- Email: [email protected]
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study
Epilepsy Research (2010) 90, 21—27
journal homepage: www.elsevier.com/locate/epilepsyres
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study Chi Heon Kim a,c,d, Bang-Bon Koo b, Chun Kee Chung a,c,d,∗, Jong-Min Lee b, June Sic Kim a,c,d, Sang Kun Lee e a
Department of Neurosurgery, Seoul National University College of Medicine, Seoul, South Korea Department of Biomedical Engineering, Graduate School of Engineering, Hanyang University, Seoul, South Korea c Neuroscience Research Institute, Seoul National University Medical Research Center, Seoul, South Korea d Clinical Research Institute, Seoul National University Hospital, Seoul, South Korea e Department of Neurology, Seoul National University Hospital, Seoul, South Korea b
Received 12 December 2009; received in revised form 25 February 2010; accepted 1 March 2010 Available online 21 March 2010
KEYWORDS Magnetic resonance imaging; Tensor; Hippocampus; Thalamus; Temporal lobe epilepsy
Summary Objective: The seizure network may be different between temporal lobe epilepsy with hippocampal sclerosis (TLE + HS) and without HS (TLE − HS). Chronic seizure activity may alter the diffusion properties of a seizure network. The thalamus is known to have an anatomical connection to the medial temporal area and to play a role in seizure modulation. This study aimed to evaluate differences in thalamic changes between TLE + HS and TLE − HS with diffusion tensor imaging (DTI). Methods: Nine patients with TLE + HS and nine patients with TLE − HS were included in the study. All patients underwent surgery with good seizure outcomes. Hippocampal sclerosis was verified pathologically. Sixteen right-handed, normal subjects were enrolled as controls. DTI was acquired using 3.0 T MRI. The mean diffusivity (MD) and fractional anisotropy (FA) were calculated in the center of the bilateral thalamus with the DTIstudio program. Results: The MD of bilateral thalami increased in both TLE groups compared to controls (p < 0.05), while FA values did not differ from controls. The MD of the thalamus ipsilateral to the epileptogenic side was higher in the TLE + HS group than in the TLE − HS group (p = 0.007). Onset age, seizure duration, seizure frequency and total seizure number were not correlated with FA and MD changes (p > 0.05). Conclusion: Bilateral thalamic diffusion properties are altered in temporal lobe epilepsy. The presence of hippocampal sclerosis enhances the change ipsilaterally. © 2010 Elsevier B.V. All rights reserved.
∗ Corresponding author at: Department of Neurosurgery, Seoul National University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, South Korea. Tel.: +82 2 2072 2352/2358; fax: +82 2 744 8459. E-mail address: [email protected] (C.K. Chung).
0920-1211/$ — see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2010.03.002
22
Introduction Chronic seizure activity may alter the diffusion properties of a seizure network (Newberg et al., 2000; Rugg-Gunn et al., 2001; Arfanakis et al., 2002; Yoo et al., 2002; Choi et al., 2003; Natsume et al., 2003; Thivard et al., 2005; Gross et al., 2006; Kimiwada et al., 2006; Mueller et al., 2006). The seizure network is wider than the seizure onset zone (Concha et al., 2005). Diffusion tensor imaging (DTI) reflects subtle changes that may not be visible in routine magnetic resonance (MR) images (Rugg-Gunn et al., 2001; Yoo et al., 2002; Concha et al., 2005). The seizure network that is involved in temporal lobe epilepsy with hippocampal sclerosis (TLE + HS) may be different from the network involved in TLE without hippocampal sclerosis (TLE − HS). The thalamus is known to have a strong anatomical connection to the medial temporal area and is known to modulate seizures. Consequently, there may be differences in thalamic changes in different seizure networks (Bertram and Zhang, 1999; Choi et al., 2003; Kim et al., 2003; Kimiwada et al., 2006; Labate et al., 2008; Riederer et al., 2008; Bertram, 2009). This study aimed to evaluate the differences in thalamic changes between TLE + HS and TLE − HS with DTI.
Materials and methods Patients We selected patients with temporal lobe epilepsy who underwent operations from May 2005 to August 2008. Included were patients who had more than 2 years of seizure history, who had taken DTI preoperatively with no seizures in the previous week, who were right-handed by the Edinburg Handedness Inventory (a score of less than −40 is indicative of right-handedness, a score of more than 40 is indicative of left-handedness, and scores between −40 and 40 are indicative of ambidexterity) and who obtained a postoperative seizure outcome of Engel class I (no seizure) or Engel class II (rare seizure). TLE + HS patients had histologically proven hippocampal sclerosis with no neocortical malformation in the temporal lobe, and TLE − HS patients had cortical dysplasia, as suggested by Palmini et al. (Oldfield, 1971; Palmini et al., 2004). Eighteen consecutive patients satisfied these criteria: 9 patients with TLE − HS and 9 patients with TLE + HS (Table 1). All patients scored less than −80 in the Edinburg Handedness Inventory (Oldfield, 1971). In TLE − HS patients, the seizure focus was lateralized to the right side in four patients and to the left side in five patients. In TLE + HS patients, the seizure focus was lateralized to the right side in four patients and to the left side in five patients. The mean subject age was 30 ± 7 years (median, 32; range, 20—40) in the TLE − HS group, 31 ± 8 years (median, 29; range, 20—44) in the TLE + HS group and 33 ± 3 years (median, 33; range, 30—39) in the control group. MR images were interpreted by expert neuroradiologists. Hippocampal atrophy was observed in MR images of all TLE + HS patients. There was no hippocampal atrophy in TLE − HS patients. Cortical dysplasia was suspected on T2-weighted fluid attenuation inversion recovery (FLAIR) MR image in 2 patients in the temporal lobe (Table 1). TLE + HS patients underwent
C.H. Kim et al. standard anterior temporal lobectomy with amygdalohippocampectomy. Hippocampal sclerosis was pathologically confirmed without associated neocortical malformation. TLE − HS patients underwent neocortical temporal lobe resection based on the results of subdural electrode monitoring. Pathological examination showed mild malformation of cortical development (MCD) in 2 patients and type IA cortical dysplasia in 7 patients. A seizure-free outcome (Engel class I) was achieved in 16 patients with TLE + HS or TLE − HS, and Engel class II status (rare seizures) was achieved in 2 patients with TLE − HS (Engel et al., 1993). The mean postoperative follow-up period was 28 months (range, 15—45) for TLE + HS patients and 31 months (range, 19—49) for TLE − HS patients. Sixteen right-handed, normal subjects with no history of neurological disorders were recruited as the control group (8 men and 8 women). All control subjects scored less than −80 on the Edinburg Handedness Inventory. This study was approved by the Institutional Review Board of Seoul National University Hospital (H-0712-056-230).
Diffusion tensor imaging protocol DTI was acquired before operation using a 3.0 T MR scanner (General Electric VH/i, Milwaukee, Wisconsin, USA) with a conventional head gradient coil. We used a single-shot spinecho EPI sequence. The B-factor was set at 1000 s/mm2 . The acquisition parameters were: FOV = 240 mm, matrix = 256 × 256, slice thickness/gap = 3.5 mm/0 mm, total slice number = 38, TR/TE = 10,000 ms/90 ms, and scan average = 1. To estimate the intensity and direction of the diffusion anisotropy, axial MR images with 25 non-collinear diffusion gradients and one without diffusion gradients were acquired. From 26 diffusion-weighted images, we obtained six diffusion tensor components, Dxx, Dyy, Dzz, Dxy, Dxz and Dyz, using multiple linear equations.
Data processing The digital images from all subjects were stored as digital imaging and communications in medicine (DICOM) files. The DICOM files were converted for analysis using the MRIcro program (version 1.40, University of South Carolina, Columbia SC 29208, USA, www.mricro.com). Diffusion parameter (apparent diffusion coefficient [ADC] and fractional anisotropy [FA]) maps were reconstructed from the diffusion tensor imaging using program DTIstudio (Johns Hopkins University, Baltimore, USA, www.mristudio.org). Region of interest (ROI) analysis was performed using the DTIstudio program for quantitative analysis. The ROI was marked as follows: mark fixed-sized ROI (10 voxels, diameter of 9.8 mm) at the center of the thalamus just above 10.5 mm (3 slices) from the lower margin of the thalamus on a 2-dimensional FA map (Oouchi et al., 2007). Another ROI was marked on the opposite homotopic site (Fig. 1). The mean diffusivity (MD) and FA were calculated in each ROI of the bilateral thalamus. To maintain consistency, all ROIs were marked by one investigator (CHK) who was blinded to the patient information.
Summary of patients.
Pt
Group
Age
Gender
Side
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS
37 23 20 33 20 29 36 40 32 44 29 33 20 28 20 39 36 26
Male Male Female Male Male Female Male Male Female Female Female Female Female Male Female Male Male Female
Right Left Left Left Right Left Right Left Right Right Left Left Right Right Left Left Right Left
Onset age (yr) 13 14 13 13 15 17 17 8 15 10 12 26 18 5 18 11 13 2
Seizure duration (yr)
Frequency/month
24 9 7 20 5 12 19 32 17 34 17 7 2 23 2 28 23 24
1.5 3.5 30 2 0.5 1.5 4 4 1 3.5 5 4.5 15 1 2 10 5.5 2.5
Total seizure number 432 378 2520 480 30 216 912 1536 204 1428 1020 378 360 276 48 3360 1518 720
Palmini classification
Location of CD in FLAIR
IA IA IA IA Mild MCD IA IA Mild MCD IA
Temporal lobe No No No Inferior temporal No No No No
CD, cortical dysplasia; MCD, malformation of cortical development; FLAIR: fluid attenuated inversion recovery magnetic resonance imaging.
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study
Table 1
23
24
Figure 1 Location of region of interest (ROI). ROI was marked in the center of both thalami on the 2dimensional FA map. The size of the ROI was fixed with 10 voxels (diameter of 9.8 mm). This axial section is at the level of 3 slices (10.5 mm), above the inferior margin of the thalamus. ROI, region of interest; FA, fractional anisotropy.
C.H. Kim et al.
Figure 2 Fractional anisotrophy values of subjects. The numbers in the graph represent p-values. The Wilcoxonrank test and Mann—Whitney U-test were used. Right FA values of patients and control subject were corrected with regard to control subjects, and only the side ipsilateral or contralateral to the seizure focus was considered; corrected right FA value = (measured right FA value of patient) × (mean left FA of control/mean right FA of control). The ips FA was lower, Wilcoxon-rank test) than the con FA in both TLE − HS (p = 0.015, Wilcoxon-rank test) and TLE + HS (p < 0.001, Wilcoxon-rank test). FA, fractional anisotropy; MD, mean diffusivity; TLE − HS, temporal lobe epilepsy without hippocampal sclerosis; TLE + HS, temporal lobe epilepsy with hippocampal sclerosis; ips , ipsilateral side to seizure focus; con , contralateral side to seizure focus.
Analysis Thirty-four subjects (16, control; 9, TLE + HS; 9, TLE − HS) participated in the present study. The subjects were divided into left TLE + HS, right TLE + HS, left TLE − HS and right TLE − HS. This study focused on comparing the differences between ipsilateral and contralateral sides to the seizure focus. To abolish inherent hemispheric differences, we corrected the right side value of both patients and control subjects with regard to control subject values. After correction, only the ipsilateral and contralateral sides to the seizure focus could be considered, therefore patients were divided into 2 groups (TLE + HS and TLE − HS). There was a hemispheric difference in FA but no difference in MD, so FA was corrected. Since there was a hemispheric difference in FA in control subjects, the mean hemispheric ratio was used for correction in patients and control subjects; the corrected right FA of a subject = (measured right FA of individual subject) × (mean left FA of control subjects/mean right FA of control subjects). The Wilcoxon-rank test was used for comparisons between hemispheres in same group. The Mann—Whitney’s U-test and Kruskal—Wallis test were used for comparisons between groups. Spearman’s test was used for correlation analysis. Comparisons were performed with the commercially available software SPSS, version 17.0 (Statistical Package for the Social Sciences, SPSS Inc., Chicago, IL). A two-tailed p value of less than 0.05 was considered as statistically significant.
Results There was no difference in age between subjects (control vs. TLE − HS vs. TLE + HS, p = 0.259, Kruskal—Wallis test). Right FA was higher than left FA in normal subjects, but MD was not different between hemispheres (p = 0.002 and 0.245, respectively, Wilcoxon-rank test). Right side FA values were corrected. The corrected right side FA value and uncorrected right side MD values of control subjects were used for comparison. First, comparisons between hemispheres were conducted; ipsilateral side to seizure focus (ips ) and contralateral side to seizure focus (con ). The ips FA was lower (p = 0.015, Wilcoxon-rank test) than the con FA, while MD was not different (p = 0.180, Wilcoxon-rank test, Table 2, Figs. 2 and 3) in TLE − HS. Similarly, the ips FA was lower (p < 0.001, Wilcoxon-rank test) than the con FA, while MD was not different (p = 0.150, Wilcoxon-rank test, Table 2, Figs. 2 and 3) in TLE + HS. Second, comparisons between three groups (TLE − HS, TLE + HS and control groups) were carried out. The comparison between the TLE − HS and control groups showed that FA values were not different (p > 0.05, Mann—Whitney U-test, Fig. 2), while ips and con MD values were increased (p = 0.001 and 0.002, respectively, Mann—Whitney U-test, Fig. 3) in the TLE − HS group. In the same way, FA values were not different between the TLE + HS and control groups (p > 0.05, Mann—Whitney U-test, Fig. 2), while ips
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study Table 2
25
Comparison between hemispheres in each study group. MD (×10−3 mm2 /s)
FA
TLE − HS* TLE + HS* Cor-Control‡
Ipsilateral
Contralateral
p-Value†
Ipsilateral
Contralateral
p-Value†
0.258 ± 0.031 0.277 ± 0.040 0.275 ± 0.017
0.279 ± 0.038 0.302 ± 0.043 0.275 ± 0.020
0.015 <0.001 0.944
2.439 ± 0.128 2.633 ± 0.140 2.214 ± 0.050
2.396 ± 0.125 2.524 ± 0.208 2.201 ± 0.038
0.180 0.150 0.245
Plus-minus values (±) are standard deviations. FA, fractional anisotropy; MD, mean diffusivity; TLE − HS, temporal lobe epilepsy without hippocampal sclerosis; TLE + HS, temporal lobe epilepsy with hippocampal sclerosis. * Right FA values of patients are corrected with regard to control subjects, and only the side ipsilateral or contralateral to the seizure focus is considered; corrected right FA value = (measured right FA value of patient) × (mean left FA of control/mean right FA of control). † Statistical analysis was performed with the Wilcoxon-rank test. ‡ Cor-Control: right side FA values were corrected, and original right side MD values of control subjects, which are presented in ipsilateral box, were used for comparison.
and con MD values were increased in the TLE + HS group compared to the control group (p < 0.001 and 0.001, respectively, Mann—Whitney U-test, Table 2, Fig. 3). The comparison between the TLE − HS and TLE + HS groups showed that the ips MD value was higher in the TLE + HS group than in the TLE − HS group (p = 0.007, Mann—Whitney U-test, Fig. 3), while other FA and MD values did not differ (Figs. 2 and 3). Onset age, seizure duration, frequency/month and total seizure number were not correlated with ips FA (p = 0.43, 0.39, 0.63 and 0.40, respectively, Spearman’s correlation test), cons FA (p = 0.29, 0.36, 0.57 and 0.29, respectively, Spearman’s correlation test), ips MD (p = 0.17, 0.75, 0.28 and 0.28, respectively, Spearman’s correlation test), cons MD (p = 0.96, 0.68, 0.43 and 0.79, respectively, Spearman’s correlation test).
Figure 3 Mean diffusivity values of subjects. The right value of control subject was used for comparison. The numbers in the graph represent p-values. The Wilcoxon-rank test and Mann—Whitney U-test were used. A unit of MD was mm2 /s. Ipsilateral and contralateral MD values were increased in both TLE − HS and TLE + HS than control group. Ipsilateral MD value in the TLE + HS group is higher than that in the TLE − HS group (p = 0.007, Mann—Whitney U-test). Abbreviations are the same as in Fig. 2.
Discussion Temporal lobe epilepsy is the most common type of localization-related epilepsy. The effect of chronic seizure activity on the adjacent brain has been the subject of numerous investigations (Newberg et al., 2000; Choi et al., 2003; Gross et al., 2006; Kimiwada et al., 2006; Labate et al., 2008; Riederer et al., 2008). The thalamus is known to have a strong anatomical connection to the medial temporal area and is known to modulate seizures (Bertram and Zhang, 1999; Natsume et al., 2003; Kimiwada et al., 2006; Keller and Roberts, 2008; Labate et al., 2008; Riederer et al., 2008; Bertram, 2009). Several reports have shown changes in the thalamus in TLE patients, suggesting that the thalamus is a part of the seizure network (Men et al., 2000; Natsume et al., 2003; Kimiwada et al., 2006; Gong et al., 2008; Keller and Roberts, 2008; Riederer et al., 2008; Bertram, 2009). In addition to an anatomical connection, physiological connections have also contributed to changes in the thalamus (Bertram and Zhang, 1999; Bertram et al., 2001; Kimiwada et al., 2006; Gong et al., 2008). Electrical stimulation revealed that the thalmus was closely connected to the hippocampus in animal study (Bertram and Zhang, 1999). These findings suggest that different involvement of the hippocampus may be associated with different changes in the thalamus between patients with TLE − HS and TLE + HS. Many reports have revealed thalamic changes in TLE. The most significant drawback of these previous studies has been that the diagnosis of TLE was not verified with surgical seizure outcome and pathological examination (Kimiwada et al., 2006; Gong et al., 2008; Keller and Roberts, 2008; Labate et al., 2008). The present study included patients whose diagnosis (TLE + HS and TLE − HS) was confirmed by both surgical outcome and pathological findings. In the present study, we compared FA and MD values between TLE + HS, TLE − HS and normal subjects. Although FA values seemed to decrease ipsilaterally to the seizure focus, there was no group difference between TLE + HS, TLE − HS and normal subjects. However, MD values showed marked differences compared to normal controls. MD values increased bilaterally in TLE − HS and TLE + HS patients. The MD increase was more significant ipsilaterally to the seizure focus in the TLE + HS group than in the TLE − HS group.
26 Thalamic changes occurring in TLE have been thought to be associated with neuronal loss, gliosis and extracellular edema (Bertram and Zhang, 1999; Diehl et al., 1999; Men et al., 2000; Bertram et al., 2001; Concha et al., 2005; Gross et al., 2006; Focke et al., 2008; Labate et al., 2008; Nilsson et al., 2008). As histological confirmation is impractical in patients, we could not verify the cause of the thalamic changes (Bertram and Zhang, 1999; Men et al., 2000; Bertram et al., 2001; Focke et al., 2008). The present study showed that abnormal thalamic changes were observed bilaterally in both TLE + HS and TLE − HS and that these changes were more prominent in TLE + HS. Bilateral MD changes in unilateral TLE suggest that the seizure network is extensive, in accordance with previous results (Gong et al., 2008; Keller and Roberts, 2008; Nilsson et al., 2008). Although the thalamic MD change was bilateral in both the TLE + HS and TLE − HS groups, there was a difference between the two groups, most likely due to the close connection of the hippocampus to the thalamus (Bertram and Zhang, 1999; Bertram et al., 2001; Kim et al., 2003; Mueller et al., 2006; Gong et al., 2008). The present study demonstrated this difference by showing increased MD in TLE + HS patients ipsilaterally, suggesting that the seizure network may be different in TLE + HS vs. TLE − HS groups. There was no FA decrease in comparison with normal subjects in the present study, in contrast to the results found by Gong et al, which showed decreased bilateral FA value in TLE + HS patients (Gong et al., 2008). Several explanations may be possible. Firstly, this discrepancy may be due to different selection criteria. In the present study, diagnosis was verified with surgical seizure outcome and pathological examination. On the other hand, in the study of Gong et al., only semiology and imaging characteristics were considered and moreover, 2 patients with bilateral HS were included (Gong et al., 2008). Some TLE − HS, dual pathology could have been included in the study by Gong et al. (Baulac et al., 1998; Ho et al., 1998; Kim et al., 2001; Gong et al., 2008). Another explanation could be the different statistical method used (Gong et al., 2008). Thirdly, small number of subjects (9 TLE + HS in the present study and 17 TLE + HS in the study of Gong et al.) or different ROI marking method may cause difference (Gong et al., 2008). Defining ROI in the whole thalamus may be a more reasonable method (Gong et al., 2008). However, contamination of the internal capsule or cerebrospinal fluid may have led to more confusing data. For this reason, we positioned the ROI in the center of the thalamus rather than including the entire structure. It is very controversial whether the thalamic changes are a result of recurrent seizures or are an inherent characteristic of a seizure network (Gong et al., 2008). Gong et al. reported that onset age is positively correlated with FA changes in the thalamus (Gong et al., 2008). This result suggests that these changes might be the result of recurrent seizures (Kimiwada et al., 2006; Gong et al., 2008). The present results showed that changes in MD and FA values were not correlated with onset age, duration of seizure, total seizure number or seizure frequency. These results imply that such changes might be an inherent characteristic of a seizure network. However, without longitudinal observation, we could not differentiate whether the thalamic changes are a result of recurrent seizures or are an
C.H. Kim et al. inherent characteristic of the network (Bertram and Zhang, 1999; Kimiwada et al., 2006; Gong et al., 2008). Considering the results from previous studies and those of the present study, we suggest that the thalamic abnormality is initially present as a bilateral gliosis or neuronal loss in both groups (TLE + HS and TLE − HS) and that repetitive electrical discharges from the hippocampus in TLE + HS further increase MD in the ipsilateral thalamus, e.g., by causing increase gliosis or cell death (Diehl et al., 1999, 2001; Eriksson et al., 2001; Rugg-Gunn et al., 2001; Arfanakis et al., 2002; Thivard et al., 2005; Kimiwada et al., 2006).
Limitations of this study Our results are subject to several limitations. First, because the number of subjects was small, we needed to correct the FA values. With this correction, we successfully managed the hemispheric difference. However, we must admit that correction itself causes insurmountable interpretation problems. Further studies with larger populations should be performed. Second, there may be a problem of consistency in marking the ROI. To overcome the inherent operator dependence of the ROI analysis method, all ROIs with fixed sizes were marked by one investigator (CHK) who was blinded to the patients’ clinical information (Kim et al., 2007; Oouchi et al., 2007). Third, although there were no overt seizure attacks in the week before the DTI examination, subclinical seizures may not have been reported. Peri-ictal diffusion change manifests itself as a restriction of diffusion, leading to decreased MD (Diehl et al., 1999, 2001). Patients in a subclinical peri-ictal state might have been included in the present study. In spite of these limitations, the present study was meaningful because it provided a method that overcame the problem of potential diagnostic inaccuracy with rigorous surgical seizure outcome and pathological criteria.
Conclusion Temporal lobe epilepsy could change thalamic diffusion properties bilaterally. The presence of hippocampal sclerosis enhances the change ipsilaterally. Thalamic diffusion properties vary with different seizure networks. Further studies are needed to investigate thalamic diffusion properties in a larger number of subjects.
Disclaimer The authors report no conflicts of interest concerning the materials or methods used in this study or the findings reported in this paper. No benefits in any form have been or will be received from a commercial party directly or indirectly related to the subject of this manuscript. This study was approved by the Institutional Review Board in Seoul National University Hospital.
Competing interests None reported.
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study
Acknowledgements This research was jointly supported by grants from the Brain Research Center of the 21st Century Frontier Research Program (2009K001280) funded by the Ministry of Education, Science and Technology, the Republic of Korea.
References Arfanakis, K., Hermann, B.P., Rogers, B.P., et al., 2002. Diffusion tensor MRI in temporal lobe epilepsy. Magn Reson Imaging 20, 511—519. Baulac, M., De Grissac, N., Hasboun, D., et al., 1998. Hippocampal developmental changes in patients with partial epilepsy: magnetic resonance imaging and clinical aspects. Ann. Neurol. 44, 223—233. Bertram, E.H., 2009. Temporal lobe epilepsy: where do the seizures really begin? Epilepsy Behav. 14 (Suppl 1), 32—37. Bertram, E.H., Mangan, P.S., Zhang, D., et al., 2001. The midline thalamus: alterations and a potential role in limbic epilepsy. Epilepsia 42, 967—978. Bertram, E.H., Zhang, D.X., 1999. Thalamic excitation of hippocampal CA1 neurons: a comparison with the effects of CA3 stimulation. Neuroscience 92, 15—26. Choi, J.Y., Kim, S.J., Hong, S.B., et al., 2003. Extratemporal hypometabolism on FDG PET in temporal lobe epilepsy as a predictor of seizure outcome after temporal lobectomy. Eur. J. Nucl. Med. Mol. Imaging 30, 581—587. Concha, L., Beaulieu, C., Gross, D.W., 2005. Bilateral limbic diffusion abnormalities in unilateral temporal lobe epilepsy. Ann. Neurol. 57, 188—196. Diehl, B., Najm, I., Ruggieri, P., et al., 1999. Periictal diffusionweighted imaging in a case of lesional epilepsy. Epilepsia 40, 1667—1671. Diehl, B., Najm, I., Ruggieri, P., et al., 2001. Postictal diffusionweighted imaging for the localization of focal epileptic areas in temporal lobe epilepsy. Epilepsia 42, 21—28. Engel, J.J., VanNess, P., Rasmussen, T., et al., 1993. Outcome with respect to epileptic seizures. In: Engel, J.J. (Ed.), Surgical Treatment of the Epilepsies. Raven press, New York, pp. 553—571. Eriksson, S.H., Rugg-Gunn, F.J., Symms, M.R., et al., 2001. Diffusion tensor imaging in patients with epilepsy and malformations of cortical development. Brain 124, 617—626. Focke, N.K., Yogarajah, M., Bonelli, S.B., et al., 2008. Voxel-based diffusion tensor imaging in patients with mesial temporal lobe epilepsy and hippocampal sclerosis. NeuroImage 40, 728—737. Gong, G., Concha, L., Beaulieu, C., et al., 2008. Thalamic diffusion and volumetry in temporal lobe epilepsy with and without mesial temporal sclerosis. Epilepsy Res. 80, 184—193. Gross, D.W., Concha, L., Beaulieu, C., 2006. Extratemporal white matter abnormalities in mesial temporal lobe epilepsy demonstrated with diffusion tensor imaging. Epilepsia 47, 1360—1363. Ho, S.S., Kuzniecky, R.I., Gilliam, F., et al., 1998. Temporal lobe developmental malformations and epilepsy: dual pathology and bilateral hippocampal abnormalities. Neurology 50, 748—754.
27
Keller, S.S., Roberts, N., 2008. Voxel-based morphometry of temporal lobe epilepsy: an introduction and review of the literature. Epilepsia 49, 741—757. Kim, C.H., Chung, C.K., Kim, J.S., et al., 2007. Use of diffusion tensor imaging to evaluate weakness. J Neurosurg 106, 111—118. Kim, Y.H., Chang, K.H., Park, S.W., et al., 2001. Hippocampal sclerosis: correlation of MR imaging findings with surgical outcome. Korean J Radiol 2, 63—67. Kim, Y.K., Lee, D.S., Lee, S.K., et al., 2003. Differential features of metabolic abnormalities between medial and lateral temporal lobe epilepsy: quantitative analysis of (18)F-FDG PET using SPM. J. Nucl. Med. 44, 1006—1012. Kimiwada, T., Juhász, C., Makki, M., et al., 2006. Hippocampal and thalamic diffusion abnormalities in children with temporal lobe epilepsy. Epilepsia 47, 167—175. Labate, A., Cerasa, A., Gambardella, A., et al., 2008. Hippocampal and thalamic atrophy in mild temporal lobe epilepsy: a VBM study. Neurology 71, 1094—1101. Men, S., Lee, D.H., Barron, J.R., et al., 2000. Selective neuronal necrosis associated with status epilepticus: MR findings. Am. J. Neuroradiol. 21, 1837—1840. Mueller, S.G., Laxer, K.D., Cashdollar, N., et al., 2006. Voxel-based optimized morphometry (VBM) of gray and white matter in temporal lobe epilepsy (TLE) with and without mesial temporal sclerosis. Epilepsia 47, 900—907. Natsume, J., Bernasconi, N., Andermann, F., et al., 2003. MRI volumetry of the thalamus in temporal, extratemporal, and idiopathic generalized epilepsy. Neurology 60, 1296—1300. Newberg, A.B., Alavi, A., Berlin, J., et al., 2000. Ipsilateral and contralateral thalamic hypometabolism as a predictor of outcome after temporal lobectomy for seizures. J. Nucl. Med. 41, 1964—1968. Nilsson, D., Go, C., Rutka, J.T., et al., 2008. Bilateral diffusion tensor abnormalities of temporal lobe and cingulate gyrus white matter in children with temporal lobe epilepsy. Epilepsy Res. 81, 128—135. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97—113. Oouchi, H., Yamada, K., Sakai, K., et al., 2007. Diffusion anisotropy measurement of brain white matter is affected by voxel size: underestimation occurs in areas with crossing fibers. Am. J. Neuroradiol. 28, 1102—1106. Palmini, A., Najm, I., Avanzini, G., et al., 2004. Terminology and classification of the cortical dysplasias. Neurology 62, S2—8. Riederer, F., Lanzenberger, R., Kaya, M., et al., 2008. Network atrophy in temporal lobe epilepsy: a voxel-based morphometry study. Neurology 71, 419—425. Rugg-Gunn, F.J., Eriksson, S.H., Symms, M.R., et al., 2001. Diffusion tensor imaging of cryptogenic and acquired partial epilepsies. Brain 124, 627—636. Thivard, L., Lehericy, S., Krainik, A., et al., 2005. Diffusion tensor imaging in medial temporal lobe epilepsy with hippocampal sclerosis. NeuroImage 28, 682—690. Yoo, S.Y., Chang, K.H., Song, I.C., et al., 2002. Apparent diffusion coefficient value of the hippocampus in patients with hippocampal sclerosis and in healthy volunteers. Am. J. Neuroradiol. 23, 809—812.
journal homepage: www.elsevier.com/locate/epilepsyres
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study Chi Heon Kim a,c,d, Bang-Bon Koo b, Chun Kee Chung a,c,d,∗, Jong-Min Lee b, June Sic Kim a,c,d, Sang Kun Lee e a
Department of Neurosurgery, Seoul National University College of Medicine, Seoul, South Korea Department of Biomedical Engineering, Graduate School of Engineering, Hanyang University, Seoul, South Korea c Neuroscience Research Institute, Seoul National University Medical Research Center, Seoul, South Korea d Clinical Research Institute, Seoul National University Hospital, Seoul, South Korea e Department of Neurology, Seoul National University Hospital, Seoul, South Korea b
Received 12 December 2009; received in revised form 25 February 2010; accepted 1 March 2010 Available online 21 March 2010
KEYWORDS Magnetic resonance imaging; Tensor; Hippocampus; Thalamus; Temporal lobe epilepsy
Summary Objective: The seizure network may be different between temporal lobe epilepsy with hippocampal sclerosis (TLE + HS) and without HS (TLE − HS). Chronic seizure activity may alter the diffusion properties of a seizure network. The thalamus is known to have an anatomical connection to the medial temporal area and to play a role in seizure modulation. This study aimed to evaluate differences in thalamic changes between TLE + HS and TLE − HS with diffusion tensor imaging (DTI). Methods: Nine patients with TLE + HS and nine patients with TLE − HS were included in the study. All patients underwent surgery with good seizure outcomes. Hippocampal sclerosis was verified pathologically. Sixteen right-handed, normal subjects were enrolled as controls. DTI was acquired using 3.0 T MRI. The mean diffusivity (MD) and fractional anisotropy (FA) were calculated in the center of the bilateral thalamus with the DTIstudio program. Results: The MD of bilateral thalami increased in both TLE groups compared to controls (p < 0.05), while FA values did not differ from controls. The MD of the thalamus ipsilateral to the epileptogenic side was higher in the TLE + HS group than in the TLE − HS group (p = 0.007). Onset age, seizure duration, seizure frequency and total seizure number were not correlated with FA and MD changes (p > 0.05). Conclusion: Bilateral thalamic diffusion properties are altered in temporal lobe epilepsy. The presence of hippocampal sclerosis enhances the change ipsilaterally. © 2010 Elsevier B.V. All rights reserved.
∗ Corresponding author at: Department of Neurosurgery, Seoul National University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, South Korea. Tel.: +82 2 2072 2352/2358; fax: +82 2 744 8459. E-mail address: [email protected] (C.K. Chung).
0920-1211/$ — see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2010.03.002
22
Introduction Chronic seizure activity may alter the diffusion properties of a seizure network (Newberg et al., 2000; Rugg-Gunn et al., 2001; Arfanakis et al., 2002; Yoo et al., 2002; Choi et al., 2003; Natsume et al., 2003; Thivard et al., 2005; Gross et al., 2006; Kimiwada et al., 2006; Mueller et al., 2006). The seizure network is wider than the seizure onset zone (Concha et al., 2005). Diffusion tensor imaging (DTI) reflects subtle changes that may not be visible in routine magnetic resonance (MR) images (Rugg-Gunn et al., 2001; Yoo et al., 2002; Concha et al., 2005). The seizure network that is involved in temporal lobe epilepsy with hippocampal sclerosis (TLE + HS) may be different from the network involved in TLE without hippocampal sclerosis (TLE − HS). The thalamus is known to have a strong anatomical connection to the medial temporal area and is known to modulate seizures. Consequently, there may be differences in thalamic changes in different seizure networks (Bertram and Zhang, 1999; Choi et al., 2003; Kim et al., 2003; Kimiwada et al., 2006; Labate et al., 2008; Riederer et al., 2008; Bertram, 2009). This study aimed to evaluate the differences in thalamic changes between TLE + HS and TLE − HS with DTI.
Materials and methods Patients We selected patients with temporal lobe epilepsy who underwent operations from May 2005 to August 2008. Included were patients who had more than 2 years of seizure history, who had taken DTI preoperatively with no seizures in the previous week, who were right-handed by the Edinburg Handedness Inventory (a score of less than −40 is indicative of right-handedness, a score of more than 40 is indicative of left-handedness, and scores between −40 and 40 are indicative of ambidexterity) and who obtained a postoperative seizure outcome of Engel class I (no seizure) or Engel class II (rare seizure). TLE + HS patients had histologically proven hippocampal sclerosis with no neocortical malformation in the temporal lobe, and TLE − HS patients had cortical dysplasia, as suggested by Palmini et al. (Oldfield, 1971; Palmini et al., 2004). Eighteen consecutive patients satisfied these criteria: 9 patients with TLE − HS and 9 patients with TLE + HS (Table 1). All patients scored less than −80 in the Edinburg Handedness Inventory (Oldfield, 1971). In TLE − HS patients, the seizure focus was lateralized to the right side in four patients and to the left side in five patients. In TLE + HS patients, the seizure focus was lateralized to the right side in four patients and to the left side in five patients. The mean subject age was 30 ± 7 years (median, 32; range, 20—40) in the TLE − HS group, 31 ± 8 years (median, 29; range, 20—44) in the TLE + HS group and 33 ± 3 years (median, 33; range, 30—39) in the control group. MR images were interpreted by expert neuroradiologists. Hippocampal atrophy was observed in MR images of all TLE + HS patients. There was no hippocampal atrophy in TLE − HS patients. Cortical dysplasia was suspected on T2-weighted fluid attenuation inversion recovery (FLAIR) MR image in 2 patients in the temporal lobe (Table 1). TLE + HS patients underwent
C.H. Kim et al. standard anterior temporal lobectomy with amygdalohippocampectomy. Hippocampal sclerosis was pathologically confirmed without associated neocortical malformation. TLE − HS patients underwent neocortical temporal lobe resection based on the results of subdural electrode monitoring. Pathological examination showed mild malformation of cortical development (MCD) in 2 patients and type IA cortical dysplasia in 7 patients. A seizure-free outcome (Engel class I) was achieved in 16 patients with TLE + HS or TLE − HS, and Engel class II status (rare seizures) was achieved in 2 patients with TLE − HS (Engel et al., 1993). The mean postoperative follow-up period was 28 months (range, 15—45) for TLE + HS patients and 31 months (range, 19—49) for TLE − HS patients. Sixteen right-handed, normal subjects with no history of neurological disorders were recruited as the control group (8 men and 8 women). All control subjects scored less than −80 on the Edinburg Handedness Inventory. This study was approved by the Institutional Review Board of Seoul National University Hospital (H-0712-056-230).
Diffusion tensor imaging protocol DTI was acquired before operation using a 3.0 T MR scanner (General Electric VH/i, Milwaukee, Wisconsin, USA) with a conventional head gradient coil. We used a single-shot spinecho EPI sequence. The B-factor was set at 1000 s/mm2 . The acquisition parameters were: FOV = 240 mm, matrix = 256 × 256, slice thickness/gap = 3.5 mm/0 mm, total slice number = 38, TR/TE = 10,000 ms/90 ms, and scan average = 1. To estimate the intensity and direction of the diffusion anisotropy, axial MR images with 25 non-collinear diffusion gradients and one without diffusion gradients were acquired. From 26 diffusion-weighted images, we obtained six diffusion tensor components, Dxx, Dyy, Dzz, Dxy, Dxz and Dyz, using multiple linear equations.
Data processing The digital images from all subjects were stored as digital imaging and communications in medicine (DICOM) files. The DICOM files were converted for analysis using the MRIcro program (version 1.40, University of South Carolina, Columbia SC 29208, USA, www.mricro.com). Diffusion parameter (apparent diffusion coefficient [ADC] and fractional anisotropy [FA]) maps were reconstructed from the diffusion tensor imaging using program DTIstudio (Johns Hopkins University, Baltimore, USA, www.mristudio.org). Region of interest (ROI) analysis was performed using the DTIstudio program for quantitative analysis. The ROI was marked as follows: mark fixed-sized ROI (10 voxels, diameter of 9.8 mm) at the center of the thalamus just above 10.5 mm (3 slices) from the lower margin of the thalamus on a 2-dimensional FA map (Oouchi et al., 2007). Another ROI was marked on the opposite homotopic site (Fig. 1). The mean diffusivity (MD) and FA were calculated in each ROI of the bilateral thalamus. To maintain consistency, all ROIs were marked by one investigator (CHK) who was blinded to the patient information.
Summary of patients.
Pt
Group
Age
Gender
Side
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE − HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS TLE + HS
37 23 20 33 20 29 36 40 32 44 29 33 20 28 20 39 36 26
Male Male Female Male Male Female Male Male Female Female Female Female Female Male Female Male Male Female
Right Left Left Left Right Left Right Left Right Right Left Left Right Right Left Left Right Left
Onset age (yr) 13 14 13 13 15 17 17 8 15 10 12 26 18 5 18 11 13 2
Seizure duration (yr)
Frequency/month
24 9 7 20 5 12 19 32 17 34 17 7 2 23 2 28 23 24
1.5 3.5 30 2 0.5 1.5 4 4 1 3.5 5 4.5 15 1 2 10 5.5 2.5
Total seizure number 432 378 2520 480 30 216 912 1536 204 1428 1020 378 360 276 48 3360 1518 720
Palmini classification
Location of CD in FLAIR
IA IA IA IA Mild MCD IA IA Mild MCD IA
Temporal lobe No No No Inferior temporal No No No No
CD, cortical dysplasia; MCD, malformation of cortical development; FLAIR: fluid attenuated inversion recovery magnetic resonance imaging.
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study
Table 1
23
24
Figure 1 Location of region of interest (ROI). ROI was marked in the center of both thalami on the 2dimensional FA map. The size of the ROI was fixed with 10 voxels (diameter of 9.8 mm). This axial section is at the level of 3 slices (10.5 mm), above the inferior margin of the thalamus. ROI, region of interest; FA, fractional anisotropy.
C.H. Kim et al.
Figure 2 Fractional anisotrophy values of subjects. The numbers in the graph represent p-values. The Wilcoxonrank test and Mann—Whitney U-test were used. Right FA values of patients and control subject were corrected with regard to control subjects, and only the side ipsilateral or contralateral to the seizure focus was considered; corrected right FA value = (measured right FA value of patient) × (mean left FA of control/mean right FA of control). The ips FA was lower, Wilcoxon-rank test) than the con FA in both TLE − HS (p = 0.015, Wilcoxon-rank test) and TLE + HS (p < 0.001, Wilcoxon-rank test). FA, fractional anisotropy; MD, mean diffusivity; TLE − HS, temporal lobe epilepsy without hippocampal sclerosis; TLE + HS, temporal lobe epilepsy with hippocampal sclerosis; ips , ipsilateral side to seizure focus; con , contralateral side to seizure focus.
Analysis Thirty-four subjects (16, control; 9, TLE + HS; 9, TLE − HS) participated in the present study. The subjects were divided into left TLE + HS, right TLE + HS, left TLE − HS and right TLE − HS. This study focused on comparing the differences between ipsilateral and contralateral sides to the seizure focus. To abolish inherent hemispheric differences, we corrected the right side value of both patients and control subjects with regard to control subject values. After correction, only the ipsilateral and contralateral sides to the seizure focus could be considered, therefore patients were divided into 2 groups (TLE + HS and TLE − HS). There was a hemispheric difference in FA but no difference in MD, so FA was corrected. Since there was a hemispheric difference in FA in control subjects, the mean hemispheric ratio was used for correction in patients and control subjects; the corrected right FA of a subject = (measured right FA of individual subject) × (mean left FA of control subjects/mean right FA of control subjects). The Wilcoxon-rank test was used for comparisons between hemispheres in same group. The Mann—Whitney’s U-test and Kruskal—Wallis test were used for comparisons between groups. Spearman’s test was used for correlation analysis. Comparisons were performed with the commercially available software SPSS, version 17.0 (Statistical Package for the Social Sciences, SPSS Inc., Chicago, IL). A two-tailed p value of less than 0.05 was considered as statistically significant.
Results There was no difference in age between subjects (control vs. TLE − HS vs. TLE + HS, p = 0.259, Kruskal—Wallis test). Right FA was higher than left FA in normal subjects, but MD was not different between hemispheres (p = 0.002 and 0.245, respectively, Wilcoxon-rank test). Right side FA values were corrected. The corrected right side FA value and uncorrected right side MD values of control subjects were used for comparison. First, comparisons between hemispheres were conducted; ipsilateral side to seizure focus (ips ) and contralateral side to seizure focus (con ). The ips FA was lower (p = 0.015, Wilcoxon-rank test) than the con FA, while MD was not different (p = 0.180, Wilcoxon-rank test, Table 2, Figs. 2 and 3) in TLE − HS. Similarly, the ips FA was lower (p < 0.001, Wilcoxon-rank test) than the con FA, while MD was not different (p = 0.150, Wilcoxon-rank test, Table 2, Figs. 2 and 3) in TLE + HS. Second, comparisons between three groups (TLE − HS, TLE + HS and control groups) were carried out. The comparison between the TLE − HS and control groups showed that FA values were not different (p > 0.05, Mann—Whitney U-test, Fig. 2), while ips and con MD values were increased (p = 0.001 and 0.002, respectively, Mann—Whitney U-test, Fig. 3) in the TLE − HS group. In the same way, FA values were not different between the TLE + HS and control groups (p > 0.05, Mann—Whitney U-test, Fig. 2), while ips
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study Table 2
25
Comparison between hemispheres in each study group. MD (×10−3 mm2 /s)
FA
TLE − HS* TLE + HS* Cor-Control‡
Ipsilateral
Contralateral
p-Value†
Ipsilateral
Contralateral
p-Value†
0.258 ± 0.031 0.277 ± 0.040 0.275 ± 0.017
0.279 ± 0.038 0.302 ± 0.043 0.275 ± 0.020
0.015 <0.001 0.944
2.439 ± 0.128 2.633 ± 0.140 2.214 ± 0.050
2.396 ± 0.125 2.524 ± 0.208 2.201 ± 0.038
0.180 0.150 0.245
Plus-minus values (±) are standard deviations. FA, fractional anisotropy; MD, mean diffusivity; TLE − HS, temporal lobe epilepsy without hippocampal sclerosis; TLE + HS, temporal lobe epilepsy with hippocampal sclerosis. * Right FA values of patients are corrected with regard to control subjects, and only the side ipsilateral or contralateral to the seizure focus is considered; corrected right FA value = (measured right FA value of patient) × (mean left FA of control/mean right FA of control). † Statistical analysis was performed with the Wilcoxon-rank test. ‡ Cor-Control: right side FA values were corrected, and original right side MD values of control subjects, which are presented in ipsilateral box, were used for comparison.
and con MD values were increased in the TLE + HS group compared to the control group (p < 0.001 and 0.001, respectively, Mann—Whitney U-test, Table 2, Fig. 3). The comparison between the TLE − HS and TLE + HS groups showed that the ips MD value was higher in the TLE + HS group than in the TLE − HS group (p = 0.007, Mann—Whitney U-test, Fig. 3), while other FA and MD values did not differ (Figs. 2 and 3). Onset age, seizure duration, frequency/month and total seizure number were not correlated with ips FA (p = 0.43, 0.39, 0.63 and 0.40, respectively, Spearman’s correlation test), cons FA (p = 0.29, 0.36, 0.57 and 0.29, respectively, Spearman’s correlation test), ips MD (p = 0.17, 0.75, 0.28 and 0.28, respectively, Spearman’s correlation test), cons MD (p = 0.96, 0.68, 0.43 and 0.79, respectively, Spearman’s correlation test).
Figure 3 Mean diffusivity values of subjects. The right value of control subject was used for comparison. The numbers in the graph represent p-values. The Wilcoxon-rank test and Mann—Whitney U-test were used. A unit of MD was mm2 /s. Ipsilateral and contralateral MD values were increased in both TLE − HS and TLE + HS than control group. Ipsilateral MD value in the TLE + HS group is higher than that in the TLE − HS group (p = 0.007, Mann—Whitney U-test). Abbreviations are the same as in Fig. 2.
Discussion Temporal lobe epilepsy is the most common type of localization-related epilepsy. The effect of chronic seizure activity on the adjacent brain has been the subject of numerous investigations (Newberg et al., 2000; Choi et al., 2003; Gross et al., 2006; Kimiwada et al., 2006; Labate et al., 2008; Riederer et al., 2008). The thalamus is known to have a strong anatomical connection to the medial temporal area and is known to modulate seizures (Bertram and Zhang, 1999; Natsume et al., 2003; Kimiwada et al., 2006; Keller and Roberts, 2008; Labate et al., 2008; Riederer et al., 2008; Bertram, 2009). Several reports have shown changes in the thalamus in TLE patients, suggesting that the thalamus is a part of the seizure network (Men et al., 2000; Natsume et al., 2003; Kimiwada et al., 2006; Gong et al., 2008; Keller and Roberts, 2008; Riederer et al., 2008; Bertram, 2009). In addition to an anatomical connection, physiological connections have also contributed to changes in the thalamus (Bertram and Zhang, 1999; Bertram et al., 2001; Kimiwada et al., 2006; Gong et al., 2008). Electrical stimulation revealed that the thalmus was closely connected to the hippocampus in animal study (Bertram and Zhang, 1999). These findings suggest that different involvement of the hippocampus may be associated with different changes in the thalamus between patients with TLE − HS and TLE + HS. Many reports have revealed thalamic changes in TLE. The most significant drawback of these previous studies has been that the diagnosis of TLE was not verified with surgical seizure outcome and pathological examination (Kimiwada et al., 2006; Gong et al., 2008; Keller and Roberts, 2008; Labate et al., 2008). The present study included patients whose diagnosis (TLE + HS and TLE − HS) was confirmed by both surgical outcome and pathological findings. In the present study, we compared FA and MD values between TLE + HS, TLE − HS and normal subjects. Although FA values seemed to decrease ipsilaterally to the seizure focus, there was no group difference between TLE + HS, TLE − HS and normal subjects. However, MD values showed marked differences compared to normal controls. MD values increased bilaterally in TLE − HS and TLE + HS patients. The MD increase was more significant ipsilaterally to the seizure focus in the TLE + HS group than in the TLE − HS group.
26 Thalamic changes occurring in TLE have been thought to be associated with neuronal loss, gliosis and extracellular edema (Bertram and Zhang, 1999; Diehl et al., 1999; Men et al., 2000; Bertram et al., 2001; Concha et al., 2005; Gross et al., 2006; Focke et al., 2008; Labate et al., 2008; Nilsson et al., 2008). As histological confirmation is impractical in patients, we could not verify the cause of the thalamic changes (Bertram and Zhang, 1999; Men et al., 2000; Bertram et al., 2001; Focke et al., 2008). The present study showed that abnormal thalamic changes were observed bilaterally in both TLE + HS and TLE − HS and that these changes were more prominent in TLE + HS. Bilateral MD changes in unilateral TLE suggest that the seizure network is extensive, in accordance with previous results (Gong et al., 2008; Keller and Roberts, 2008; Nilsson et al., 2008). Although the thalamic MD change was bilateral in both the TLE + HS and TLE − HS groups, there was a difference between the two groups, most likely due to the close connection of the hippocampus to the thalamus (Bertram and Zhang, 1999; Bertram et al., 2001; Kim et al., 2003; Mueller et al., 2006; Gong et al., 2008). The present study demonstrated this difference by showing increased MD in TLE + HS patients ipsilaterally, suggesting that the seizure network may be different in TLE + HS vs. TLE − HS groups. There was no FA decrease in comparison with normal subjects in the present study, in contrast to the results found by Gong et al, which showed decreased bilateral FA value in TLE + HS patients (Gong et al., 2008). Several explanations may be possible. Firstly, this discrepancy may be due to different selection criteria. In the present study, diagnosis was verified with surgical seizure outcome and pathological examination. On the other hand, in the study of Gong et al., only semiology and imaging characteristics were considered and moreover, 2 patients with bilateral HS were included (Gong et al., 2008). Some TLE − HS, dual pathology could have been included in the study by Gong et al. (Baulac et al., 1998; Ho et al., 1998; Kim et al., 2001; Gong et al., 2008). Another explanation could be the different statistical method used (Gong et al., 2008). Thirdly, small number of subjects (9 TLE + HS in the present study and 17 TLE + HS in the study of Gong et al.) or different ROI marking method may cause difference (Gong et al., 2008). Defining ROI in the whole thalamus may be a more reasonable method (Gong et al., 2008). However, contamination of the internal capsule or cerebrospinal fluid may have led to more confusing data. For this reason, we positioned the ROI in the center of the thalamus rather than including the entire structure. It is very controversial whether the thalamic changes are a result of recurrent seizures or are an inherent characteristic of a seizure network (Gong et al., 2008). Gong et al. reported that onset age is positively correlated with FA changes in the thalamus (Gong et al., 2008). This result suggests that these changes might be the result of recurrent seizures (Kimiwada et al., 2006; Gong et al., 2008). The present results showed that changes in MD and FA values were not correlated with onset age, duration of seizure, total seizure number or seizure frequency. These results imply that such changes might be an inherent characteristic of a seizure network. However, without longitudinal observation, we could not differentiate whether the thalamic changes are a result of recurrent seizures or are an
C.H. Kim et al. inherent characteristic of the network (Bertram and Zhang, 1999; Kimiwada et al., 2006; Gong et al., 2008). Considering the results from previous studies and those of the present study, we suggest that the thalamic abnormality is initially present as a bilateral gliosis or neuronal loss in both groups (TLE + HS and TLE − HS) and that repetitive electrical discharges from the hippocampus in TLE + HS further increase MD in the ipsilateral thalamus, e.g., by causing increase gliosis or cell death (Diehl et al., 1999, 2001; Eriksson et al., 2001; Rugg-Gunn et al., 2001; Arfanakis et al., 2002; Thivard et al., 2005; Kimiwada et al., 2006).
Limitations of this study Our results are subject to several limitations. First, because the number of subjects was small, we needed to correct the FA values. With this correction, we successfully managed the hemispheric difference. However, we must admit that correction itself causes insurmountable interpretation problems. Further studies with larger populations should be performed. Second, there may be a problem of consistency in marking the ROI. To overcome the inherent operator dependence of the ROI analysis method, all ROIs with fixed sizes were marked by one investigator (CHK) who was blinded to the patients’ clinical information (Kim et al., 2007; Oouchi et al., 2007). Third, although there were no overt seizure attacks in the week before the DTI examination, subclinical seizures may not have been reported. Peri-ictal diffusion change manifests itself as a restriction of diffusion, leading to decreased MD (Diehl et al., 1999, 2001). Patients in a subclinical peri-ictal state might have been included in the present study. In spite of these limitations, the present study was meaningful because it provided a method that overcame the problem of potential diagnostic inaccuracy with rigorous surgical seizure outcome and pathological criteria.
Conclusion Temporal lobe epilepsy could change thalamic diffusion properties bilaterally. The presence of hippocampal sclerosis enhances the change ipsilaterally. Thalamic diffusion properties vary with different seizure networks. Further studies are needed to investigate thalamic diffusion properties in a larger number of subjects.
Disclaimer The authors report no conflicts of interest concerning the materials or methods used in this study or the findings reported in this paper. No benefits in any form have been or will be received from a commercial party directly or indirectly related to the subject of this manuscript. This study was approved by the Institutional Review Board in Seoul National University Hospital.
Competing interests None reported.
Thalamic changes in temporal lobe epilepsy with and without hippocampal sclerosis: A diffusion tensor imaging study
Acknowledgements This research was jointly supported by grants from the Brain Research Center of the 21st Century Frontier Research Program (2009K001280) funded by the Ministry of Education, Science and Technology, the Republic of Korea.
References Arfanakis, K., Hermann, B.P., Rogers, B.P., et al., 2002. Diffusion tensor MRI in temporal lobe epilepsy. Magn Reson Imaging 20, 511—519. Baulac, M., De Grissac, N., Hasboun, D., et al., 1998. Hippocampal developmental changes in patients with partial epilepsy: magnetic resonance imaging and clinical aspects. Ann. Neurol. 44, 223—233. Bertram, E.H., 2009. Temporal lobe epilepsy: where do the seizures really begin? Epilepsy Behav. 14 (Suppl 1), 32—37. Bertram, E.H., Mangan, P.S., Zhang, D., et al., 2001. The midline thalamus: alterations and a potential role in limbic epilepsy. Epilepsia 42, 967—978. Bertram, E.H., Zhang, D.X., 1999. Thalamic excitation of hippocampal CA1 neurons: a comparison with the effects of CA3 stimulation. Neuroscience 92, 15—26. Choi, J.Y., Kim, S.J., Hong, S.B., et al., 2003. Extratemporal hypometabolism on FDG PET in temporal lobe epilepsy as a predictor of seizure outcome after temporal lobectomy. Eur. J. Nucl. Med. Mol. Imaging 30, 581—587. Concha, L., Beaulieu, C., Gross, D.W., 2005. Bilateral limbic diffusion abnormalities in unilateral temporal lobe epilepsy. Ann. Neurol. 57, 188—196. Diehl, B., Najm, I., Ruggieri, P., et al., 1999. Periictal diffusionweighted imaging in a case of lesional epilepsy. Epilepsia 40, 1667—1671. Diehl, B., Najm, I., Ruggieri, P., et al., 2001. Postictal diffusionweighted imaging for the localization of focal epileptic areas in temporal lobe epilepsy. Epilepsia 42, 21—28. Engel, J.J., VanNess, P., Rasmussen, T., et al., 1993. Outcome with respect to epileptic seizures. In: Engel, J.J. (Ed.), Surgical Treatment of the Epilepsies. Raven press, New York, pp. 553—571. Eriksson, S.H., Rugg-Gunn, F.J., Symms, M.R., et al., 2001. Diffusion tensor imaging in patients with epilepsy and malformations of cortical development. Brain 124, 617—626. Focke, N.K., Yogarajah, M., Bonelli, S.B., et al., 2008. Voxel-based diffusion tensor imaging in patients with mesial temporal lobe epilepsy and hippocampal sclerosis. NeuroImage 40, 728—737. Gong, G., Concha, L., Beaulieu, C., et al., 2008. Thalamic diffusion and volumetry in temporal lobe epilepsy with and without mesial temporal sclerosis. Epilepsy Res. 80, 184—193. Gross, D.W., Concha, L., Beaulieu, C., 2006. Extratemporal white matter abnormalities in mesial temporal lobe epilepsy demonstrated with diffusion tensor imaging. Epilepsia 47, 1360—1363. Ho, S.S., Kuzniecky, R.I., Gilliam, F., et al., 1998. Temporal lobe developmental malformations and epilepsy: dual pathology and bilateral hippocampal abnormalities. Neurology 50, 748—754.
27
Keller, S.S., Roberts, N., 2008. Voxel-based morphometry of temporal lobe epilepsy: an introduction and review of the literature. Epilepsia 49, 741—757. Kim, C.H., Chung, C.K., Kim, J.S., et al., 2007. Use of diffusion tensor imaging to evaluate weakness. J Neurosurg 106, 111—118. Kim, Y.H., Chang, K.H., Park, S.W., et al., 2001. Hippocampal sclerosis: correlation of MR imaging findings with surgical outcome. Korean J Radiol 2, 63—67. Kim, Y.K., Lee, D.S., Lee, S.K., et al., 2003. Differential features of metabolic abnormalities between medial and lateral temporal lobe epilepsy: quantitative analysis of (18)F-FDG PET using SPM. J. Nucl. Med. 44, 1006—1012. Kimiwada, T., Juhász, C., Makki, M., et al., 2006. Hippocampal and thalamic diffusion abnormalities in children with temporal lobe epilepsy. Epilepsia 47, 167—175. Labate, A., Cerasa, A., Gambardella, A., et al., 2008. Hippocampal and thalamic atrophy in mild temporal lobe epilepsy: a VBM study. Neurology 71, 1094—1101. Men, S., Lee, D.H., Barron, J.R., et al., 2000. Selective neuronal necrosis associated with status epilepticus: MR findings. Am. J. Neuroradiol. 21, 1837—1840. Mueller, S.G., Laxer, K.D., Cashdollar, N., et al., 2006. Voxel-based optimized morphometry (VBM) of gray and white matter in temporal lobe epilepsy (TLE) with and without mesial temporal sclerosis. Epilepsia 47, 900—907. Natsume, J., Bernasconi, N., Andermann, F., et al., 2003. MRI volumetry of the thalamus in temporal, extratemporal, and idiopathic generalized epilepsy. Neurology 60, 1296—1300. Newberg, A.B., Alavi, A., Berlin, J., et al., 2000. Ipsilateral and contralateral thalamic hypometabolism as a predictor of outcome after temporal lobectomy for seizures. J. Nucl. Med. 41, 1964—1968. Nilsson, D., Go, C., Rutka, J.T., et al., 2008. Bilateral diffusion tensor abnormalities of temporal lobe and cingulate gyrus white matter in children with temporal lobe epilepsy. Epilepsy Res. 81, 128—135. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97—113. Oouchi, H., Yamada, K., Sakai, K., et al., 2007. Diffusion anisotropy measurement of brain white matter is affected by voxel size: underestimation occurs in areas with crossing fibers. Am. J. Neuroradiol. 28, 1102—1106. Palmini, A., Najm, I., Avanzini, G., et al., 2004. Terminology and classification of the cortical dysplasias. Neurology 62, S2—8. Riederer, F., Lanzenberger, R., Kaya, M., et al., 2008. Network atrophy in temporal lobe epilepsy: a voxel-based morphometry study. Neurology 71, 419—425. Rugg-Gunn, F.J., Eriksson, S.H., Symms, M.R., et al., 2001. Diffusion tensor imaging of cryptogenic and acquired partial epilepsies. Brain 124, 627—636. Thivard, L., Lehericy, S., Krainik, A., et al., 2005. Diffusion tensor imaging in medial temporal lobe epilepsy with hippocampal sclerosis. NeuroImage 28, 682—690. Yoo, S.Y., Chang, K.H., Song, I.C., et al., 2002. Apparent diffusion coefficient value of the hippocampus in patients with hippocampal sclerosis and in healthy volunteers. Am. J. Neuroradiol. 23, 809—812.