- Email: [email protected]
Cerebellar white matter changes in patients with newly diagnosed partial epilepsy of unknown etiology
Clinical Neurology and Neurosurgery 138 (2015) 25–30
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Cerebellar white matter changes in patients with newly diagnosed partial epilepsy of unknown etiology Kang Min Park a,1 , Yong Hee Han b,1 , Tae Hyung Kim c , Chi Woong Mun b,c , Kyong Jin Shin a , Sam Yeol Ha a , JinSe Park a , Yun Jung Hur d , Hae Yu Kim e , Si Hyung Park f , Sung Eun Kim a,∗ a
Department of Neurology, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea Department of Biomedical Engineering/u-HARC, Inje University, Gimhae, South Korea c Department of Health Science and Technology, Inje University, Gimhae, South Korea d Department of Pediatrics, Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea e Department of Neurosurgery, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea f Department of Internal Medicine, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea b
a r t i c l e
i n f o
Article history: Received 1 October 2014 Received in revised form 10 July 2015 Accepted 24 July 2015 Available online 29 July 2015 Keywords: Epilepsy Cerebellum Atrophy
a b s t r a c t Objective: We hypothesize that pre-existing susceptible structures in the brain may be associated with the development of newly diagnosed partial epilepsy of unknown etiology. Methods: Twenty-two patients with newly diagnosed partial epilepsy of unknown etiology and 36 healthy controls were enrolled in this study. In addition, we included 24 patients with chronic partial epilepsy of unknown etiology as a disease control group. We analyzed whole-brain T1-weighted MRIs using FreeSurfer 5.1. The volumes of the hippocampus, amygdala, thalamus, caudate, putamen, pallidum, brainstem, cerebellar gray and white matter, as well as cerebral gray and white matter were compared between the groups. We also analyzed the changes in brain volumes associated with the chronicity of epilepsy in the patients with chronic epilepsy compared to newly diagnosed epilepsy. Results: The volume of cerebellar white matter in patients with newly diagnosed epilepsy was significantly smaller than that which was observed in the healthy controls (p = 0.0001). This finding was also observed in patients with chronic epilepsy (p < 0.0001). Cerebral white matter volume was negatively correlated with the duration of epilepsy (r = −0.4, p = 0.04). Conclusion: These findings support our hypothesis that cerebellar white matter changes may constitute a pre-existing susceptible structure in the brain that is associated with the development of partial epilepsy of unknown etiology. In addition, cerebral white matter was the structure that was the most vulnerable to the progression of epilepsy. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Epilepsy is one of the most common chronic neurological disorders, with partial epilepsy accounting for 70% of all epilepsy cases [1]. Structural lesions that are observed in brain magnetic resonance imaging (MRI) upon visual inspection can be found in approximately half of all epilepsy patients [1]. Moreover, there is
∗ Corresponding author at: Department of Neurology, Haeundae Paik Hospital, Inje University College of Medicine, Haeundae-ro 875, Haeundae-gu, Busan 612 896, South Korea. Tel.: +82 51 797 1195; fax: +82 51 797 1196. E-mail address: [email protected] (S.E. Kim). 1 These two authors contributed equally to this work. http://dx.doi.org/10.1016/j.clineuro.2015.07.017 0303-8467/© 2015 Elsevier B.V. All rights reserved.
growing evidence that the brains of patients with partial epilepsy have structural lesions and anatomic abnormalities beyond the lesions that are visually observable in MRI. Several studies have demonstrated, using voxel-based morphometry (VBM), a significant volume reduction in the hippocampus as well as in the extra-temporal gray matter of patients with mesial temporal lobe epilepsy and hippocampal sclerosis [2–6]. Previously studies have also found alterations in the extra-temporal white matter using diffusion tensor imaging (DTI) [5–7]. Although there are anatomic abnormalities beyond the visual lesions observed in MRI in patients with partial epilepsy with structural lesions, the pathogenesis of epilepsy might still be directly related to these visual lesions.
26
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
However, the mechanism of partial epilepsy of unknown etiology is unclear. Epilepsy of unknown etiology is presumed to be symptomatic in nature and of an unidentified cause. Both genetic and environmental factors likely play a role to varying extents in individual patients. Nevertheless, a recent genome-wide association study with a large cohort did not identify common variants that influence the risk for epilepsy among patients with partial epilepsy of unknown etiology [8]. Meanwhile, there have been reports that there are anatomic abnormalities in patients with partial epilepsy of unknown etiology. Several reports have demonstrated a volume reduction in the thalamus, cerebellum, hippocampus, caudate, and cerebral white matter [6,9–11]; whereas other studies did not find any anatomic abnormalities in patients with partial epilepsy of unknown etiology [12,13]. These discrepancies are most likely due to methodological differences, such as the method of analysis, the correction for multiple comparisons, and the thresholds set for cluster extent. Furthermore, previous studies shared the problem of enrolling heterogeneous patient groups, thereby mixing patients with newly diagnosed epilepsy (NDE) with those who experienced chronic epilepsy (CHE) [2–7,9–13]. Thus, they could not determine whether the anatomic abnormalities existed before the onset of epilepsy or were the cumulative effects of seizure-induced damage. Therefore, studies that only include patients with NDE are needed. From these studies, we may obtain information concerning the role of pre-existing anatomic abnormalities before the onset of epilepsy in the development of epilepsy of unknown etiology. In this present study, we only enrolled patients with NDE to increase homogeneity, and we included patients with CHE as a disease control group. VBM analysis does a voxel-by-voxel comparison to search for differences in volumes, whereas analysis using FreeSurfer image analysis suite can segment the brain into individual structures and directly measure the volumes of these structures [14–18]. There have been no studies that investigate the anatomic abnormalities in patients with newly diagnosed partial epilepsy of unknown etiology, especially using FreeSurfer image analysis suite. The aim of this study is to clarify whether the anatomic abnormalities existed before the onset of epilepsy in patients with newly diagnosed partial epilepsy of unknown etiology. We hypothesize that pre-existing anatomic abnormalities may be associated with the development of partial epilepsy of unknown etiology.
2. Materials and methods 2.1. Patient subjects This study was conducted with the approval of the Institutional Review Board of our institution. This study was consecutively performed in a single tertiary hospital. We prospectively enrolled 22 patients with a clinical diagnosis of newly diagnosed partial epilepsy of unknown etiology. The diagnosis of epilepsy was made on the basis of clinical histories and electroencephalographic (EEG) findings. We excluded patients that had already displayed only auras or febrile convulsions before their first episodes of unprovoked epileptic seizures. All patients had normal MRI results upon visual inspection using conventional brain MRI protocols, including T1- and T2 weighted images and fluid attenuated inversion recovery (FLAIR) images. All patients had cryptogenic partial epilepsy using the International League Against Epilepsy (ILAE) classification [19] and had epilepsy of an unknown cause according to the current ILAE classification [20]. In addition, we included 24 patients with chronic partial epilepsy of unknown etiology as a disease control group. The definition for CHE in this study was that the epilepsy persisted for more than 2 years. Patients did not have any history of other neurological or psychiatric diseases due to the potential influence of these conditions on the brain atrophy. We collected
demographic data such as age, sex, age of onset, and the duration of epilepsy from these patients at the time of the MRI. 2.2. Normal controls The control group consisted of 36 age- and sex-matched healthy subjects. All subjects had a normal neurological examination and no history of cardiovascular, neurological or psychiatric disease or of diabetes, hypertension, or dyslipidemia. All healthy controls had a normal MRI upon visual inspection. 2.3. MRI data acquisition All patients with NDE had an MRI protocol performed at the initial diagnosis, whereas the patients with CHE had an MRI protocol performed at least 2 years after diagnosis. All scans were performed on a 3.0T MRI scanner (AchievaTx, Phillips Healthcare, Best, The Netherlands) that was equipped with an 8-channel head coil. All subjects underwent conventional brain MRI protocols, including axial and coronal 2D T2-weighted images, which were obtained with a turbo spin echo (TSE) sequence (repetition time (TR)/echo time (TE) = 3000/80 ms, slice thickness = 5 mm, echo train length = 14, field of view (FOV) = 210 mm, matrix size = 512 × 512) as well as axial and coronal 2D T1-weighted images, which were obtained with an inversion recovery (IR) sequence (inversion time (TI) = 800 ms, TR/TE = 2000/10 ms, slice thickness = 5 mm, echo train length = 7, FOV = 210 mm, and matrix size = 512 × 512).). In addition, 2D axial-oriented FLAIR images were obtained to evaluate the lesions in these images (TI = 2800 ms, TR/TE = 10,000/120 ms, slice thickness = 5 mm, echo train length = 26, FOV = 210 mm, matrix size = 512 × 512). Sagittal-oriented high-resolution contiguous 3D T1-weighted images were obtained. The 3D T1-weighted images were obtained using a turbo-field echo (TFE) sequence with the following parameters: TI = 1300 ms, TR/TE = 8.6/3.96 ms, Flip angle (FA) = 8◦ , and 1 mm3 isotropic voxel size. To speed-up data acquisition, SENSE (SENSitivity Encoding) parallel imaging was applied with an acceleration factor of two. 2.4. MRI data processing and analysis using FreeSurfer Volumetric analysis was performed based on the 3D T1weighted images using FreeSurfer image analysis suite (version 5.1; http://surfer.nmr.mgh.harvard.edu/), on a 64-bit Linux CentOS 5. The automated procedures for volumetric measures of the different brain structures are described by Fischl et al. [21,22]. Briefly, we first performed image preprocessing, including linear registration, B1 field correction, and non-linear registration. For linear registration, each volume is rigid registered with a specific atlas, such as the Talairach space. Next, any non-homogenous signal intensity due to the B1 bias field is corrected. Then, high dimensional, non-linear morphing to the atlas is performed. After this image preprocessing, the volume is labeled. To label both the cortical and the subcortical volumes in a segmented fashion, we use three pieces of information to disambiguate the labels: (1) the prior probability of a given tissue class occurring at a specific atlas location, (2) the likelihood of the image intensity given that tissue class, and (3) the probability of the local spatial configuration of labels given the tissue class. The technique has previously been shown to be comparable in accuracy to manual labeling [21]. In addition, all segmentations were visually inspected for accuracy prior to inclusion in the group analysis to correct for a potential error in the automated procedure. All images were visually inspected by two neurologists, in which the extent of spatial overlap was identified between overlaid segmented gray or white matter structures. In addition, we used raw T1-weighted images to ensure that obvious errors in skull stripping and tissue segmentation did not occur. As no cases had errors,
p-Value
as determined by visual inspection, all scans were retained in the analysis.
27
0.6266 0.0974 0.0625 0.0804 0.3432 0.0001 0.8405 0.6537 0.9723 0.7248
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
4494.0 ± 449.0 1626.6 ± 216.6 3317.0 ± 598.3 5030.5 ± 553.9 1529.6 ± 135.9 7323.0 ± 736.7 53475.7 ± 5786.8 15481.2 ± 1668.0 230990.9 ± 23051.1 238603.6 ± 25735.0 Mean ± SD, mm3 .
Left side
4443.0 ± 435.7 1550.8 ± 162.3 3571.5 ± 540.8 5388.8 ± 567.3 1566.6 ± 189.1 8397.9 ± 828.3 53202.3 ± 5694.3 15314.8 ± 1459.2 230800.8 ± 23209.0 236477.6 ± 25302.8 0.1013 0.3506 0.2629 0.2732 0.2226 0.0038 0.3589 0.3904 0.8924 0.7076 Hippocampus Amygdala Caudate Putamen Pallidum Thalamus Cerebellar gray matter Cerebellar white matter Cerebral gray matter Cerebral white matter
0.5452 0.2278 0.1961 0.7591 0.7916 0.0016 0.4214 0.8665 0.9677 0.8138
4056.5 ± 616.7 1541.6 ± 280.6 3600.5 ± 391.0 5135.2 ± 742.3 1563.9 ± 349.3 8234.7 ± 907.1 51071.1 ± 5924.1 13519.5 ± 1573.4 229925.3 ± 27508.1 217197.1 ± 24863.8 4424.7 ± 445.7 1622.5 ± 213.1 3422.1 ± 512.8 5115.4 ± 842.3 1516.2 ± 273.4 7556.3 ± 1037.3 54074.9 ± 7483.5 13656.1 ± 1896.8 225372.5 ± 24169.8 238674.3 ± 24655.0 4341.0 ± 464.1 1550.1 ± 177.6 3614.5 ± 456.9 5193.2 ± 829.8 1539.7 ± 312.4 8857.7 ± 1132.4 52306.7 ± 6950.4 13560.7 ± 1844.6 225087.3 ± 24329.4 237005.3 ± 24148.8
4336.3 ± 540.1 1614.9 ± 257.6 3467.3 ± 422.4 4901.3 ± 719.2 1450.9 ± 279.8 7374.1 ± 1045.4 52683.9 ± 6131.7 13112.8 ± 1675.0 231015.1 ± 28017.1 219908.3 ± 24912.9
Healthy controls (n = 36)
p-Value Left side p-Value
Right side
Patients with chronic epilepsy (n = 24)
Right side Left side
Of the 36 healthy controls, 14 patients (39%) were men and 22 patients (61%) were women. The mean age was 35.7 ± 10.9 years. In comparisons of the asymmetry in brain volumes between the left and right sides, the volumes of the hippocampus, amygdala, caudate, putamen, pallidum, cerebellar gray and white matter, and the cerebral gray and white matter were not significantly different. However, the volume of the left thalamus was significantly larger than that of the right thalamus (8397.9 vs. 7323.0 mm3 , p < 0.0001) (Table 1). In addition, there was a negative correlation between age and the putamen (r = −0.5992, p = 0.0001), the pallidum (r = −0.4134, p = 0.0128), the left thalamus (r = −0.4076, p = 0.0136), the right thalamus (r = −0.3926, p = 0.0179), the cerebral gray matter (r = −0.5857, p = 0.0002) and the cerebellar gray matter (r = −0.6123, p = 0.0001). Of the 22 patients with NDE, 13 patients (59%) were men and nine (41%) were women. The mean age was 35.5 ± 13.8 years. No patients with NDE were on antiepileptic drugs (AEDs) at the time of the MRI protocol being performed. The volume of the left thalamus was also larger than that of the right thalamus in patients with NDE (8857.7 vs. 7556.3 mm3 , p = 0.0016) (Table 1). Of the 24 patients with CHE, 11 patients (46%) were men and 13 patients (54%) were women. The mean age was 30.1 ± 9.2 years. The mean age of onset was 16.8 ± 10.4 years. The median duration of epilepsy was 150 months (95% CI 96–198 months, range
Patients with newly diagnosed epilepsy (n = 22)
3.1. Demographics and brain volumes in the study groups
Structure
3. Results
Table 1 A comparison of asymmetry in brain volumes between the left and right hemispheres in patients with epilepsy and in healthy controls.
The volumes of the hippocampus, amygdala, thalamus, caudate, putamen, pallidum, brainstem, cerebellar gray and white matter, as well as cerebral gray and white matter were compared among the three groups. There was significant asymmetry between the right and left sides in the thalamic volumes; thus, we did not sum the thalamic volumes of the right and left sides. However, because there were no volumetric differences between the right and left sides of the other structures, the volumes of both sides were summed. Additionally, we quantified correlations between clinical variables such as age, duration of epilepsy, and brain volumes using Pearson’s correlation test in patients with CHE. Moreover, we investigated the differences of relative structural volumes in the each measure of abnormal volumes among the three groups. The relative volumes were calculated using the following equation: the relative volumes of each structure (%) = (volume of each structure/total intracranial volume) × 100. Comparisons were analyzed using the Chi-square test for categorical variables and Student’s t-test or Mann–Whitney U-test for numerical variables. Categorical variables were presented as the frequency and percentage. Numerical variables with normal distribution were presented as the mean ± standard deviation (SD), and those without normal distribution were described as the median with 95% confidence intervals and the range. Because we compared the volumes of the 12 structures (i.e., the hippocampus, amygdala, caudate, putamen, pallidum, brainstem, cerebellar gray and white matter, cerebral gray and white matter, and the left and right thalamus) between the experimental groups, the statistical significance was set to a p-value <0.004 (0.05/12, Bonferroni correction) with multiple corrections in the analyses of the volumetric differences. However, a p-value of less than 0.05 was considered statistically significant for the other calculations. All statistical tests were performed using MedCalc®.
Right side
2.5. Statistics of clinical factors and radiological findings
28
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
healthy controls (p = 0.0060), although without statistical significance after multiple corrections. 3.3. Vulnerable structures related with chronicity compared to newly diagnosed epilepsy
Fig. 1. Box and whisker plots showing the volume of cerebellar white matter among the three groups. The volume of cerebellar white matter in patients with newly diagnosed epilepsy or chronic epilepsy is significantly smaller than that of the healthy controls. NDE, newly diagnosed epilepsy; CHE, chronic epilepsy.
24–420 months). Twelve patients were considered to have drugresistant epilepsy according to the ILAE recommendation [23], and the remaining patients had epilepsy that was well controlled by AEDs. The volume of the left thalamus was also larger than that of the right thalamus in patients with CHE (8234.9 vs. 7374.1 mm3 , p = 0.0038) (Table 1). There were no significant differences in the male gender and the mean age between the patients with NDE and normal controls (13/22 vs. 14/36, p = 0.2204; 35.5 vs. 35.7 years, p = 0.9705, respectively) and between the patients with CHE and normal controls (11/24 vs. 14/36, p = 0.7893; 30.5 vs. 35.7 years, p = 0.1164, respectively). 3.2. Reduced cerebellar white matter volume indicates the cerebellar white matter as being a vulnerable structure in patients with partial epilepsy of unknown etiology The volumes of the hippocampus (p = 0.4655), amygdala (p = 0.9600), caudate (p = 0.6088), putamen (p = 0.7586), pallidum (p = 0.7278), left thalamus (p = 0.0803), right thalamus (p = 0.0685), brainstem (p = 0.4943), cerebellar gray matter (p = 0.9309), cerebral gray matter (p = 0.1361) and cerebral white matter (p = 0.4523) were not significantly different between patients with NDE and healthy controls after multiple corrections. However, the volume of cerebellar white matter in patients with NDE was significantly smaller than that of the healthy controls after multiple corrections (p = 0.0001). In addition, these findings were consistent with the comparison between patients with CHE and healthy controls. Only the volume of cerebellar white matter in patients with CHE was significantly smaller than that of the healthy controls after multiple corrections (p < 0.0001) (Fig. 1). In addition, there were significant differences in the relative volumes of cerebellar white matter between patients with NDE and healthy controls (1.7347 vs. 2.0575%, p = 0.0001). In addition, there was often a difference in the relative volumes of cerebellar white matter between patients with CHE and healthy controls (1.8708 vs. 2.0575%, p = 0.0272). There were no volume differences between patients with CHE and healthy controls in the other structures, including the hippocampus (p = 0.0266), amygdala (p = 0.8475), caudate (p = 0.5000), putamen (p = 0.2376), pallidum (p = 0.4960), left thalamus (p = 0.4747), right thalamus (p = 0.8250), brainstem (p = 0.0994), cerebellar gray matter (p = 0.3415), and cerebral gray matter (p = 0.9487). However, the cerebral white matter was smaller in patients with CHE than in
We compared the structures between the patients with NDE and CHE of unknown etiology and found that the left thalamus and cerebral white matter tended to be smaller in patients with CHE than in patients with NDE (8234.7 vs. 8857.6 mm3 , p < 0.0446; 437105.4 vs. 464601.8 mm3 , p < 0.0720, respectively) despite not being statistically significant. However, there were no volumetric differences between patients with NDE and CHE in the other structures, including the hippocampus (p = 0.1847), amygdala (p = 0.9010), caudate (p = 0.9041), putamen (p = 0.5475), pallidum (p = 0.8148), right thalamus (p = 0.2204), brainstem (p = 0.4760), cerebellar gray matter (p = 0.5001), cerebellar white matter (p = 0.5558), and cerebral gray matter (p = 0.1995). After identifying these two structures (the left thalamus and cerebral white matter), we tried to correlate the differences with the duration of CHE with the hope of a positive association between these two structures and duration. The volume of the left thalamus (r = −0.4181, p = 0.0421) and cerebral white matter (r = −0.4195, p = 0.0413) were significantly negatively correlated with the duration of epilepsy in patients with CHE (Fig. 2). 4. Discussion This study investigated the anatomic abnormalities in patients with newly diagnosed partial epilepsy of unknown etiology using FreeSurefer image analysis suite. The main finding of this study was that there was significant volume reduction of cerebellar white matter in patients with NDE. We can assume that pre-existing cerebellar abnormalities in the brain may be associated with the development of partial epilepsy of unknown etiology in this study. These findings were replicated in patients with CHE. Furthermore, we found that the volume of cerebral white matter was negatively correlated with the duration of epilepsy. Moreover, we found that the volume of the left thalamus was larger than that of the right thalamus in the healthy control subjects as well as in the patients with epilepsy. Traditionally, epileptic seizures have been thought of as cerebrocortical phenomena, but there is growing evidence that there is an association between the cerebellum and epileptic seizures [24]. The presence of cerebellar volume reduction was already demonstrated in patients with CHE using quantitative MRI investigation [10,12,25]. In addition, cerebellar atrophy was reported in a pathological investigation of the epileptic cerebellum. Here, researchers observed a profound loss of Purkinje cells and granule cells, a preservation of basket cells, and gliosis [26]. Another study also revealed altered white matter integrity in the cerebellum using DTI [27]. However, debates regarding the cause of cerebellar volume reduction continue. Several hypotheses about the mechanism of volume reduction in the cerebellum have been postulated. These include seizure-induced damage (e.g., hypoxia), the effects of AEDs (e.g., phenytoin), and the potential effects of initial etiological insults and early neurodevelopmental factors [10,28,29]. However, the studies that include patients with CHE did not show a causal relationship between the pre-existing anatomic abnormalities in the cerebellum with the onset of epilepsy. Our findings indicate that there was significant cerebellar volume reduction in patients with NDE, which could exclude the confounding effects of recurrent seizures and long-term AEDs on the brain. We can assume that the pre-existing cerebellar volume reductions in patients with partial epilepsy of unknown etiology may be associated with the underlying epileptic pathogenesis that led to the onset of epilepsy.
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
A
mm3
B
r=-0.4181, p=0.0421
Cerebral white maer
CorticalWhiteMatterVol
Le thalamus
10000 9500 9000 8500 8000 7500 7000 6500 0
100
200
300
Duraon of epilepsy
400
500
months
r=-0.4195, p=0.0413
mm3
10500
29
520000 500000 480000 460000 440000 420000 400000 380000 360000 340000 320000 0
100
200
300
400
500
months
Duraon of epilepsy
Fig. 2. A negative correlation between the left thalamus (A) and cerebral white matter (B) and the duration of epilepsy.
Our assumption can be supported by several reports. First, a previous study used DTI to demonstrate abnormal white matter in the cerebellum of patients with NDE [30]. Second, anatomical connections can theoretically support our assumption. Much evidence suggests that the cerebral–cerebellar connections are bidirectional and that the cerebellum exerts an inhibitory effect over seizure activity on the cortex by means of the release of the inhibitory transmitter gamma amino butyric acid (GABA) from Purkinje cells [31,32]. The cortico-ponto-cerebellar tract is a well-known connection between the cerebellum and the frontal, parietal, and occipital lobes of the cortex [33] and ascending connections between the cerebellum and the hippocampus, amygdala, and the temporal lobe have also been identified in both anatomical and electrophysiological studies [34,35]. Moreover, there have been recent studies concerning the functional connectivity between the cerebral cortex and the cerebellum using resting functional MRI [36,37]. Thus, cerebellar volume reduction may contribute to increased seizure activity in the cerebral cortex through these connections. Third, cerebellar stimulation improved or shortened seizure activity in a previous study [38], and total cerebellectomy increased seizure length in a study of kindled cats [39]. In addition, a recent report identified an association between cerebellar volume reduction and reduced seizure control following anterior temporal lobectomy in patients with temporal lobe epilepsy [40]. Fourth, another study demonstrated that children with NDE exhibited mild diffuse cognitive impairment when compared to controls [41]. The pre-existing cerebellar abnormalities might underlie this cognitive impairment as the volumes of the cerebellum were negatively correlated with impaired cognition [41–43]. Lastly, previous studies concerning patients with NDE demonstrated that there were already volume reductions in other structures, such as the hippocampus [44], or micro-anatomic abnormalities in the cerebral white matter [45]. Together, these studies and our results suggest that anatomic abnormalities existed before the onset of epilepsy and that preexisting susceptible structures in the brain may be associated with the development of partial epilepsy of unknown etiology. We also found that the volumes of the left thalamus and the cerebral white matter were negatively correlated with the duration of epilepsy. A potential confounding factor while examining the duration of epilepsy was the subjects’ chronologic age, as there is typically a negative correlation between the volumes of brain structures and age. We also confirmed that increasing age was associated with progressive volume reductions in the cerebral and cerebellar gray matter as well as in the thalamus in healthy controls. Furthermore, this finding had been previously reported [11]. However, in the present study, the volume of cerebral white matter
was independent of age in healthy controls as well as in patients with CHE. Thus, decreasing the volume of cerebral white matter in proportion to the duration of epilepsy was not a consequence of the aging process. In addition, we analyzed partial correlations with age as covariates and found that strongly negative correlation between the volume of cerebral white matter and the duration of epilepsy persisted (r = −0.4717, p = 0.0231). These findings suggested that the cerebral white matter was a more vulnerable structure during the progression of epilepsy. This assumption was supported by a previous report that demonstrated that alterations in the cerebral white matter was more prominent in patients with active epilepsy than in those with remitted epilepsy [46] and that white matter, especially in the periventricular area, was highly susceptible to ischemic injury presumably caused by recurrent seizures [47]. Another interesting finding of our study was that the volume of the left thalamus was larger than that of the right thalamus in the control subjects as well as in the patients with epilepsy. The literature is inconclusive as to the significance of asymmetry within the thalamus. Some studies revealed a significant leftward asymmetry in the thalamus similar to our findings, whereas other study reported opposite results [48,49]. These discrepancies are most likely due to differences of the analysis methods, disease conditions and the handedness of the subjects. Further controlled evaluation is needed. These are several limitations to this study. First, the number of subjects was relatively small. Second, as none of the patients had prolonged video-EEG records, specific seizure localization was not available in all patients with epilepsy. In most cases, we could not be certain as to the specific seizure localization by using interictal scalp EEG recording and semiology alone in patients with partial epilepsy of unknown etiology. Third, we did not investigate the intra-personal follow-up MRI but only analyzed the MRI in the different groups of patients (i.e., NDE or CHE compared to healthy controls). To confirm our finding that cerebral white matter is a vulnerable structure related to chronicity, further studies with intra-personal follow-up MRI might be needed. 5. Conclusion These findings support our hypothesis that cerebellar white matter changes make the cerebellum a pre-existing, susceptible structure in the brain associated with the development of partial epilepsy of unknown etiology. In addition, cerebral white matter was a vulnerable structure throughout the progression of epilepsy.
30
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
References [1] C.P. Panayiotopoulos, A Clinical Guide to Epileptic Syndromes and Their Treatment, second ed., Springer, London, 2007, pp. 14–15. [2] M. Pail, M. Brazdil, R. Marecek, M. Mikl, An optimized voxel-based morphometric study of gray matter changes in patients with left-sided and right-sided mesial temporal lobe epilepsy and hippocampal sclerosis (MTLE/HS), Epilepsia 51 (2010) 511–518. [3] S.S. Keller, U.C. Wieshmann, C.E. Mackay, C.E. Denby, J. Webb, N. Roberts, Voxel based morphometry of grey matter abnormalities in patients with medically intractable temporal lobe epilepsy: effects of side of seizure onset and epilepsy duration, J. Neurol. Neurosurg. Psychiatry 73 (2002) 648–655. [4] A. Labate, A. Cerasa, A. Gambardella, U. Aguglia, A. Quattrone, Hippocampal and thalamic atrophy in mild temporal lobe epilepsy: a VBM study, Neurology 71 (2008) 1094–1101. [5] L. Bonilha, J.C. Edwards, S.L. Kinsman, P.S. Morgan, J. Fridriksson, C. Rorden, et al., Extrahippocampal gray matter loss and hippocampal deafferentation in patients with temporal lobe epilepsy, Epilepsia 51 (2010) 519–528. [6] F. Riederer, R. Lanzenberger, M. Kaya, D. Prayer, W. Serles, C. Baumgartner, Network atrophy in temporal lobe epilepsy: a voxel-based morphometry study, Neurology 71 (2008) 419–425. [7] D.W. Gross, L. Concha, C. Beaulieu, Extratemporal white matter abnormalities in mesial temporal lobe epilepsy demonstrated with diffusion tensor imaging, Epilepsia 47 (2006) 1360–1363. [8] D. Kasperaviciute, C.B. Catarino, E.L. Heinzen, C. Depondt, G.L. Cavalleri, L.O. Caboclo, et al., Common genetic variation and susceptibility to partial epilepsies: a genome-wide association study, Brain 133 (2010) 2136–2147. [9] S.G. Mueller, K.D. Laxer, J. Barakos, I. Cheong, D. Finlay, P. Garcia, et al., Involvement of the thalamocortical network in TLE with and without mesiotemporal sclerosis, Epilepsia 51 (2010) 1436–1445. [10] B.P. Hermann, K. Bayless, R. Hansen, J. Parrish, M. Seidenberg, Cerebellar atrophy in temporal lobe epilepsy, Epilepsy Behav. 7 (2005) 279–287. [11] K. Dabbs, T. Becker, J. Jones, P. Rutecki, M. Seidenberg, B. Hermann, Brain structure and aging in chronic temporal lobe epilepsy, Epilepsia 53 (2012) 1033–1043. [12] S.G. Mueller, K.D. Laxer, N. Cashdollar, S. Buckley, C. Paul, M.W. Weiner, Voxelbased optimized morphometry (VBM) of gray and white matter in temporal lobe epilepsy (TLE) with and without mesial temporal sclerosis, Epilepsia 47 (2006) 900–907. [13] F.G. Woermann, S.L. Free, M.J. Koepp, J. Ashburner, J.S. Duncan, Voxelby-voxel comparison of automatically segmented cerebral gray matter—a rater-independent comparison of structural MRI in patients with epilepsy, Neuroimage 10 (1999) 373–384. [14] J. Ashburner, K.J. Friston, Voxel-based morphometry—the methods, Neuroimage 11 (2000) 805–821. [15] A.M. Dale, B. Fischl, M.I. Sereno, Cortical surface-based analysis. I. Segmentation and surface reconstruction, Neuroimage 9 (1999) 179–194. [16] B. Fischl, M.I. Sereno, A.M. Dale, Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system, Neuroimage 9 (1999) 195–207. [17] A.M. Winkler, P. Kochunov, J. Blangero, L. Almasy, K. Zilles, P.T. Fox, et al., Cortical thickness or grey matter volume? The importance of selecting the phenotype for imaging genetics studies, Neuroimage 53 (2010) 1135–1146. [18] M.S. Panizzon, C. Fennema-Notestine, L.T. Eyler, T.L. Jernigan, E. PromWormley, M. Neale, et al., Distinct genetic influences on cortical surface area and cortical thickness, Cereb. Cortex 19 (2009) 2728–2735. [19] Proposal for revised classification of epilepsies and epileptic syndromes, Commission on Classification and Terminology of the International League Against Epilepsy, Epilepsia (1989) (30) 389–399. [20] A.T. Berg, S.F. Berkovic, M.J. Brodie, J. Buchhalter, J.H. Cross, W. van Emde Boas, et al., Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009, Epilepsia 51 (2010) 676–685. [21] B. Fischl, D.H. Salat, E. Busa, M. Albert, M. Dieterich, C. Haselgrove, et al., Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain, Neuron 33 (2002) 341–355. [22] B. Fischl, A. van der Kouwe, C. Destrieux, E. Halgren, F. Segonne, D.H. Salat, et al., Automatically parcellating the human cerebral cortex, Cereb. Cortex 14 (2004) 11–22. [23] P. Kwan, A. Arzimanoglou, A.T. Berg, M.J. Brodie, W. Allen Hauser, G. Mathern, et al., Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies, Epilepsia 51 (2010) 1069–1077.
[24] A. Kandel, G. Buzsaki, Cerebellar neuronal activity correlates with spike and wave EEG patterns in the rat, Epilepsy Res. 16 (1993) 1–9. [25] C.R. McDonald, D.J. Hagler Jr., M.E. Ahmadi, E. Tecoma, V. Iragui, A.M. Dale, et al., Subcortical and cerebellar atrophy in mesial temporal lobe epilepsy revealed by automatic segmentation, Epilepsy Res. 79 (2008) 130–138. [26] J.H. Margerison, J.A. Corsellis, Epilepsy and the temporal lobes. A clinical, electroencephalographic and neuropathological study of the brain in epilepsy, with particular reference to the temporal lobes, Brain 89 (1966) 499–530. [27] J.D. Riley, D.L. Franklin, V. Choi, R.C. Kim, D.K. Binder, S.C. Cramer, et al., Altered white matter integrity in temporal lobe epilepsy: association with cognitive and clinical profiles, Epilepsia 51 (2010) 536–545. [28] G.C. Ney, G. Lantos, W.B. Barr, N. Schaul, Cerebellar atrophy in patients with long-term phenytoin exposure and epilepsy, Arch. Neurol. 51 (1994) 767–771. [29] M. Dam, T. Bolwig, M. Hertz, J. Bajorec, P. Lomax, A.M. Dam, Does seizure activity produce Purkinje cell loss? Epilepsia 25 (1984) 747–751. [30] E. Widjaja, A. Kis, C. Go, C. Raybaud, O.C. Snead, M.L. Smith, Abnormal white matter on diffusion tensor imaging in children with new-onset seizures, Epilepsy Res. 104 (2013) 105–111. [31] C. Takayama, Y. Inoue, GABAergic signaling in the developing cerebellum, Anat. Sci. Int. 79 (2004) 124–136. [32] P. Brodal, J.G. Bjaalie, Salient anatomic features of the cortico-ponto-cerebellar pathway, Prog. Brain Res. 114 (1997) 227–249. [33] A. Brodal, Cerebrocerebellar pathways. Anatomical data and some functional implications, Acta Neurol. Scand. Suppl. 51 (1972) 153–195. [34] R.G. Heath, C.W. Dempesy, C.J. Fontana, W.A. Myers, Cerebellar stimulation: effects on septal region, hippocampus, and amygdala of cats and rats, Biol. Psychiatry 13 (1978) 501–529. [35] R.G. Heath, J.W. Harper, Ascending projections of the cerebellar fastigial nucleus to the hippocampus, amygdala, and other temporal lobe sites: evoked potential and histological studies in monkeys and cats, Exp. Neurol. 45 (1974) 268–287. [36] C. Habas, N. Kamdar, D. Nguyen, K. Prater, C.F. Beckmann, V. Menon, et al., Distinct cerebellar contributions to intrinsic connectivity networks, J. Neurosci. 29 (2009) 8586–8594. [37] F.M. Krienen, R.L. Buckner, Segregated fronto-cerebellar circuits revealed by intrinsic functional connectivity, Cereb. Cortex 19 (2009) 2485–2497. [38] R. Davis, S.E. Emmonds, Cerebellar stimulation for seizure control: 17-year study, Stereotact. Funct. Neurosurg. 58 (1992) 200–208. [39] T. Tatlisumak, M. Fisher, Handbook of Experimental Neurology, Cambridge University Press, 2006. [40] U. Specht, T. May, R. Schulz, M. Rohde, A. Ebner, R.C. Schmidt, et al., Cerebellar atrophy and prognosis after temporal lobe resection, J. Neurol. Neurosurg. Psychiatry 62 (1997) 501–506. [41] B. Hermann, J. Jones, R. Sheth, C. Dow, M. Koehn, M. Seidenberg, Children with new-onset epilepsy: neuropsychological status and brain structure, Brain 129 (2006) 2609–2619. [42] B. Hermann, M. Seidenberg, L. Sears, R. Hansen, K. Bayless, P. Rutecki, et al., Cerebellar atrophy in temporal lobe epilepsy affects procedural memory, Neurology 63 (2004) 2129–2131. [43] S. Hellwig, V. Gutmann, M.R. Trimble, L.T. van Elst, Cerebellar volume is linked to cognitive function in temporal lobe epilepsy: a quantitative MRI study, Epilepsy Behav. 28 (2013) 156–162. [44] A. Saukkonen, R. Kalviainen, K. Partanen, P. Vainio, P. Riekkinen, A. Pitkanen, Do seizures cause neuronal damage? A MRI study in newly diagnosed and chronic epilepsy, Neuroreport 6 (1994) 219–223. [45] E. Hutchinson, D. Pulsipher, K. Dabbs, Myers, A. Gutierrez, R. Sheth, J. Jones, et al., Children with new-onset epilepsy exhibit diffusion abnormalities in cerebral white matter in the absence of volumetric differences, Epilepsy Res. 88 (2010) 208–214. [46] I. Amarreh, K. Dabbs, D.C. Jackson, J.E. Jones, M.E. Meyerand, C.E. Stafstrom, et al., Cerebral white matter integrity in children with active versus remitted epilepsy 5 years after diagnosis, Epilepsy Res. 107 (2013) 263–271. [47] R.P. Skoff, D.A. Bessert, J.D. Barks, D. Song, M. Cerghet, F.S. Silverstein, Hypoxicischemic injury results in acute disruption of myelin gene expression and death of oligodendroglial precursors in neonatal mice, Int. J. Dev. Neurosci. 19 (2001) 197–208. [48] Y. Xie, Y.A. Chen, M.D. De Bellis, The relationship of age, gender, and IQ with the brainstem and thalamus in healthy children and adolescents: a magnetic resonance imaging volumetric study, J. Child Neurol. 27 (2012) 325–331. [49] R.L. Ahsan, R. Allom, I.S. Gousias, H. Habib, F.E. Turkheimer, S. Free, et al., Volumes, spatial extents and a probabilistic atlas of the human basal ganglia and thalamus, Neuroimage 38 (2007) 261–270.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Cerebellar white matter changes in patients with newly diagnosed partial epilepsy of unknown etiology Kang Min Park a,1 , Yong Hee Han b,1 , Tae Hyung Kim c , Chi Woong Mun b,c , Kyong Jin Shin a , Sam Yeol Ha a , JinSe Park a , Yun Jung Hur d , Hae Yu Kim e , Si Hyung Park f , Sung Eun Kim a,∗ a
Department of Neurology, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea Department of Biomedical Engineering/u-HARC, Inje University, Gimhae, South Korea c Department of Health Science and Technology, Inje University, Gimhae, South Korea d Department of Pediatrics, Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea e Department of Neurosurgery, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea f Department of Internal Medicine, Haeundae Paik Hospital, Inje University College of Medicine, Busan, South Korea b
a r t i c l e
i n f o
Article history: Received 1 October 2014 Received in revised form 10 July 2015 Accepted 24 July 2015 Available online 29 July 2015 Keywords: Epilepsy Cerebellum Atrophy
a b s t r a c t Objective: We hypothesize that pre-existing susceptible structures in the brain may be associated with the development of newly diagnosed partial epilepsy of unknown etiology. Methods: Twenty-two patients with newly diagnosed partial epilepsy of unknown etiology and 36 healthy controls were enrolled in this study. In addition, we included 24 patients with chronic partial epilepsy of unknown etiology as a disease control group. We analyzed whole-brain T1-weighted MRIs using FreeSurfer 5.1. The volumes of the hippocampus, amygdala, thalamus, caudate, putamen, pallidum, brainstem, cerebellar gray and white matter, as well as cerebral gray and white matter were compared between the groups. We also analyzed the changes in brain volumes associated with the chronicity of epilepsy in the patients with chronic epilepsy compared to newly diagnosed epilepsy. Results: The volume of cerebellar white matter in patients with newly diagnosed epilepsy was significantly smaller than that which was observed in the healthy controls (p = 0.0001). This finding was also observed in patients with chronic epilepsy (p < 0.0001). Cerebral white matter volume was negatively correlated with the duration of epilepsy (r = −0.4, p = 0.04). Conclusion: These findings support our hypothesis that cerebellar white matter changes may constitute a pre-existing susceptible structure in the brain that is associated with the development of partial epilepsy of unknown etiology. In addition, cerebral white matter was the structure that was the most vulnerable to the progression of epilepsy. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Epilepsy is one of the most common chronic neurological disorders, with partial epilepsy accounting for 70% of all epilepsy cases [1]. Structural lesions that are observed in brain magnetic resonance imaging (MRI) upon visual inspection can be found in approximately half of all epilepsy patients [1]. Moreover, there is
∗ Corresponding author at: Department of Neurology, Haeundae Paik Hospital, Inje University College of Medicine, Haeundae-ro 875, Haeundae-gu, Busan 612 896, South Korea. Tel.: +82 51 797 1195; fax: +82 51 797 1196. E-mail address: [email protected] (S.E. Kim). 1 These two authors contributed equally to this work. http://dx.doi.org/10.1016/j.clineuro.2015.07.017 0303-8467/© 2015 Elsevier B.V. All rights reserved.
growing evidence that the brains of patients with partial epilepsy have structural lesions and anatomic abnormalities beyond the lesions that are visually observable in MRI. Several studies have demonstrated, using voxel-based morphometry (VBM), a significant volume reduction in the hippocampus as well as in the extra-temporal gray matter of patients with mesial temporal lobe epilepsy and hippocampal sclerosis [2–6]. Previously studies have also found alterations in the extra-temporal white matter using diffusion tensor imaging (DTI) [5–7]. Although there are anatomic abnormalities beyond the visual lesions observed in MRI in patients with partial epilepsy with structural lesions, the pathogenesis of epilepsy might still be directly related to these visual lesions.
26
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
However, the mechanism of partial epilepsy of unknown etiology is unclear. Epilepsy of unknown etiology is presumed to be symptomatic in nature and of an unidentified cause. Both genetic and environmental factors likely play a role to varying extents in individual patients. Nevertheless, a recent genome-wide association study with a large cohort did not identify common variants that influence the risk for epilepsy among patients with partial epilepsy of unknown etiology [8]. Meanwhile, there have been reports that there are anatomic abnormalities in patients with partial epilepsy of unknown etiology. Several reports have demonstrated a volume reduction in the thalamus, cerebellum, hippocampus, caudate, and cerebral white matter [6,9–11]; whereas other studies did not find any anatomic abnormalities in patients with partial epilepsy of unknown etiology [12,13]. These discrepancies are most likely due to methodological differences, such as the method of analysis, the correction for multiple comparisons, and the thresholds set for cluster extent. Furthermore, previous studies shared the problem of enrolling heterogeneous patient groups, thereby mixing patients with newly diagnosed epilepsy (NDE) with those who experienced chronic epilepsy (CHE) [2–7,9–13]. Thus, they could not determine whether the anatomic abnormalities existed before the onset of epilepsy or were the cumulative effects of seizure-induced damage. Therefore, studies that only include patients with NDE are needed. From these studies, we may obtain information concerning the role of pre-existing anatomic abnormalities before the onset of epilepsy in the development of epilepsy of unknown etiology. In this present study, we only enrolled patients with NDE to increase homogeneity, and we included patients with CHE as a disease control group. VBM analysis does a voxel-by-voxel comparison to search for differences in volumes, whereas analysis using FreeSurfer image analysis suite can segment the brain into individual structures and directly measure the volumes of these structures [14–18]. There have been no studies that investigate the anatomic abnormalities in patients with newly diagnosed partial epilepsy of unknown etiology, especially using FreeSurfer image analysis suite. The aim of this study is to clarify whether the anatomic abnormalities existed before the onset of epilepsy in patients with newly diagnosed partial epilepsy of unknown etiology. We hypothesize that pre-existing anatomic abnormalities may be associated with the development of partial epilepsy of unknown etiology.
2. Materials and methods 2.1. Patient subjects This study was conducted with the approval of the Institutional Review Board of our institution. This study was consecutively performed in a single tertiary hospital. We prospectively enrolled 22 patients with a clinical diagnosis of newly diagnosed partial epilepsy of unknown etiology. The diagnosis of epilepsy was made on the basis of clinical histories and electroencephalographic (EEG) findings. We excluded patients that had already displayed only auras or febrile convulsions before their first episodes of unprovoked epileptic seizures. All patients had normal MRI results upon visual inspection using conventional brain MRI protocols, including T1- and T2 weighted images and fluid attenuated inversion recovery (FLAIR) images. All patients had cryptogenic partial epilepsy using the International League Against Epilepsy (ILAE) classification [19] and had epilepsy of an unknown cause according to the current ILAE classification [20]. In addition, we included 24 patients with chronic partial epilepsy of unknown etiology as a disease control group. The definition for CHE in this study was that the epilepsy persisted for more than 2 years. Patients did not have any history of other neurological or psychiatric diseases due to the potential influence of these conditions on the brain atrophy. We collected
demographic data such as age, sex, age of onset, and the duration of epilepsy from these patients at the time of the MRI. 2.2. Normal controls The control group consisted of 36 age- and sex-matched healthy subjects. All subjects had a normal neurological examination and no history of cardiovascular, neurological or psychiatric disease or of diabetes, hypertension, or dyslipidemia. All healthy controls had a normal MRI upon visual inspection. 2.3. MRI data acquisition All patients with NDE had an MRI protocol performed at the initial diagnosis, whereas the patients with CHE had an MRI protocol performed at least 2 years after diagnosis. All scans were performed on a 3.0T MRI scanner (AchievaTx, Phillips Healthcare, Best, The Netherlands) that was equipped with an 8-channel head coil. All subjects underwent conventional brain MRI protocols, including axial and coronal 2D T2-weighted images, which were obtained with a turbo spin echo (TSE) sequence (repetition time (TR)/echo time (TE) = 3000/80 ms, slice thickness = 5 mm, echo train length = 14, field of view (FOV) = 210 mm, matrix size = 512 × 512) as well as axial and coronal 2D T1-weighted images, which were obtained with an inversion recovery (IR) sequence (inversion time (TI) = 800 ms, TR/TE = 2000/10 ms, slice thickness = 5 mm, echo train length = 7, FOV = 210 mm, and matrix size = 512 × 512).). In addition, 2D axial-oriented FLAIR images were obtained to evaluate the lesions in these images (TI = 2800 ms, TR/TE = 10,000/120 ms, slice thickness = 5 mm, echo train length = 26, FOV = 210 mm, matrix size = 512 × 512). Sagittal-oriented high-resolution contiguous 3D T1-weighted images were obtained. The 3D T1-weighted images were obtained using a turbo-field echo (TFE) sequence with the following parameters: TI = 1300 ms, TR/TE = 8.6/3.96 ms, Flip angle (FA) = 8◦ , and 1 mm3 isotropic voxel size. To speed-up data acquisition, SENSE (SENSitivity Encoding) parallel imaging was applied with an acceleration factor of two. 2.4. MRI data processing and analysis using FreeSurfer Volumetric analysis was performed based on the 3D T1weighted images using FreeSurfer image analysis suite (version 5.1; http://surfer.nmr.mgh.harvard.edu/), on a 64-bit Linux CentOS 5. The automated procedures for volumetric measures of the different brain structures are described by Fischl et al. [21,22]. Briefly, we first performed image preprocessing, including linear registration, B1 field correction, and non-linear registration. For linear registration, each volume is rigid registered with a specific atlas, such as the Talairach space. Next, any non-homogenous signal intensity due to the B1 bias field is corrected. Then, high dimensional, non-linear morphing to the atlas is performed. After this image preprocessing, the volume is labeled. To label both the cortical and the subcortical volumes in a segmented fashion, we use three pieces of information to disambiguate the labels: (1) the prior probability of a given tissue class occurring at a specific atlas location, (2) the likelihood of the image intensity given that tissue class, and (3) the probability of the local spatial configuration of labels given the tissue class. The technique has previously been shown to be comparable in accuracy to manual labeling [21]. In addition, all segmentations were visually inspected for accuracy prior to inclusion in the group analysis to correct for a potential error in the automated procedure. All images were visually inspected by two neurologists, in which the extent of spatial overlap was identified between overlaid segmented gray or white matter structures. In addition, we used raw T1-weighted images to ensure that obvious errors in skull stripping and tissue segmentation did not occur. As no cases had errors,
p-Value
as determined by visual inspection, all scans were retained in the analysis.
27
0.6266 0.0974 0.0625 0.0804 0.3432 0.0001 0.8405 0.6537 0.9723 0.7248
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
4494.0 ± 449.0 1626.6 ± 216.6 3317.0 ± 598.3 5030.5 ± 553.9 1529.6 ± 135.9 7323.0 ± 736.7 53475.7 ± 5786.8 15481.2 ± 1668.0 230990.9 ± 23051.1 238603.6 ± 25735.0 Mean ± SD, mm3 .
Left side
4443.0 ± 435.7 1550.8 ± 162.3 3571.5 ± 540.8 5388.8 ± 567.3 1566.6 ± 189.1 8397.9 ± 828.3 53202.3 ± 5694.3 15314.8 ± 1459.2 230800.8 ± 23209.0 236477.6 ± 25302.8 0.1013 0.3506 0.2629 0.2732 0.2226 0.0038 0.3589 0.3904 0.8924 0.7076 Hippocampus Amygdala Caudate Putamen Pallidum Thalamus Cerebellar gray matter Cerebellar white matter Cerebral gray matter Cerebral white matter
0.5452 0.2278 0.1961 0.7591 0.7916 0.0016 0.4214 0.8665 0.9677 0.8138
4056.5 ± 616.7 1541.6 ± 280.6 3600.5 ± 391.0 5135.2 ± 742.3 1563.9 ± 349.3 8234.7 ± 907.1 51071.1 ± 5924.1 13519.5 ± 1573.4 229925.3 ± 27508.1 217197.1 ± 24863.8 4424.7 ± 445.7 1622.5 ± 213.1 3422.1 ± 512.8 5115.4 ± 842.3 1516.2 ± 273.4 7556.3 ± 1037.3 54074.9 ± 7483.5 13656.1 ± 1896.8 225372.5 ± 24169.8 238674.3 ± 24655.0 4341.0 ± 464.1 1550.1 ± 177.6 3614.5 ± 456.9 5193.2 ± 829.8 1539.7 ± 312.4 8857.7 ± 1132.4 52306.7 ± 6950.4 13560.7 ± 1844.6 225087.3 ± 24329.4 237005.3 ± 24148.8
4336.3 ± 540.1 1614.9 ± 257.6 3467.3 ± 422.4 4901.3 ± 719.2 1450.9 ± 279.8 7374.1 ± 1045.4 52683.9 ± 6131.7 13112.8 ± 1675.0 231015.1 ± 28017.1 219908.3 ± 24912.9
Healthy controls (n = 36)
p-Value Left side p-Value
Right side
Patients with chronic epilepsy (n = 24)
Right side Left side
Of the 36 healthy controls, 14 patients (39%) were men and 22 patients (61%) were women. The mean age was 35.7 ± 10.9 years. In comparisons of the asymmetry in brain volumes between the left and right sides, the volumes of the hippocampus, amygdala, caudate, putamen, pallidum, cerebellar gray and white matter, and the cerebral gray and white matter were not significantly different. However, the volume of the left thalamus was significantly larger than that of the right thalamus (8397.9 vs. 7323.0 mm3 , p < 0.0001) (Table 1). In addition, there was a negative correlation between age and the putamen (r = −0.5992, p = 0.0001), the pallidum (r = −0.4134, p = 0.0128), the left thalamus (r = −0.4076, p = 0.0136), the right thalamus (r = −0.3926, p = 0.0179), the cerebral gray matter (r = −0.5857, p = 0.0002) and the cerebellar gray matter (r = −0.6123, p = 0.0001). Of the 22 patients with NDE, 13 patients (59%) were men and nine (41%) were women. The mean age was 35.5 ± 13.8 years. No patients with NDE were on antiepileptic drugs (AEDs) at the time of the MRI protocol being performed. The volume of the left thalamus was also larger than that of the right thalamus in patients with NDE (8857.7 vs. 7556.3 mm3 , p = 0.0016) (Table 1). Of the 24 patients with CHE, 11 patients (46%) were men and 13 patients (54%) were women. The mean age was 30.1 ± 9.2 years. The mean age of onset was 16.8 ± 10.4 years. The median duration of epilepsy was 150 months (95% CI 96–198 months, range
Patients with newly diagnosed epilepsy (n = 22)
3.1. Demographics and brain volumes in the study groups
Structure
3. Results
Table 1 A comparison of asymmetry in brain volumes between the left and right hemispheres in patients with epilepsy and in healthy controls.
The volumes of the hippocampus, amygdala, thalamus, caudate, putamen, pallidum, brainstem, cerebellar gray and white matter, as well as cerebral gray and white matter were compared among the three groups. There was significant asymmetry between the right and left sides in the thalamic volumes; thus, we did not sum the thalamic volumes of the right and left sides. However, because there were no volumetric differences between the right and left sides of the other structures, the volumes of both sides were summed. Additionally, we quantified correlations between clinical variables such as age, duration of epilepsy, and brain volumes using Pearson’s correlation test in patients with CHE. Moreover, we investigated the differences of relative structural volumes in the each measure of abnormal volumes among the three groups. The relative volumes were calculated using the following equation: the relative volumes of each structure (%) = (volume of each structure/total intracranial volume) × 100. Comparisons were analyzed using the Chi-square test for categorical variables and Student’s t-test or Mann–Whitney U-test for numerical variables. Categorical variables were presented as the frequency and percentage. Numerical variables with normal distribution were presented as the mean ± standard deviation (SD), and those without normal distribution were described as the median with 95% confidence intervals and the range. Because we compared the volumes of the 12 structures (i.e., the hippocampus, amygdala, caudate, putamen, pallidum, brainstem, cerebellar gray and white matter, cerebral gray and white matter, and the left and right thalamus) between the experimental groups, the statistical significance was set to a p-value <0.004 (0.05/12, Bonferroni correction) with multiple corrections in the analyses of the volumetric differences. However, a p-value of less than 0.05 was considered statistically significant for the other calculations. All statistical tests were performed using MedCalc®.
Right side
2.5. Statistics of clinical factors and radiological findings
28
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
healthy controls (p = 0.0060), although without statistical significance after multiple corrections. 3.3. Vulnerable structures related with chronicity compared to newly diagnosed epilepsy
Fig. 1. Box and whisker plots showing the volume of cerebellar white matter among the three groups. The volume of cerebellar white matter in patients with newly diagnosed epilepsy or chronic epilepsy is significantly smaller than that of the healthy controls. NDE, newly diagnosed epilepsy; CHE, chronic epilepsy.
24–420 months). Twelve patients were considered to have drugresistant epilepsy according to the ILAE recommendation [23], and the remaining patients had epilepsy that was well controlled by AEDs. The volume of the left thalamus was also larger than that of the right thalamus in patients with CHE (8234.9 vs. 7374.1 mm3 , p = 0.0038) (Table 1). There were no significant differences in the male gender and the mean age between the patients with NDE and normal controls (13/22 vs. 14/36, p = 0.2204; 35.5 vs. 35.7 years, p = 0.9705, respectively) and between the patients with CHE and normal controls (11/24 vs. 14/36, p = 0.7893; 30.5 vs. 35.7 years, p = 0.1164, respectively). 3.2. Reduced cerebellar white matter volume indicates the cerebellar white matter as being a vulnerable structure in patients with partial epilepsy of unknown etiology The volumes of the hippocampus (p = 0.4655), amygdala (p = 0.9600), caudate (p = 0.6088), putamen (p = 0.7586), pallidum (p = 0.7278), left thalamus (p = 0.0803), right thalamus (p = 0.0685), brainstem (p = 0.4943), cerebellar gray matter (p = 0.9309), cerebral gray matter (p = 0.1361) and cerebral white matter (p = 0.4523) were not significantly different between patients with NDE and healthy controls after multiple corrections. However, the volume of cerebellar white matter in patients with NDE was significantly smaller than that of the healthy controls after multiple corrections (p = 0.0001). In addition, these findings were consistent with the comparison between patients with CHE and healthy controls. Only the volume of cerebellar white matter in patients with CHE was significantly smaller than that of the healthy controls after multiple corrections (p < 0.0001) (Fig. 1). In addition, there were significant differences in the relative volumes of cerebellar white matter between patients with NDE and healthy controls (1.7347 vs. 2.0575%, p = 0.0001). In addition, there was often a difference in the relative volumes of cerebellar white matter between patients with CHE and healthy controls (1.8708 vs. 2.0575%, p = 0.0272). There were no volume differences between patients with CHE and healthy controls in the other structures, including the hippocampus (p = 0.0266), amygdala (p = 0.8475), caudate (p = 0.5000), putamen (p = 0.2376), pallidum (p = 0.4960), left thalamus (p = 0.4747), right thalamus (p = 0.8250), brainstem (p = 0.0994), cerebellar gray matter (p = 0.3415), and cerebral gray matter (p = 0.9487). However, the cerebral white matter was smaller in patients with CHE than in
We compared the structures between the patients with NDE and CHE of unknown etiology and found that the left thalamus and cerebral white matter tended to be smaller in patients with CHE than in patients with NDE (8234.7 vs. 8857.6 mm3 , p < 0.0446; 437105.4 vs. 464601.8 mm3 , p < 0.0720, respectively) despite not being statistically significant. However, there were no volumetric differences between patients with NDE and CHE in the other structures, including the hippocampus (p = 0.1847), amygdala (p = 0.9010), caudate (p = 0.9041), putamen (p = 0.5475), pallidum (p = 0.8148), right thalamus (p = 0.2204), brainstem (p = 0.4760), cerebellar gray matter (p = 0.5001), cerebellar white matter (p = 0.5558), and cerebral gray matter (p = 0.1995). After identifying these two structures (the left thalamus and cerebral white matter), we tried to correlate the differences with the duration of CHE with the hope of a positive association between these two structures and duration. The volume of the left thalamus (r = −0.4181, p = 0.0421) and cerebral white matter (r = −0.4195, p = 0.0413) were significantly negatively correlated with the duration of epilepsy in patients with CHE (Fig. 2). 4. Discussion This study investigated the anatomic abnormalities in patients with newly diagnosed partial epilepsy of unknown etiology using FreeSurefer image analysis suite. The main finding of this study was that there was significant volume reduction of cerebellar white matter in patients with NDE. We can assume that pre-existing cerebellar abnormalities in the brain may be associated with the development of partial epilepsy of unknown etiology in this study. These findings were replicated in patients with CHE. Furthermore, we found that the volume of cerebral white matter was negatively correlated with the duration of epilepsy. Moreover, we found that the volume of the left thalamus was larger than that of the right thalamus in the healthy control subjects as well as in the patients with epilepsy. Traditionally, epileptic seizures have been thought of as cerebrocortical phenomena, but there is growing evidence that there is an association between the cerebellum and epileptic seizures [24]. The presence of cerebellar volume reduction was already demonstrated in patients with CHE using quantitative MRI investigation [10,12,25]. In addition, cerebellar atrophy was reported in a pathological investigation of the epileptic cerebellum. Here, researchers observed a profound loss of Purkinje cells and granule cells, a preservation of basket cells, and gliosis [26]. Another study also revealed altered white matter integrity in the cerebellum using DTI [27]. However, debates regarding the cause of cerebellar volume reduction continue. Several hypotheses about the mechanism of volume reduction in the cerebellum have been postulated. These include seizure-induced damage (e.g., hypoxia), the effects of AEDs (e.g., phenytoin), and the potential effects of initial etiological insults and early neurodevelopmental factors [10,28,29]. However, the studies that include patients with CHE did not show a causal relationship between the pre-existing anatomic abnormalities in the cerebellum with the onset of epilepsy. Our findings indicate that there was significant cerebellar volume reduction in patients with NDE, which could exclude the confounding effects of recurrent seizures and long-term AEDs on the brain. We can assume that the pre-existing cerebellar volume reductions in patients with partial epilepsy of unknown etiology may be associated with the underlying epileptic pathogenesis that led to the onset of epilepsy.
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
A
mm3
B
r=-0.4181, p=0.0421
Cerebral white maer
CorticalWhiteMatterVol
Le thalamus
10000 9500 9000 8500 8000 7500 7000 6500 0
100
200
300
Duraon of epilepsy
400
500
months
r=-0.4195, p=0.0413
mm3
10500
29
520000 500000 480000 460000 440000 420000 400000 380000 360000 340000 320000 0
100
200
300
400
500
months
Duraon of epilepsy
Fig. 2. A negative correlation between the left thalamus (A) and cerebral white matter (B) and the duration of epilepsy.
Our assumption can be supported by several reports. First, a previous study used DTI to demonstrate abnormal white matter in the cerebellum of patients with NDE [30]. Second, anatomical connections can theoretically support our assumption. Much evidence suggests that the cerebral–cerebellar connections are bidirectional and that the cerebellum exerts an inhibitory effect over seizure activity on the cortex by means of the release of the inhibitory transmitter gamma amino butyric acid (GABA) from Purkinje cells [31,32]. The cortico-ponto-cerebellar tract is a well-known connection between the cerebellum and the frontal, parietal, and occipital lobes of the cortex [33] and ascending connections between the cerebellum and the hippocampus, amygdala, and the temporal lobe have also been identified in both anatomical and electrophysiological studies [34,35]. Moreover, there have been recent studies concerning the functional connectivity between the cerebral cortex and the cerebellum using resting functional MRI [36,37]. Thus, cerebellar volume reduction may contribute to increased seizure activity in the cerebral cortex through these connections. Third, cerebellar stimulation improved or shortened seizure activity in a previous study [38], and total cerebellectomy increased seizure length in a study of kindled cats [39]. In addition, a recent report identified an association between cerebellar volume reduction and reduced seizure control following anterior temporal lobectomy in patients with temporal lobe epilepsy [40]. Fourth, another study demonstrated that children with NDE exhibited mild diffuse cognitive impairment when compared to controls [41]. The pre-existing cerebellar abnormalities might underlie this cognitive impairment as the volumes of the cerebellum were negatively correlated with impaired cognition [41–43]. Lastly, previous studies concerning patients with NDE demonstrated that there were already volume reductions in other structures, such as the hippocampus [44], or micro-anatomic abnormalities in the cerebral white matter [45]. Together, these studies and our results suggest that anatomic abnormalities existed before the onset of epilepsy and that preexisting susceptible structures in the brain may be associated with the development of partial epilepsy of unknown etiology. We also found that the volumes of the left thalamus and the cerebral white matter were negatively correlated with the duration of epilepsy. A potential confounding factor while examining the duration of epilepsy was the subjects’ chronologic age, as there is typically a negative correlation between the volumes of brain structures and age. We also confirmed that increasing age was associated with progressive volume reductions in the cerebral and cerebellar gray matter as well as in the thalamus in healthy controls. Furthermore, this finding had been previously reported [11]. However, in the present study, the volume of cerebral white matter
was independent of age in healthy controls as well as in patients with CHE. Thus, decreasing the volume of cerebral white matter in proportion to the duration of epilepsy was not a consequence of the aging process. In addition, we analyzed partial correlations with age as covariates and found that strongly negative correlation between the volume of cerebral white matter and the duration of epilepsy persisted (r = −0.4717, p = 0.0231). These findings suggested that the cerebral white matter was a more vulnerable structure during the progression of epilepsy. This assumption was supported by a previous report that demonstrated that alterations in the cerebral white matter was more prominent in patients with active epilepsy than in those with remitted epilepsy [46] and that white matter, especially in the periventricular area, was highly susceptible to ischemic injury presumably caused by recurrent seizures [47]. Another interesting finding of our study was that the volume of the left thalamus was larger than that of the right thalamus in the control subjects as well as in the patients with epilepsy. The literature is inconclusive as to the significance of asymmetry within the thalamus. Some studies revealed a significant leftward asymmetry in the thalamus similar to our findings, whereas other study reported opposite results [48,49]. These discrepancies are most likely due to differences of the analysis methods, disease conditions and the handedness of the subjects. Further controlled evaluation is needed. These are several limitations to this study. First, the number of subjects was relatively small. Second, as none of the patients had prolonged video-EEG records, specific seizure localization was not available in all patients with epilepsy. In most cases, we could not be certain as to the specific seizure localization by using interictal scalp EEG recording and semiology alone in patients with partial epilepsy of unknown etiology. Third, we did not investigate the intra-personal follow-up MRI but only analyzed the MRI in the different groups of patients (i.e., NDE or CHE compared to healthy controls). To confirm our finding that cerebral white matter is a vulnerable structure related to chronicity, further studies with intra-personal follow-up MRI might be needed. 5. Conclusion These findings support our hypothesis that cerebellar white matter changes make the cerebellum a pre-existing, susceptible structure in the brain associated with the development of partial epilepsy of unknown etiology. In addition, cerebral white matter was a vulnerable structure throughout the progression of epilepsy.
30
K.M. Park et al. / Clinical Neurology and Neurosurgery 138 (2015) 25–30
References [1] C.P. Panayiotopoulos, A Clinical Guide to Epileptic Syndromes and Their Treatment, second ed., Springer, London, 2007, pp. 14–15. [2] M. Pail, M. Brazdil, R. Marecek, M. Mikl, An optimized voxel-based morphometric study of gray matter changes in patients with left-sided and right-sided mesial temporal lobe epilepsy and hippocampal sclerosis (MTLE/HS), Epilepsia 51 (2010) 511–518. [3] S.S. Keller, U.C. Wieshmann, C.E. Mackay, C.E. Denby, J. Webb, N. Roberts, Voxel based morphometry of grey matter abnormalities in patients with medically intractable temporal lobe epilepsy: effects of side of seizure onset and epilepsy duration, J. Neurol. Neurosurg. Psychiatry 73 (2002) 648–655. [4] A. Labate, A. Cerasa, A. Gambardella, U. Aguglia, A. Quattrone, Hippocampal and thalamic atrophy in mild temporal lobe epilepsy: a VBM study, Neurology 71 (2008) 1094–1101. [5] L. Bonilha, J.C. Edwards, S.L. Kinsman, P.S. Morgan, J. Fridriksson, C. Rorden, et al., Extrahippocampal gray matter loss and hippocampal deafferentation in patients with temporal lobe epilepsy, Epilepsia 51 (2010) 519–528. [6] F. Riederer, R. Lanzenberger, M. Kaya, D. Prayer, W. Serles, C. Baumgartner, Network atrophy in temporal lobe epilepsy: a voxel-based morphometry study, Neurology 71 (2008) 419–425. [7] D.W. Gross, L. Concha, C. Beaulieu, Extratemporal white matter abnormalities in mesial temporal lobe epilepsy demonstrated with diffusion tensor imaging, Epilepsia 47 (2006) 1360–1363. [8] D. Kasperaviciute, C.B. Catarino, E.L. Heinzen, C. Depondt, G.L. Cavalleri, L.O. Caboclo, et al., Common genetic variation and susceptibility to partial epilepsies: a genome-wide association study, Brain 133 (2010) 2136–2147. [9] S.G. Mueller, K.D. Laxer, J. Barakos, I. Cheong, D. Finlay, P. Garcia, et al., Involvement of the thalamocortical network in TLE with and without mesiotemporal sclerosis, Epilepsia 51 (2010) 1436–1445. [10] B.P. Hermann, K. Bayless, R. Hansen, J. Parrish, M. Seidenberg, Cerebellar atrophy in temporal lobe epilepsy, Epilepsy Behav. 7 (2005) 279–287. [11] K. Dabbs, T. Becker, J. Jones, P. Rutecki, M. Seidenberg, B. Hermann, Brain structure and aging in chronic temporal lobe epilepsy, Epilepsia 53 (2012) 1033–1043. [12] S.G. Mueller, K.D. Laxer, N. Cashdollar, S. Buckley, C. Paul, M.W. Weiner, Voxelbased optimized morphometry (VBM) of gray and white matter in temporal lobe epilepsy (TLE) with and without mesial temporal sclerosis, Epilepsia 47 (2006) 900–907. [13] F.G. Woermann, S.L. Free, M.J. Koepp, J. Ashburner, J.S. Duncan, Voxelby-voxel comparison of automatically segmented cerebral gray matter—a rater-independent comparison of structural MRI in patients with epilepsy, Neuroimage 10 (1999) 373–384. [14] J. Ashburner, K.J. Friston, Voxel-based morphometry—the methods, Neuroimage 11 (2000) 805–821. [15] A.M. Dale, B. Fischl, M.I. Sereno, Cortical surface-based analysis. I. Segmentation and surface reconstruction, Neuroimage 9 (1999) 179–194. [16] B. Fischl, M.I. Sereno, A.M. Dale, Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system, Neuroimage 9 (1999) 195–207. [17] A.M. Winkler, P. Kochunov, J. Blangero, L. Almasy, K. Zilles, P.T. Fox, et al., Cortical thickness or grey matter volume? The importance of selecting the phenotype for imaging genetics studies, Neuroimage 53 (2010) 1135–1146. [18] M.S. Panizzon, C. Fennema-Notestine, L.T. Eyler, T.L. Jernigan, E. PromWormley, M. Neale, et al., Distinct genetic influences on cortical surface area and cortical thickness, Cereb. Cortex 19 (2009) 2728–2735. [19] Proposal for revised classification of epilepsies and epileptic syndromes, Commission on Classification and Terminology of the International League Against Epilepsy, Epilepsia (1989) (30) 389–399. [20] A.T. Berg, S.F. Berkovic, M.J. Brodie, J. Buchhalter, J.H. Cross, W. van Emde Boas, et al., Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009, Epilepsia 51 (2010) 676–685. [21] B. Fischl, D.H. Salat, E. Busa, M. Albert, M. Dieterich, C. Haselgrove, et al., Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain, Neuron 33 (2002) 341–355. [22] B. Fischl, A. van der Kouwe, C. Destrieux, E. Halgren, F. Segonne, D.H. Salat, et al., Automatically parcellating the human cerebral cortex, Cereb. Cortex 14 (2004) 11–22. [23] P. Kwan, A. Arzimanoglou, A.T. Berg, M.J. Brodie, W. Allen Hauser, G. Mathern, et al., Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies, Epilepsia 51 (2010) 1069–1077.
[24] A. Kandel, G. Buzsaki, Cerebellar neuronal activity correlates with spike and wave EEG patterns in the rat, Epilepsy Res. 16 (1993) 1–9. [25] C.R. McDonald, D.J. Hagler Jr., M.E. Ahmadi, E. Tecoma, V. Iragui, A.M. Dale, et al., Subcortical and cerebellar atrophy in mesial temporal lobe epilepsy revealed by automatic segmentation, Epilepsy Res. 79 (2008) 130–138. [26] J.H. Margerison, J.A. Corsellis, Epilepsy and the temporal lobes. A clinical, electroencephalographic and neuropathological study of the brain in epilepsy, with particular reference to the temporal lobes, Brain 89 (1966) 499–530. [27] J.D. Riley, D.L. Franklin, V. Choi, R.C. Kim, D.K. Binder, S.C. Cramer, et al., Altered white matter integrity in temporal lobe epilepsy: association with cognitive and clinical profiles, Epilepsia 51 (2010) 536–545. [28] G.C. Ney, G. Lantos, W.B. Barr, N. Schaul, Cerebellar atrophy in patients with long-term phenytoin exposure and epilepsy, Arch. Neurol. 51 (1994) 767–771. [29] M. Dam, T. Bolwig, M. Hertz, J. Bajorec, P. Lomax, A.M. Dam, Does seizure activity produce Purkinje cell loss? Epilepsia 25 (1984) 747–751. [30] E. Widjaja, A. Kis, C. Go, C. Raybaud, O.C. Snead, M.L. Smith, Abnormal white matter on diffusion tensor imaging in children with new-onset seizures, Epilepsy Res. 104 (2013) 105–111. [31] C. Takayama, Y. Inoue, GABAergic signaling in the developing cerebellum, Anat. Sci. Int. 79 (2004) 124–136. [32] P. Brodal, J.G. Bjaalie, Salient anatomic features of the cortico-ponto-cerebellar pathway, Prog. Brain Res. 114 (1997) 227–249. [33] A. Brodal, Cerebrocerebellar pathways. Anatomical data and some functional implications, Acta Neurol. Scand. Suppl. 51 (1972) 153–195. [34] R.G. Heath, C.W. Dempesy, C.J. Fontana, W.A. Myers, Cerebellar stimulation: effects on septal region, hippocampus, and amygdala of cats and rats, Biol. Psychiatry 13 (1978) 501–529. [35] R.G. Heath, J.W. Harper, Ascending projections of the cerebellar fastigial nucleus to the hippocampus, amygdala, and other temporal lobe sites: evoked potential and histological studies in monkeys and cats, Exp. Neurol. 45 (1974) 268–287. [36] C. Habas, N. Kamdar, D. Nguyen, K. Prater, C.F. Beckmann, V. Menon, et al., Distinct cerebellar contributions to intrinsic connectivity networks, J. Neurosci. 29 (2009) 8586–8594. [37] F.M. Krienen, R.L. Buckner, Segregated fronto-cerebellar circuits revealed by intrinsic functional connectivity, Cereb. Cortex 19 (2009) 2485–2497. [38] R. Davis, S.E. Emmonds, Cerebellar stimulation for seizure control: 17-year study, Stereotact. Funct. Neurosurg. 58 (1992) 200–208. [39] T. Tatlisumak, M. Fisher, Handbook of Experimental Neurology, Cambridge University Press, 2006. [40] U. Specht, T. May, R. Schulz, M. Rohde, A. Ebner, R.C. Schmidt, et al., Cerebellar atrophy and prognosis after temporal lobe resection, J. Neurol. Neurosurg. Psychiatry 62 (1997) 501–506. [41] B. Hermann, J. Jones, R. Sheth, C. Dow, M. Koehn, M. Seidenberg, Children with new-onset epilepsy: neuropsychological status and brain structure, Brain 129 (2006) 2609–2619. [42] B. Hermann, M. Seidenberg, L. Sears, R. Hansen, K. Bayless, P. Rutecki, et al., Cerebellar atrophy in temporal lobe epilepsy affects procedural memory, Neurology 63 (2004) 2129–2131. [43] S. Hellwig, V. Gutmann, M.R. Trimble, L.T. van Elst, Cerebellar volume is linked to cognitive function in temporal lobe epilepsy: a quantitative MRI study, Epilepsy Behav. 28 (2013) 156–162. [44] A. Saukkonen, R. Kalviainen, K. Partanen, P. Vainio, P. Riekkinen, A. Pitkanen, Do seizures cause neuronal damage? A MRI study in newly diagnosed and chronic epilepsy, Neuroreport 6 (1994) 219–223. [45] E. Hutchinson, D. Pulsipher, K. Dabbs, Myers, A. Gutierrez, R. Sheth, J. Jones, et al., Children with new-onset epilepsy exhibit diffusion abnormalities in cerebral white matter in the absence of volumetric differences, Epilepsy Res. 88 (2010) 208–214. [46] I. Amarreh, K. Dabbs, D.C. Jackson, J.E. Jones, M.E. Meyerand, C.E. Stafstrom, et al., Cerebral white matter integrity in children with active versus remitted epilepsy 5 years after diagnosis, Epilepsy Res. 107 (2013) 263–271. [47] R.P. Skoff, D.A. Bessert, J.D. Barks, D. Song, M. Cerghet, F.S. Silverstein, Hypoxicischemic injury results in acute disruption of myelin gene expression and death of oligodendroglial precursors in neonatal mice, Int. J. Dev. Neurosci. 19 (2001) 197–208. [48] Y. Xie, Y.A. Chen, M.D. De Bellis, The relationship of age, gender, and IQ with the brainstem and thalamus in healthy children and adolescents: a magnetic resonance imaging volumetric study, J. Child Neurol. 27 (2012) 325–331. [49] R.L. Ahsan, R. Allom, I.S. Gousias, H. Habib, F.E. Turkheimer, S. Free, et al., Volumes, spatial extents and a probabilistic atlas of the human basal ganglia and thalamus, Neuroimage 38 (2007) 261–270.