional neuroimaging (Sisodiya et al., 1995; Richardson et al., 1998). The in vivo characterization of MCD as the source of seizures, malfunctioning and normal function might be addressed by `multi modality imaging', involving a range of structural and functional data (Krakow et al., 1999).
Metabolic abnormalities in white matter surrounding MCD were almost as frequent and as heterogeneous as lesional abnormalities. A small degree of contamination among adjacent voxels due to the imperfect volume selection using MRSI would be expected to contribute to this effect; however, it cannot explain the extent of the perilesional abnormality, which sometimes exceeded that in the lesions. Perilesional increases of NAA might represent microscopic structural abnormalities outside the visible lesion, such as microdysgenesis that is not evident on MRI. Such an abnormality might be expected to give this white matter a metabolite profile that more resembled grey matter: increases in other grey-matter-correlated metabolites (Cr and Glx) were also seen in these perilesional regions. Perilesional decreases of NAA were also common, and always accompanied lesional decreases, suggesting malfunction beyond the visible lesion. Metabolic abnormalities were commonly found in visually normal grey and white matter, contralateral to MCD. These suggest widespread abnormalities of neuronal function or microdysgenesis and may explain why patients with apparently focal MCD are less likely to become seizure-free after surgical removal of the visible lesion than patients with other pathologies (Sisodiya et al., 1995). Further studies are needed to determine whether patients with focal MCD and the absence of widespread metabolic abnormality have a better prognosis for seizure remission post-surgically.
Quantitative 1H-MRSI demonstrated abnormalities beyond macroscopic MCD in areas normal on visual MRI inspection. The method may therefore also be useful in the evaluation of MRI-negative epilepsies, to identify focal abnormalities that may underlie epileptic foci and occult MCD. Such studies will be particularly useful with the advent of multislice 1H-MRSI, which will give greater coverage of the brain.
|
Acknowledgments |
|---|
The authors wish thank The National Society for Epilepsy, Chalfont St Peter, UK (F.G.W., P.A.B. and J.S.D.), SmithKline Beecham (M.A.M.) and the Multiple Sclerosis Society of Great Britain and Northern Ireland (G.J.B.) for their support of this research.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Aasly J, Silfvenius H, Aas TC, Sonnewald U, Olivecrona M, Juul R, et al. Proton magnetic resonance spectroscopy of brain biopsies from patients with intractable epilepsy. Epilepsy Res 1999; 35: 211–7.[Web of Science][Medline]
Battaglia G, Arcelli P, Granata T, Selvaggio M, Andermann F, Dubeau F, et al. Neuronal migration disorders and epilepsy: a morphological analysis of three surgically treated patients. Epilepsy Res 1996; 26: 49–58.[Web of Science][Medline]
Brand A, Richter-Landsberg C, Leibfritz D. Multinuclear NMR studies on the energy metabolism of glial and neuronal cells. Dev Neurosci 1993; 15: 289–98.[Web of Science][Medline]
Castillo M, Kwock L, Scatliff, Gudeman S, Greenwood R. Proton MR spectroscopic characteristics of a presumed giant subcortical heterotopia. AJNR Am J Neuroradiol 1993; 14: 426–9.[Abstract]
Cendes F, Andermann F, Dubeau F, Matthews PM, Arnold DL. Normalization of neuronal metabolic dysfunction after surgery for temporal lobe epilepsy. Evidence from proton MR spectroscopic imaging. Neurology 1997; 49: 1525–33.
Doyle TJ, Bedell BJ, Narayana PA. Relative concentrations of proton MR visible neurochemicals in gray and white matter in human brain. Magn Reson Med 1995; 33: 755–9.[Web of Science][Medline]
Hanefeld F, Kruse B, Holzbach U, Christen HJ, Merboldt KD, Hänicke W, et al. Hemimegalencephaly: localized proton magnetic resonance spectroscopy in vivo. Epilepsia 1995; 36: 1215–24.[Web of Science][Medline]
Hetherington HP, Mason GF Pan JW, Ponder SL, Vaughan JT, Twieg DB, et al. Evaluation of cerebral gray and white matter metabolite differences by spectroscopic imaging at 4.1 T. Magn Reson Med 1994; 32: 565–71.[Web of Science][Medline]
Hetherington HP, Pan JW, Mason GF, Adams D, Vaughn MJ, Twieg DB, et al. Quantitative 1H spectroscopic imaging of human brain at 4.1 T using image segmentation. Magn Reson Med 1996; 36: 21–9.[Web of Science][Medline]
Hugg JW, Kuzniecky RI, Gilliam FG, Morawetz RB, Fraught RE, Hetherington HP. Normalization of contralateral metabolic function following temporal lobectomy demonstrated by 1H magnetic resonance spectroscopic imaging. Ann Neurol 1996; 40: 236–9.[Web of Science][Medline]
Krakow K, Wieshmann UC, Woermann FG, Symms MR, McLean MA, Lemieux L, et al. Multimodal MR imaging: Functional, diffusion tensor, and chemical shift imaging in a patient with localization-related epilepsy. Epilepsia 1999; 40: 1459–62.[Web of Science][Medline]
Kuzniecky R, Hetherington H, Pan J, Hugg J, Palmer C, Gilliam F, et al. Proton spectroscopic imaging at 4.1 tesla in patients with malformations of cortical development and epilepsy. Neurology 1997; 48: 1018–24.[Abstract]
Li LM, Cendes F, Bastos AC, Andermann F, Dubeau F, Arnold DL. Neuronal metabolic dysfunction in patients with cortical developmental malformations: a proton magnetic resonance spectroscopic imaging study. Neurology 1998; 50: 755–9.
Lynch J, Peeling J, Auty A, Sutherland GR. Nuclear magnetic resonance study of cerebrospinal fluid from patients with multiple sclerosis. Can J Neurol Sci 1993; 20: 194–8.[Web of Science][Medline]
Marsh L, Lim KO, Sullivan EV, Lane B, Spielman D. Proton magnetic resonance spectroscopy of a gray matter heterotopia. Neurology 1996; 47: 1571–4.
Mason GF, Pan JW, Chu WJ, Newcomer BR, Zhang Y, Orr R, et al. Measurement of the tricarboxylic acid cycle rate in human grey and white matter in vivo by 1H-[13C] magnetic resonance spectroscopy at 4.1T. J Cereb Blood Flow Metab 1999; 19: 1179–88.[Web of Science][Medline]
McLean MA, Barker GJ, Tofts P, Duncan JS. Interactive ROI analysis in 1H-CSI [abstract]. Proceedings of the 7th ISMRM 1999. p. 1562.
McLean MA, Woermann FG, Barker GJ, Duncan JS. Quantitative analysis of short echo time 1H-MRSI of cerebral grey and white matter. Magn Reson Med 2000; 44: 401–11.[Web of Science][Medline]
Moffett JR, Namboodiri MA, Cangro CB, Neale JH. Immunohistochemical localization of N-acetylaspartate in rat brain. Neuroreport 1991; 2: 131–4.[Web of Science][Medline]
Noworolski SM, Nelson SJ, Henry RG, Day MR, Wald LL, Star-Lack J, et al. High spatial resolution 1H-MRSI and segmented MRI of cortical gray matter and subcortical white matter in three regions of the human brain. Magn Reson Med 1999; 41: 21–9.[Web of Science][Medline]
Palmini A, Andermann F, Olivier A, Tampieri D, Robitaille Y, Andermann E, et al. Focal neuronal migration disorders and intractable partial epilepsy: a study of 30 patients. Ann Neurol 1991a; 30: 741–9.[Web of Science][Medline]
Palmini A, Andermann F, Olivier A, Tampieri D, Robitaille Y. Focal neuronal migration disorders and intractable partial epilepsy: results of surgical treatment. [Review]. Ann Neurol 1991b; 30: 750–7.[Web of Science][Medline]
Pan JW, Twieg DB, Hetherington HP. Quantitative spectroscopic imaging of the human brain. Magn Reson Med 1998; 40: 363–9.[Web of Science][Medline]
Peeling J, Sutherland G. 1H magnetic resonance spectroscopy of extracts of human epileptic neocortex and hippocampus. Neurology 1993; 43: 589–94.
Petroff OA, Pleban LA, Spencer DD. Symbiosis between in vivo and in vitro NMR spectroscopy: the creatine, N-acetylaspartate, glutamate, and GABA content of the epileptic human brain. Magn Reson Imaging 1995; 13: 1197–211.[Web of Science][Medline]
Pfefferbaum A, Adalsteinsson E, Spielman D, Sullivan EV, Lim KO. In vivo spectroscopic quantification of the N-acetyl moiety, creatine, and choline from large volumes of brain gray and white matter: effects of normal aging. Magn Reson Med 1999; 41: 276–84.[Web of Science][Medline]
Provencher SW. Estimation of metabolite concentrations from localized in vivo proton NMR spectra. Magn Reson Med 1993; 30: 672–9.[Web of Science][Medline]
Provencher SW. LC model [cited 2000 Sep 29]. Available from: http://S-provencher.com.
Raymond AA, Fish DR, Sisodiya SM, Alsanjari N, Stevens JM, Shorvon SD. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features in 100 adult patients. [Review]. Brain 1995; 118: 629–60.
Richardson MP, Koepp MJ, Brooks DJ, Coull JT, Grasby P, Fish DR, et al. Cerebral activation in malformations of cortical development. Brain 1998; 121: 1295–304.
Rosenow F, Lüders HO, Dinner DS, Prayson RA, Mascha E, Wolgamuth BR, et al. Histopathological correlates of epileptogenicity as expressed by electrocorticographic spiking and seizure frequency. Epilepsia 1998; 39: 850–6.[Web of Science][Medline]
Simone IL, Federico F, Tortorella C, De Blasi R, Bellomo R, Lucivero V, et al. Metabolic changes in neuronal migration disorders: evaluation by combined MRI and proton MR spectroscopy. Epilepsia 1999; 40: 872–9.[Web of Science][Medline]
Sisodiya SM. Surgery for malformations of cortical development causing epilepsy. [Review]. Brain 2000; 123: 1075–91.
Sisodiya SM, Free SL, Stevens JM, Fish DR, Shorvon SD. Widespread cerebral structural changes in patients with cortical dysgenesis and epilepsy. Brain 1995; 118: 1039–50.
Soher BJ, van Zijl PC, Duyn JH, Barker PB. Quantitative proton MR spectroscopic imaging of the human brain. Magn Reson Med 1996; 35: 356–63.[Web of Science][Medline]
Tedeschi G, Bertolino A, Righini A, Campbell G, Raman R, Duyn JH, et al. Brain regional distribution pattern of metabolite signal intensities in young adults by proton magnetic resonance spectroscopic imaging. Neurology 1995; 45: 1384–91.[Abstract]
Wang Y, Li SJ. Differentiation of metabolic concentrations between gray matter and white matter of human brain by in vivo 1H magnetic resonance spectroscopy. Magn Reson Med 1998; 39: 28–33.[Web of Science][Medline]
Woermann FG, Free SL, Koepp MJ, Ashburner J, Duncan JS. Voxel-by-voxel comparison of automatically segmented cerebral gray matter—a rater-independent comparison of structural MRI in patients with epilepsy. Neuroimage 1999a; 10: 373–84.[Web of Science][Medline]
Woermann FG, McLean MA, Bartlett PA, Parker GJ, Barker GJ, Duncan JS. Short echo time single-voxel IH magnetic resonance spectroscopy in magnetic resonance imaging-negative temporal lobe epilepsy: different biochemical profile compared with hippocampal sclerosis. Ann Neurol 1999b; 45: 369–76.[Web of Science][Medline]
Received January 20, 2000. Revised July 7, 2000. Accepted September 28, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. P Richardson Epilepsy and surgical mapping Br. Med. Bull., March 1, 2003; 65(1): 179 - 192. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||