FULL-LENGTH ORIGINAL RESEARCH
Mesial temporal lobe epilepsy with hippocampal sclerosis is a network disorder with altered cortical hubs *†Seung-Hyun Jin, *‡Woorim Jeong, and *†‡§¶Chun Kee Chung Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
Seung-Hyun Jin is a principal researcher in Neuroscience Research Institute at Seoul National University.
Objective: Electrophysiologic hubs within the large-scale functional networks in mesial temporal lobe epilepsy (mTLE) with hippocampal sclerosis (HS) have not been investigated. We hypothesized that mTLE with HS has different resting-state network hubs in their large-scale functional networks compared to the hubs in healthy controls (HC). We also hypothesized that the hippocampus would be a functional hub in mTLE patients with HS. Methods: Resting-state functional networks, identified by using magnetoencephalography (MEG) signals in the theta, alpha, beta, and gamma frequency bands, were evaluated. Networks in 44 mTLE patients with HS (left mTLE = 22; right mTLE = 22) were compared with those in 46 age-matched HC. We investigated betweenness centrality at the source-level MEG network. Results: The main network hubs were at the pole of the left superior temporal gyrus in the beta band, the pole of the left middle temporal gyrus in the beta and gamma bands, left hippocampus in the theta and alpha bands, and right posterior cingulate gyrus in all four frequency bands in mTLE patients; all of which were different from the main network hubs in HC. Only patients with left mTLE showed profound differences from HC at the left hippocampus in the alpha band. Significance: Our analysis of resting-state MEG signals shows that altered electrophysiologic functional hubs in mTLE patients reflect pathophysiologic brain network reorganization. Because we detected network hubs in both hippocampal and extrahippocampal areas, it is probable that mTLE is a large-scale network disorder rather than a focal disorder. The hippocampus was a network hub in left mTLE but not in right mTLE patients, which may be due to intrinsic functional and structural asymmetries between left and right mTLE patients. The evaluation of cortical hubs, even in the spike-free resting-state, could be a clinical diagnostic marker of mTLE with HS. KEY WORDS: Mesial temporal lobe epilepsy, Hippocampal sclerosis, Resting-state magnetoencephalography, Large-scale functional network, Electrophysiologic cortical hubs.
Temporal lobe epilepsy (TLE) is the most common drugresistant epilepsy in adults.1 Because epilepsy is a disorder of distributed neural networks,2 recent advances in brain network analysis have been used to explore network abnormalities in patients with TLE. Evidence for altered brain networks in TLE has been accumulating through both structural3–6 and functional7–9 network analyses. Graph-theoretic analysis of brain networks enables us to describe the topologic organization of connectivity,10,11 and allows for development of a network perspective on pathophysiology.12 Network hubs are brain regions that occupy a central position within the network, and they can be assessed by using
Accepted February 11, 2015. *Department of Neurosurgery, Seoul National University Hospital, Seoul, Korea; †Neuroscience Research Institute, Seoul National University Medical Research Center, Seoul, Korea; ‡Interdisciplinary Program in Neuroscience, Seoul National University College of Natural Science, Seoul, Korea; §Department of Neurosurgery, Seoul National University College of Medicine, Seoul, Korea; and ¶Department of Brain and Cognitive Sciences, Seoul National University College of Natural Sciences, Seoul, Korea Address correspondence to Chun Kee Chung, Department of Neurosurgery, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Korea. E-mail: [email protected]
Wiley Periodicals, Inc. © 2015 International League Against Epilepsy
2 S.-H. Jin et al. graph-theoretic techniques among various measures for characterizing large-scale networks.11 Network hubs have a privileged role in organizing network dynamics; thus, assessing hubs may allow improved description of their potential roles in pathologic processes.13 In TLE, network hubs have been investigated with structural6 and functional7 magnetic resonance imaging (MRI). Electrophysiologic network analysis allows the precise assessment of brain dynamics by virtue of its high temporal resolution compared to that provided by functional MRI (fMRI). However, the electrophysiologic hubs of the largescale (whole-brain) functional networks in TLE, and particularly in mesial TLE (mTLE) with hippocampal sclerosis (HS), have not been investigated. In this study, we focused on mTLE patients with histopathologically proven HS and postsurgical seizure freedom in order to obtain patient population homogeneity in terms of main pathology and surgical outcome. We investigated the large-scale functional networks in different frequency bands in resting-state mTLE patients with HS and in healthy controls (HC) to elucidate their functional hubs and their group-wise differences. We anticipated that mTLE patients with HS would have different resting-state functional hubs compared to those in HC. Because we included only mTLE patients who had HS that was surgically removed, and who achieved postsurgical seizure freedom, we suspected that the hippocampus would be a leading candidate for the epileptogenic focus. We thus hypothesized that the hippocampus would comprise functional hubs in the resting-state large-scale brain networks of mTLE patients with HS.
Of the 44 enrolled patients, 42 (left mTLE = 22; right mTLE = 20) underwent neuropsychological testing before performing surgical procedures. The neuropsychological tests included the Korean version of the Rey Auditory Verbal Learning Test (K-AVLT) for verbal memory, and the Korean version of the Rey Complex Figure Test (K-CFT) for visual memory.14 Forty-six right-handed HC voluntarily participated in this study. The MEG data from all 46 control subjects were also used in our previous study.15 No participant had neurologic abnormalities or MRI lesions. Written informed consent was submitted by all subjects. The study protocol was approved by the Seoul National University Hospital Institutional Review Board (IRB H-0607-029-178).
MRI evaluation Preoperative MRI was performed on either a GE 1.5 T or 3 T MRI unit (GE Horizon Echospeed; GE Healthcare, Pollards Wood, United Kingdom) or on a Siemens 1.5 T scanner (MAGNETOM Avanto; Siemens, Berlin, Germany). The standard MRI protocol included T1-weighted, T2, fluid-attenuated inversion recovery (FLAIR) axial, T2 and FLAIR oblique coronal, fast inversion recovery with myelin suppression, and three-dimensional (3D) gradient echo coronal T1 images with whole-brain coverage. The 3D gradient echo T1 images were reconstructed to a 1 mm slice thickness, whereas the T2 images were acquired by using a 3 mm thickness with a 1 mm interslice gap. Preoperative MRI scans were reviewed separately and results confirmed by two neuroradiologists specializing in epilepsy and blinded to seizure focus.
Subjects Forty-four mTLE with HS patients (right-handed = 41; left-handed = 1; ambidextrous = 2) were enrolled in this retrospective study of 87 patients who underwent epilepsy surgery between 2005 and 2011 at Seoul National University Hospital and who had been followed for >2 years. We included mTLE patients with histopathologically proven HS and postsurgical seizure freedom in order to obtain patient population homogeneity in terms of main pathology and surgical outcome. Twenty-two of the 44 patients were mTLE with left HS (left mTLE), and the other 22 patients were mTLE with right HS (right mTLE). Patients with no magnetoencephalography (MEG) before surgery (n = 10), no seizure-free period (n = 18), aged 2 SD were defined as functional hubs with a high BC for each subject and in each frequency band.15 To construct a ranked distribution at each frequency band, the extracted hubs per subject were added across subjects, and transformed into percentage ranks at each node. Then, the ranked distributions of hubs irrespective of frequency bands were constructed based on the aggregated ranking percentage across subjects (>50%, Fig. 1) in both HC and mTLE groups. This approach allowed us to determine the electrophysiologic hubs of the large-scale functional networks within each group.15,18 For the identified hubs in the mTLE patients through those previous steps, a Kruskal-Wallis test was performed to assess the differences between the HC and mTLE groups. If the test result was significant, the BC differences at hubs among groups (HC vs. left mTLE vs. right mTLE), were tested by using a post hoc Mann-Whitney test for groupwise comparisons. After correction for multiple comparisons, p-values < 0.05 were accepted as significant. The Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
4 S.-H. Jin et al. A
Figure 1. Ranked distribution of hubs. Shown are the hubs selected on the basis of the aggregated ranking percentage across subjects (>50% of ranking percent; left panels) and their topologic maps (A = axial view; B = axial, sagittal, and coronal view, respectively) projected onto a cortical surface derived from the betweenness centrality (BC) measure (A, healthy controls [HC]; B, mTLE patients with hippocampal sclerosis [mTLE]) irrespective of frequency band. Network hub color indicates the number of frequency bands identified at the hubs at the same location (magenta, four frequency bands; green, three frequency bands; blue, two frequency bands; yellow, one frequency band). Hub size corresponds to the largest aggregated ranking percentage of the hub location. The horizontal axes in each sub–bargraph indicate percentage (%). F1, dorsolateral superior frontal gyrus; HIP, hippocampus; T1P, superior temporal gyrus (temporal pole); T2P, middle temporal gyrus (temporal pole); MCIN, middle cingulate gyrus; PCIN, posterior cingulate gyrus; _L or _R, left or right. Epilepsia ILAE
relationships between the BC and clinical variables (i.e., age at seizure onset and disease duration) were tested by using Spearman correlations (p < 0.05). Because a lateralized difference in verbal delayed memory among mTLE patients was previously reported,14 the K-AVLT delayed verbal recall and recognition scores were compared between the left mTLE and right mTLE groups by using Mann-Whitney tests.
MCIN_L, and PCIN_L were 85%, 70%, and 67%, respectively. In the mTLE group, the main network hubs were located at the left temporal pole of the superior temporal gyrus (T1P_L) in the beta band, left temporal pole of the middle temporal gyrus (T2P_L) in the beta and gamma bands, left hippocampus (HIP_L) in the theta and alpha bands, and right posterior cingulate gyrus (PCIN_R) in all four frequency bands. The highest aggregated rank percentages at T1P_L, T2P_L, HIP_L, and PCIN_R were 89%, 61%, 59%, and 50%, respectively.
Functional hubs Figure 1 shows the ranked distribution of all hubs irrespective of the frequency bands in the HC and mTLE groups. In the HC group, the main network hubs were located at the left dorsolateral superior frontal gyrus (F1_L) in all four frequency bands, left middle cingulate gyrus (MCIN_L) in all four frequency bands, and left posterior cingulate gyrus (PCIN_L) in the alpha, beta, and gamma bands. The highest aggregated rank percentages at F1_L,
Betweenness centrality differences at functional hubs Table 2 and Figure 2 show the BC differences between the HC and mTLE groups, and provide post hoc test results at the functional hubs identified in each group. Of the nine identified hubs (Fig. 1B), seven were significantly different between groups. Compared to the HC, left mTLE patients had significantly greater BC values at all seven significantly different hubs (T1P_L: beta [p < 0.01]; T2P_L: beta [p < 0.05]; HIP_L: alpha [p < 0.01]; PCIN_R: theta
Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
5 Altered Functional Hubs in mTLE with HS Table 2. Mean and standard error values for betweenness centrality in healthy controls and mesial temporal lobe epilepsy patients with left and right hippocampal sclerosisa Location T1P_L T2P_L HIP_L PCIN_R
Frequency band Beta Beta Gamma Theta Alpha Theta Alpha Beta Gamma
Left mTLE #
2.51 (0.43) 3.54 (0.38)# 3.92 (0.44) 3.52 (0.55) 4.00 (0.59)# 2.23 (0.33)# 2.55 (0.34)# 2.02 (0.27)# 2.09 (0.26)#
Right mTLE ††
5.43 (0.53) 6.49 (0.73)† 5.38 (0.69) 4.94 (0.62) 6.58 (0.65)†† 8.64 (0.78)†† 7.64 (0.94)†† 10.49 (1.05)†† 8.03 (0.77)††
4.76 (0.48)§§ 4.93 (0.53) 5.15 (0.59) 3.94 (0.50) 5.72 (0.99) 8.52 (0.73)§§ 8.34 (0.80)§§ 9.22 (0.78)§§ 6.81 (0.70)§§
HC, healthy controls; HIP, hippocampus; mTLE, mesial temporal lobe epilepsy with hippocampal sclerosis; T1P, superior temporal gyrus (temporal pole); T2P, middle temporal gyrus (temporal pole); PCIN, posterior cingulate gyrus; _L or _R, left or right. a Data are presented as mean (standard error). # p < 0.05 (corrected, between HC and mTLE); †p < 0.05 (corrected, between HC and left mTLE); ††p < 0.01 (corrected, between HC and left mTLE); §§ p < 0.01 (corrected, between HC and right mTLE).
Figure 2. Group differences in betweenness centrality at patients’ hubs. Shown are BC values (mean and standard error, in arbitrary units) at hubs identified in mTLE patients, and their group-wise differences. Topologic map is identical to that in Figure 1B (axial view). Network hub color indicates the number of frequency bands identified at the hubs at the same location (magenta, four frequency bands; green, three frequency bands; blue, two frequency bands; yellow, one frequency band). Hub size corresponds to the largest aggregated ranking percentage of the hub location. HIP, hippocampus; HC, healthy controls; mTLE, mesial temporal lobe epilepsy with hippocampal sclerosis; T1P, superior temporal gyrus (temporal pole); T2P, middle temporal gyrus (temporal pole); PCIN, posterior cingulate gyrus; _L or _R, left or right. #p < 0.05 (corrected, between HC and mTLE); †p < 0.05 (corrected, between HC and left mTLE); ††p < 0.01 (corrected, between HC and left mTLE); §§p < 0.01 (corrected, between HC and right mTLE). Epilepsia ILAE Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
6 S.-H. Jin et al. [p < 0.01], alpha [p < 0.01], beta [p < 0.01], gamma [p < 0.01]), whereas right mTLE patients had significantly greater BC at five of the seven significantly different hubs (T1P_L: beta [p < 0.01]; PCIN_R: theta [p < 0.01], alpha [p < 0.01], beta [p < 0.01], gamma [p < 0.01]). There were no statistically significant differences in BC values between left and right mTLE patients. Of interest, only left mTLE patients had significantly greater BC values than those of HC at T2P_L in the beta band and HIP_L in the alpha band. There were no significant correlations between the BC values and clinical variables. Memory function differences between left mTLE and right mTLE Twenty-two left mTLE and 20 right mTLE patients underwent the neuropsychological testing. Left mTLE patients (mean SD = 4.64 2.44) had significantly lower delayed verbal recall memory performance (p = 0.02) than that in right mTLE patients (mean SD = 6.50 2.24). Moreover, left mTLE patients (mean SD = 6.27 1.98) had significantly lower delayed verbal recognition memory performance (p = 0.01) than that in right mTLE patients (mean SD = 8.10 2.45).
Discussion To characterize pathophysiologic changes in brain topology in mTLE, we investigated electrophysiologic functional hubs in mTLE patients who had histopathologically proven HS and were postoperatively seizure-free. We hypothesized that the hippocampus would consist of functional hubs in the resting-state large-scale brain network in mTLE patients with HS. Based on the definition of BC as a center of information integration,21 we assumed that the epileptogenic zone might be functional hubs. An epileptogenic zone is responsible for the generation and propagation of an epileptic seizure, suggesting the presence of an easy route for electrophysiologic access to the rest of the brain network through the hub. Hippocampal changes result in seizures and correspond to an epileptogenic zone.24 Previous studies have shown an association between the hippocampus and seizure onset, and rapid seizure propagation has been reported,25,26 and increased hippocampal connectivity involving several areas in the cortex in TLE.9 Our hippocampal hypothesis turned out to be only valid in left mTLE patients. Left mTLE patients showed a significantly increased BC at HIP_L in the alpha band compared to that in HC. It suggested that HIP_L in the alpha band might be an electrophysiologic functional network biomarker of an epileptogenic zone in left mTLE patients, a compelling indication for using network properties as clinical diagnostic markers. We interpret this result as indicative of HIP_L being a highly central region in left mTLE, and a region that provides the easiest accessibility within the large-scale functional network and, consequently, is responsible for Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
epileptic propagation within the network. This network is mediated primarily through alpha-band oscillatory brain activities. Along with the recently highlighted relevance of using graph-theoretic analyses to study TLE for pathophysiology,12 our results provide direct evidence for the role of the hippocampus as a hub in the functional network of left mTLE patients. In contrast, HIP_R was not identified as a functional hub in the mTLE patients. This contrasting HIP_R result could indicate different underlying electrophysiologic network reorganization between left and right mTLE patients because of intrinsic functional and structural asymmetries between groups.9,27–30 Based on the predominance of left hemisphere abnormalities,9,27–30 it is probable that a greater vulnerability in the left hemisphere would result in greater network reorganization, leading to the presence of HIP_L as a network hub. Because we investigated the whole-brain resting-state functional network, there is a possibility that a relatively less dominant network abnormality, which might exist in HIP_R, could not be detected. It should be noted that when seed-based connectivity was applied with a seed in HIP_R, significantly strong connection strength in the theta, beta, and gamma bands was observed in right mTLE compared to that in HC (Fig. S1), whereas there was no significant difference between left mTLE and HC. It implies that HIP_R in right mTLE is a network node that has a strong connection across the rest of the network. It also could indicate that a seed-based approach might be a preferred approach when studying abnormal connectivity of the right hippocampus in right mTLE, whereas a whole-brain network approach is adequate when investigating the altered nodal characteristics of the left hippocampus in left mTLE. Distinct approaches to patients with left and right mTLE were reported in a resting-state fMRI study.31 Differences in memory dysfunction between left and right mTLE patients could be another conceivable reason for hemispheric differences in hippocampal “hubness.” Left mTLE patients generally show greater impairment in memory than right mTLE patients, both before and after surgery.28,32 Our result of more impaired verbal delayed memory function in left mTLE patients rather than in right mTLE patients is in agreement with previous studies. However, further indepth studies will be required to unveil the fundamental causes of the hemisphere-dependent role of the affected hippocampus in functional networks. Herein, we detected network hubs in the extrahippocampal regions, indicating that mTLE is a disorder of neuronal networks that is not limited to the hallmark lesion, but a disorder that involves the large-scale functional network. An abundance of previous electroclinical work suggests that the epileptogenic network in TLE is broad.1 A diffusion tensor imaging (DTI) study reported the possible compensatory nonlocal connections of the hippocampal network.33 The influence of this compensatory network reorganization
7 Altered Functional Hubs in mTLE with HS could be attributable to the presence of extrahippocampal network hubs, as detected in this study. T1P_L and T2P_L were identified as functional hubs in our mTLE patients, which is in agreement with a previous fMRI study.34 The appearance of T1P_L and T2P_L as hubs could indicate the presence of different intrinsic restingstate functional hubs in mTLE patients compared to HC because the temporal pole could be a manifestation of self during rest.15,35,36 The PCIN was also identified as a functional hub. Previously, reduced regional homogeneity in the Default Mode Network (DMN), including the PCIN, was reported in adult unilateral mTLE patients as a result of the long-term deleterious effects of epileptic activity,34 and significantly decreased functional and structural connectivity between the PCIN and the bilateral mesial temporal lobes in mTLE patients was also reported.37 In contrast, we showed increased BC in both left and right mTLE patients in the PCIN, ranging from the theta to the gamma bands. This difference between our study and previous studies34,37 could be attributed to the difference in modality. Our results may indicate the presence of different influences of brain oscillatory activities on the large-scale functional network in a relatively high frequency range. It should be emphasized that our mTLE patients exhibited altered functional hubs in the spike-free resting state. Measurement of functional hubs with sufficiently high temporal resolution, even if performed during the resting state, provides valuable information about hub functionality.38 Thus, our results provide evidence for using network properties as clinical diagnostic markers, even in the spike-free resting state. To take full advantage of the high temporal resolution of MEG, a more in-depth investigation on the specific role of cortical hubs at each frequency band is necessary. Comparisons of functional hubs across different subtypes of epileptic brains and associated with different surgical outcomes would be an interesting future topic for further research. Because the antiepileptic drugs could influence brain networks, which are a limitation of this study, studies on the influence of antiepileptic drugs on functional networks and their hubs will be necessary. The weak detectability of deep brain structure activities is a well-known disadvantage of MEG due to a low spatial resolution for source localization compared to fMRI. However, MEG can localize the generators in hippocampus with good accuracy,39 and the contribution of hippocampal theta to cognition could be investigated by means of MEG.40 In addition, functional cortical hubs in the resting state have been studied15,17,18 with the same approach used in this study. In conclusion, we showed altered electrophysiologic functional hubs in mTLE patients reflecting pathophysiologic brain network reorganization. Because we detected network hubs in both hippocampal and extrahippocampal areas, it is probable that mTLE is a large-scale network
disorder rather than a focal disorder. The hippocampus is a hub within the large-scale functional network in left mTLE patients, but not in right mTLE patients. The evaluation of cortical hubs, even in the spike-free resting-state, could be a clinical diagnostic marker of mTLE with HS.
Acknowledgments This research was supported by a grant of the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (Grant no. HI11C1360), and the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (MSIP) (Grant no. 2010-0028631).
Disclosures None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Bernhardt BC, Hong S, Bernasconi A, et al. Imaging structural and functional brain networks in temporal lobe epilepsy. Front Hum Neurosci 2013;7:624. 2. Engel J Jr, Thompson PM, Stern JM, et al. Connectomics and epilepsy. Curr Opin Neurol 2013;26:186–194. 3. Ahmadi ME, Hagler DJ Jr, McDonald CR, et al. Side matters: diffusion tensor imaging tractography in left and right temporal lobe epilepsy. AJNR Am J Neuroradiol 2009;30:1740–1747. 4. Concha L, Kim H, Bernasconi A, et al. Spatial patterns of water diffusion along white matter tracts in temporal lobe epilepsy. Neurology 2012;79:455–462. 5. McDonald CR, Ahmadi ME, Hagler DJ, et al. Diffusion tensor imaging correlates of memory and language impairments in temporal lobe epilepsy. Neurology 2008;71:1869–1876. 6. Bernhardt BC, Chen Z, He Y, et al. Graph-theoretical analysis reveals disrupted small-world organization of cortical thickness correlation networks in temporal lobe epilepsy. Cereb Cortex 2011;21: 2147–2157. 7. Liao W, Zhang Z, Pan Z, et al. Altered functional connectivity and small-world in mesial temporal lobe epilepsy. PLoS ONE 2010;5: e8525. 8. Waites AB, Briellmann RS, Saling MM, et al. Functional connectivity networks are disrupted in left temporal lobe epilepsy. Ann Neurol 2006;59:335–343. 9. Haneef Z, Lenartowicz A, Yeh HJ, et al. Functional connectivity of hippocampal networks in temporal lobe epilepsy. Epilepsia 2014;55:137–145. 10. Bassett DS, Gazzaniga MS. Understanding complexity in the human brain. Trends Cogn Sci 2011;15:200–209. 11. Bullmore E, Sporns O. Complex brain networks: graph theoretical analysis of structural and functional systems. Nat Rev Neurosci 2009;10:186–198. 12. Chiang S, Haneef Z. Graph theory findings in the pathophysiology of temporal lobe epilepsy. Clin Neurophysiol 2014;125:1295–1305. 13. Bullmore E, Sporns O. The economy of brain network organization. Nat Rev Neurosci 2012;13:336–349. 14. Shin MS, Seol SH, Lee SK, et al. The lateralization and localization of memory and neurocognitive functioning in patients with temporal lobe epilepsy. J Epilepsy Res 2012;16:14–25. 15. Jin SH, Jeong W, Seol J, et al. Functional cortical hubs in the eyesclosed resting human brain from an electrophysiological perspective using magnetoencephalography. PLoS ONE 2013;8:e68192. 16. Taulu S, Hari R. Removal of magnetoencephalographic artifacts with temporal signal-space separation: demonstration with single-trial auditory-evoked responses. Hum Brain Mapp 2009;30:1524–1534. Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
8 S.-H. Jin et al. 17. Jin SH, Seol J, Kim JS, et al. How reliable are the functional connectivity networks of MEG in resting states? J Neurophysiol 2011;106:2888–2895. 18. Jin SH, Jeong W, Lee DS, et al. Preserved high centrality hubs but efficient network reorganization during eyes-open state compared to eyes-closed resting state: An MEG study. J Neurophysiol 2014;111:1455–1465. 19. Oostenveld R, Fries P, Maris E, et al. FieldTrip: open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data. Comput Intell Neurosci 2011;2011:156869. 20. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. NeuroImage 2002;15:273–289. 21. Buckner RL, Sepulcre J, Talukdar T, et al. Cortical hubs revealed by intrinsic functional connectivity: mapping, assessment of stability, and relation to Alzheimer’s disease. J Neurosci 2009;29:1860– 1873. 22. Wang L, Yu C, Chen H, et al. Dynamic functional reorganization of the motor execution network after stroke. Brain 2010;133:1224– 1238. 23. Xia M, Wang J, He Y. BrainNet viewer: a network visualization tool for human brain connectomics. PLoS ONE 2013;8:e68910. 24. Chatzikonstantinou A. Epilepsy and the hippocampus. Front Neurol Neurosci 2014;34:121–142. 25. Javidan M. Electroencephalography in mesial temporal lobe epilepsy: a review. Epilepsy Res Treat 2012;2012:637430. 26. Perucca P, Dubeau F, Gotman J. Intracranial electroencephalographic seizure-onset patterns: effect of underlying pathology. Brain 2014;137:183–196. 27. Bonilha L, Rorden C, Halford JJ, et al. Asymmetrical extrahippocampal grey matter loss related to hippocampal atrophy in patients with medial temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 2007;78:286–294. 28. Keller SS, Roberts N. Voxel-based morphometry of temporal lobe epilepsy: an introduction and review of the literature. Epilepsia 2008;49:741–757. 29. Kemmotsu N, Girard HM, Bernhardt BC, et al. MRI analysis in temporal lobe epilepsy: cortical thinning and white matter disruptions are related to side of seizure onset. Epilepsia 2011;52:2257–2266. 30. Lu J, Li W, He H, et al. Altered hemispheric symmetry found in leftsided mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE/HS) but not found in right-sided MTLE/HS. Magn Reson Imaging 2013;31:53–59.
Epilepsia, **(*):1–8, 2015 doi: 10.1111/epi.12966
31. Pereira FR, Alessio A, Sercheli MS, et al. Asymmetrical hippocampal connectivity in mesial temporal lobe epilepsy: evidence from resting state fMRI. BMC Neurosci 2010;11:66. 32. Helmstaedter C, Elger CE. Chronic temporal lobe epilepsy: a neurodevelopmental or progressively dementing disease? Brain 2009;132:2822–2830. 33. Bonilha L, Nesland T, Martz GU, et al. Medial temporal lobe epilepsy is associated with neuronal fibre loss and paradoxical increase in structural connectivity of limbic structures. J Neurol Neurosurg Psychiatry 2012;83:903–909. 34. Zeng H, Pizarro R, Nair VA, et al. Alterations in regional homogeneity of resting-state brain activity in mesial temporal lobe epilepsy. Epilepsia 2013;54:658–666. 35. Fransson P. Spontaneous low-frequency BOLD signal fluctuations: an fMRI investigation of the resting-state default mode of brain function hypothesis. Hum Brain Mapp 2005;26:15–29. 36. Frith U, Frith CD. Development and neurophysiology of mentalizing. Philos Trans R Soc Lond B Biol Sci 2003;358:459–473. 37. Liao W, Zhang Z, Pan Z, et al. Default mode network abnormalities in mesial temporal lobe epilepsy: a study combining fMRI and DTI. Hum Brain Mapp 2011;32:883–895. 38. Martino J, Honma SM, Findlay AM, et al. Resting functional connectivity in patients with brain tumors in eloquent areas. Ann Neurol 2011;69:521–532. 39. Attal Y, Schwartz D. Assessment of subcortical source localization using deep brain activity imaging model with minimum norm operators: a MEG study. PLoS ONE 2013;8:e59856. 40. Cornwell BR, Arkin N, Overstreet C, et al. Distinct contributions of human hippocampal theta to spatial cognition and anxiety. Hippocampus 2012;22:1848–1859.
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. List of anatomic regions of interest (Label, Montreal Neurological Institute (MNI) coordinates of left side, Brodmann area (BA)). Figure S1. Group differences in nodal degree at a seed location in the right hippocampus.