Masaryk University Faculty of Medicine Epilepsy Surgery Habilitation thesis (Collection of previously published scholarly works with commentary) Irena Doležalová M.D., Ph.D. The First Department of Neurology Masaryk University and St. Anne´s University Hospital Brno 2020 CONTENT: Acknowledgement…..…………………………………………………………………………….6 Abstract ……………………………………………………………………………………….…..7 Abbreviations…………………………………………………………………………..…………8 Introduction………………………………………………………………………………………9 The basic concept of epilepsy surgery………………………………….……………………….11 MRI and related methods………………………………………………….……………………14 EEG and its evaluation ………….………………………………………………………………20 Scalp EEG……………………………………………………………………..………….21 IEEG…………………………………………………………….………………………..24 High-frequency oscillations (HFOs)…………………………………...…………31 Clinical semiology……………………………………………………………………………….35 Clinical semiology in nonepileptic seizures……………………………………………....35 Clinical semiology in epileptic seizures………………………………..………………....38 Neuropsychology………………………………………………………………………………...41 Positron emission tomography (PET)……………………….………………………………….45 Interictal/ictal SPECT and SISCOM…………………………………………………………..48 The decision-making process within epilepsy surgery………….……………………………..49 Clinical situation: MRI-negative TLE – differences between mesial and neocortical types………………………………………………………………………………………50 1 Clinical situation: Bitemporal epilepsy…………………………………………………...51 Neurostimulation methods……………………………………………………………………...53 Vagus nerve stimulation (VNS)………………………………………………………......55 VNS efficacy………………………………………………………...……...……..55 Limitations of VNS therapy…………………………………...…………………...59 Deep brain stimulation of the anterior nucleus of the thalamus (ANT-DBS) in epilepsy….69 DBS – efficacy and adverse events……………………….………………………..70 DBS – electrophysiological aspects……………..………………………………...74 Responsive neurostimulation (RNS)…………………………………..….………...…….78 Benefits of surgery………………………….……………………………………………………79 Impact on seizure frequency…………………….…………………………………...……79 Impact on the quality of life, employment status, and other variables………………......…80 Future directions………………………………………………………………………………...82 The development of the prediction paradigm for neurostimulation………………..……...82 Seizure detectors…………………………………..……………………………………...85 EEG-based detectors……………………………………………………..………85 Non-EEG-based detector…………………………………………………………86 Benefits of seizure detectors……………..……………………….……………….88 Conclusion……………………………………………………………………………………….91 Annexes:…………………………………………………………………………………………92 2 Annex 1: Kojan M, Gajdoš M, Říha P, Doležalová I, Řehák Z, Rektor I. Arterial spin labeling is a useful MRI method for presurgical evaluation in nonlesional epilepsy. (submitted)………..……………………………………………………………………………....95 Annex 2: Dolezalova I, Brazdil M, Hermanova M, Janousova E, Kuba R. Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges and its relationship to surgical outcome in patients with hippocampal sclerosis. Epilepsy Research 2014;108(8):1406- 1416…………………………………………..…………………………………………………108 Annex 3: Frauscher B, von Ellenrieder N, Zelmann R, Dolezalova I, Minotti L, Olivier A, Hall J, Hoffmann D, Nguyen DK, Kahane P, Dubeau F, Gotman J. Atlas of the normal intracranial electroencephalogram: neurophysiological awake activity in different cortical areas. Brain. 2018;141:4………………………………...……………….……………………………………120 Annex 4: Dolezalova I, Brazdil M, Hermanova M, Horakova I, Rektor I, Kuba R. Intracranial seizure onset patterns in unilateral temporal lobe epilepsy and their relationship to other variables. Clinical Neurophysiology. 2013;124(6):1079-1088………….…………………………………136 Annex 5: Pail M, Rehulka P, Cimbalnik J, Dolezalova I, Chrastina J, Brazdil M. Frequencyindependent characteristics of high-frequency oscillations in epileptic and non-epileptic regions. Clinical Neurophysiology. 2017;128(1):106-114……………………...………………………..147 Annex 6: Dolezalova I, Brazdil M, Rektro I, Tyrlikova I, Kuba R. Syncope with atypical trunk convulsions in a patient with malignant arrhythmia. Epileptic Disorders. 2013;15(2):171- 174………...……………………………………………………………………..……………...157 Annex 7: Dolezalova I, Brazdil M, Kahane P. Temporal lobe epilepsy? Things are not always what they seem. Epileptic Disorders. 2017;19(1):59-66……………….…………..……………162 Annex 8: Dolezalova I, Schachter S, Chrastina J, Hemza J, Hermanova M, Rektor I, Pazourkova M, Brazdil M. Atypical handedness in mesial temporal lobe epilepsy. Epilepsy & Behavior. 2017;72:78-81…………...………………………………………………...…………………….171 3 Annex 9: Kojan M, Dolezalova I, Koritakova E, Marecek R, Rehak Z, Hermanova M, Brazdil M, Rektor I. Predictive value of preoperative statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy with hippocampal sclerosis. Epilepsy & Behavior. 2018;79:46-52……………………………...……………………………………..……………..176 Annex 10: Dolezalova I, Brazdil M, Chrastina J, Hemza J, Hermanova M, Janousova E, Pazourkova M, Kuba R. Differences between mesial and neocortical magnetic-resonance-imagingnegative temporal lobe epilepsy. Epilepsy & Behavior. 2016;62:21-26……………..…………..184 Annex 11: Rehulka P, Dolezalova I, Janousova E, Tomasek M, Marusic P, Brazdil M, Kuba R. Ictal and postictal semiology in patients with bilateral temporal lobe epilepsy. Epilepsy & Behavior. 2014;41:40-46…………………………………………………..……..……………..191 Annex 12: Chrastina J, Novak Z, Zeman T, Kocvarova J, Pail M, Dolezalova I, Jarkovsky J, Brazdil M. Single-center long-term results of vagus nerve stimulation for epilepsy: A 10-17 year follow-up study. Seizure – European Journal of Epilepsy. 2018;59:41-47………………………199 Annex 13: Chrastina J, Kocvarova J, Novak Z, Dolezalova I, Svoboda M, Brazdil M. Older age and longer epilepsy duration do not predict worse seizure reduction outcome after vagus nerve stimulation. Journal of neurological surgery part A – Central European neurosurgery. 2018;79(2):152-158……………………..…………………………………………………..…..207 Annex 14: Chrastina J, Doležalová I, Novák Z, Pešlová E, Brázdil M. Pregnancy outcomes in refractory epilepsy with vagus nerve stimulation: long-term single center experience. (submitted)………………………………………………………………………………………215 Annex 15: Brázdil M, Doležalová I, Koriťáková E, Chládek J, Roman R, Pail M, Jurák P, Shaw DJ, Chrastina J. EEG reactivity predicts individual efficacy of vagal nerve stimulation in intractable epileptics. Frontiers in Neurology. 2019;10:392-392……………………………………………229 Annex 16: Plešinger F, Halámek J, Chládek J, Jurák P, Doležalová I, Chrastina J, Brázdil M. Response to vagal stimulation by heart-rate feature in drug-resistant epileptic patients. (submitted)…………………………………………………………………..…………..………241 4 Annex 17: Doležalová I, Kunst J, Kojan M, Chrastina J, Baláž M, Brázdil M. Anterior thalamic deep brain stimulation in epilepsy and persistent psychiatric side effects following discontinuation. Epilepsy Behavior Reports 2019;12:100344……………………………………………….…....246 Annex 18: Rektor I, Dolezalova I, Chrastina J, Jurak P, Halamek J, Balaz M, Brazdil M. Highfrequency oscillations in the human anterior nucleus of the thalamus. Brain Stimulation. 2016;9(4):629-631……………………..…………………………….………………………….250 Annex 19: Doležalova I., Pešlová E, Michnová M, Nečasová T, Kočvarová J, Musilová K, Rektor I. Brázdil M. Epileptochirurgická léčba zlepšuje kvalitu života – výsledky dotazníkové studie. Ceská a Slovenská Neurologie a Neurochirurgie. 2016;79/112(4):430- 439……………………………………………………………………………………………....254 References:..……………………………………………………………………………………265 5 Acknowledgments: I would like to thank prof. Robert Kuba MD, Ph.D. and prof. Milan Brázdil MD, Ph.D., who taught me both clinical epileptology and research basics. I also want to thank all my clinical collaborators from the First Department of Neurology and from the Department of Neurosurgery at St. Anne’s University Hospital. I would like to particularly mention prof. Ivan Rektor MD, Ph.D., Jitka Kočvarová MD, Martin Pail MD, Ph.D., Ondřej Strýček MD, and Klára Štillová MD, Ph.D. from the First Department of Neurology, and assoc. prof. Jan Chrastina MD, PhD., assoc. prof. Eva Brichtová MD, Ph.D., and Jan Hemza MD. Ph.D. et Ph.D. from the Department of Neurosurgery. I would also like to thank my research collaborators Pavel Jurák MSc., Ph.D., Petr Klimeš MSc., Ph.D., Jan Chládek MSc., Jan Cimbalník MSc., Ph.D., and Filip Plešinger MSc., Ph.D. from the Institute of Scientific Instruments, and Pavel Říha MSc. and Martin Kojan MSc. from Masaryk University. Finally, my thanks to my whole family for their lifelong support. 6 Drug-resistant epilepsy represents a significant health problem. The ideal therapy, with a high probability of complete seizure cessation, is epilepsy surgery with a resection of areas responsible for seizure genesis. Despite numerous studies, resective brain surgery can currently be offered only to some patients for whom it is possible to formulate a rational hypothesis about the location of the epileptic generator. Other patients can be implanted with a neurostimulator. The habilitation thesis is divided into three main parts. The first part is focused on individual diagnostic steps that are described in more detail. We pay attention to descriptions of magnetic resonance imaging (MRI) and its post-processing, EEG, and invasive EEG (IEEG), clinical semiology, neuropsychology, and positron emission tomography (PET). The last section of this part concerns two relatively common clinical situation: bitemporal epilepsy and MRI-negative temporal lobe epilepsy (TLE). The second part discusses neurostimulation methods: vagus nerve stimulation (VNS) and deep brain stimulation (DBS). A critical portion of this part is devoted to the development of a predictive algorithm for VNS based on pre-implantation data. The third part is dedicated to our future research: the evaluation of seizure detectors in real life and the validation of our statistic classifier for predicting neurostimulator efficacy. Abstract: 7 Abbreviations ADHD – attention deficit hyperactivity syndrome, AEDs – antiepileptic drugs, ANT – nucleus anterior thalami, ASL – arterial spin labeling, AT – attenuation of background activity, CT – computed tomography, DBS – deep brain stimulation, BOLD – blood oxygenation level dependent, DKI – diffusional kurtosis imaging, DTI – diffusion tensor imaging, ECG – electrocardiography, EEG– electroencephalography, EMG – electromyography, ETLE – extratemporal lobe epilepsy, FCD – focal cortical dysplasia, FIA – fast ictal activity, fMRI – function magnetic resonance, HDEEG – high-density EEG, HFOs – high-frequency oscillations, HRV – heart-rate variability, HS – hippocampal sclerosis, IEDs – interictal epileptiform discharges, IEEG – invasive electroencephalography, ILAE – International League Against Epilepsy, MEG – magnetoencephalography, MRI – magnetic resonance imaging, MRR – median of RR interval, NMRR – normalized median of RR interval, PET – positron emission tomography, PNES – psychogenic non-epileptic seizures, PSEs – psychiatric side effects, QoL – quality of life, ReHo – regional homogeneity, RR – RR interval, SANTE – stimulation of the anterior nuclei of thalamus, SIA – slow ictal activity, SPM – statistic parametric mapping, SPM-PET – statistic parametric mapping of positron emission tomography, RNS – responsive neurostimulation, SUDEP – sudden unexpected death in epilepsy, TLE – temporal lobe epilepsy, vHFOs – very high-frequency oscillations, VNS – vagus nerve stimulation, WHO – World Health Organization, 18F-FDG-PET – fluorine-18fluorodeoxyclucose positron emission tomography 8 Introduction Epilepsy is a multicausal and often devastating chronic disorder characterized by recurrent spontaneous seizures, affecting around 50 million people worldwide. According to the World Health Organization (WHO), epilepsy is the most common serious neurological condition in Europe and a source of significant long-term disability. It is thought to be directly responsible for about 33,000 deaths in Europe every year, one-third of which could have been avoided if proper standards of care were available. The burden of epilepsy stems from a multitude of issues, including pharmacoresistance (30% of patients) and various comorbidities. This global burden translates into huge socioeconomic costs, estimated at 0.2% of the GDP of developed countries (3 billion euros in the European Union) (Charlson et al., 2016; Neurological Disorders Collaborator Group, 2017; Thakur et al., 2016). Drug-resistant epilepsy is a major health problem. Currently, the most effective treatment for focal drug-resistant epilepsy is resective surgery. Every year worldwide, thousands of patients with focal drug-resistant epilepsy undergo resective brain surgery to stop their seizures (Engel, 2008). The last 15 years have seen a remarkable evolution in the treatment indications for epilepsy surgery. Notably, the increasing worldwide use of invasive EEG (IEEG) has likely been driven by high numbers of patients referred with extratemporal lobe epilepsies (ETLE) and magnetic resonance imaging (MRI)-negative epilepsies; these categories pose particular challenges for presurgical evaluation (Jehi et al., 2015). Over the last 50 years, the success rate for surgical interventions in patients with drug-resistant epilepsy has significantly increased. At the moment, it is possible to achieve long-term seizure control in 80% of patients with mesial or neocortical temporal lobe epilepsy (TLE) with 9 hippocampal sclerosis (HS) or focal cortical dysplasia (FCD) type 2 and for up to two-thirds of patients with ETLE. This achievement is obtained with minimal perioperative risks, which have decreased to approximately 1% (Ryvlin and Rheims, 2016). Despite this progress in resective surgery and its obvious benefits, surgery has failed to control seizures for a large number of epileptic patients (Krucoff et al., 2017; Téllez-Zenteno et al., 2005). The reasons for these failures are varied, but are thought to be in large part linked to the suboptimal identification of the epileptogenic zone responsible for seizure origin and, consequently, the nonresection or inadequate modulation of the critical nodes and pathways that fundamentally characterize the epileptogenic network (Harroud et al., 2012; McIntosh et al., 2004; Ryvlin and Kahane, 2005; Salanova et al., 2005). Based on these findings, there is still a significant research gap in identifying ideal surgical candidates and optimizing the methods used in the presurgical evaluation. Despite numerous efforts, many patients with drug-resistant epilepsy are still not indicated for “classical” resective brain surgery. This is attributed to many variables, e.g., epilepsy multifocality, relevant patient co-morbidities, severe mental retardation, personal options, or previous resective surgery failure. These patients can be offered implantation of a neurostimulator: vagus nerve stimulation (VNS), deep brain stimulation (DBS), and responsive neurostimulation (RNS) (Englot et al., 2017). In general, these methods do not offer a high chance for a complete seizure cessation, but they can lead to a significant decrease in seizure frequency, severity, and duration with minimal additional risks (De Herdt et al., 2007; Kuba et al., 2009; Labar, 2004; Renfroe and Wheless, 2002; Vonck et al., 2004). 10 The basic concept of epilepsy surgery Epilepsy surgery is related to identifying the epileptogenic zone that is responsible for seizure generation. Complete removal of epileptogenic zone is essential for sustained seizure freedom. Unfortunately, it is not possible to measure the epileptogenic zone directly by any modality. Its borders can be demarcated only indirectly based on the localization of the following areas: (1) epileptogenic lesion, (2) irritative zone, (3) seizure onset zone, (4) symptomatogenic zone, and (5) functional deficit zone (Rosenow and Lüders, 2001). The organization of the epileptogenic network can be simple, well-organized, and easily understandable (Figure 1a). However, some patients have very complex and complicated epileptogenic zone organization (Figure 1b). Epileptogenic lesion: It is a visible lesion, usually identified by MRI. The epileptogenic lesion can be easily identified and has clear-cut borders. Lesional cases are more likely to become surgical candidates and have more satisfying surgical results. Some patients have no clear MRI lesion; these are called nonlesional. Figure 1: Epileptogenic zone organization 11 The relation between MRI lesions and the epileptogenic zone is complex. Some visible MRI lesions are not epileptogenic, i.e., they remain “silent” in terms of epilepsy. For this reason, the results of all methods must be evaluated in context. There are limited correlations between the extent of MRI lesions and the borders of the epileptogenic zone. In some cases, partial removal of an epileptogenic lesion is sufficient for complete seizure cessation. In contrast, complete resection can fail to terminate seizures in the others. Two possible theories may explain this observation. According to the first theory, the surrounding tissue is probably responsible for epilepsy. If this surrounding tissue is not removed together with the epileptogenic lesion, epilepsy will persist even after the operation. In the second theory, the failure is conditioned by the limits of neuroimaging, i.e., only “the tip of the iceberg” is visible on MRI and a substantial portion of the epileptogenic lesion remains hidden. Irritative zone: This zone is responsible for the generation of interictal epileptiform discharges (IEDs). Seizure onset zone: The seizure onset zone is defined as an area delivering epileptic seizures. This zone can be identified by the results of ictal scalp EEG or IEEG, or by ictal single-photon emission tomography (SPECT) and ictal SPECT co-registered to MRI (SISCOM). In terms of epilepsy, there are two more zones, the symptomatogenic zone and the functional deficit zone, for which relations to the epileptogenic zone are even less clear. Symptomatogenic zone: The symptomatogenic zone is activated by the ictal activity during seizure propagation. The anatomical localization of the epileptogenic zone and the symptomatogenic zone can be remote, which is conditioned by the fact that large parts of the brain 12 represent functionally silent areas, and ictal signs appear following the activation of the eloquent cortex. Functional deficit zone: The functional deficit zone is the area of the cortex that is functionally abnormal during the interictal period. This dysfunction may be a direct result of the destructive effect of an epileptogenic lesion or it may be functionally mediated, i.e., caused by abnormal neuronal transmission. The functional deficit zone can be identified by neurological examination, neuropsychological testing, PET, or interictal SPECT. Epilepsy surgery is a stepwise process in which patients undergo whole batteries of examinations to precisely identify the epileptogenic zone. These batteries can be divided into three levels. The basic level, which is compulsory for every patient, usually includes MRI, scalp video EEG, PET, and neuropsychology. Some advanced evaluation should be done in patients in whom a hypothesis about the epileptogenic zone cannot be formulated or is unclear based on the results of the previous examination. This advanced level is represented by SISCOM, high-density EEG (HD-EEG) with electrical source imaging (ESI), or some other imaging techniques. IEEG represents the third and highest level, which is reserved only for pre-selected patients with a plausible hypothesis about epileptogenic zone localization that requires further confirmation before surgery. 13 MRI and related methods MRI is an almost obligatory examination for all patients with epilepsy. Immediately after the first seizure, i.e., in the acute phase, MRI is not required; the performance of standard non-contrasted computed tomography (CT) scans is sufficient to exclude acute brain disorder. However, MRI examinations should be scheduled even in these patients (Bernasconi et al., 2019). The main goal of MRI is the identification of an epileptogenic lesion; this is essential for further management of patients. For drug-resistant epilepsy, the chances for lesional cases to become seizure-free after surgery are apparently higher than for nonlesional cases (Téllez-Zenteno et al., 2010). The International League Against Epilepsy (ILAE) recommended a standard MRI protocol for epilepsy patients; this protocol includes the use of 3T MRI and involves the following sequences: isotropic, millimetric 3D T1 and FLAIR images, and high-resolution 2D submillimetric T2 images. The MRI should be evaluated by an experienced neuroradiologist and an epileptologist who has access to the patient’s further clinical data (Bernasconi et al., 2019). Despite all efforts, a significant number of patients remain nonlesional. In the past, nonlesional MRI was described in up to 40% of reviewed MRI scans (Bien et al., 2009). We expect this number to be lower now due to technical progress, specifically due to improved MRI quality and the use of a standard protocol for data acquisition. Our group is focusing on neuroimaging in nonlesional cases. This focus is connected with a grant related to the post-processing of MRI data, functional MRI (fMRI), and tractography, which are evaluated in the context of other clinical data and other results. For this grant, we defined nonlesional cases as (1) patients without any lesions detected by conventional MRI and (2) patients with some discrete changes that made it impossible to process them directly to the surgery. 14 We work with the structural and functional MRI techniques detailed below. Additional information is provided by PET, SISCOM, and HD-EEG with ESI; these methods are detailed in individual sections. Voxel-based morphometry (VBM) VBM is an automated MRI technique used to study differences in brain morphometry based on changes in the volume or concentration of white and gray matter associated with the presence of FCD or HS. In our research, we focused on the evaluation of T1 voxel-based methods: gray matter volume, gray matter concentration, and junction maps. Junction maps point out the blurring between white and gray matter that is present in FCD (Martin et al., 2017). fMRI The function of fMRI is based on the fluctuation in the concentration of oxyhemoglobin and deoxyhemoglobin in the blood (blood oxygenation level dependent [BOLD] signal). IEDs cause alterations in the blood oxygenation that can be transcribed by fMRI. This knowledge led to the development of EEG-fMRI, but its use is complicated by several factors, including the necessity to record EEG during MRI acquisition and the absence of IEDs during a limited time interval. The discovery of spontaneous neuronal activities helped to overcome these limitations. Spontaneous neuronal activities are coherent low-frequency fluctuations in BOLD signals. IEDs modify the amplitudes of these fluctuations; this modification led to the development of amplitude of lowfrequency fluctuations (ALF) as an fMRI technique (Zhang et al., 2010). 15 Another novel method working with fMRI is ReHo (regional homogeneity) which explores the regional coherence of the fMRI time course and is capable of establishing the synchronization of cerebral activity between different brain regions (Zang et al., 2004). Arterial spin labeling (ASL) ASL is a method for evaluating cerebral blood flow noninvasively, without the use of any tracer, only by magnetic labeling of blood. In epileptology, it seems that ASL will have comparable value to 18F-FDG-PET or interictal SPECT (Haller et al., 2016). Diffusion MRI Epilepsy, like other brain disorders, is a network disease with alterations of connections between brain areas. These alternations of brain connectivity are evaluated by diffusion tensor imaging (DTI) and diffusional kurtosis imaging (DKI), which post-proceed diffusion MRI. DTI works on the macroscopic millimeter scale, which is insufficient for imaging tracks in the areas with high representation (corona radiata, tractus opticus). In these areas, DKI can be useful because of its higher resolution (Lee et al., 2014). The current aim in our research is to identify individual structural and functional characteristics of individual pathologies and to integrate them in a model individualized to each patient. We anticipate some structural and functional overlap, which will be helpful in the delineation of the epileptogenic zone. This approach is illustrated by the case report of our patient (Figure 2). 16 Figure 2: The results of advanced MRI techniques, local synchrony, and positron emission tomography (PET) in a patient The patient was a 42-year-old female with drug-resistant nonlesional epilepsy. She underwent a resection in the right parieto-occipital region (the boundaries are marked in 17 green). Advanced MRI techniques show several regions of interest. However, some overlap can be found in the operated area. The correlation between arterial spin labeling (ASL) and statistic parametric mapping of positron emission tomography (SPM-PET) is remarkable in this patient. Diffusion tensor imaging (DTI) and diffusional kurtosis imaging (DKI) were not done in this case. We recently submitted a paper for publication comparing ASL and 18F-FDG-PET (Kojan et al., submitted; Annex 1). PET with fluorine-18 fluorodeoxyglucose (18F-FDG-PET) used as a tracer is a well-established method within epilepsy surgery that is helpful in the identification of the epileptogenic zone. However, its use has several disadvantages, including the higher financial costs, lack of availability in some areas, and the use of radioactive tracers. ASL is an MRI technique that is based on magnetic labeling of blood; ASL can be acquired quickly during routine MRI with minimal data acquisition time and no tracer needs to be applied. It seems that 18F-FDG-PET and ASL provide similar types of information for the epileptogenic zone based on the localization of hypometabolism or hypoperfusion (Boscolo Galazzo et al., 2016; Kim et al., 2016; Oner et al., 2015; Wang et al., 2018). Several studies have compared 18F-FDG-PET and ASL in epilepsy surgery patients, but to our knowledge this study is the first to confirm its results in nonlesional patients by a tailored surgical procedure. The main aim of our work was to compare the utility of 18F-FDG-PET and ASL in nonlesional patients with drug-resistant epilepsy. We selected 8 patients (7 patients with no lesions, 1 patient with discrete changes on MRI). All patients underwent 18F-FDG-PET and ASL; these methods were post-processed using a voxel-based approach, i.e., we determined the level of metabolism for 18 18F-FDG-PET or perfusion for ASL in each voxel using statistic parametric mapping (SPM). Voxels with significant hypometabolism or hypoperfusion were chosen. This process revealed areas (clusters) with significant hypometabolism or hypoperfusion. The distribution of these clusters was compared to the resection mask. If a cluster was located within the resection mask, it was categorized as a true positive. If a cluster was located outside the resection mask, it was categorized as a false positive. We next calculated the positive predictive value (PPV) of either 18F-FDG-PET or ASL in the individual patient: 𝑃𝑃𝑉 = 𝑇𝑃 𝑇𝑃 + 𝐹𝑃 The PPV of 18F-FDG-PET and ASL in all patients were statistically compared. We revealed the hypometabolic or hypoperfused clusters in each patient. The PPV was higher for ASL in 5 patients; the PPV was the same for both methods in 1 patient. In the remaining 2 patients, the PPV value was higher for 18F-FDG-PET. We did not find any statistically significant differences when comparing the PPV of 18F-FDG-PET and ASL (p=0.89); i.e., 18F-FDG-PET is not superior to ASL and vice versa. 18F-FDG-PET and ASL provide similar information for localization of the epileptogenic zone. Considering the benefits of ASL (lower costs, lower acquisition time, and no need for radioactive tracers), ASL will probably be able to replace 18F-FDG-PET in epilepsy surgery in the future. However, our results need evaluation in large patient cohorts. PPV – Positive predictive value TP – True positive FP – False positive 19 EEG and its evaluation Hans Berger recorded the first scalp EEG in 1924 (Kugler, 1991; Reif et al., 2016). In a short time, EEG was being used for intraoperative monitoring of cerebral activity from the brain surface (Foerster and Altenburger, 1935). In 1949, Robert Hayne and Russell Meyers reported the first attempts at signal recording from deep structures using stereotactically implanted electrodes (Hayne and Meyers, 1950). EEG provides unique information about cerebral activity; it enables physicians and researchers to map and assess brain functions, connections, and processes in real time on a millisecond scale. From this point of view, EEG cannot be replaced by MRI or by other neuroimaging techniques that can give a detailed high-quality structural perspective. Figure 3: High-density EEG 20 The essential limitation of standard scalp EEG is its low spatial resolution. HD-EEG with higher numbers of electrodes was initially used to overcome this limitation (Figure 3). The data obtained by high-density EEG requires the determination of their likely generator within the brain, which could be mathematically determined by ESI (Nemtsas et al., 2017). We employed HD-EEG in our project, focusing on neuroimaging of nonlesional epilepsy. Scalp EEG Scalp EEG is used in all patients with epilepsy as well as in patients with nonepileptic events. Many authors have worked on the evaluation of scalp EEG and its contribution to decision-making processes or prognostication within epilepsy surgery programs. We used scalp EEG to analyze the prognostic value of IEDs for patients with TLE associated with HS (HS-TLE) (Doležalová et al., 2014; Annex 2). Among patients with TLE, the distribution of IEDs, specifically the lateralization of IEDs between the hemispheres, has prognostic value for surgical outcome. Patients with strictly unilateral IEDs have a better prognosis than patients with blateral IEDs (Chung et al., 1991; Radhakrishnan et al., 1998; Villanueva et al., 2004). This division does not apply for the subgroup of patients with HS-TLE. Three studies analyzing the distribution of IEDs in patients with HS-TLE proved no differences in surgical outcomes between patients with strictly unilateral IEDs and patients with bilateral IEDs (Hardy et al., 2003; Janszky et al., 2005; Krendl et al., 2008). However, some studies found that unilateral IEDs correlate with better surgical outcome (Aull-Watschinger et al., 2008; Schulz et al., 2000). 21 It was reported that AED reduction does not have any impact on IED frequency or localization, but the influence of AED reduction on IED distribution in terms of lateralization between the two hemispheres was not systematically proven (Gotman and Koffler, 1989; Gotman and Marciani, 1985; Marciani and Gotman, 1986; So and Gotman, 1990; Spencer et al., 1981). Our study focused on HS-TLE patients. We evaluated: (1) whether there is a change in IED lateralization conditioned by AED reduction, (2) whether there is a change of EEG prognostic value for surgical outcome between periods with full medication and reduced medication. We retrospectively analyzed the distribution of IEDs between the two hemispheres in patients with HS-TLE. The EEG was analyzed in two periods: full medication (no AED reduction) and reduced medication period (AED dose reduced). We collected 20 minutes of EEG recording in waking and sleeping states in each period (four segments of EEG were collected for each patient). The patients were designated as having unilateral IEDs (more than 90% IEDs on the side with HS) or bilateral IEDs (less than 90% IEDs on the side with HS). The distribution of IEDs was correlated with surgical outcome, defined as excellent (Engel I) or poor (Engel II-IV). We identified 43 HS-TLE patients. In the full medication period, 38 out of 43 (88%) patients had unilaterall IEDs in the waking state, and 37 (86%) patients had unilaterall IEDs in the sleeping state. In the reduced medication period, the number of patients with unilateral IEDs decreased to 32 (74%) patients in the waking state, and 25 (58%) patients in the sleeping state. We proved a statistically significant change in IED distribution between the full medication and reduced medication periods in the sleeping state (p=0.003), but not in the waking state (p=0.114). The lateralization of IEDs had a predictive value for surgical outcome only in the full medication period during the sleeping state (p=0.020). There was a borderline statistical significance of IED 22 distribution to surgical outcome in the full medication period during the waking state (p=0.073). We found no relation between IED distribution and surgical outcome in the reduced medication period in the waking state (p=0.698) or in the sleeping state (p=0.717). Based on our results, we can summarize that EEG can provide valuable information for surgical outcomes in HS-TLE patients. This information is present exclusively in the period with full medication, i.e., before the initiation of AED reduction. It seems that sleeping EEG is crucial for surgical efficacy prediction. When AEDs are reduced, the EEG predictive value is lost, and there is no clear-cut relation to the surgical results. 23 IEEG IEEG is used only in carefully selected patients with drug-resistant epilepsy in the presurgical evaluation. IEEG is used as an “open window” to the brain, enabling the detailed analysis of physiological and pathological phenomena occurring in the brain tissue. Currently, it represents the most accurate method for the characterization of the epileptogenic zone. Many papers have described the pathological, mainly epileptiform, activity of the brain. This is in contrast with the sparse information about physiological brain rhythms; the majority of such studies were formulated in a pre-MRI period (Gastaut, 1949; Graf et al., 1984; Jasper and Penfield, 1949; Petersen et al., 1953); this could be problematic in interpreting intracerebral recordings as the physiological activity in a given region could be evaluated as pathological and vice versa. Such misinterpretations could have a negative impact on the tailoring of surgery and potentially on surgical results. We tried to overcome this limitation with a common project of intracranial EEG data (Frauscher et al., 2018; Annex 3). The main aim of this work was to characterize, in detail, physiological awake brain activity in invasive EEG recording. IEEG data from both intracerebral and subdural electrodes were retrospectively collected in three participating institutes. The “normal” contacts for each patient were selected. These were defined as contacts outside pathological lesion, seizure onset zone, or irritative zone without IEDs or pathological slow waves. All contacts were localized into stereotactic space, and anatomical structures were segmented. Subsequently, the power spectral densities for each anatomical structure were estimated using Welsch’s method. A total of 106 patients, with 1785 contacts, were included in the study. The positions of contacts are shown in Figure 4. 24 Figure 4: The localization of brain contacts over brain surface Blue – intracerebral electrodes, yellow – subdural electrodes. Republished with permission. The individual brain lobes were characterized by different peaks of power spectral density (Figure 5.1). The occipital lobe showed a peak in the alpha band (7.75-10.25 Hz), the frontal lobe in the beta and gamma band (16.75-80 Hz), the parietal lobe in the alpha and beta band (8.75-9.75 Hz, and 16.75-31.75 Hz), and the temporal lobe in the alpha and delta band (7.25-9.25 Hz, and 0.75- 2.25 Hz). Some brain regions were characterized by a specific signature (peak present in ≥ 60% of contacts, Figure 5.2). These regions were the precentral gyrus (lateral aspect 20-24 Hz, medial 25 aspect 24-31 Hz), the opercular part of the inferior frontal gyrus (20-24 Hz), the cuneus (8 Hz), and the hippocampus (1 Hz). Figure 5: Power spectra in individual brain lobes (5.1) and in regions with a specific signature (5.2) – semi-logarithmic graph 26 Red line – the median spectral density of all contacts; pink shaded region – the 25th and 75th percentiles; dotted red lines – the upper and lower bound; vertical black lines – the frequency band; colored horizontal segments in the upper part of each graph – the presence of peaks (the color of the line indicates the percentage of contacts with the peak). Republished with permission This work is the first comprehensive atlas of physiological intracerebral EEG. It could be used in surgery planning and for physiological studies of normal brain activity. We plan to continue to collaborate with the other authors of this study in making a physiological intracranial atlas of sleep stages. The data for this work will be collected prospectively, and our goal is to describe the activity of individual brain regions during sleep. The changes associated with seizure onsets in intracranial EEG have been systematically studied by many authors because of their significance for our understanding of epileptogenesis (Lee et al., 2000; Spanedda et al., 1997; Spencer et al., 1992). Some patterns were shown to be associated with underlying histopathological changes (Bragin et al., 2009; Williamson et al., 1995); others were associated with the anatomical organization in a given region (Lee et al., 2000; Spencer et al., 1992). It has been proposed that some patterns at ictal onset have different values for the localization of seizure onset. It seems that fast ictal activity at seizure onset is associated with excellent surgical outcomes (Faught et al., 1992; Weinand et al., 1992). The slower frequencies, mainly from the theta range, at seizure onset indicate the propagation of ictal activity and are associated with unsatisfactory results (Schiller et al., 1998). In one study, we focused on the predictive value of seizure onset patterns in IEEG for surgery in patients with TLE (Doležalová et al., 2013; Annex 4). 27 The main aim of this project was to identify whether there is a correlation between the type of seizure pattern present in IEEG in patients with unilateral TLE and surgical outcome, localization of seizure onset zone, or histopathological findings. We retrospectively identified the patients with TLE who underwent IEEG and subsequent surgery for drug-resistant epilepsy. We analyzed the frequency at seizure onset using the Fast Fourier Transform algorithm. Based on this frequency, patients were subdivided into three groups: (1) fast ictal activity (FIA) – defined by frequency ≥ 8 Hz; (2) slow ictal activity (SIA) – defined by frequency < 8 Hz; and (3) attenuation of background activity (AT) – defined as a diffuse change of EEG signal (decrease in amplitude, increase in frequency) with delayed development of clear-cut rhythmic activity (Figure 6). In the next step, FIA was subdivided into two groups: FIA > 15 Hz and FIA 8-15 Hz. SIA was subdivided similarly into two groups: SIA ≤ 2 Hz and SIA 2 - 8 Hz. The patients were categorized based on the surgical outcome (Engel I vs. Engel II-IV), on the localization of the seizure onset zone (lateral, mesial, polar), and on the histopathological findings (HS, FCD, negative, or other types). We managed to identify a total of 51 patients: 34 (67%) patients exhibited FIA, 13 (25%) patients exhibited SIA, and 4 (8%) patients exhibited AT. We proved the prognostic value of seizure onset patterns for surgical outcomes. Of the 34 patients with FIA, 29 (85%) had an Engel I outcome, in comparison with 4 (31%) out of 13 with SIA, and 0 (0%) of the 4 with AT (p<0.001). When FIA was subdivided into FIA>15 Hz and FIA 8-15 Hz, we did not find any statistical significance for the surgical outcome (p=0.164). When SIA was subdivided into SIA≤2 Hz and SIA 2-8 Hz, we found borderline statistical significance for the surgical outcome (p=0.070); patients with SIA 2-8 Hz tended to have a worse prognosis than patients with SIA≤2 Hz. 28 We did not find any relationship between seizure onset pattern and localization of seizure onset zone (p=0.878) or between seizure onset patterns and histopathological findings (p=0.362). Based on our work, we can hypothesize that the seizure onset pattern reflects the relation of the electrode to the localization of the seizure onset zone. If the electrode is directly in the seizure onset zone or in its close neighborhood, it is possible to record fast ictal activity as the first chance of the electrical signal. If the electrode is remote to the seizure onset zone and we record a propagated pattern, the first EEG change is from the slower frequency band. 29 Figure 6: The individual seizure onset patterns in invasive EEG (IEEG) A – electrode position, B – intracranial EEG 30 High-frequency oscillations (HFOs) EEG was traditionally evaluated within Berger’s scale, with cutoff values for the highest frequencies of about 50 Hz. A critical change in the conventional clinical and research thinking took place in the early 1990s, when HFOs were described. HFOs were first proven to be present in the hippocampal CA1 region in the rat model (Buzsáki et al., 1992), and subsequently in the area of seizure onset (Allen et al., 1992; Fisher et al., 1992). This started an entirely new era in electrophysiology and clinical epileptology, in which attention was transferred to higher frequencies. HFOs are the category of brain activity between 80 and 500 Hz, subdivided into ripples (80-250 Hz) and fast ripples (250-500 Hz) (Bragin et al., 1999). In the beginning, HFOs were regarded as the “holy grail” of epilepsy surgery. Several authors showed that resection of areas with high HFO rates is associated with better surgical outcome than resection of areas with low HFO rates (Fujiwara et al., 2012; Haegelen et al., 2013; Jacobs et al., 2009; Usui et al., 2011; van 't Klooster et al., 2011). Importantly, it was shown that the HFO rates present in presurgical electrocorticography do not correlate with surgical outcome, but the HFO rates in postsurgical electrocorticography do correlate, i.e., if there is a high number of remaining HFOs after resection, the patients tend to have worse outcomes (van 't Klooster et al., 2015; van 't Klooster et al., 2017). However, it is not possible to reliably state whether a patient becomes seizure-free in case of HFO high-resection rate (Frauscher et al., 2017). The occurrence of physiological HFOs in several brain regions even complicates the situation (Axmacher et al., 2008; Melani et al., 2013; Nagasawa et al., 2012). Physiological HFOs play an essential role in some processes, such as memory consolidation (Buzsáki et al., 1992; Curio et al., 1997). The frequency of HFOs is not sufficient for clear-cut delineation between physiological and pathological ones (Bragin et al., 2007). 31 In our study, we tried to determine the existence of other frequency-independent parameters of HFOs, namely rates, duration, and amplitude. This could help differentiate between seizure onset zone, irritative zone, and other areas (Pail et al., 2017; Annex 5). The study sample size included 31 patients who underwent IEEG for drug-resistant epilepsy. The HFOs were automatically detected in all these patients and divided into three groups based on their localization: (1) HFOs within the seizure onset zone, (2) HFOs within the irritative zone, and (3) HFOs outside both the seizure onset zone and the irritative zone (the outer zone). The duration and amplitude of HFOs were estimated. Subsequently, the three zones were compared with respect to HFO rates, duration, and amplitude. The rates of HFOs within the seizure onset zone were significantly higher in the seizure onset zone than in areas outside (p=0.038 for ripples, p=0.018 for fast ripples). When analyzing the duration of HFOs, the ripples had shorter duration in the seizure onset zone than in the irritative zone and in the outer zone (p<0.001). The duration of fast ripples in the seizure onset zone was also shorter in the irritative zone (p<0.001) and in the outer zone (p<0.001). The amplitude of ripples was higher in the seizure onset zone than in the irritative zone and the outer zone (p<0.001). When focusing on fast ripple amplitude, we did not find any differences in amplitude between the three zones. To conclude, frequency-independent characteristics of HFOs can be helpful in differentiating between physiological and pathological HFOs. The higher rates of both ripples and fast ripples are present in the seizure onset zone. Moreover, both ripples and fast ripples have a shorter duration in the seizure onset zone. The ripples within the seizure onset zone also have a higher amplitude. 32 Most studies have considered the topic of “classical” HFOs, determined by 500 Hz cut-off value. Descriptions of activities and processes above the 500 Hz limit are rare in the literature and remain beyond the main research stream. Usui et al. (2010) reported the existence of very high-frequency oscillations (vHFOs) defined as the activity above 1000 Hz (Usui et al., 2010). In a later study, they managed to find a correlation between the presence of vHFOs and surgical outcomes in a group of 12 patients with drug-resistant epilepsy (Usui et al., 2015). Our group has also succeeded in describing vHFOs (500-1000 Hz); moreover, we identified a novel electrophysiological biomarker of epilepsy called ultrafast high-frequency oscillations (ufHFOs, 1000-2000 Hz, Figure 7) (Brázdil et al., 2017). These were identified in an interictal state in the hippocampus, amygdala, Figure 7: Very fast (500-1000 Hz) and ultrafast (1000-2000 Hz) high-frequency oscillation (vHFOs, ufHFOs) Examples of different types of oscillations recorded from standard depth macroelectrodes (A) at high frequencies. vHFOs and ufHFOs strongly correlate with positive resection outcome (B). The correlation coefficient was 0.19 in ripples, 0.44 in fast ripples, 0.70 in vHFOs, and 0.90 in ufHFOs. 33 and entorhinal cortex. Based on the article by Usui et al (2015), we suggest that their group and our group opened an utterly new research field within electrophysiology and epileptology. This field concerns the topic of vHFOs/ufHFOs and their clinical application. We hope we will be able to expand this topic in the near future. 34 Clinical semiology The evaluation of clinical semiology is an indisputable part of standard video EEG monitoring. The assessment is essential for both the differentiation between epileptic seizures and nonepileptic events and for the categorization of epilepsy. Clinical semiology in nonepileptic seizures The fundamental question is the distinction between psychogenic nonepileptic seizures (PNES) and epilepsy. Certain signs are typical of PNES: long duration, fluctuating course, asynchronous movements, pelvic thrusting, side-to-side head or body movements, forced eye closure, ictal crying, and memory recall (Avbersek and Sisodiya, 2010). PNES and epileptic seizures cannot be reliably distinguished based on tongue biting and urinary incontinence (Brigo et al., 2013). Abrupt onset, ictal eye-opening, and postictal confusion or sleep are highly associated with epileptic seizures (Syed et al., 2011). The differentiation between other types of nonepileptic attacks and epileptic seizures is less frequent in video EEG monitoring but can be extremely important, even life-saving, in patients with cardiac arrhythmias. The correct diagnosis of syncope can be straightforward in patients with typical signs (sudden loss of consciousness without any additional movements, ECG abnormality). However, the presence of convulsions in “convulsive syncopes” is challenging and leads to misinterpretations. The incidence of convulsive syncope is highly variable in the literature; it ranges from 12% in blood donors experiencing vasovagal syncope to 45% in patients with malignant cardiac arrhythmia (Aminoff et al., 1988; Lin et al., 1982). The movements associated with syncopes are usually described as arrhythmic jerks of the extremities, generalized myoclonus, 35 and some additional movements (head turns, oral automatisms, and righting movements) (Crompton and Berkovic, 2009). We described a patient with atypical syncopal manifestation with the most prominent signs of abdominal/thoracoabdominal contractions (Doležalová et al., 2013; Annex 6). We published a case report of a 71-year-old female patient who developed attacks of unconsciousness at the age of 70 years. Their frequency was approximately 1 per month in the beginning. The attacks were categorized as epileptic seizures based on motor signs and unremarkable results of an internal examination, including 24-hour ECG. After approximately one year, the frequency dramatically increased to 1 per day. The patient was admitted to our hospital for video EEG monitoring. We recorded several attacks accompanied by asystoles ranging from 20 to 30 seconds. A pacemaker was quickly implanted in the patient. The attacks were triggered by cardiac asystole; head and eye deviation appeared in the beginning, followed by brief jerks of the upper and lower extremities. However, the most prominent feature was contractions of the abdominal or thoracoabdominal wall. We recorded one attack, in which these abdominal/thoracoabdominal contractions were the only noticeable feature (Figure 8). The patient regained consciousness spontaneously after the restoration of heart activity; it was sometimes accompanied by nausea and vomiting. The precise mechanism of convulsive movements in syncopes is not fully understood. The most probable hypothesis is that the central pattern generators are responsible for their origin. The central pattern generators are distributed on the subcortical level, mainly the brain stem and spinal cord. They provide the rhythmic behavior, but normally they are inhibited by neocortical control. They can manifest only in cases of cortical control suppression; this situation leads to the 36 disinhibition of central pattern generators and subsequent rhythmic movements (Tassinari et al., 2009). Figure 8: Atypical clinical manifestation of convulsive syncope triggered by heart asystole The most prominent clinical feature was contraction of the abdominal/thoracoabdominal wall. The movements are marked by red circles. 37 Clinical semiology in epileptic seizures Seizure semiology is helpful in distinguishing the type of epilepsy (generalized vs. focal, temporal vs. extratemporal). Semiology was described in detail in patients with TLE. Temporal seizures usually have a longer duration and are characterized by a typical sequence of ictal signs. A patient usually reports an aura (epigastric, déjà vu) at the seizure beginning, followed by loss of consciousness. Afterward, patients develop oroalimentary automatisms, hand automatisms, dystonic posturing, and head version with a possible transition to a bilateral tonic-clonic seizure. Some signs have high lateralizing value (e.g., dystonic posture lateralizes seizure onset to the contralateral hemisphere; perseveration of ictal speech lateralizes seizure onset to the nondominant hemisphere) (Maillard et al., 2004). Frontal lobe seizures are characterized by shorter duration; the semiology is more diverse than with TLE. However, semiological categorization is possible and correlates with the anatomic organization along the rostrocaudal axis (Bonini et al., 2014). When differentiating between ETLE and TLE, it necessary to remember “pseudo-temporal” epilepsy. Pseudo-temporal epilepsy is epilepsy with seizures electrophysiologically starting in extratemporal areas but spreading to the temporal lobe, typically with temporal-like semiology (Andermann, 2003). Orbito-frontal, insular, and parieto-occipital, especially posterior cingulate, epilepsies can manifest as pseudo-temporal (Alkawadri et al., 2013; Enatsu et al., 2014; Isnard et al., 2000; Liava et al., 2014; Palmini et al., 1993; Shihabuddin et al., 2001; Williamson et al., 1992). The differentiation in nonlesional cases between temporal and pseudo-temporal epilepsy represents a diagnostic challenge. The distribution of IEDs seems to be helpful. Extratemporal IEDs represent a relevant red flag because they are atypical for pure TLE (Rémi et al., 2011). 38 We published a case report of a 33-year-old female patient who developed epilepsy at the age of 14 (Doležalová et al., 2017; Annex 7). The patient reported epigastric constriction or gustatory hallucination as an aura. Subsequently, temporal-like seizures developed; early oroalimentary automatisms, upper limb automatisms (mainly left-sided), non-lateralizing nose-wiping, and face-rubbing were present. IEDs were mainly over the left fronto-temporal region, but independent right and left frontal IEDs were also recorded. The ictal pattern slightly differed between individual seizures, but usually the ictal activity appeared over the left anterior fronto-temporal area, then it spread over the left hemisphere and to the contralateral side. MRI was described as normal, with an exception for left-sided thalamic infarct. Neuropsychological testing showed bilateral temporal functional impairment (verbal and visual encoding difficulties). 18F-FDG-PET revealed bitemporal hypometabolism (leftsided involvement was more pronounced). Based on the inconsistent results of non-invasive investigations, IEEG was suggested as an inevitable further step. An extensive exploration of the temporal areas was performed; the extratemporal regions known for spreading ictal activity to the temporal lobe were also covered, namely the fronto-orbital cortex, insula, and posterior cingulate gyrus. IEDs were independently present in mesial temporal structures and in the posterior cingulate gyrus. The ictal onsets were recorded in all seizures in the posterior cingulate gyrus. The MRI was reviewed, and an inconspicuous change in the gray matter was revealed in the posterior cingulate gyrus in the proximity of the recorded seizure onset. The patient underwent surgery; a limited cortectomy was performed in the left posterior cingulate gyrus (Figure 9). The patient has been seizure-free since surgery. 39 This case report highlights the problematic differential diagnosis in nonlesional TLE. These patients should be referred for IEEG, which has to cover both temporal and extratemporal areas. This approach has to allow the differential between pseudo-temporal and temporal seizure onset. Figure 9: The resection in posterior cingulate gyrus This figure illustrates the borders of confined resection in the left posterior cingulate gyrus (yellow circle). The red circle represents the area where low-voltage fast ictal activity was recorded. Republished with permission. 40 Neuropsychology Neuropsychological testing is a standard procedure within epilepsy surgery programs. Neuropsychological deficits can be conditioned by structural lesions and by the functional derangement of different brain networks (Baxendale et al., 1998; Durwen et al., 1989; Jokeit et al., 2000). Historically, the most information is available about functional deficits associated with TLE because TLE patients have formed the majority of surgical candidates. TLE affects mainly memory networks. The relation between left-sided TLE with verbal memory impairment is indisputable. The association between right TLE and figural or visuospatial memory is weaker because patients with left-sided TLE also exhibit very often falsely-lateralizing figural or visuospatial memory difficulties (Helmstaedter et al., 1994; Helmstaedter et al., 1995; Loring et al., 1999). TLE patients also have naming difficulties (Malow et al., 1996). The knowledge about neuropsychological deficits associated with extratemporal epilepsies is less extensive. Patients with frontal lobe epilepsy often have problems with executive functions, working memory, choice of appropriate behavior, the establishment of emotional valences, and evaluating and balancing consequences (Bechara et al., 2000; Rolls, 2000; Sarazin et al., 1998). There is even less information about deficits in insular, parietal, and occipital epilepsies. These “posterior” epilepsies can manifest with aphasia, alexia, agraphia, acalculia, agnosia, or neglect syndrome, but these are not very common and usually well-compensated. They often mimic frontal epilepsies or TLE with their types of dysfunction, depending on the pattern of the spread of ictal activity (Kasowski et al., 2003) . 41 We know that early-onset epilepsy can alter brain networks and condition the significant redistribution of brain functions. We pursued this topic in our study (Doležalová. et al., 2017; Annex 8) There are significant differences between the function of the left (dominant) and right (nondominant) hemispheres. The left hemisphere is responsible for language, verbal, and motor (handedness) function. The clinical variable conditioning this hemispheric shift is the age of insult, i.e., lower age means a higher likelihood of functional reorganization. We know that earlyonset epilepsy can condition such a critical shift and cause the redistribution of brain functions. It was demonstrated that left-sided mesial TLE is associated with atypical lateralization of verbal memory and language function (Adcock et al., 2003; Alessio et al., 2013; Powell et al., 2007; Richardson et al., 2003). However, we had no specific information on whether the “pure” temporal lesion as present in mesial TLE is sufficient to cause atypical handedness because of the core function of the frontal lobe for motor functions. In our study, we investigated the occurrence of left-handedness in mesial TLE and clinical variables associated with this left-handedness, We investigated 73 patients with mesial TLE associated with hippocampal sclerosis. The patients underwent the Edinburgh Handedness Inventory for handedness lateralization. Patients with laterality index from 0 to 100 were categorized as right-handed; patients with laterality index from -100 to 0 were categorized as left-handed. 42 Figure 10: The percent representation of left-handedness and right-handedness in patients with left-sided and right-sided mesial TLE Republished with permission. There were 32 (43.8%) patients with right-sided mesial TLE and 42 (56.2%) patients with leftsided mesial TLE. Fifty-four (74%) patients were right-handed. The other 19 (26%) patients were left-handed. The representation of left-handed patients was significantly higher among patients with left-sided mesial TLE (p=0.030, Figure 10). When assessing the clinical variable correlating with the change of handedness in left-sided mesial TLE, we found that left-handed patients had an earlier epilepsy onset (median 8 years of age, minmax 0.5-17 years) compared to right-handed patients (median 15 years of age, min-max 3-30 years; p<0.001, Figure 11). 43 Figure 11: Left-sided mesial TLE - the difference in age at epilepsy onset between left-handed and right-handed patients Republished with permission Even a confined lesion like hippocampal sclerosis can alter remote brain function and cause a change of handedness. The probability of this change is age-dependent. 44 Positron emission tomography (PET) PET is a functional neuroimaging method that is essential for tailoring surgical procedures. 18FFDG-PET is routinely performed in an interictal state with a glucose analog. 18F-FDG-PET is an indirect marker of neuronal activity and allows the absolute quantification of cerebral glucose metabolism. Ictal 18F-FDG-PET is not routinely performed because of the short half-life of the tracer. The seizure onset zone is characterized by a decrease in glucose metabolism in the interictal state. 18F-FDG-PET can be evaluated by both visual (qualitative) and mathematical (quantitative) analysis. Qualitative analysis was found to be reliable in both lateralization of the seizure onset zone and prediction of postoperative outcome in patients with temporal lobe epilepsy (Delbeke et al., 1996; Radtke et al., 1993), but its proper interpretation is influenced by high levels of variability among investigators. To overcome this limitation, methods of mathematical, i.e., quantitative, PET analysis have been developed (Signorini et al., 1999; Takahashi et al., 2012; Van Bogaert et al., 2000). We tried to establish the benefits of quantitative analysis for epilepsy surgery in our work (Kojan et al., 2018; Annex 9). In this work, we compared 18F-FDG-PET using SPM SPM-PET in patients with HS-TLE and control groups in order to assess the differences in the distribution of glucose hypometabolism. In the next step, we tried to identify the differences between HS-TLE subgroups. We identified a group of 49 HS-TLE patients and 24 control individuals who underwent 18F-FDGPET. Statistical parametric maps were calculated for each subject. First, SPM-PET was 45 mathematically compared between patients and controls. Next, we tried to identify the differences in the HS-TLE group in terms of epilepsy duration, HS side, histopathological findings, history of insult, and surgical outcome. When comparing HS-TLE patients to controls, we found profound bilateral hypometabolism in HSTLE patients. On the HS side, the hypometabolism was present over both temporal areas (pole, Figure 12: The difference in hypometabolism between patients with temporal lobe epilepsy associated with hippocampal sclerosis (HS-TLE) and a control group using statistic parametric mapping of positron emission tomography (SPM-PET). Republished with permission. 46 amygdala, hippocampus, parahippocampal gyrus) and extratemporal areas (orbitofrontal cortex, insula, thalamus, posterior cingulate gyrus). On the contralateral side, the hypometabolism was present in the thalamus, anterior cingulate gyrus, and middle cingulate gyrus (Figure 12). When comparing individual subgroups within HS-TLE, we observed differences between patient subcategories: (1) HS class I and II versus HS class III and IV, (2) excellent surgical outcome (Engel I and II) versus poor surgical outcome (Engel III and IV), and (3) presence versus absence of potential insults (encephalitis, meningitis, febrile seizure). We did not find any significant differences between patients with respect to the HS side and to FCD in the temporal pole. There are significant differences between HS-TLE patients and controls regarding glucose metabolism. Hypometabolism is bilateral and more widespread on the HS side. There are differences in the hypometabolism distribution even among HS-TLE patients depending on the HS type, surgical outcome, and potential insult. In general, the sensitivity of 18F-FDG-PET depends on the localization of epileptogenic zone. It is higher for TLE than for ETLE (70-85% for TLE in comparison to 30-60% for ETLE) (Arnold et al., 1996; Drzezga et al., 1999; Juhász et al., 2001; Ryvlin et al., 1998; Won et al., 1999). PET is essential in patients with clinical and electrophysiological characteristics of TLE but without any suspected lesion on MRI. The well-expressed hypometabolism over the temporal lobe could be found in some patients who form MRI-negative, PET-positive TLE groups (Bell et al., 2009; Carne et al., 2004; Carne et al., 2007a; Carne et al., 2007b; Kuba et al., 2011). 47 Interictal/ictal SPECT and SISCOM Interictal/ictal SPECT and SISCOM are other functional techniques widely applied within the epilepsy surgery program. They use tracers marked by radioactive technetium (99m Tc-HMPAO) that are applied interictally for interictal SPECT or ictally for ictal SPECT. SISCOM is obtained by mathematical postprocessing of interictal and ictal SPECT; the results are displayed on MRI. The highest limitation of ictal SPECT is the necessity of applying a radioactive tracer early during the seizure evolution to visualize the seizure onset rather than the propagation of seizure activity to different brain areas (Lee et al., 2006b; Patil et al., 2007). SPECT/SISCOM are mentioned here only briefly because we paid only marginal attention to it. 48 The decision-making process within epilepsy surgery Clinicians focusing on epilepsy surgery have to address several questions when managing a patient to find an optimal therapeutical approach. The fundamental problem concerns the localization of the epileptogenic zone. First, a rational hypothesis about its localization and its extent must be formulated. This hypothesis must be based on the results of all the methods used in the presurgical evaluation and on the localization of zones involved in seizure genesis. There is currently no widely accepted “tool” for precise epileptogenic zone delineation. Some methods for epileptogenic zone characterization have been proposed in the literature, including the epileptogenic index (Bartolomei et al., 2008). The epileptogenic index is a semi-automated method for quantifying the epileptogenicity of different brain areas. This index is based on two variables: (1) the ability of individual brain structures to generate fast discharges, and (2) the time delay between seizure onset and the development of fast discharges in a given structure. A high epileptogenicity index is present in structures that generate fast rhythms early in the onset, meaning that the structure is highly epileptogenic. The epileptogenic index has been used in several studies, but there are doubts about its usefulness (Aubert et al., 2009; Neal et al., 2020; Pizzo et al., 2017). In our research, we focused on two clinicaly relevant situations: MRI-negative TLE epilepsy (Doležalová et al., 2016a; Annex 10), and bitemporal epilepsy (Řehulka et al., 2014; Annex 11). 49 Clinical situation: MRI-negative TLE – differences between mesial and neocortical types In MRI-negative TLE patients, it is essential to differentiate between those with mesial seizure onsets and those with neocortical seizure onsets for precise seizure tailoring and minimization of adverse events. In most cases, the differentiation can be made based on chronic intracerebral recording. We tried to delineate these two groups, i.e., mesial and neocortical MRI-negative TLE, based on a non-invasive analysis. Our retrospective study included a total of 20 patients with MRI-negative TLE who underwent intracerebral recording before surgery. According to the localization of the seizure onset zone in IEEG, the patients were subdivided into mesial (amygdala, hippocampus, parahippocampal gyrus) and neocortical MRI-negative TLE groups. We tried to find differences between the following variables: demographic data, FDG-PET findings, interictal and ictal semi-invasive EEG, seizure semiology, results of surgery, and histopathological findings. Within 20 MRI-negative TLE patients, 13 (65%) patients had mesial seizure onset zones and 7 (35%) patients had neocortical seizure onset zones. The statistically significant differences between mesial and neocortical MRI-negative TLE patients were present in the IED distribution (neocortical MRI-negative TLE patients exhibit extratemporal IEDs more often, p=0.031) and seizure lateralization (all neocortical MRI-negative TLE patients had at least one non-lateralized or falsely lateralized seizure, p=0.044). When analyzing seizure semiology, we managed to identify some trends, but a statistical comparison was not possible because of the low numbers in each group. No differences were found between other variables. 50 In conclusion, two-thirds of patients with MRI-negative TLE have mesial seizure onset zones and one-third of patients with MRI-negative TLE have neocortical seizure onset zones. Differentiation based on non-invasive data is challenging despite some distinctions in IED distribution and seizure lateralization. However, the surgical results are satisfactory in both groups, comparable to MRIpositive TLE cases. Clinical situation: Bitemporal epilepsy Our next study focused on the analysis of bitemporal epilepsy (Řehulka et al., 2014; Annex 11). It is necessary to clarify the definition of bitemporal epilepsy. The definition differs substantially among publications. In some publications, the definition of bitemporal is based on independent seizures arising from both temporal lobes in scalp EEG (Holmes et al., 2003). Some authors have used various combinations of different criteria (scalp or sphenoidal EEG, MRI, and neuropsychological) (Spanedda et al., 1997). We adapted the definition of Hirsh et al. (1991), who defined bitemporal epilepsy as independent seizures originating from both temporal lobes based on IEEG analysis (Hirsch et al., 1991). We also included patients in whom habitual seizures were recorded from one temporal lobe and elicited by electrical stimulation from the contralateral side. In our study, we tried to characterize the bitemporal group in more detail and to establish differences when comparing to strictly unitemporal ones. We retrospectively identified a group of patients with bitemporal epilepsy; these patients were matched to the patients with unilateral temporal lobe epilepsy. We then compared bitemporal and unitemporal patients in terms of demographic data, scalp EEG findings, and ictal semiology. From the electrophysiological point of view, we evaluated the presence or the absence of early rhythmic theta/delta activity, time to development of rhythmic theta/alpha activity, bitemporal propagation 51 time, and ictal activity duration. We evaluated the presence of ictal signs: early oroalimentary automatisms, ictal motor signs (early nonversive head-turning, lateralized ictal immobility of the upper limb, ictal dystonia, rhythmic ictal nonclonic hand motions [RINCH], peri-ictal vegetative symptoms (retching with/without vomiting, cough, urinary urge, and water drinking), peri-ictal motor signs (peri-ictal nose-wiping, peri-ictal bed leaving), and duration of peri-ictal unresponsiveness (short ≤ 3 min, medium 3-5 min, long > 5 min), defined as the period from the end of EEG ictal activity to the restitution of both fluent speech and orientation. We found statistically significant differences between bitemporal and unitemporal epilepsy when analyzing seizure semiology in the presence of ictal motor signs and the duration of postictal unresponsiveness. At least one ictal motor sign was observed in 18 (20.7%) out of 87 seizures in bitemporal epilepsy vs. in 47 (61%) out of 77 seizures in unitemporal epilepsy (p<0.001). Patients with bitemporal epilepsy tend to have a longer duration of postictal unresponsiveness (p=0.002). Patients with bitemporal epilepsy had postictal unresponsiveness < 3 min in 18 (28.6%) seizures and > 5 min in 27 (42.9%) out of 63 seizures; patients with unitemporal epilepsy had postictal unresponsiveness < 3 min in 34 (57.6%) and > 5 min in 11 (18.6%) out of 59 seizures. No other statistically significant differences were found when analyzing other variables. In our study, we managed to demonstrate differences in ictal semiology between patients with unitemporal and bitemporal epilepsy. Patients with bitemporal epilepsy have poorer ictal semiology in terms of ictal motor signs and need a longer time to reach complete recovery. 52 Neurostimulation methods In patients who could not be offered “classical” resective surgery or in whom epilepsy surgery failed to abolish seizures, neurostimulation methods represent a possible therapeutic option. Although complete seizure cessation after implantation is relatively rare, neurostimulation can bring significant seizure reduction and improvement of QOL (Fisher et al., 2010; Harroud et al., 2012; Ma and Rao, 2018; McIntosh et al., 2004; Orosz et al., 2014; Ryvlin et al., 2014; Ryvlin and Kahane, 2005; Salanova et al., 2005). The efficacy can be measured by average seizure reduction or responder rate. The responder rate is traditionally defined as a percentage representation of patients with at least 50% seizure reduction (≥50% seizure reduction). The responder rate of all neurostimulation methods is approximately 50% in the first year after implantation with subsequent increases (Fisher et al., 2010; Harroud et al., 2012; Ma and Rao, 2018; McIntosh et al., 2004; Orosz et al., 2014; Ryvlin et al., 2014; Ryvlin and Kahane, 2005; Salanova et al., 2005). At the moment, three neurostimulators are approved for the treatment of drug-resistant epilepsy: (1) VNS, (2) DBS, and (3) RNS (not approved in Europe). The precise antiepileptic mechanism of neurostimulation methods is not fully understood. It is based on the historical observation that the modulation of some cortical or subcortical structures can influence brain excitability and subsequently lead to seizure reduction. The following structures were chosen as possible stimulation targets: cerebellum, caudate nucleus, thalamus (centromedian, anterior, subthalamic nuclei), vagus nerve, and the cortical area of the seizure onset itself (Theodore and Fisher, 2004). 53 The first attempts with neurostimulation in the treatment of drug-resistant epilepsy were made with the cerebellum (Cooper et al., 1973; Reimer et al., 1967). Despite the promising results of preliminary and uncontrolled studies, the outcomes of a controlled randomized study were disappointing (Davis and Emmonds, 1992; Van Buren et al., 1978; Wright et al., 1984). The efficacy of cerebellar stimulation was not proven. Despite the apparent limitations, specifically the small patient sample (only 12 patients were included), the idea of cerebellar stimulation in the treatment of epilepsy treatment was abandoned and attention was focused on the identification of novel targets. In 1990, preliminary results with stimulation of the vagus nerve were introduced (Penry and Dean, 1990; Uthman et al., 1990; Uthman et al., 1993). The efficacy and safety of this method were subsequently proven (Ben-Menachem et al., 1994; Handforth et al., 1998), and VNS became the first authorized neurostimulation system for the treatment of drug-resistant epilepsy. 54 Vagus nerve stimulation (VNS) VNS is the most widespread neurostimulation method for the treatment of drug-resistant epilepsy; it has been implanted in over 100,000 people, a third of which are children, around the world (BenMenachem, 2002; Englot et al., 2017). VNS was approved for the treatment of drug-resistant focal epilepsy based on the results of two randomized, double-blind controlled trials, EO3 (multicenter international) and EO5 (multicenter USA) (Ben-Menachem et al., 1994; Handforth et al., 1998). These studies compared the efficacy and safety of high-level stimulation (therapeutic; the stimulation frequency ranged between 20 and 50 Hz) and low-level stimulation (sham; the stimulation frequency was 1 or 2 Hz). The mean reduction in seizure frequency in EO3 was 24.5% for active (high-frequency) stimulation and 6.1% for sham (low-frequency) stimulation; in EO5 it was 24.5% for active stimulation and 15% for sham stimulation (Ben-Menachem et al., 1994; Handforth et al., 1998). VNS efficacy Further studies focused on the long-term effect of VNS; they proved an increase in VNS effectiveness over time. Révész et al. (2018) observed 130 consecutive patients with epilepsy. They found significant seizure reduction; the responder rate increased from 22.1% in the first year to 43.8% in the fifth year regardless of changes in AEDs (Révész et al., 2018). However, there is still limited information about patient responses to VNS with more than a 10-year perspective (Elliott et al., 2011; Révész et al., 2018; Serdaroglu et al., 2016). We tried to address this limitation in our study (Chrastina et al., 2018b; Annex 12). 55 The main aim of this study was to evaluate our long-term (10-17 years) experience with VNS in a group of drug-resistant epilepsy patients. The second aim of this study was to find the predictors for long-term VNS response based on clinical data analysis. We retrospectively analyzed the response to VNS in patients who were implanted at least 10 years ago. Based on this response, patients were categorized as follows: non-responder (seizure reduction < 50%), responder (seizure reduction ≥ 50%), and 90% responders (reduction ≥ 90%). We found 74 patients who were implanted with VNS for at least 10 years; their response to VNS is summarized in Figure 13. The responders’ rates were as follows: 38.4% in year 1, 51.4% in year 2, 63.6% in year 10, and 77.8% in year 17. The 90% responders’ rates were: 1.4% in year 1, 5.6% in year 2, 15.1% in year 10, and 11.1% in year 17. Changes were made in AEDs in most patients. We did not find any statistical predictors for long-term patient responses to VNS. Based on our results, we can summarize that VNS is a safe treatment option in patients with drugresistant epilepsy with long-term efficacy. In our study, we proved an excellent response to VNS, even in patients treated for 17 years. 56 Figure 13: Temporal changes in representation of responders (R) and 90% responders (90% R) in year 1-9 and year 10-17 Republished with permission. 57 Based on improved surgical care, the low risk of complications, and accessibility, VNS is also indicated for older patients with chronic internal disease, and patients with severe intellectual disability are referred for palliative surgery such as VNS (Andriola and Vitale, 2001; Danielsson et al., 2008). The main aim of our further work was to assess whether even these older patients and patients with severe intellectual disability can profit from VNS implantation (Chrastina et al., 2018a; Annex 13). We retrospectively collected data about patients implanted with VNS. The patients were categorized as responders (≥ 50% seizure reduction), 90% responders (≥ 90% seizure reduction), and non-responders (< 50% seizure reduction). The patients were evaluated 1 year after surgery and at their most recent visit after at least 2 years of follow-up care. The patients were divided into groups depending on age (< 40 years versus ≥40 years; < 50 years versus ≥ 50 years; respectively), epilepsy duration (< 20 years versus ≥ 20 years; < 30 years versus ≥ 30 years; < 40 years versus ≥ 40 years; respectively), and presence of intellectual disability (present versus absent). We identified 103 patients with VNS who had at least one follow-up visit, and 94 patients treated with VNS longer than 2 years. When analyzing the response to VNS therapy, we did not prove significant statistical differences with respect to patient age, epilepsy duration, or presence of intellectual disability, with only two exceptions. First, the patients with epilepsy duration ≥ 20 years tended to be classified more often as responders than patients with epilepsy duration < 20 years at the last follow-up visit (p=0.046). Second, the patients with epilepsy duration ≥ 30 years were more often 90% responders than patients with epilepsy duration < 30 years at the follow-up 58 visit 1 year after implantation (p=026). Nevertheless, this difference was lost at the last follow-up visit. It is possible to claim than VNS can be used as a method of therapeutical choice even in patients with higher age, longer epilepsy duration, and severe intellectual disability because all these groups can profit from this treatment. Limitations of VNS therapy There are limitations associated with VNS therapy, namely (1) adverse events of VNS therapy, and (2) inability to predict efficacy at the individual patient level. Adverse events of VNS therapy In general, adverse events conditioned by VNS are moderate and can be divided into two groups: (a) adverse events related to surgery, and (b) adverse events associated with long-term use. Adverse events related to surgery include common infections, vocal-cord palsy, lower facial weakness, and bradycardia or systole in the operating room in 0.1% cases. The reason for this rare heart rhythm abnormality is not fully understood; it can be conditioned by abnormal electrode placement, indirect stimulation of the cervical cardiac nerves, technical malfunction, or an idiosyncratic reaction (Ben-Menachem, 2002; Handforth et al., 1998). The adverse events associated with long-term use include coughing, throat pain, and hoarseness; they tend to improve over time (Handforth et al., 1998; Morris et al., 1999). VNS does not influence blood counts or blood chemistry, including liver function, which is essential information for many patients (Ben-Menachem, 2002). 59 VNS was implanted in more than 100,000 people around the world in all age groups (BenMenachem, 2002; Englot et al., 2017). These 100,000 patients include women of child-bearing age, and there are anecdotal reports about pregnancies, deliveries, and further child development in these patients. VNS may influence gravidity in both its afferent and efferent fibers. Afferent fibers (forming 80% of the vagus nerve) have connections with cortical and subcortical structures. These fibers, among other things, influence the functioning of the central autonomic system and excretion of neuroendocrine hormones with particular importance of oxytocin for gravidity and child-bearing. Efferent fibers (forming 20% of the vagus nerve) influence mainly the heart functioning, but a small portion projects to the uterus. These fibers may have an impact on uterus functioning during pregnancy and delivery (Sabers et al., 2018; Sato et al., 1996). However, it is necessary to highlight that gravidity, child-development, and delivery are not influenced only by VNS. AEDs, usually in polytherapy, seizures, and concomitant medication play essential roles, and it is impossible to separate their influence. We collected data about gravidity in patients treated with VNS for drug-resistant epilepsy (Chrastina, submitted; Annex 14). From a group of 71 women of child-bearing age who were implanted with VNS in the Brno Epilepsy Center between 1999 and 2018, we identified 4 patients who gave birth. The individual cases are presented below. 60 Patient 1: Patient 1, treated for bitemporal epilepsy, with a history of IEEG and temporal lobe resection, became pregnant at the age of 26 years. She was treated with lamotrigine, levetiracetam, carbamazepine, and perampanel. The VNS was implanted 81 months before pregnancy; the patient was a non-responder with a 40% seizure reduction. She reported a mild decrease in seizure frequency during gravidity (no changes were made to AEDs or stimulation parameters). The birth was induced because of epilepsy; she delivered a healthy son (49 cm, 3500 g) at the 39th gestational week. The post-birth adaptation was without any problems. Breastfeeding lasted for 6 months. At the moment, the boy is 2 years old and developmentally normal. Patient 2: Patient 2 (33 years) was treated by levetiracetam and lamotrigine for focal epilepsy with high seizure frequency (several seizures per day) not accompanied by any falls or bilateral tonic-clonic seizures. The patient had a history of 4 spontaneous abortions due to cervical insufficiency. VNS was implanted 10 months before conception. Pregnancy was uneventful with no change of AEDs or VNS parameters. A cesarean section was indicated because of epilepsy at the 34th gestation week. The patient gave birth to a son (47 centimeters, 2150 g) who spent 1 week in an incubator for worse post-birth adaptation. Breastfeeding was not initiated. The patient spontaneously reported a week-long seizure-free period after birth (the most prolonged seizure-free period in her life since epilepsy development). Now, the boy is 10 years, treated with Attention Deficit Hyperactivity Disorder (ADHD). He is in elementary school, requiring an assistant because of ADHD. 61 Patient 3 Patient 3, 38 years old, was treated for focal epilepsy. She had a history of an uncomplicated pregnancy, which led to the birth of a healthy child before VNS implantation. VNS was implanted 63 months before the second pregnancy and had no impact on seizure frequency (VNS was explanted later). The patient took 4 AEDs: valproic acid, primidone, zonisamide, and clonazepam. She reported a slight increase in seizure frequency. However, a placental separation developed in the 31st week of gravidity, which led to an acute cesarean section. A boy (33 cm, 1315 g) spent 2 months in an incubator. Breastfeeding was not initiated. At the moment, he is 2 years old developmentally normal for his age. Patient 4: Patient 4 was 22 years old at the time of pregnancy. She had focal epilepsy treated by carbamazepine, lamotrigine, and valproic acid. VNS was implanted 23 months before conception, and seizure reduction was 60%. The gravidity was uneventful with a decrease of seizure frequency (no change of VNS parameters or AEDs) terminated by spontaneous vaginal birth. A son was delivered at the 39th gestational week (48 cm, 2700 g), post-birth adaptation was without any problems. At the moment, the boy is 10 years old, diagnosed with mental retardation with an autistic spectrum disorder; he attends a special school with assistant support. To our knowledge, this is the second-largest single-center report about the influence of VNS on pregnancy. We documented high numbers of obstetric interventions in our patients. The pregnancies in our patients led to the delivery of 4 children; 2 of them require special schooling. This number is relatively high, which deserves particular attention in the future. 62 Prediction of VNS efficacy The most significant problem associated with VNS therapy is the inability to predict VNS efficacy at the individual patient level. At the moment, it is possible to recognize which patient groups can profit more from VNS treatment. Englot et al., in a meta-analysis of more than 3,321 patients, found that patients with generalized epilepsy and children, especially children younger than six years, exhibit the highest seizure reduction. Regarding etiology, patients with posttraumatic epilepsy and tuberous sclerosis profit the most from VNS (Englot et al., 2011). However, it is not possible to reliably predict VNS efficacy at the individual level, i.e., based on pre-implantation data. The pre-implantation identification of patients profiting from VNS is crucial for several reasons. Importantly, it can decrease the need for ineffective surgery, improving the allocation of financial costs. Moreover, it can increase the confidence of physicians and patients in neurostimulation. At the moment, four studies have dealt with pre-implantation VNS efficacy prediction (BabajaniFeremi et al., 2018; Ibrahim et al., 2017; Mithani et al., 2019). Ibrahim et al. (2017) published a prediction model based on the analysis of resting-state fMRI in a group of 21 children with drug-resistant epilepsy. They demonstrated that thalamocortical connectivity to the anterior cingulate and insular cortices is stronger in responders than in nonresponders. They reached an accuracy of 86% in their study. When validating their model on an external dataset of 8 children, the accuracy was almost 90% (Ibrahim et al., 2017). The study by Babajani-Feremi et al. (2018) predicts VNS efficacy by analyzing resting-state magnetoencephalography (MEG). Resting-state MEG can be used to study and quantify the impact 63 of epilepsy on the brain networks and the changes of brain networks conditioned by different therapeutical interventions. In this study, the authors made calculations on recorded resting-state MEG using three global graph measures: modularity, transitivity, and characteristic path length, in which they found statistically significant differences between responders and non-responders. Subsequently, they formulated a classifier that was able to predict VNS response with an accuracy of 87% (Babajani-Feremi et al., 2018). The study by Mithani et al. (2019) proved the differences in white matter microstructures in DTI between responders and non-responders; these differences were subsequently used for the construction of a classifier for VNS response prediction (Mithani et al., 2019). In our study, we tried to develop a statistical classifier for prediction of VNS response based on the EEG reactivity to external stimuli (Brázdilet al., 2019; Annex 15). Our work is based on the finding that synchronization and desynchronization of EEG are responsible for VNS efficacy (Jaseja, 2010; Marrosu et al., 2005). We presumed that interindividual variability in the EEG ability to synchronize/desynchronize in response to external stimuli conditions a type of response to VNS (responders vs. non-responders). As the next step, we tried to formulate a statistical classifier for predicting response to VNS in an individual patient based on pre-implantation EEG. We collected data in a group of patients treated with VNS for drug-resistant epilepsy. Patients were categorized as responders or non–responders based on the criteria by McHugh et al. (2007). In each patient, a standard pre-implantation EEG with photic stimulation and hyperventilation was identified. The EEG was segmented to several time-intervals and filtered to the individual 64 frequency bands (theta, alpha, beta, gamma). A Hilbert transform was employed to estimate the envelopes of pre-defined pass-band frequency oscillations as a function of time. Subsequently, absolute and relative mean powers were calculated in time intervals for each frequency band. The responders and non-responders were compared using these relative mean powers. Based on the data obtained in the previous steps, we developed a statistical classifier for prediction of VNS efficacy. This model was validated on an independent patient dataset. Figure 14: Differences in relative mean powers between responders and non-responders The differences between responders and non-responders were present mainly in the alpha and gamma frequency during photic stimulation and hyperventilation (white dots). Black dots represent individual electrodes. Republished with permission. 65 The study included a total of 60 patients treated with VNS: 35 (58%) responders and 25 (42%) non-responders. When relative mean powers were calculated, the differences between responders and non-responders were revealed. These differences were present mainly in the alpha and gamma frequency ranges during photic stimulation and hyperventilation (Figure 14). In the next step, the statistic classifiers based on the relative mean power changes in (1) areas defined by a group of electrodes or (2) a single electrode were created using automatic machine learning. By this approach, we managed to achieve 86.7% accuracy (88.6% sensitivity and 84% specificity). We then validated this statistical model on the independent patient data (25 patients), Figure 15: The individual areas (a) and electrodes (b) used for statistic classifiers in each time interval for individual frequency bands Republished with permission. 66 where we confirmed our results with 86.4% accuracy (83.3% sensitivity, 90% specificity). The groups of electrodes and the individual electrodes chosen for the development of a statistical classifier are shown in Figure 15. We found differences in the cerebral reactivity to external stimuli between responders and nonresponders. These differences were present mainly in photic stimulation and hyperventilation. In the next step, a statistical classifier for predicting VNS efficacy was developed. The accuracy of the classifier was confirmed in an independent patient dataset. We plan to validate our results in another project (see the Future directions section). Our previous work concerned VNS response prediction based on EEG analysis. However, there are also reports about links between VNS efficacy and heart-rate variability (HRV) based on ECG analysis. Liu et al. (2017) demonstrated that non-responders to VNS have more pronounced impairment of cardiac autonomic functions (Liu et al., 2017). We tried to confirm this theory in our manuscript (Plesinger et al., submitted; Annex 16). The patients were recruited from our previous study based on the availability of quality ECG accompanying EEG recording. The study was done in the following steps: (1) an individual EEG was sub-segmented into 9 time intervals: Rest 1, Eyes-opening 1, Rest 2, Photic stimulation, Hyperventilation by mouth, Hyperventilation by nose, Rest 4, Eyes-opening 2; (2) the individual heartbeats were detected, and heartbeat intervals (RR intervals) were estimated; (3) median RR (MRR) were counted in each time interval; (4) MRRs were normalized to baseline at Rest 4, which led to the estimation of normalized MRR (NMRR); and (5) statistical differences in MRR and in NMRR between responders and non-responders were counted; the Area Under The ReceiverOperator Curve (AUC) was determined. 67 We included a total of 66 patients from the source study. We did not find any statistical differences between the responders and non-responders in MRR. However, when normalization was performed, the differences in NMRR were present in 5 time intervals: Rest 1, Rest 3, Hyperventilation by nose, Hyperventilation by mouth, and Eyes-opening 2 (Figure 16). Figure 16. The separation of responders and non-responders based on normalization of median RR (NMRR) interval - 1st and 3rd quartile of specific boxplots marked by blue and orange stripes - significant features marked in gray ECG can provide additional information in the prediction of VNS efficacy. It will probably be helpful in the construction of a statistical classifier. We plan to employ this method in our future research. 68 Deep brain stimulation of the anterior nucleus of the thalamus (ANT-DBS) in epilepsy DBS is a common and effective therapeutic method for patients with movement disorders, particularly Parkinson’s disease. Despite the extensive experience with DBS in this field, there is substantially less knowledge about DBS as a treatment for refractory epilepsy. ANT-DBS is currently approved as a therapeutic option for the treatment of drug-resistant epilepsy in Europe, the USA, Canada, and Australia (Li and Cook, 2018). The ANT was chosen as a suitable target for its anatomical and functional position. The ANT is part of the “Papez circuit” and has dense afferent connections from the hippocampus both directly via the fornix and indirectly via the mammillothalamic tract. There are other afferent fibers to the ANT from the anterior cingulate cortex, posterior cingulate cortex, retrosplenial cortex, and inferior parietal lobule. The efferent fibers project mainly to the hippocampus (Child and Benarroch, 2013). Despite this detailed information about ANT anatomy and function, the precise mechanism of action of ANT-DBS is not fully understood (Child and Benarroch, 2013). The first ideas of thalamic stimulation as a suitable treatment for neurological diseases, including epilepsy, were introduced in the 1970s and 1980s (Cooper, et al., 1980). The importance of ANT for seizure genesis/suppression was first proven in experimental animal studies (Mirski and Ferrendelli, 1984, 1986; Mirski et al., 1986; Mirski et al., 1997). This led to pilot studies in humans (Andrade et al., 2006; Hodaie et al., 2002; Kerrigan et al., 2004; Lee et al., 2006a; Lim et al., 2007; Osorio et al., 2007). 69 DBS – efficacy and adverse events Hodaie et al. (2002) implanted a group of 5 patients with ANT-DBS, which caused significant seizure reduction (the seizure reduction ranged from 33% to 89%; mean seizure reduction was 54%), the implantation and stimulation did not elicit any significant adverse events or any behavioral changes (Hodaie et al., 2002). Other preliminary studies also had satisfying results (Andrade et al., 2006; Kerrigan et al., 2004; Lee et al., 2006a; Lim et al., 2007; Osorio et al., 2007). Based on this, the study of Stimulation of the Anterior Nuclei of Thalamus (SANTE study) was initiated (Fisher et al., 2010). The SANTE study was a multicenter, double-blind, randomized trial of bilateral ANT stimulation for focal epilepsy. The SANTE study was divided into two phases: the blinded phase (the first three months) and the unblinded phase (subsequent 21 months). In the blinded phase, patients were randomized into a stimulated group and a non-stimulated group. In the next phase, all patients received stimulation. At the end of the blinded phase, the stimulated group reported significantly higher seizure reduction than the control group: a 29% greater reduction in seizure frequency in the stimulated group. During the two years of follow-up care, the authors proved the long-term efficacy of ANT-DBS (there was a 56% median reduction in seizure frequency, 54% patients were responders [≥50% seizure reduction], and 12% patients were seizure-free for at least six months) [30]. The follow-up care of patients within the SANTE study continued, and Salanova et al. (2015) reported about the ANT-DBS efficacy in 5 years (Salanova et al., 2015). Seventy-five (68%) patients out of the 110 initially implanted patients remained in the study. In the 5th year, the median seizure reduction was 69%, the responder rate was 68%, and 16% of patients were seizure-free at 70 least for six months. A similar trend is present with VNS; the percent representation of patients with favorable responses to stimulation increases over time, though the precise reason for this phenomenon remains still speculative. The most common adverse events of ANT-DBS during the first year of the SANTE study were as follows: paresthesias (18% patients), pain in the implant site (10.9% patients), and infection in implant site (9.1% patients) (Fisher et al., 2010). During the five years of follow-up care, device-related adverse events and other adverse events were reported (Salanova et al., 2015). The device-related adverse events were pain in the implant site (23.6% patients), paresthesias in the implant site (22.7% patients), implant site infection (12.7% patients), therapeutic product ineffective (10% patients), discomfort (9.1% patients), lead(s) not within target (8.2% patients), sensory disturbance (8.2% patients), memory impairment (7.3% patients), implant site inflammation (7.3% patients), dizziness (6.4% of patients), postprocedural pain (6.4% patients), extension fracture (5.5% of patients), and neurostimulator migration (5.5% patients). Infection led to the explantation of the system in nine patients (five patients experienced complete system explantation, four had partial explantation). Depression, memory impairment, suicidal ideation, and deaths ranked among other adverse events. Depression was present in 37.3% of patients (3 events in 3 subjects were considered to be devicerelated). Memory impairment was present in 27.3% patients (approximately a third of memory impairment was confirmed by neuropsychological examination). Suicidal ideation was reported by 11.8% (1 subject committed suicide; the suicide was not believed to be device-related). Seven patients died during the study – 1 probable sudden unexpected death in epilepsy (SUDEP), 2 definite SUDEP, and 1 possible SUDEP. 71 The explanation for psychiatric side effects is supported by the role of the ANT in the “Papez circuit,” which is crucial for emotional and cognitive control. In general, there are no reliable data regarding psychiatric side effects. Järvenpääet al. (2018) reported remarkable results when analyzing a group of 22 patients with ANT-DBS (Järvenpääet al., 2018). The authors did not find any changes in psychological tests at the whole group level. However, transient mood disorders were present in four patients – two patients reported sudden-onset depressive and melancholic symptoms, and the other two reported slowly-progressive psychiatric symptoms, including irritation, sleep disturbances, anxiety, fear, and paranoid symptoms. All previously mentioned symptoms disappeared after adjusting the stimulation current or changing the stimulation parameters. Novais et al. (2019) reported that ANT-DBS is associated with impaired psychological functions after surgery, as well as a bilateral epileptogenic zone and a lack of remission of disabling seizures (Novais et al., 2019). In our center, we have experience with the development of de novo psychopathology related to ANT-DBS (Doležalováet al., 2019; Annex 17). The patient is a female born in 1970 with a family history of mesial temporal lobe epilepsy. The epilepsy was characterized as focal bitemporal based on scalp EEG. There was a right-sided HS on MRI, but the Wada test showed the crucial role of the right-sided hippocampus for memory. The patient was implanted with VNS, which was inefficient and was explanted. The patient was offered ANT-DBS implantation at the age of 41 years. She was not treated for psychiatric illness, which was confirmed by neuropsychological examination. However, psychology tests revealed echolalia, perseveration, and the global deterioration of cognitive function (IQ 72, the most severe alteration in execution and memory). After implantation, the monopolar stimulation of the most proximal contacts (contact 3 on the left and contact 11 on the right, Figure 17) was initialized. After 72 implantation, the patient did not report the decrease of seizure frequency, but ANT-DBS influenced seizure severity and duration positively. The PSEs started to appear 3 months after stimulation initialization. The patient’s family started to complain about irritability, hostility, and aggressiveness, which were accompanied by paranoid production. The patient reported that ANT-DBS influenced her behavior and forced her to walk backward. The situation worsened when anyone was speaking about DBS. The psychological and psychiatric examinations confirmed incoherent thoughts, behavioral and emotional alteration, and disordered personality and behavior. The stimulation was discontinued, but the patient has not yet Figure 17: The relation between the ANT and DBS electrodes Panel a shows the relation between the ANT (violet) and electrodes in a coronal section. Panel b illustrates the estimated distribution of the electrical field (yellow) during stimulation and its links to the ANT (blue). We chose a correct contact on the right, but a more distal contact would be in a more suitable position on the left. L – left, R – right; SureTune software (Medtronic, Minneapolis, MN, USA). Republished with permission. 73 returned to her pre-implantation level, as supported by repeated psychiatric hospitalizations. In this patient, we suspect that the implantation of DBS electrodes caused some alteration in thalamic circuits that had long-term consequences. DBS – electrophysiological aspects From the electrophysiological point of view, it is remarkable to record the changes of scalp EEG in association with DBS or the intracerebral activity recorded directly from the ANT via intracerebral electrodes. Kerrigan et al. (2004) reported the presence of changes, so-called driving response, of scalp EEG conditioned by the stimulation of the ANT. This response was time and frequency locked to ANT stimulation (Kerrigan et al., 2004b). We recorded the cerebral signal from electrodes implanted in the ANT (Rektor et al., 2016, Annex 18). We reported the interictal and ictal recordings from the ANT in six patients. The recordings from ANT-DBS was obtained in a similar way as recordings from DBS in patients with Parkinson’s disease, i.e., DBS was implanted and the electrodes were left externalized for recording. After several days, the internalization was performed. Interictal recording No IEDs were found in the ANT. However, HFOs up to 240 Hz. were recorded in all patients. We even recorded HFOs up to 500 Hz in one patient. The interictal HFOs are shown in Figure 18. 74 Ictal recording We recorded ictally an early broadband increase of power in all seizures (Figure 19). Moreover, a specific ictal activity was present in one seizure; this ictal activity preceded the clinical onset in a given seizure (Figure 20). To our knowledge, this is the first described occurrence of HFOs in subcortical structures. Figure 18: The High-frequency oscillations (HFOs) recorded in nucleus anterior thalami (ANT) The figure demonstrates HFOs recorded in ANT, raw signal, sampling frequency 5kHz. Two types of HFOs are shown: (1) periodical HFOs – longer rectangle (lasting >20 ms, R1); (2) single peaks (burst) – shorten rectangle (lasting <10 ms, R2). Republished with permission. 75 Figure 19: Preictal recording in the ANT – preictal broadband increase in all power spectra A: Enlargement of raw signal from 30 ms to 22 ms before seizure onset B: Raw signal – time window 90 ms before and 50 ms after seizure onset C: Normalized Time Frequency Maps with preictal broadband increase in all power spectra. Yellow vertical line – clinical seizure onset Republished with permission. 76 Figure 20: The clinical seizure onset (vertical yellow line) was preceded by clear-cut rhythmic activity in the left ANT, contacts L2-L3 Upper panel: The rhythmic activity could be seen first in the left ANT (red arrow); it then spreads to the right ANT. Bottom panel: Normalized Time Frequency Maps. Assumed from Rektor et al., 2016; published with permission. 77 Responsive neurostimulation (RNS) RNS was approved for clinical use in the US in 2013. This method differs widely from VNS and DBS, which are based on an open-loop approach, i.e., they derive pulses on a fixed schedule. RNS derives stimulation only in the presence of ictal discharges, i.e., it works as a closed loop. RNS is surgically implanted in the cranium and is connected to the skull. The device contains leads (subdural strips or depth electrodes) for monitoring brain activity. These leads are implanted through the burr holes or craniotomy to the proximity of the presumed seizure onset zone. The unequivocal condition for implantation of RNS is a hypothesis about seizure onset zone localization. For successful resective surgery, it is necessary to have a precise delineation of the epileptogenic zone. The hypothesis for RNS can be weaker; it is sufficient to implant RNS somewhere in the epileptogenic network to obtain a sufficient clinical response (Geller, 2018). 78 Benefits of surgery Impact on seizure frequency With classical resective surgery, the best modality for measuring success is seizure freedom. The seizure freedom rates are the highest in TLE; approximately 70% of patients are seizure-free after surgery. There are differences between lesional and nonlesional TLE: in patients with lesional TLE, 75% of cases are seizure-free; the representation of seizure-free patients decreases to 51% in nonlesional cases. The numbers of seizure-free patients in ETLE are lower than in TLE; in this group, 46% of patients report complete seizure freedom after surgery. In these ETLE, the difference between lesional and nonlesional cases is high. The number of seizure-free patients is almost comparable to TLE in lesional cases (seizure freedom in 60% of lesional ETLE). However, the reported numbers in nonlesional ETLE are much lower; only one-third of patients experienced complete seizure cessation in this subgroup (Téllez-Zenteno et al., 2010). The use of any neurostimulation methods only rarely leads to seizure freedom. The efficacy of these methods can be measured by responder rates (patients experiencing≥ 50% seizure reduction). The responder rates are comparable among all neurostimulation methods; approximately 50% of patients are classified as responders after the first year of stimulation (a more detailed description is provided in the relevant chapters) (Ben-Menachem et al., 1994; Fisher et al., 2010; Handforth et al., 1998; Salanova et al., 2015). 79 Impact on the quality of life, employment status, and other variables Importantly, successful neurosurgery constitutes an effective therapy for epileptic seizures and may also reduce comorbidities (e.g., depression and cognitive deficits), improve quality of life (QOL), and increase social reinsertion, including employment; moreover, it reduces long-term morbidity and mortality (Picot et al., 2016). For the impact of epilepsy surgery on QOL, Seiam et al. (2011) identified 32 studies published between 1950 and 2008 focusing on this topic. Twenty-nine (90.6%) out of these 32 reports described a significant positive effect. The remaining three studies did not prove an association of surgery with the improvement of QOL. Seiam et al. (2011) identified some methodological insufficiencies in those studies, such as an inappropriate choice of a questionnaire or a small sample size. The determinants predicting the improvement of QOL were both preoperative and postoperative. Preoperative anxiety and depression or poor baseline psychological functions are associated with lower postoperative QOL. The increase of QOL can also be attributed to seizure control, a robust surgical placebo effect, or the reduction of adverse events related to AEDs in cases with daily dose reduction. The complete cessation of “disabling” seizures is crucial for the improvement of QOL. The patients with “only” seizure reduction reported five times less improvement of QOL than seizure-free patients. On the other hand, the persistence of auras did not condition worse QOL (Seiam et al., 2011). In our study, we tried to determine the impact of surgery on daily life, including QOL (Doležalová et al., 2016b; Annex 19). We designed a questionnaire with 13 items. The items were divided into four main topics: (1) demographic data, (2) information regarding surgery, (3) social issues (employment status and 80 driving license), (4) subjective evaluation of surgery benefits. The questionnaires were sent to 137 patients who underwent surgery in our center with at least one year of follow-up care. We included 91 respondents who correctly completed the questionnaire (56 men, 35 women). Of those, 59 (64.8%) were seizure-free after surgery. After surgery, there was a slight increase in the number of employed patients (46 [50.5%] patients were employed before surgery in comparison to 50 [54.9%] patients after surgery). This increase was dependent on gender, i.e., present only in the male subgroup. Nineteen (20.9%) patients obtained driving permission after surgery (p<0.001). Forty-nine (53.8%) patients reported an increase of QOL after surgery. The only variable that was related to the increment of QOL after surgery was employment status (the patients who were employed after surgery reported a higher QOL, p=0.037). To conclude our results, the surgery can bring a chance for seizure cessation and improvement of QOL, but in general, it has a low impact on employment status. This fact offers an opportunity for social interventions or occupational therapy with the goal of employment status improvement. 81 Future directions In our future research, we would like to extend our attention in two directions. The first direction is related to the development of the prediction paradigm for neurostimulation (VNS, DBS). The second direction concerns seizure detectors and proof of their clinical utility. The development of the prediction paradigm for neurostimulation Neurostimulation methods offer a high probability of substantial seizure reduction (Fisher et al., 2010; Harroud et al., 2012; Ma and Rao, 2018; McIntosh et al., 2004; Orosz et al., 2014; Ryvlin et al., 2014; Ryvlin and Kahane, 2005; Salanova et al., 2005). Despite large groups of implanted patients and many attempts, it is not possible to predict the effectiveness or benefits for an individual patient (Englot et al., 2011; Englot et al., 2016). This fact brings uncertainty into the patient decision-making and may decrease the credibility of neurostimulation as a treatment for drug-resistant epilepsy. We have developed a statistical classifier that can predict the benefit of VNS treatment with high accuracy (sensitivity and specificity) (Brázdil et al., 2019). Despite this success, we have to challenge the limitations of our previous work. The gaps in our previous work could be summarized as follows: (1) the age-limited validity; (2) the absence of confirmation in independent datasets; and (3) the absence of prospective study. We want to close these gaps using a grant already started in 2019 (Ministry of Health of the Czech Republic, grant 19-04-00343). First, we based the statistical model on data recorded only in adult patients. It means that this model could be applied only for predictions in adults, which contrasts with current praxis when one-third of all implanted patients are children. Moreover, children seem to be better candidates for VNS 82 therapy because they have higher responder rates than adults (Englot et al., 2011; Soleman et al., 2018). Based on these findings, we want to try to develop a similar prediction algorithm that could be applied even to children. We expect that this work will be more challenging because of the higher variability of children’s EEG conditioned by brain maturation (Benninger et al., 1984). Second, the absence of current model validation on an external independent patient dataset is presumably the most significant limitation of our study. We want to overcome this limitation through international collaboration with a foreign epilepsy center, where we plan to retrospectively recruit patients already treated with VNS. Third, we plan to design and conduct a prospective multicenter study for the prediction of VNS efficacy. We want to initiate a study called PRECISE (Prediction of vagus nerve stimulation EfficaCy In drug-reSistant Epilepsy, Figure 21). The principal aim of PRECISE is to verify the predictability of VNS efficacy by analyzing the pre-implantation routine EEG. Patients will be classified according to their predicted outcome (predicted responders versus predicted nonresponders). After the first and the second year of the study, the real-life outcomes (real-life responder versus real-life non-responder) will be determined. The predicted outcomes and real-life outcomes will be compared in terms of accuracy, specificity, and sensitivity. In the meantime, the patients will be managed according to the best clinical practice to obtain the best therapeutical response. Fourth, we would like to verify whether it is possible to extrapolate our experience with VNS to ANT-DBS. It seems feasible based on neuroanatomical, experimental, and clinical knowledge. A vagus nerve consists of 80% afferent fibers that transmit information from visceral organs to the nucleus tractus solitarius. There are dense connections between the nucleus tractus solitarius and 83 other higher centers, namely, the hypothalamus, dorsal raphe, nucleus ambigus, dorsal motor nucleus of the vagus nerve, amygdala, and thalamus (Rutecki, 1990). The influence of VNS on Figure 21: Design of prospective study PRECISE (Prediction of vagal nerve stimulation EfficaCy In drug-reSistant Epilepsy The figure represent patient flow within planned prospective study PRECISE higher brain structures was demonstrated by both PET and fMRI (Ko et al., 1996; Rutecki, 1990). PET studies showed increased perfusion in the inferior cerebellum, hypothalamus, and thalamus; the decreased perfusion was present in the hippocampus, amygdala, and posterior cingulate gyrus. In subsequent studies, the authors proved the increased bilateral thalamic cerebral blood flow, 84 which correlated with decreased seizure frequency. To conclude, it seems that thalamus and limbic structures are actively involved in VNS therapy, i.e., we can expect the involvement of similar circuits as during ANT-DBS (Ben-Menachem et al., 1994). These theories could be supported by a recently published clinical study that proved a similar pattern of response in VNS and DBS treatment (Kulju et al., 2018). Based on these statements, we can expect that a similar algorithm for efficacy prediction can be postulated even for ANT-DBS. The work with ANT-DBS will be influenced by relatively low numbers of implanted patients in each center, so international collaboration will be needed. Seizure detectors The next area of our future interest is real-life seizure detection. We want to center our attention on commercial detection systems and test them in ordinary patient environments, i.e., outside of the video EEG monitoring units. Our goal is to examine their utility and their benefits for patients and their families. The seizure detection systems could be divided into EEG-based detectors and non-EEG-based detectors (van Andel et al., 2016). EEG-based detectors EEG-based detectors are not routinely used, and at the moment there is no commercial EEG-based detector. This fact is conditioned mainly by technical limitations (van Andel et al., 2016). In the case of “classic” scalp EEG, the limitations are associated with the presence of electrodes on the patients’ heads; the electrodes are thought to be non-aesthetic and bothersome during long-term use. This restraint could be at least partially overcome by the reduction of electrode size, e.g., by ear electrodes (Looney et al., 2012), which seem to be suitable for recording epileptic activity 85 (Mikkelsen et al., 2015; Zibrandtsen et al., 2017). Earlier studies had to combat the disruption of EEG signals with artifacts, mainly from movements (Pauri et al., 1992); more recently developed algorithms for artifact rejection has improved the accuracy of these detections (Duun-Henriksen et al., 2012; Hopfengartner et al., 2014; Ramgopal et al., 2014). This progress increases the chance of developing commercial EEG-based detectors in the future. Non-EEG-based detectors Non-EEG-based detectors are currently used in daily praxis. These detectors analyze the physiological parameters that change at the beginning of seizures. They work as sensors analyzing movements (accelerometers, gyroscopes, magnetometers, electromyography [EMG], mattressbased sensors, or cameras), electrocardiography (ECG; changes of heart rate or heart-rate variability), and electrodermal activity. It seems that mainly the combination of individual sensors enables them to reach high accuracy (Osorio and Schachter, 2011). Figure 22: Bracelet-like detector Embrace 2 produced by Empatica The detectors, except mattress-based sensors and cameras, are bracelet-like or watch-like, so they are suitable for daily wear and are not intrusive (Figure 22). The mechanism of the detector function 86 is illustrated in Figure 23. The detector analyzes physiological parameters. If there is a relevant change indicating a seizure onset, the detector sends a signal to the patient’s mobile phone via wireless technology. The mobile phone then transmits a message to the patient’s family members or caregivers. The patient can personally stop the warning message at several levels to reduce the number of incorrect reports. Figure 23: The mechanism of EEG-based detector functions The figure illustrates the mechanism of functions of non-EEG detectors in several steps. The information about detector accuracy is valuable for both physicians and patients. The accuracy of detectors is described in terms of sensitivity (the percent representation of correctly detected seizures) and by false alarm/false detection rates (the number of false detections during a defined time interval, usually 8 or 24 hours). High detection accuracy is required; patients and their families do not want to be disturbed by false messages. In the most recent review of literature focusing on the accuracy of seizure detection, the authors identified five studies reporting data about the accuracy of seizure detectors (Leijten and Dutch TeleEpilepsy Consortium, 2018). Almost all the studies took place in the environment of a video 87 EEG monitoring unit. At the moment, there is very little data about seizure detector efficacy in a “normal” real-life environment. The best results were published by Onorati et al. (2017), who reported a 95% sensitivity and a false alarm rate of 0.07 per 8 hours (Onorati et al., 2017), which seems to be promising. Benefits of seizure detectors We believe that seizure detectors can benefit patients and everyone involved in their care, including caregivers, physicians, researchers, and pharmaceutical companies. Seizure detectors can decrease the risk of sudden unexpected death in epilepsy (SUDEP) or status epilepticus, which are rare but dreaded complications of epilepsy. The lifelong risk of SUDEP in a patient treated with drug-resistant epilepsy is around 10% (Thurman et al., 2014). SUDEP is most often sleep-related and unwitnessed. Nocturnal supervision is recommended in high risk patients (Ryvlin et al., 2013; Shorvon and Tomson, 2011). However, whole-night supervision means the loss of privacy for both patients and their families. Moreover, it can be exhausting. It is possible that seizure detectors represent an acceptable form of nocturnal supervision for patients at risk for SUDEP (van Andel et al., 2016). The risk of status epilepticus correlates negatively with intervention time-lag between seizure onset and interventions, i.e., patients who are quickly given rescue medication are at a lower risk of status epilepticus than patients whose treatment is delayed or not treated (Neligan and Shorvon, 2011). Seizure detectors can shorten this time lag, which could have a positive impact on the status of epilepticus incidence. We cannot expect the decline of injuries associated with seizure, because injuries are usually time-related to seizure onset. They happen immediately after a seizure starts and probably cannot be prevented (Nguyen and TéllezZenteno, 2009). 88 Seizures, status epilepticus, and SUDEP mean stress for the patients and their families and caregivers. This stress increases anxiety and has a negative influence on the quality of life. The detectors could potentially decrease the fear of ongoing seizures, and subsequently improve the quality of life (van Andel et al., 2009). Moreover, the quality of patient lives can be negatively influenced by the attitude of some professional care providers. Epilepsy is, in some individuals, associated with mental retardation, which implies the necessity of whole day professional care and institutionalization. Some social institutes refuse people with active epilepsy because of their fear of seizures, associated injuries, SUDEP, and legal consequences. At the moment, there are no standards for surveillance in patients in social and nursing institutions. We anticipate that the detectors will represent a tool or an alternative for surveillance in these institutionalized patients (van Andel et al., 2016). Finally, the detectors can provide precise information about seizure frequency, severity, and duration. This information could be helpful for physicians in tailoring medication or for pharmaceutical companies developing medication. It is currently necessary to rely on the information in patient seizure diaries, which can be misinterpreted (van Andel et al., 2016). We plan to initiate a prospective study in which patients will be equipped with seizure detectors for a defined period. At the beginning and the end of the study, patients will be given a questionnaire regarding seizure frequency, quality of life, and depression. The patient responses will be compared, and the impact of detectors on patient lives will be evaluated. 89 Conclusion Epileptology is an exciting branch of clinical neurology. It has been undergoing a revolution since the late 20th century. This revolution is conditioned by the shift from therapy with sodium bromide to the use of modern AEDs, MRI scanners, IEEG, MRI post-processing methods, and other techniques. I believe that we can still anticipate a significant paradigmatic shift towards the broader use of genetic therapy, individualized virtual brains for tailoring surgical procedures, optogenetics, and other methods that are currently limited to research laboratories. It may sound like science fiction today, but the future holds possibilities that will make our work even more remarkable and will improve the level of care for our patients. 90 Annexes: Annex 1: Kojan M, Gajdoš M, Říha P, Doležalová I, Řehák Z, Rektor I. Arterial spin labeling is a useful MRI method for presurgical evaluation in nonlesional epilepsy. (submitted) Annex 2: Dolezalova I, Brazdil M, Hermanova M, Janousova E, Kuba R. Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges and its relationship to surgical outcome in patients with hippocampal sclerosis. Epilepsy Research 2014;108(8):1406-1416. Annex 3: Frauscher B, von Ellenrieder N, Zelmann R, Dolezalova I, Minotti L, Olivier A, Hall J, Hoffmann D, Nguyen DK, Kahane P, Dubeau F, Gotman J. Atlas of the normal intracranial electroencephalogram: neurophysiological awake activity in different cortical areas. Brain. 2018;141:4. Annex 4: Dolezalova I, Brazdil M, Hermanova M, Horakova I, Rektor I, Kuba R. Intracranial seizure onset patterns in unilateral temporal lobe epilepsy and their relationship to other variables. Clinical Neurophysiology. 2013;124(6):1079-1088. Annex 5: Pail M, Rehulka P, Cimbalnik J, Dolezalova I, Chrastina J, Brazdil M. Frequencyindependent characteristics of high-frequency oscillations in epileptic and non-epileptic regions. Clinical Neurophysiology. 2017;128(1):106-114. Annex 6: Dolezalova I, Brazdil M, Rektro I, Tyrlikova I, Kuba R. Syncope with atypical trunk convulsions in a patient with malignant arrhythmia. Epileptic Disorders. 2013;15(2):171-174. Annex 7: Dolezalova I, Brazdil M, Kahane P. Temporal lobe epilepsy? Things are not always what they seem. Epileptic Disorders. 2017;19(1):59-66. 91 Annex 8: Dolezalova I, Schachter S, Chrastina J, Hemza J, Hermanova M, Rektor I, Pazourkova M, Brazdil M. Atypical handedness in mesial temporal lobe epilepsy. Epilepsy & Behavior. 2017;72:78-81. Annex 9: Kojan M, Dolezalova I, Koritakova E, Marecek R, Rehak Z, Hermanova M, Brazdil M, Rektor I. Predictive value of preoperative statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy with hippocampal sclerosis. Epilepsy & Behavior. 2018;79:46-52. Annex 10: Dolezalova I, Brazdil M, Chrastina J, Hemza J, Hermanova M, Janousova E, Pazourkova M, Kuba R. Differences between mesial and neocortical magnetic-resonance-imagingnegative temporal lobe epilepsy. Epilepsy & Behavior. 2016;62:21-26. Annex 11: Rehulka P, Dolezalova I, Janousova E, Tomasek M, Marusic P, Brazdil M, Kuba R. Ictal and postictal semiology in patients with bilateral temporal lobe epilepsy. Epilepsy & Behavior. 2014;41:40-46. Annex 12: Chrastina J, Novak Z, Zeman T, Kocvarova J, Pail M, Dolezalova I, Jarkovsky J, Brazdil M. Single-center long-term results of vagus nerve stimulation for epilepsy: A 10-17 year follow-up study. Seizure – European Journal of Epilepsy. 2018;59:41-47. Annex 13: Chrastina J, Kocvarova J, Novak Z, Dolezalova I, Svoboda M, Brazdil M. Older age and longer epilepsy duration do not predict worse seizure reduction outcome after vagus nerve stimulation. Journal of neurological surgery part A – Central European neurosurgery. 2018;79(2):152-158. 92 Annex 14: Chrastina J, Doležalová I, Novák Z, Pešlová E, Brázdil M. Pregnancy outcomes in refractory epilepsy with vagus nerve stimulation: long-term single center experience. (submitted) Annex 15: Brázdil M, Doležalová I, Koriťáková E, Chládek J, Roman R, Pail M, Jurák P, Shaw DJ, Chrastina J. EEG reactivity predicts individual efficacy of vagal nerve stimulation in intractable epileptics. Frontiers in Neurology. 2019;10:392-392. Annex 16: Plešinger F, Halámek J, Chládek J, Jurák P, Doležalová I, Chrastina J, Brázdil M. Response to vagal stimulation by heart-rate feature in drug-resistant epileptic patients. (submitted) Annex 17: Doležalová I, Kunst J, Kojan M, Chrastina J, Baláž M, Brázdil M. Anterior thalamic deep brain stimulation in epilepsy and persistent psychiatric side effects following discontinuation. Epilepsy Behavior Reports 2019;12:100344. Annex 18: Rektor I, Dolezalova I, Chrastina J, Jurak P, Halamek J, Balaz M, Brazdil M. Highfrequency oscillations in the human anterior nucleus of the thalamus. Brain Stimulation. 2016;9(4):629-631. Annex 19: Doležalova I., Pešlová E, Michnová M, Nečasová T, Kočvarová J, Musilová K, Rektor I. Brázdil M. Epileptochirurgická léčba zlepšuje kvalitu života – výsledky dotazníkové studie. Ceská a Slovenská Neurologie a Neurochirurgie. 2016;79/112(4):430-439. 93 Annex 1: Kojan M, Gajdoš M, Říha P, Doležalová I, Řehák Z, Rektor I. Arterial spin labeling is a useful MRI method for presurgical evaluation in nonlesional epilepsy. (submitted) 94 Brief communication Arterial spin labeling is a useful MRI method for presurgical evaluation in nonlesional epilepsy Martin Kojan 1,2 , Martin Gajdoš 2 , Pavel Říha 1,2 , Irena Doležalová 1 , Zdeněk Řehák 3 , Ivan Rektor 1,2 1 Brno Epilepsy Center, Department of Neurology, St. Anne’s University Hospital and Medical Faculty of Masaryk University, Brno, Czech Republic 2 CEITEC – Central European Institute of Technology, Neuroscience Center, Masaryk University, Brno, Czech Republic 3 Department of Nuclear Medicine, Masaryk Memorial Cancer Institute, Brno, Czech Republic Word count: 1508 Number of text pages: 7 Abstract: 270 References: 15 Tables: 1 Supplementary Figures: 8 Study funding: Supported by Ministry of Health of the Czech Republic, grant No. 16- 33395AZV. All rights reserved.; MH CZ – DRO (MMCI, 00209805), LO 1413. We acknowledge the core facility CF MAFIL supported by the Czech-BioImaging large RI project (LM2015062 funded by MEYS CR) for their support with obtaining scientific data presented in this paper. Corresponding author: Ivan Rektor, Prof., M.D., CSc, FANA, FEAN, Masaryk University, CEITEC and Department of Neurology, St. Anne’s University Hospital, Pekarska 53, 656 91 Brno, Czech Republic, irektor@med.muni.cz, +420 603 864 046 Running title: ASL in epilepsy surgery Key-words: PET – ASL – epilepsy surgery 95 Objective: This study aimed to identify the usefulness of arterial spin labeling (ASL) for presurgical evaluation in pharmacoresistant nonlesional epilepsy as compared with statistical parametric mapping of interictal [18 F] fluorodeoxyglucose PET (FDG-PET). Methods: ASL was compared with FDG-PET in eight epilepsy surgery candidates (4 female, 4 male; aged 22 to 53 years, median 25 years; 7 MRI-negative epilepsy and one suspected lesion) who underwent a complete presurgical evaluation followed by resective surgery. We compared the results of presurgical FDG-PET and ASL scans with their resection masks. For this purpose, we used asymmetry index (AI) maps for both modalities. From AI map values, we calculated the 95th percentile and created clusters from the whole brain which were separated into two groups – clusters with intersection with resection mask which were considered as true positive (TP), and clusters outside the resection mask marked as false positive (FP). From the number of TP and FP, we calculated the positive predictive value (PPV). PPV results from both the PET and ASL methods were compared using the nonparametric two-sided Wilcoxon signed rank test. Results: The comparison of the PPV values in each subject between ASL and PET showed better PPV in 5 out of 8 patients using the ASL method. In 2 out of 8 patients, a better PPV value was obtained by PET. In one patient, both methods showed the same value. According to the results of the Wilcoxon signed rank test p=0.89, we cannot reject the null hypothesis and assume that these methods have the same predictive power. Conclusion: We conclude that ASL is a useful method for presurgical evaluation also in nonlesional epilepsy. 96 Key-words: PET – ASL – epilepsy surgery Abbreviations ASL – arterial spin labeling; PET – positron emission tomography, SPM – statistical parametric mapping; CBF – cerebral blood flow; TLE – temporal lobe epilepsy; ROI – region of interest 97 Introduction This aim of this study is to explore the clinical utility of the arterial spin labeling (ASL) MRI technique. ASL is a generally accessible non-invasive technique measuring brain perfusion using magnetically labeled blood as a tracer. To explore its clinical utility for presurgical exploration in pharmacoresistant epilepsy, ASL was compared with [18 F]fluorodeoxyglucose (FDG) PET in patients in whom the location of the seizure onset zone was confirmed by the subsequent resective surgery. FDG-PET is a well-established method in the presurgical evaluation of patients with pharmacoresistant epilepsy1 . FDG-PET is used in combination with other imaging, electrophysiological, and clinical methods. It is of particular interest in patients with MR negative (nonlesional) focal epilepsy2 . It can lead to surgery or can help in targeting intracranial EEG3 . We chose to compare ASL with PET because although PET is frequently used in presurgical evaluations, there are several limitations to this method that are not present when using ASL. The main drawbacks of FDG-PET are radiation exposure for patients, high costs, and limited availability in some centers4 . Methods Voxel-based approaches such as statistical parametric mapping (SPM) and the asymmetry index (AI)5 used as a control/healthy group comparison are methods of quantitative analysis and do not depend on the selection and size of the ROI. The SPM method assesses the null hypothesis at each voxel with univariate statistics and constructs an image from its results. SPM does not require a priori hypotheses about the location and extent of effects. We conducted this study in order to compare perfusion measured by MR scanner with interictal FDG-PET using SPM analysis. For this purpose, we used the pseudo-continuous arterial spin labeling (pCASL) technique. The perfusion-weighted image (PWI) is computed 98 from differences between magnetically labelled images and control images acquired without labeling of the blood. In addition to the labeled and control images, there is usually a calibration scan (M0), which is used for quantification of the perfusion in order to obtain a cerebral blood flow (CBF) map. Patient population Eight epilepsy surgery candidates (4 female, 4 male; aged 22 to 53 years, median 25 years; 7 MRI-negative epilepsy and one suspected lesion) who underwent a complete presurgical evaluation followed by resective surgery at the Brno Epilepsy Center were examined. Anatomical and PCASL MR scans and FDG-PET scans were performed. An anatomical MR scan performed after the surgery identified the location and extension of the resection. This was used to create the mask of the resection. Patient data, including epilepsy zone location and histopathology, are shown in Table 1. FDG-PET image acquisition and preprocessing Patients were imaged as outpatients in the interictal state. Subjects prepared by fasting for six hours before the scan and resting in a quiet, darkened room for 50-60 minutes after FDG administration. EEG monitoring was not performed during the scan and patients were asked to report any seizures experienced on the day of the scan. The dose of FDG administered was 170-200 MBq per subject with no weight differentiation. The PET images were acquired using a Siemens mCT Flow PET/CT scanner in 3-D mode using the “Brain” protocol. Spatial pre-processing was performed using SPM12 (Wellcome Department of Cognitive Neurology, Institute of Neurology, University College London, UK) and MATLAB version 2011b (MathWorks Inc., Natick, MA, USA). All FDG-PET images were spatially normalized into the standardized stereotactic Montreal Neurological Institute space using our in-house 99 FDG-PET template described in our previous study6 . A three-dimensional isotropic Gaussian kernel with 8 mm FWHM was used for smoothing all spatially normalized images. MRI data acquisition and preprocessing We used the 3.0 T Siemens Prisma MR machine (MAGNETOM Prisma, Siemens Healthcare, Erlangen, Germany) for the image acquisition. The patients had no seizures within the 24 h period prior to the examination, and no seizures were observed during the scanning. We acquired 41 pairs of control/label images with pCASL using single band EPI readout sequence7 . We acquired M0 scans for CBF quantification purposes. Preprocessing of pCASL data consisted of motion correction, registration to MNI space, and quantification of CBF maps based on the approach published by Alsop8 . Comparison of metabolic/perfusion abnormalities using the asymmetry index The normalization of PET and ASL datasets into a common space enabled a voxelwise comparison between modalities. In order to identify left/right asymmetries in metabolism and perfusion, we calculated the AI for both modalities following 9 AI = 100*(Right - Left) / (Right + Left) Statistical comparison of the methods and visualization The patient group consisted of successful surgery patients with positive histological finding in the resected area. From AI map values, we calculated the 95th percentile and used it as a threshold value. Subsequently we chose only clusters with volumes larger than 2cm3 . These clusters from the whole brain were separated into two groups – clusters with intersection with resection mask which were considered as true positive (TP), and clusters outside the resection 100 mask marked as false positive (FP). From the number of TP and FP, we calculated the positive predictive value (PPV) using following formula: 𝑃𝑃𝑉 = 𝑇𝑃 𝑇𝑃+𝐹𝑃 ; Finally, PPV results from both the PET and ASL methods were compared using the nonparametric two-sided Wilcoxon signed rank test. Results In terms of sensitivity to metabolic and perfusion changes, both methods showed lower metabolism and lower perfusion in all 8 patients on the FDG and ASL AI maps. Numbers of significant clusters from thresholded AI maps with volume larger than 2cm3 for each method and each patient with individual values of PPV and FDR are shown in Table 2. Overall numbers of PPV are low because the calculation and evaluation were performed on the whole brain for comparison purposes. The direct comparison of the PPV values in each subject between ASL and PET showed better PPV in 5 out of 8 patients using the ASL method. In 2 out of 8 patients, a better PPV value was obtained by PET. In patient number 3, both methods showed the same value. According to the results of the Wilcoxon signed rank test p=0.89, we cannot reject the null hypothesis and assume that these methods have the same predictive power. A visualization of significant clusters is shown on the supplementary images 1a-h. Discussion In this study we compared regional distributions of FDG-PET and ASL in patients with drug resistant epilepsy. Several studies reported ASL as a method that could be useful for localizing the epileptogenic zone9–11 . This study is the first in which the localization is confirmed by the resective surgery. ASL is a non-invasive method with only one disadvantage as compared to FDG-PET, which is lower SNR. The results show decreased AI values of 101 glucose metabolism in FDG-PET and perfusion in ASL in resected areas. It appears that the sensitivity of both methods is high. It was possible to find a TP cluster in every patient in at least one method. In one patient, the ASL method did not find any TP cluster. The specificity is lower with similar numbers of FP results as the two methods display also regions of the spread of the interictal activity12 . The two methods show two different processes – the CBF in ASL and glucose metabolism in PET – and it is not expected that the distributions will overlap exactly12 . In concordance with recent studies that reported the usefulness of ASL in both TLE13 and extratemporal focal epilepsy10,11,14 , we proved that ASL is also effective in localizing the SOZ in nonlesional epilepsy. Recent studies comparing simultaneous PET and ASL acquisitions on hybrid PET/MR scanners10,11 where the acquisitions occur under the same conditions reported high concordance (r = 0.49, p < 0.0005) between them; our results are in agreement with these findings. One recent study13 employed the SEEG technique to assess the effectiveness of ASL in lateralizing the SOZ in MTLE with significant correlation between the AI values from PET and ASL (r = 0.72, p < 0.05). The authors analyzed only one region; the hippocampus was the only ROI in this study. The FDG-PET method for measuring brain metabolism does not have the wide clinical application that one might expect, partly because of its high cost and the complexity of the quantification procedure. The main benefits of ASL over PET are that it avoids radiation exposure for patients, and it offers lower costs, higher availability, and better time efficiency15 . Due to shorter acquisition time, it should be possible to include it in the clinical MRI protocol. The limitation of our 102 study is the small sample size of patients with epilepsy after resection surgery with favorable outcome. Conclusion We conclude that ASL is a useful method for presurgical evaluation also in nonlesional epilepsy. Larger studies are needed to establish whether it could be used as a complementary procedure to FDG-PET or could replace FDG-PET in the future. Statement: The work described in this article is consistent with the Journal’s guidelines for ethical publication Disclosure of Conflicts of Interest: None of the authors has any conflict of interest to disclose. The authors: 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.” 103 Table 1 Patients – clinical description ASL PET Surgery Histology Sex Age TP FP PPV TP FP PPV 1 R FLE FCD IIB M 25 1 5 0.17 1 10 0.09 2 LTLE FCD I F 22 1 4 0.20 1 9 0.10 3 R TLE FCD IIA M 25 2 10 0.17 1 5 0.17 4 L TLE HS I F 24 1 8 0.11 1 3 0.25 5 L TLE HS I M 53 2 8 0.20 1 10 0.09 6 R TLE Gliosis M 33 1 3 0.25 1 5 0.17 7 R parietal Gliosis F 44 1 7 0.13 1 10 0.09 8 L TLE HS I F 24 0 7 0.00 1 3 0.25 Table 1: Patient population with both ASL and PET results. Description of the calculation of TP (true positive), FP (false positive), and PPV (positive predictive value) parameters is in the methodology section. 104 1. Van Bogaert P, Massager N, Tugendhaft P, et al. Statistical Parametric Mapping of Regional Glucose Metabolism in Mesial Temporal Lobe Epilepsy. NeuroImage. 2000; 12(2):129–38. 2. Kuba R, Tyrlíková I, Chrastina J, et al. “MRI-negative PET-positive” temporal lobe epilepsy: Invasive EEG findings, histopathology, and postoperative outcomes. Epilepsy & Behavior. 2011; 22(3):537–41. 3. Rathore C, Dickson JC, Teotónio R, et al. The utility of 18F-fluorodeoxyglucose PET (FDG PET) in epilepsy surgery. Epilepsy Research. 2014; 108(8):1306–14. 4. Kashyap R, Dondi M, Paez D, et al. Hybrid Imaging Worldwide—Challenges and Opportunities for the Developing World: A Report of a Technical Meeting Organized by IAEA. Seminars in Nuclear Medicine. 2013; 43(3):208–23. 5. Signorini M, Paulesu E, Friston K, et al. Rapid Assessment of Regional Cerebral Metabolic Abnormalities in Single Subjects with Quantitative and Nonquantitative [18F]FDG PET: A Clinical Validation of Statistical Parametric Mapping. NeuroImage. 1999; 9(1):63–80. 6. Kojan M, Doležalová I, Koriťáková E, et al. Predictive value of preoperative statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy with hippocampal sclerosis. Epilepsy & Behavior. 2018; 79:46–52. 7. Li X, Wang D, Auerbach EJ, et al. Theoretical and experimental evaluation of multi-band EPI for high-resolution whole brain pCASL Imaging. NeuroImage. 2015; 106:170–81. 8. Alsop DC, Detre JA, Golay X, et al. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: A consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magnetic Resonance in Medicine. 2015; 73(1):102–16. 9. Lim Y-M, Cho Y-W, Shamim S, et al. Usefulness of pulsed arterial spin labeling MR imaging in mesial temporal lobe epilepsy. Epilepsy Research. 2008; 82(2):183–9. 10. Boscolo Galazzo I, Mattoli MV, Pizzini FB, et al. Cerebral metabolism and perfusion in MRnegative individuals with refractory focal epilepsy assessed by simultaneous acquisition of 18FFDG PET and arterial spin labeling. NeuroImage: Clinical. 2016; 11:648–57. 11. Galazzo IB, Storti SF, Felice AD, et al. Patient-Specific Detection of Cerebral Blood Flow Alterations as Assessed by Arterial Spin Labeling in Drug-Resistant Epileptic Patients. PLOS ONE. 2015; 10(5):e0123975. 12. Fink GR, Pawlik G, Stefan H, et al. Temporal lobe epilepsy: Evidence for interictal uncoupling of blood flow and glucose metabolism in temporomesial structures. Journal of the Neurological Sciences. 1996; 137(1):28–34. 13. Wang Y-H, An Y, Fan X-T, et al. Comparison between simultaneously acquired arterial spin labeling and 18F-FDG PET in mesial temporal lobe epilepsy assisted by a PET/MR system and SEEG. NeuroImage: Clinical. 2018; 19:824–30. 14. Kim BS, Lee S-T, Yun TJ, et al. Capability of arterial spin labeling MR imaging in localizing seizure focus in clinical seizure activity. European Journal of Radiology. 2016; 85(7):1295–303. 105 15. Wolf RL, Detre JA. Clinical Neuroimaging Using Arterial Spin-Labeled Perfusion MRI. Neurotherapeutics. 2007; 4(3):346–59. 106 Annex 2: Dolezalova I, Brazdil M, Hermanova M, Janousova E, Kuba R. Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges and its relationship to surgical outcome in patients with hippocampal sclerosis. Epilepsy Research 2014;108(8):1406-1416. 107 Epilepsy Research (2014) 108, 1406—1416 journal homepage: www.elsevier.com/locate/epilepsyres Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges and its relationship to surgical outcome in patients with hippocampal sclerosis Irena Doleˇzalováa,b,∗ , Milan Brázdila,b , Markéta Hermanovác , Eva Janouˇsovád , Robert Kubaa,b a Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic b Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic c First Department of Pathological Anatomy, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic d Institute of Biostatistics and Analyses, Masaryk University, Brno, Czech Republic Received 9 November 2013; received in revised form 20 May 2014; accepted 13 June 2014 Available online 5 July 2014 KEYWORDS Epilepsy; Hippocampal sclerosis; Outcome; Interictal epileptiform discharges (IEDs); Unitemporal IEDs; Bitemporal IEDs Summary Objective: To assess changes in the relative lateralization of interictal epileptiform discharges (IEDs) and interictal EEG prognostic value in terms of surgical outcome between periods with full medication (FMP) and reduced medication (RMP) in patients with temporal lobe epilepsy (TLE) associated with hippocampal sclerosis (HS). Methods: Interictal scalp EEGs of 43 patients were evaluated for the presence of IEDs separately in a waking state (WS) and sleeping state (SS) during FMP and RMP. In each period, patients were categorized as having unitemporal or bitemporal IEDs. Surgical outcome was classified at year 1 after surgery and at last follow-up visit as Engel I or Engel II—IV; and alternatively as completely seizure-free or not seizure-free. Results: There were significant changes in relative IED lateralization between FMP and RMP during SS. The representation of patients with unitemporal IEDs declined from 37 (86%) in FMP during SS to 25 (58%) in RMP during SS (p = 0.003). At year 1 after surgery, the relative IED lateralization is a predictive factor for surgical outcome defined as Engel I vs. Engel II—IV in ∗ Corresponding author at: Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Pekaˇrská 53, 656 91 Brno, Czech Republic. Tel.: +420 543 182 645; fax: +420 543 182 624. E-mail address: irena.dolezalova@fnusa.cz (I. Doleˇzalová). http://dx.doi.org/10.1016/j.eplepsyres.2014.06.009 0920-1211/© 2014 Elsevier B.V. All rights reserved. 108 Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges 1407 both FMP during WS (p = 0.037) and during SS (p = 0.007), and for surgical outcome defined as completely seizure-free vs. not seizure-free in FMP during SS (p = 0.042). At last follow up visit, the relative IED lateralization is a predictor for outcome defined as Engel I vs. Engel II—IV in FMP during SS (p = 0.020), and for outcome defined as completely seizure-free vs. not seizure-free in both FMP during WS (p = 0.043) and in FMP during SS (p = 0.015). When stepwise logistic regression analysis was applied, only FMP during SS was found to be an independent predictor for surgical outcome at year 1 after surgery (completely seizure-free vs. not seizure-free p = 0.032, Engel I vs. Engel II—IV p = 0.006) and at last follow-up visit (completely seizure-free vs. not seizurefree p = 0.024, Engel I vs. Engel II—IV p = 0.017). Gender was found to be independent predictor for surgical efficacy at year 1 if the outcome was defined as completely seizure-free vs. not seizure-free (p = 0.036). Conclusion: The predictive value of relative IED lateralization with respect to surgical outcome in interictal EEG is present only during FMP; the predictive value decreases with the reduction of AEDs caused by the change of relative IED lateralization. © 2014 Elsevier B.V. All rights reserved. Introduction Temporal lobe epilepsy (TLE) is the most common form of drug-resistant epilepsy (Wiebe et al., 2001). Surgery was proved to be an effective therapeutic method in cases where pharmacological treatment failed (Williamson, 1998). In approximately two-thirds of patients, surgery leads to seizure freedom (Engel, 1996; Janszky et al., 2005a). Many investigators focused on the evaluation of interictal EEG and its significance in predicting post-surgical outcome. In general, in patients with TLE, the relative lateralization of interictal epileptiform discharges (IEDs), i.e. the lateralization ratio of interictal spikes between the two hemispheres (unitemporal or bitemporal), appears to have prognostic value. Unitemporal IEDs are thought to be a favorable prognostic factor, while bitemporal IEDs are associated with worse outcomes (Chung et al., 1991; Radhakrishnan et al., 1998; Villanueva et al., 2004). This division fails in the subgroup of patients with mesial TLE associated with unilateral HS proved by histopathological examination or unilateral hippocampal atrophy on magnetic resonance imaging (MRI). Three studies focused on this subgroup proved no differences in surgical outcome between patients with unitemporal and bitemporal IEDs (Hardy et al., 2003; Janszky et al., 2005b; Krendl et al., 2008). Schulz et al. (2000) proved significant differences in outcome between these two groups, but their study included both mesial TLE associated with HS and non-lesional cases. Aull-Watschinger et al. (2008) found the relative IED lateralization to be a predictive factor for short-term (at year 1 and 2 after surgery) outcomes where an excellent outcome was defined as no seizures, with or without nondisabling auras, during the 12 month-period prior to the assessment, but there were no statistically significant differences between patients with unitemporal and bitemporal IED if the excellent outcome was defined as complete seizure freedom at any time after surgery. Antiepileptic drug (AED) reduction influences neither the frequency of IEDs nor the localization of seizure origin, but the influence of AED reduction on relative IED lateralization has not yet been systematically studied (Gotman and Marciani, 1985; Gotman and Koffler, 1989; Marciani and Gotman, 1986; So and Gotman, 1990; Spencer et al., 1981). The main goal of our study was to determine if there is a change in relative IED lateralization and/or interictal EEG prognostic value related to the surgical outcome caused by AED reduction and the effect of sleep in patients with mesial TLE associated with unilateral HS. Methods Patient selection This retrospective study included all the patients with unilateral HS who had surgery at the Brno Epilepsy Center of St. Anne’s University Hospital between 2005 and 2011 and who reached at least 1 year follow-up after the surgery. All included patients had unilateral HS on MRI, the presence of which was subsequently proven by histopathological evaluation. We excluded patients with other possible epileptogenic lesion than HS present on MRI, except blurring of border between gray and white matter in temporal pole and subsequently detected focal cortical dysplasia (FCD) in histopathological specimen of resected tissue of temporal pole. These criteria were applied because it seems that FCD in the temporal pole does not influence surgical outcome and because standard anteromedial temporal lobe resection was performed in all patients (Kuba et al., 2012) We analyzed interictal EEG during semi-invasive videoEEG monitoring (scalp EEG + sphenoidal electrodes) in both waking state (WS) and sleeping state (SS). The SS analysis was performed in the non-rapid eye movement (non-REM) sleep stage II-III in two periods. These two periods were defined by AED daily dosage, as a full medication period (FMP) and a reduced medication period (RMP). Only patients with both WS and SS recordings during FMP as well as during RMP were included in this study (see below—–EEG criteria) (Fig. 1). Patient demographic data (age at seizure onset, epilepsy duration, age at time of surgery, time after surgery), other characteristics [number of habitual seizures per month, occurrence of habitual generalized tonic—clonic seizures (GTCS) in two years preceding surgery], and data describing semi-invasive EEG monitoring (duration and number of seizures) were assessed by reviewing patient charts. 109 1408 I. Doleˇzalová et al. Fig. 1 The schema of interictal EEG evaluation during FMP and RMP. (FMP—–full medication period, RMP—–reduced medication period.) We included a total of 43 patients who fulfilled these criteria. All patients gave their informed consent. The study was approved by the ethics committee of St. Anne’s University Hospital. Presurgical evaluation All 43 patients underwent a comprehensive evaluation, including detailed history and neurological examination, neuropsychological testing, MRI, and semi-invasive video-EEG monitoring with sphenoidal electrodes. Fluorodeoxyglucose-positron emission tomography (FDGPET) was performed in all patients; other functional neuroimaging techniques such as ictal SPECT and subtraction ictal SPECT co-registered to MRI (SISCOM) were performed in 15 patients out of 43 (35%) patients. Invasive EEG was additionally performed in 4 out of 43 (9%) patients, because their non-invasive data were found to be insufficiently conclusive to proceed directly to surgery. EEG analysis All patients were investigated using semi-invasive videoEEG monitoring. Semi-invasive EEG was recorded with the international 10—20 system with anterotemporal (T1, T2), supraorbital (SO1, SO2), and sphenoidal electrodes (Sp1, Sp2). The EEG recording was performed on the 64-channel Alien Deymed system. Monopolar recordings (a reference electrode on the mastoid process) and special bipolar montages were used to evaluate the EEG activity. EEG was amplified with a bandwidth of 0.4—70 Hz at a sampling rate of 128 Hz. The EEG was analyzed independently for the presence of IEDs by two authors of this study (ID and RK) both authors were blinded and discrepancies were resolved by consensus. IEDs were considered to be spikes, sharp waves, and spike-wave complexes. A spike or sharp wave had duration of <200 ms, was distinguished by its morphology and/or amplitude from normal background activity, and was usually followed by a slow wave. Both FMP and RMP were characterized by the analysis of a 20-min WS and a 20-min SS interictal EEG recording sample (Fig. 1). These EEG samples were chosen from our archive, where interictal EEG specimens blindly selected by our technical staff from 24-h recordings are stored longterm (we do not store whole 24-h samples), on the basis of several criteria. First, interictal EEG analysis in FMP was always made without any antiepileptic drug (AEDs) reduction. Second, interictal EEG analysis in RMP was always made after at least 48 h after AED reduction initialization. The AED reduction velocity was individualized to the patient (depending on the frequency of habitual seizures, presence/absence of GTCS, etc.) on the basis of physician decision. We excluded the patients, in whom the AED reduction was not made. Third, the SS analysis was performed during non-REM sleep stages II—III; the minimal representation of stage III as the most important inductor of IED was 30% in a single SS recording (Malow et al., 1997; Sammaritano et al., 1991). The sleep stages were scored using the American Academy of Sleep Medicine Manual (Iber et al., 2007). Fourth, we did not include the EEG trace if (1) the recording was made 5 min before the seizure and/or 60 min after the seizure, (2) analysis was prevented by artifact, or (3) the recording duration was shorter than 20 min. The 5-min pre-seizure limit was chosen because no changes in spiking prior to seizures were found in previously published studies (Gotman and Marciani, 1985; Gotman and Koffler, 1989; Katz et al., 1992). The 60-min post-seizure limit was selected as sufficient for practical purposes, even though the changes following seizures can persist for hours and even days (Gotman and Marciani, 1985; Gotman and Koffler, 1989). Fifth, if more than one interictal EEG sample was available for analysis, we always chose the first one which fulfilled previously described criteria. We described the relative lateralization of IEDs between the temporal lobe with HS (HS side) and the contralateral temporal lobe (non-HS side) by laterality index (LI). LI was defined as: LI = IEDS on HS side − IEDS on non HS side IEDS on HS side + IEDS on non HS side LI can range from 1 to −1. If the IEDs are all localized on the HS side, LI equals 1. If the IEDs are equally distributed between the HS side and non-HS side, LI equals 0. If the IEDs are all localized on the non-HS side, LI equals −1. According to LI value, we divided our patients into two groups: - unitemporal IEDs (LI ≥ 0.8)—–the group of patients with strictly or predominantly unitemporal IEDs, i.e. more than 90% of IEDs localized on HS side - bitemporal IEDs (LI < 0.8)—–the group of patients with independent bitemporal IEDs, i.e. less than 90% of IEDs localized on HS side. The 90% cut-off value was based on the results of Chung et al. (1991), who proved that the patients with more than 90% of IEDs lateralized to the operated temporal lobe have comparable surgical results to the patients with strictly unitemporal IEDs. Surgical procedure and outcome Anteromedial temporal lobe resection was performed on all patients. Surgical efficacy was evaluated at regularly scheduled visits usually every 3 months, only in minority 110 Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges 1409 Table 1 Patient demographic data and characteristics. Demographic data Age at epilepsy onset in years (mean ± SD, ranges from-to) 12.6 ± 10.6 0.5—42 Duration of epilepsy in years (mean ± SD, ranges from-to) 26.3 ± 12.5 4—44 Age at time of surgery in years (mean ± SD, ranges from-to) 38.9 ± 9.4 19—56 Patients’ characteristics The number of habitual seizures in a patient per month (median, ranges from-to) 5 1—30 The number of seizures in a patient during semi-invasive EEG (median, ranges from-to) 4 1—10 FCD in the temporal pole in histopathological specimen in number (%) of patients 7 (16) FCD—–focal cortical dysplasia. of patients the efficacy was reported by their neurologist in their place residence, and it was categorized according to Engel classification (Engel, 1987) and simultaneously according to the classification proposed by the International League Against Epilepsy (ILAE) (Wieser et al., 2001). Patients were categorized as Engel I (no seizures, with or without nondisabling auras, during the 12-month period prior to the assessment) or as Engel II—IV (Engel, 1987). They were then categorized as completely seizure-free (i.e. complete seizure freedom and absence of nondisabling auras at any time after surgery; correlates class Ia according to ILAE classification) or not seizure-free. For further analysis, two time milestones — year 1 after surgery, and last follow-up visit — was chosen because of the change of surgical responsiveness over time (Janszky et al., 2005b; Salanova et al., 1996). Neuropathological examination Evaluable formalin-fixed paraffin-embedded tissues of anterior temporal resection specimens including the hippocampal complex were available from all patients. The paraffin-embedded tissue specimens, slides, and histopathology reports were retrieved from the files of the First Department of Pathological Anatomy of St. Anne’s University Hospital and re-evaluated by a histopathologist experienced in the evaluation of specimens obtained from epilepsy surgery patients (MH). All examined resected tissues were identically treated, fixed in 10% neutral buffered formalin, grossly inspected, carefully oriented, and measured. Temporal pole resection specimens were cut so as to obtain 2—3 mm thick tissue slices perpendicular to the cortical surface. Hippocampal en block resections were dissected into 2—3 mm thick tissue slices along the anteriorposterior axis. Representative tissue samples were routinely processed and paraffin embedded. Five ␮m thick tissue sections were stained by hematoxylin and eosin, evaluated under light microscope, and reported. NeuN immunohistochemistry (using mouse monoclonal anti-NeuN antibody, dilution 1:100, clone A-60, Millipore), was performed on preselected tissue sections if there was an inconclusive picture in hematoxylin and eosin. The classification system for HS reported by Blümcke et al. (2007) was applied. If FCD was present in the temporo-polar region, the classification system reported by Blümcke et al. (2011) was used. Statistics McNemar’s test was used to test the LI change between FMP and RMP during both WS and SS. Fisher’s exact test was applied to analyze the relation between LI and surgical outcome. The Mann—Whitney or Fisher’s exact tests were used to analyze the relation between LI and patient demographic data and characteristics or data describing semi-invasive EEG monitoring. A univariate logistic regression analysis was applied to assess the influence of other variables (age at time of surgery, age at epilepsy onset, duration of epilepsy, time from surgery, sex, side of surgery, presence of FCD, frequency of habitual seizures, presence of habitual GTCS) on surgical outcome. Moreover, a stepwise logistic regression analysis using a forward selection procedure was performed to assess the independent influence of relative IED lateralization in FMP and RMP during WS and SS, and of patients’ demographic data and characteristics on surgical outcome. The odds ratios were calculated and the Wald statistics applied to test whether the regression coefficients were statistically significantly different from zero. In all statistical tests, p-value <0.05 was considered to be statistically significant. Results Demographic data, surgery and outcome The study included 43 patients: 21 females and 22 males. Patient demographic data and characteristics are summarized in Table 1. Surgery was performed on the right side in 21 out of 43 (49%) patients and on the left side in 22 out of 43 (51%) patients. The duration of follow-up ranged from 1 to 8 years (with an average 4.5 ± 2.0 years). At year 1 after surgery, 35 out of 43 (81%) patients were classified as Engel I and 8 out of 43 (19%) as Engel II—IV (2 as Engel II, 2 as Engel III, and 4 as Engel IV). At this time milestone, 31 out of 43 (72%) were classified as completely seizure-free and 12 (28%) as not seizure-free. At the last follow-up visit using Engel’s classification, 33 out of 43 (77%) patients were classified as Engel I and 10 out of 43 (23%) patients as Engel II—IV (3 as Engel II, 4 as Engel III, and 4 as Engel IV). At the last follow-up visit, 28 out of 43 (65%) patients were classified 111 1410 I. Doleˇzalová et al. Table 2 The relative IED distribution between HS side and non-HS side and its relation to surgical outcome at year 1 after surgery classified as Engel I or Engel II—IV during FMP and RMP for different LI values. FMP RMP WS SS WS SS uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs Engel I n (%) 33 (87) 2 (40) 33 (89) 2 (33) 27 (84) 8 (73) 21 (84) 14 (78) Engel II—IV n (%) 5 (13) 3 (60) 4 (11) 4 (67) 5 (16) 3 (27) 4 (16) 4 (22) Total number 38 5 37 6 32 11 25 18 p-Value p = 0.037 p = 0.007 p = 0.401 p = 0.701 bi IEDs—–bitemporal interictal epileptiform discharges, FMP—–full medication period, uni IEDs—–unitemporal interictal epileptiform discharges, RMP—–reduced medication period, SS—–sleeping state, WS—–waking state. as completely seizure-free and 15 out of 43 (35%) patients as not seizure-free. AEDs treatment At the time of semi-invasive EEG monitoring, monotherapy was used in 3 out of 43 (7%) patients; the other 40 out of 43 (93%) patients were treated with combinations of AEDs: 30 (70%) patients with a combination of two AEDs and 10 (23%) with a combination of three AEDs. Interictal EEG during FMP was analyzed in all patients before the occurrence of seizures; 12 out of 43 (28%) patients exhibited seizures before EEG analysis in RMP. The other 31 out of 43 (82%) patients had seizures after the analysis of interictal EEG in RMP. The relative IED lateralization between HS side and non-HS side During FMP, 38 out of 43 (88%) patients had unitemporal IEDs in WS, the number of patients with unitemporal IEDs decreased to 37 (86%) in SS. During RMP, 32 out of 43 (74%) patients had unitemporal IEDs in WS in contrast to only 25 (58%) in SS. In all evaluated periods, 24 out of 43 (56%) patients had exclusively unitemporal IEDs, the other 19 out of 43 (44%) patients had bitemporal IEDs in at least one period. No patient had unitemporal IEDs in SS and simultaneously bitemporal IEDs in WS during either FMP or RMP. When we evaluated the change in the relative IED lateralization between FMP and RMP, a significant difference was present in SS (p = 0.003) but not in WS (p = 0.114). The change in relative IED lateralization during SS between FMP and RMP could have been caused in 12 patients with the occurrence of seizures preceding EEG analysis in RMP, therefore we also tested separately the change in relative IED lateralization between FMP and RMP in patients in whom the EEG analysis during RMP was not preceded by seizures (n = 31). In this smaller group, 27 out of 31 (87%) patients had unitemporal IEDs in FMP during WS, and 24 out of 31 (77%) patients in FMP during SS. During RMP, 27 out of 31 (87%) patients had unitemporal IEDs in WS, in contrast to 20 (65%) in SS. The difference in SS between FMP and RMP was significant even in this smaller group (p = 0.023). Three out of 43 (7%) patients had independent extratemporal IEDs—–the IEDs were present over ipsilateral central region in 1 patient, over ipsilateral frontal region in 1 patient, and over both frontal regions in 1 patient. Other analysis was not performed with extratemporal spikes because of small sample-size. To conclude, there was a significant decrease in the number of patients with unitemporal IEDs in relation to AED reduction. This decline is present during sleep, but not in wakefulness, and is caused by activation of IEDs on the nonHS side. Relative IED lateralization between HS side and non-HS side and surgical outcome at year 1 after surgery At year 1 after surgery, relative IED lateralization has a predictive value for surgical outcome defined as Engel I vs. Engel II—IV in FMP during both WS and SS (Table 2). In FMP during WS, respectively, during SS, 33 out of 38 (87%), respectively, 33 out of 37 (89%) patients with unitemporal IEDs were classified as Engel I in comparison to 2 out of 5 (40%), respectively, 2 out of 6 (33%) patients with bitemporal IEDs (p = 0.037, respectively, p = 0.007). There were no differences in surgical outcome between patients with unitemporal IEDs and bitemporal IEDs in RMP during both WS and SS (p = 0.401; p = 0.701). When patients were categorized according to ILAE classification as completely seizure-free and not seizure-free at year 1 after surgery, the statistical significant differences were found only in FMP during SS (Table 3). In FMP during SS, 29 out of 38 (78%) patients with unitemporal IEDs were categorized as completely seizure-free in comparison to 2 out of 6 (33%) patients with bitemporal IEDs (p = 0.042). There were no significant differences between patients with unitemporal IEDs and bitemporal IEDs in FMP during WS, and in RMP during both WS and SS (p = 0.123; p = 1.000; p = 0.516). Relative IED lateralization between HS side and non-HS side and surgical outcome at last follow-up visit Relative IED lateralization has a predictive value for surgical outcome defined as Engel I or Engel II—IV in FMP during SS (Table 4). During SS, 31 out of 38 (82%) patients with unitemporal IEDs were classified as Engel I in comparison to 2 out of 5 (33%) patients with bitemporal IEDs (p = 0.020). In FMP during WS, the results were borderline statistical 112 Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges 1411 Table 3 The relative IED distribution between HS side and non-HS side and its relation to surgical outcome at year 1 after surgery classified as completely seizure-free or not seizure-free after surgery during FMP and RMP. FMP RMP WS SS WS SS uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs CSF n (%) 29 (76) 2 (40) 29 (79) 2 (33) 23 (72) 8 (73) 19 (76) 12 (67) NSF n (%) 9 (24) 3 (60) 8 (22) 4 (67) 9 (28) 3 (27) 6 (24) 6 (33) Total number 38 5 37 6 32 11 25 18 p-Value p = 0.123 p = 0.042 p = 1.000 p = 0.516 bi IEDs—–bitemporal interictal epileptiform discharges, CSF—–completely seizure-free, FMP—–full medication period, NSF—–not seizurefree, RMP—–reduced medication period, uni IEDs—–unitemporal interictal epileptiform discharges, SS—–sleeping state, WS—–waking state. Table 4 The relative IED distribution between HS side and non-HS side and its relation to surgical outcome at last follow-up visit classified as Engel I or Engel II-IV during FMP and RMP. FMP RMP WS SS WS SS uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs Engel I n (%) 31 (82) 2 (40) 31 (84) 2 (33) 25 (78) 8 (73) 20 (80) 13 (72) Engel II—IV n (%) 7 (18) 3 (60) 6 (16) 4 (67) 7 (22) 3 (27) 5 (20) 5 (28) Total number 38 5 37 6 32 11 25 18 p-Value p = 0.073 p = 0.020 p = 0.698 p = 0.717 bi IEDs—–bitemporal interictal epileptiform discharges, FMP—–full medication period, uni IEDs—–unitemporal interictal epileptiform discharges, RMP—–reduced medication period, SS—–sleeping state, WS—–waking state. non-significant (p = 0.073). There were no differences in surgical outcome between patients with unitemporal IEDs and bitemporal IEDs in RMP during both WS and SS (p = 0.698; p = 0.717). When patients were classified as completely seizure-free or not seizure-free after surgery, we found a significant difference in relative IED lateralization in FMP during both WS and SS (Table 5). In FMP during WS, 27 out of 38 (71%) patients with unitemporal IEDs were categorized as completely seizure-free after surgery in comparison to 1 out of 5 (20%) patients with bitemporal IEDs (p = 0.043). Similarly, in FMP during SS, 27 out of 37 (73%) patients with unitemporal IEDs were completely seizure-free after surgery in comparison to 1 patient out of 6 (17%) with bitemporal IEDs (p = 0.015). We did not find any differences between these two groups in RMP during WS or SS. Relative IED lateralization between HS side and non-HS side and other variables No statistically significant differences were found according to relative IED lateralization in patient demographic data and characteristics or in data describing semi-invasive EEG (Table 6). Surgical outcome and other variables A univariate logistic regression analysis was performed to assess the influence of patients’ demographic data or characteristics (age at time of surgery, age at epilepsy onset, duration of epilepsy, time from surgery, gender, side of surgery, presence of FCD, frequency of habitual seizures, presence of habitual GTCS), and relative IED lateralization in FMP and RMP during WS and SS on surgical outcome at year 1 after surgery (Table 7) and at last follow-up visit (Table 8). To assess the independent influence of the previously mentioned variables on surgical outcome, a stepwise logistic regression was performed. According this stepwise logistic regression, only the relative IED lateralization in FMP during SS was found to be an independent predictor for surgical outcome at either year 1 after surgery (completely seizure-free vs. not seizure-free p = 0.032, Engel I vs. Engel II—IV p = 0.006) or at last follow-up visit (completely seizurefree vs. not seizure-free p = 0.024, Engel I vs. Engel II—IV p = 0.017). Gender was found to be independent predictor for surgical efficacy at year 1 if the outcome was defined as completely seizure-free vs. not seizure-free (p = 0.036). Discussion In our study, 44% of patients with TLE associated with HS exhibit bitemporal IEDs, defined by LI < 0.8, during semiinvasive video-EEG monitoring. These findings correlate well with results of previously published works. In studies focusing on the whole group of TLE, approximately 20—30% of patients were reported to have bitemporal IEDs (Chung et al., 1991; Radhakrishnan et al., 1998; Villanueva et al., 2004). In studies focusing on TLE associated with HS or unilateral hippocampal atrophy, the representation of patients 113 1412 I. Doleˇzalová et al. Table 5 The relative IED distribution between HS side and non-HS side and its relation to surgical outcome at last follow-up visit classified as completely seizure-free or not seizure-free after surgery during FMP and RMP. FMP RMP WS SS WS SS uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs uni IEDs bi IEDs CSF n (%) 27 (71) 1 (20) 27 (73) 1 (17) 20 (63) 8 (73) 17 (68) 11 (61) NSF n (%) 11 (29) 4 (80) 10 (27) 5 (83) 12 (37) 3 (27) 8 (32) 7 (39) Total number 38 5 37 6 32 11 25 18 p-Value p = 0.043 p = 0.015 p = 0.719 p = 0.750 bi IEDs—–bitemporal interictal epileptiform discharges, CSF—–completely seizure-free, FMP—–full medication period, NSF—–not seizurefree, RMP—–reduced medication period, uni IEDs—–unitemporal interictal epileptiform discharges, SS—–sleeping state, WS—–waking state. with bitemporal IEDs was higher; approximately 40% (AullWatschinger et al., 2008; Hardy et al., 2003; Krendl et al., 2008). The highest representation of bitemporal IEDs was reported by Ergene et al. (2000), who found them in approximately 60% of patients with non-lesional TLE during long-term EEG monitoring, although all patients in this study had only unitemporal IEDs on serial routine EEGs. In our work, we found a change in relative IED lateralization during semi-invasive EEG monitoring that has not been previously described. The representation of patients with unitemporal IEDs declined between FMP and RMP from 88% to 74% in WS, and from 86% to 58% in SS, but only the difference in SS reached statistical significance. In our opinion, the change of relative IED lateralization between FMP and RMP in our study could be potentially conditioned by two factors: AED reduction and/or seizure occurrence. As we managed to prove significant changes in relative IED lateralization between FMP and RMP even in a subgroup of patients without seizures preceding RMP EEG analysis, we can assume that AED reduction could be sufficient to cause a relative IED lateralization change, but it seems that it needs the added effect of sleeping, which is generally accepted as a strong inductor of IEDs (Malow et al., 1997; Sammaritano et al., 1991). As mentioned above, there are no studies focusing on the change of relative IED lateralization with respect to fast AEDs withdrawal during long-term EEG monitoring or seizure occurrence. Only two studies of the impact of fast AED reduction on IEDs have thus far been published, and both of them focused only on the change in IED frequency. AED reduction was found not to have any impact on IED frequency (Gotman and Marciani, 1985; Gotman and Koffler, 1989). IED frequency increase was related only to seizure occurrence followed them in hours and days in regions showing preictally the presence of IEDs and contemporary involved in seizure generation or propagation (Gotman and Marciani, 1985; Gotman and Koffler, 1989). It is necessary to point out that almost all the patients in these two studies had TLE; however, both studies used only invasive EEG and therefore their results may not apply for scalp or semi-invasive EEG. Moreover, minor IED frequency changes can cause significant changes in IED lateralization, and significant changes in IEDs frequency may not have any impact on IED lateralization. The incidence of patients with bitemporal IEDs increased during semi-invasive EEG monitoring; this change is substantial mainly during sleep, and even AED reduction is a sufficient causative factor. Patients with unitemporal IEDs tend to have excellent prognosis within the whole group of TLE, i.e. within the group of TLE associated with different etiology (Chung et al., 1991; Radhakrishnan et al., 1998; Villanueva et al., 2004). In a study by Chung et al. (1991), 89% of patients with more than 90% of IEDs localized to the side of operation had good surgical outcomes (seizure free, or significant improvement), in comparison with 35% of patients with less than 90% of IEDs localized to the operated side. In a study by Radhakrishnan et al. (1998), patients with 100% of IEDs on the operated side had excellent outcomes (seizure freedom, or only no disabling seizures) in 74%, compared with 44% in those who had less than 100% of IEDs on operated side. In spite of these convincing findings within the group of TLE as a whole, the results about the prognostic value of IED lateralization in the subgroup of mesial TLE are more diverse (Aull-Watschinger et al., 2008; Hardy et al., 2003; Krendl et al., 2008; Schulz et al., 2000). Only AullWatschinger et al. (2008) and Schulz et al. (2000) proved significant differences in relation to relative IED lateralization. Aull-Watschinger et al. (2008) found unilateral IEDs to be associated with excellent short-term (i.e. at year 1 and 2 after surgery) outcome where this outcome was defined on the basis of Engel classification as no seizures, with or without nondisabling auras, during a 12 month-period prior to the assessment. The difference between patients with unitemporal and bitemporal IEDs was lost in long-term (i.e. at year 5 after surgery) follow-up, and when the definition of complete seizure-freedom (i.e. no seizures plus no auras at any time after surgery), which seems to be a more appropriate indicator of patient quality of life, was applied. In the study of Schulz et al. (2000), almost 85% of patients with strictly unilateral temporal IEDs are postoperatively categorized as completely seizure free, compared with 52% of patients with bilateral or ipsilateral extratemporal IEDs, but they included mesial TLE associated with HS and non-lesional cases. Hardy et al. (2003) found no significant difference in surgical outcome in 118 patients with HS in relation to IED lateralization. In their study, 65% of patients with ipsilateral temporal IEDs, 50% with ipsilateral temporal and simultaneous extratemporal IEDs, and 61% with bilateral IEDs were postoperatively seizure free. Janszky et al. (2005b) proved no statistically significant differences in short-term or in long-term surgical 114 Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges 1413 Table6DifferencesbetweenLIgroupsanddemographicdata,patientandsemi-invasiveEEGmonitoringcharacteristics. FMPRMP WSSSWSSS uniIEDsbiIEDsuniIEDsbiIEDsuniIEDsbiIEDsuniIEDsbiIEDs Demographicdata Meanageatepilepsy onset 12.2±9.916.0±16.012.0±1016.2±14.311.4±8.816.0±14.711.9±9.213.6±12.4 pValuep=0.595p=0.390p=0.577p=0.941 Meandurationof epilepsy 26.2±12.527.0±13.926.6±12.424.5±13.927.8±12.221.7±12.927.4±12.324.7±13.0 pValuep=0.835p=0.792p=0.159p=0.522 Meanageatsurgery38.3±9.043±12.339.6±9.040.1±12.439.3±9.137.7±10.739.3±8.838.3±10.4 pValuep=0.255p=0.587p=0.717p=0.768 Patientscharacteristic FCDn(%)6(16)1(20)6(16)1(17)6(19)1(9)4(16)3(17) pValuep=1.000p=1.000p=0.656p=1.000 Habitualseizure frequency ≤5/monthn(%)30(79)3(60)30(81)3(50)24(75)9(82)19(76)14(78) >5/monthn(%)8(21)2(40)7(19)3(50)8(25)2(18)6(24)4(22) pValuep=0.575p=0.127p=0.575p=1.000 HabitualGTCSin number(%)ofpatients 17(45)3(60)17(46)3(50)17(53)3(27)11(44)9(50) pValuep=0.650p=1.000p=0.175p=0.763 Semi-invasiveEEG Meanduration,days6.0±1.85.2±1.86.0±2.25.0±1.75.9±2.05.8±2.66.1±2.25.5±2.2 pValuep=0.570p=0.318p=0.676p=0.680 Numberofseizures ≤5n(%)33(87)5(100)32(86)6(100)27(84)11(100)21(84)17(94) >5n(%)5(13)0(0)5(14)0(0)5(16)0(0)4(16)1(6) pValuep=1.000p=1.000p=0.306p=0.380 biIEDs—–bitemporalinterictalepileptiformdischarges,FCD—–focalcorticaldysplasia,FMP—–fullmedicationperiod,GTCS—–generalizedtonicclonicseizures,RMP—–reducedmedication period,SS—–sleepingstate,uniIEDs—–unitemporalinterictalepileptiformdischarges,WS—–wakingstate. 115 1414 I. Doleˇzalová et al. Table 7 The influence of demographic data and LI index on surgical outcome at year 1 after surgery defined as Engel I or Engel II-IV and as completely seizure-free or not seizure-free - results of the univariate logistic regression analysis. Variable Risk category/baseline category Outcome—–CSF vs. NSF Outcome—–Engel I vs. Engel II—IV OR (95% CI) p-Value OR (95% CI) p-Value Gender Male/female 4.75 (1.07—21.14) 0.041 4.00 (0.71—22.67) 0.117 Age at epilepsy onset — 1.01 (0.95—1.08) 0.695 1.02 (0.96—1.10) 0.514 Duration of epilepsy — 1.00 (0.95—1.06) 0.950 0.99 (0.93—1.05) 0.713 Age at surgery — 1.02 (0.95—1.10) 0.601 1.01 (0.93—1.10) 0.803 Side of surgery Right/left 1.07 (0.28—4.04) 0.924 1.06 (0.23—4.92) 0.942 FCD Yes/no 1.04 (0.17—6.26) 0.966 0.69 (0.07—6.70) 0.749 Habitual seizure frequency ≤5/month/>5/month 1.74 (0.31—9.69) 0.528 0.89 (0.15—5.29) 0.897 Habitual GTCS Yes/no 0.76 (0.20—2.93) 0.692 1.19 (0.26—5.52) 0.827 FMP—WS bi IEDs/uni IEDs 4.83 (0.70—33.61) 0.111 9.90 (1.31—74.73) 0.026 FMP—SS bi IEDs/uni IEDs 7.25 (1.12—47.00) 0.038 16.50 (2.26—120.64) 0.006 RMP—WS bi IEDs/uni IEDs 0.96 (0.21—4.45) 0.957 2.02 (0.40—10.38) 0.397 RMP—SS bi IEDs/uni IEDs 1.58 (0.41—6.06) 0.502 1.50 (0.32—7.01) 0.606 bi IED—–bitemporal interictal epileptiform discharges, CSF—–completely seizure-free, FCD—–focal cortical dysplasia, FMP—–full medication period, GTCS—–generalized tonic clonic seizure, NSF—–not seizure-free, RMP—–reduced medication period, SS—–sleeping state, WS—–waking state, uni IED—–unitemporal interictal epileptiform discharges. outcomes (i.e. at year 0.5, 2, 3, and 5) related to relative IED lateralization in their study of 171 patients with HS. In the study by Krendl et al. (2008) focusing on mesial TLE associated with unilateral hippocampal atrophy, the relative lateralization of IEDs was not found to be surgical predictor; 71% of patients with unitemporal IEDs and 60% of patients with bitemporal IEDs were postoperatively seizure free. In our study, the relation between relative IED lateralization and excellent surgical outcome defined as both Engel I and completely seizure-free exists, but is present only in FMP; moreover, only FMP in SS was an independent outcome predictor when stepwise logistic regression model was applied. The contradictory results of our study and studies by Aull-Watschinger et al. (2008), Hardy et al. (2003), Janszky et al. (2005b) and Krendl et al. (2008) could be explained by different study design. We analyzed separately 20 min of waking and 20 min of sleeping interictal EEG recordings during two periods defined by relative AED level withdrawal. Aull-Watschinger et al. (2008), Janszky et al. (2005b), and Krendl et al. (2008) evaluated 10, 2, and 5 min samples, respectively, every hour. Hardy et al. (2003) assessed long-term EEG monitoring, but did not specify the evaluation method more precisely in their article. Table 8 The influence of demographic data and LI index on surgical outcome at last follow-up visit defined as Engel I or Engel II—IV and as completely seizure-free or not seizure-free—–results of the univariate logistic regression analysis. Variable Risk category/baseline category Outcome—–CSF vs. NSF Outcome—–Engel I vs. Engel II—IV OR (95% CI) p-Value OR (95% CI) p-Value Gender Male/female 2.00 (0.56—7.16) 0.287 3.17 (0.69—14.46) 0.137 Age at epilepsy onset — 1.01 (0.95—1.07) 0.686 1.02 (0.95—1.08) 0.647 Duration of epilepsy — 1.00 (0.95—1.05) 0.989 1.01 (0.96—1.08) 0.624 Age at surgery — 1.01 (0.95—1.07) 0.663 1.05 (0.97—1.14) 0.251 Time from surgery in years — 1.08 (0.79—1.47) 0.639 1.12 (0.79—1.60) 0.518 Side of surgery Right/left 1.32 (0.38—4.64) 0.666 1.80 (0.43—7.59) 0.423 FCD Yes/no 0.71 (0.12—4.18) 0.703 1.40 (0.23—8.63) 0.717 Habitual seizure frequency ≤5/month/>5/month 1.33 (0.29—6.14) 0.712 0.63 (0.13—3.08) 0.566 Habitual GTCS Yes/no 1.01 (0.29—3.55) 0.988 1.20 (0.29—4.95) 0.801 FMP—WS bi IEDs/uni IEDs 9.82 (1.01—98.00) 0.049 6.64 (0.93—47.55) 0.059 FMP—SS bi IEDs/uni IEDs 13.50 (1.40—130.19) 0.024 10.33 (1.53—69.73) 0.017 RMP—WS bi IEDs/uni IEDs 0.63 (0.14—2.82) 0.541 1.34 (0.28—6.43) 0.715 RMP—SS bi IEDs/uni IEDs 1.35 (0.38—4.80) 0.640 1.54 (0.37—6.38) 0.553 bi IED—–bitemporal interictal epileptiform discharges CSF—–completely seizure-free, FCD—–focal cortical dysplasia, FMP—–full medication period, GTCS—–generalized tonic clonic seizure, NSF—–not seizure-free, RMP—–reduced medication period, SS—–sleeping state, WS—–waking state, uni IED—–unitemporal interictal epileptiform discharges. 116 Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges 1415 The interictal EEG evaluation seems to be the most reliable in sleeping, when AED reduction has not yet been initiated. During RMP, the differences between individual LI groups with respect to surgical outcome are lost because of the changes in relative IED lateralization described in the previous part of discussion. In the literature, there are some discrepancies in the value of sleeping recordings. Sammaritano et al. (1991) believed that evaluating interictal EEG during waking or rapid eye movement (REM) sleep is more reliable than evaluating slow wave sleep. In contrast, Adachi et al. (1996) proved that sleep never produced new misleading (i.e. incorrectly lateralizing) information, which is in concurrence with the results of our study during FMP. In this period, only 1 patient with LI ≥ 0.8 in the waking state developed LI < 0.8 in the sleeping state, and in this patient, the change probably reflected bitemporal impairment, because the patient was postoperatively categorized as Engel III. During RMP, the situation was rather different: 22% of patients categorized as having waking state LI ≥ 0.8 developed sleeping state LI < 0.8. We could hypothesize that results of Adachi et al. (1996) reflect FMP rather than RMP, because they evaluated initial sleeping EEG, and at the beginning of EEG monitoring the AED reduction, if already started, is generally minimal. Adachi et al. (1996) also published that IEDs during sleep could lead to correct lateralization in patients with bitemporal IEDs in waking state. In our study, there was no patient with bitemporal IEDs in WS and simultaneously unitemporal IEDs in SS. This difference is caused by a different cut-off value for defining bitemporal IEDs; we used a 90% cut-off value, and Adachi et al. (1996) used a 75% cut-off value. Ergene et al. (2000) found that patients with bitemporal IEDs tend to have longer epilepsy duration, but this difference did not reach statistical significance. Janszky et al. (2003) published that patients with bitemporal IEDs have higher seizure frequency and later epilepsy onset. In our study, no dissimilarities between patients with unitemporal and bitemporal IEDs with respect to epilepsy duration, seizure frequency, age at epilepsy onset, or other demographic data, patient characteristics, or video EEG monitoring characteristics were present. When focusing on other factors except LI which could be supposed to influence surgical outcome, we found only gender to be statistically significant predictor at year 1 after if the excellent outcome was defined on the basis of complete seizure-freedom. It is interesting that similar results were present in study of Aull-Watschinger et al. (2008), they found male gender to be the only predictor of postoperatively complete seizure-freedom at year 2 after surgery. The suggestion of male gender as a positive predictor for shortterm outcome is difficult to interpret and should be tested in larger series of patients. A study by Janszky et al. (2005a) identified several distinct predictors for outcome at years 2, 3, and 5 after surgery. The following factors were associated with worse surgical outcome in an individual year: the presence of sGTCS and ictal dystonia at year 2, longer epilepsy duration and ictal dystonia at year 3, and longer epilepsy duration at year 5. We see three basic limitations of our work. First, we described AED reduction only on the basis of relative daily dose reduction, not by serum levels. Second, only relatively short and preselected periods (80 min) were analyzed in each patient, which could influence our results. Third, we did not use EEG polysomnography, which would enable us to more precisely identify sleeping stages. On the other hand, neither AED serum level measuring nor EEG polysomnography are methods common in the daily praxis of epilepsy centers during video-EEG monitoring, and the main purpose of this work was to provide a helpful tool for clinicians, both epileptologists and neurophysiologists, in patient evaluation. It is a challenge to perform an analogous prospective study with more detailed patient investigations, including AED serum levels, and 5 or 10 min samples every hour, or may be better continuous 24-h lasting interictal EEG evaluation. With respect to our conclusions, it would be meaningful to establish a cut-off value of AED reduction determining the change in relative IED lateralization. We were not able to find such cut-off value because in each individual patient we used only two recordings—–one with full AED medication and one with marked AED reduction. Conclusion The relative IED lateralization in interictal EEG in patients with TLE with HS could be a predictor for surgical outcome, defined as both Engel I or Engel II—IV, and seizure-free or not seizure-free, only before initialization of AED reduction. Its predictive value is diminished with AED reduction because of changes in relative IED lateralization. Acknowledgments We thank Anne Johnson for grammatical assistance. This work was supported by the project ‘‘CEITEC—–Central European Institute of Technology’’ (CZ.1.05/1.1.00/02.0068) from European Regional Development Fund. References Adachi, N., Alarcon, G., Binnie, C.D., Elwes, R.D.C., Polkey, C.E., Reynolds, E.H., 1996. Predictive value of interictal epileptiform discharges during non-REM sleep on scalp EEG recording for the lateralization of epileptogenesis. Epilepsia 139, 628—632. Aull-Watschinger, S., Pataraia, E., Czech, T., Baumgartner, C., 2008. Outcome predictors for surgical treatment of temporal lobe epilepsy with hippocampal sclerosis. Epilepsia 49, 1308—1316. Blümcke, I., Pauli, E., Clusmann, H., Schramm, J., Becker, A., Elger, C., Merschhemke, M., Meencke, H.J., Lehmann, T., von Deimling, A., Scheiwe, C., Zentner, J., Volk, B., Romstöck, J., Stefan, H., Hildebrandt, M., 2007. A new clinico-pathological classification system for mesial temporal sclerosis. Acta Neuropathol. 113, 235—244. Blümcke, I., Thom, M., Aronica, E., Armstrong, D.D., Vinters, H.V., Palmini, A., 2011. The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52, 158—174. Chung, M.Y., Walczak, T.S., Lewis, D.W., Dawson, D.V., Radtke, R., 1991. Temporal lobectomy and independent bitemporal activity: what degree of lateralization is sufficient? Epilepsia 32, 195—201. 117 1416 I. Doleˇzalová et al. Engel Jr., J., 1987. Outcome with respect to epileptic seizures. In: Engel Jr., Jr. (Ed.), Surgical Treatment of the Epilepsies. , second ed. Raven Press, New York, NY, pp. 553—571. Engel Jr., J., 1996. Surgery for seizures. N. Engl. J. Med. 334, 647—652. Ergene, E., Shih, J.J., Blum, D.E., So, N.K., 2000. Frequency of bitemporal independent interictal epileptiform discharges in temporal lobe epilepsy. Epilepsia 41, 213—218. Gotman, J., Marciani, M.G., 1985. Electroencephalographic spiking activity, drug levels, and seizure occurrence in epileptic patients. Ann. Neurol. 17, 597—603. Gotman, J., Koffler, D.J., 1989. Interictal spiking increases after seizures but does not after decrease in medication. Electroencephalogr. Clin. Neurophysiol. 72, 7—15. Hardy, S.G., Miller, J.W., Holmes, M.D., Born, D.E., Ojemann, G.A., Dodrill, C.B., Hallam, D.K., 2003. Factors predicting outcome of surgery for intractable epilepsy with pathologically verified mesial temporal sclerosis. Epilepsia 44, 565—568. Iber, C., Ancoli-Israel, S., Chesson, A., Quan, S.F., 2007. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. American Academy of Sleep Medicine, Westchester, NY. Janszky, J., Rásonyii, G., Clemens, Z., Schulz, R., Hoppe, M., Barsi, P., Fogarasi, A., Halász, P., Ebner, A., 2003. Clinical differences in patients with unilateral hippocampal sclerosis and unitemporal or bitemporal epileptiform discharges. Seizure 12, 550—554. Janszky, J., Pannek, H.W., Janszky, I., Schulz, R., Behne, F., Hoppe, M., Ebner, A., 2005a. Failed surgery for temporal lobe epilepsy: predictors of long-term seizure-free course. Epilepsy Res. 64, 35—44. Janszky, J., Janszky, I., Schulz, R., Hoppe, M., Behne, F., Pannek, H.W., 2005b. Temporal lobe epilepsy with hippocampal sclerosis: predictors for long-term surgical outcome. Brain 128, 395—404. Katz, A., Marks, D.A., McCarthy, G., Spencer, S.S., 1992. Does interictal spiking change prior to seizures? Electroencephalogr. Clin. Neurophysiol. 79, 153—156. Krendl, R., Lurger, S., Baumgartner, C., 2008. Absolute spike frequency predicts surgical outcome in TLE with unilateral hippocampal atrophy. Neurology 71, 413—418. Kuba, R., Tyrlíková, I., Paˇzourková, M., Hermanová, M., Horáková, I., Brázdil, M., Rektor, I., 2012. Grey-white matter abnormalities in temporal lobe epilepsy associated with hippocampal sclerosis: inter-observer analysis, histopathological findings, and correlation with clinical variables. Epilepsy Res. 102, 78—85. Malow, B.A., Kushwaha, R., Lin, X., Morton, K.J., Aldrich, M.S., 1997. Relationship of interictal epileptiform discharges to sleep depth in partial epilepsy. Electroencephalogr. Clin. Neurophysiol. 102, 20—26. Marciani, M.G., Gotman, J., 1986. Effect of drug withdrawal on location of seizure onset. Epilepsia 27, 131—423. Radhakrishnan, K., So, E.L., Silbert, P.L., Jack, C.R., Cascino, G.D., Sharbrough, F.W., O’Brien, P.C., 1998. Predictors of outcome of anterior temporal lobectomy for intractable epilepsy: a multivariate study. Neurology 51, 465—471. Salanova, V., Andermann, F., Rasmussen, T., Olivier, A., Quesney, L., 1996. The running down phenomenon in temporal lobe epilepsy. Brain 119, 989—996. Sammaritano, M., Gigli, G.L., Gotman, J., 1991. Interictal spiking during wakefulness and sleep and the localization of foci in temporal lobe epilepsy. Neurology 41, 290—297. So, N., Gotman, J., 1990. Changes in seizure activity following anticonvulsant drug withdrawal. Neurology 40, 407—413. Schulz, R., Luders, H.O., Hoppe, M., Tuxhorn, I., May, T., Ebner, A., 2000. Interictal EEG and ictal scalp EEG propagation are highly predictive of surgical outcome in mesial temporal lobe epilepsy. Epilepsia 41, 564—570. Spencer, S.S., Spencer, D.D., Williamson, P.D., Manson, R.H., 1981. Ictal effects of anticonvulsant medication withdrawal in epileptic patients. Epilepsia 22, 297—307. Villanueva, V., Peral, E., Albisua, J., de Felipe, J., Serratosa, J.M., 2004. Prognostic factors in temporal lobe epilepsy surgery. Neurologia 19, 92—98. Wiebe, S., Blume, W.T., Girvin, J.P., Eliasziw, M., Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group, 2001. A randomized, controlled trial of surgery for temporallobe epilepsy. N. Engl. J. Med. 345, 311—318. Wieser, H.G., Blume, W.T., Fish, D., Goldensohn, E., Hufnagel, A., King, D., Sperling, M.R., Luders, H., Pedley, T.A., 2001. ILAE Commission Report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 42, 282—286. Williamson, P., 1998. Anatomic classification of localization-related epilepsies. In: Engel Jr., J. (Ed.), Epilepsy: A Comprehensive Textbook. , first ed. Lippincott-Raven, New York, NY, pp. 2465—2647. 118 Annex 3: Frauscher B, von Ellenrieder N, Zelmann R, Dolezalova I, Minotti L, Olivier A, Hall J, Hoffmann D, Nguyen DK, Kahane P, Dubeau F, Gotman J. Atlas of the normal intracranial electroencephalogram: neurophysiological awake activity in different cortical areas. Brain. 2018;141:4. 119 Atlas of the normal intracranial electroencephalogram: neurophysiological awake activity in different cortical areas Birgit Frauscher,1,2 Nicolas von Ellenrieder,1 Rina Zelmann,1,3 Irena Dolezˇalova´,4 Lorella Minotti,5 Andre´ Olivier,1 Jeffery Hall,1 Dominique Hoffmann,5 Dang Khoa Nguyen,6 Philippe Kahane,5 Franc¸ois Dubeau1 and Jean Gotman1 In contrast to scalp EEG, our knowledge of the normal physiological intracranial EEG activity is scarce. This multicentre study provides an atlas of normal intracranial EEG of the human brain during wakefulness. Here we present the results of power spectra analysis during wakefulness. Intracranial electrodes are placed in or on the brain of epilepsy patients when candidates for surgical treatment and non-invasive approaches failed to sufficiently localize the epileptic focus. Electrode contacts are usually in cortical regions showing epileptic activity, but some are placed in normal regions, at distance from the epileptogenic zone or lesion. Intracranial EEG channels defined using strict criteria as very likely to be in healthy brain regions were selected from three tertiary epilepsy centres. All contacts were localized in a common stereotactic space allowing the accumulation and superposition of results from many subjects. Sixty-second artefact-free sections during wakefulness were selected. Power spectra were calculated for 38 brain regions, and compared to a set of channels with no spectral peaks in order to identify significant peaks in the different regions. A total of 1785 channels with normal brain activity from 106 patients were identified. There were on average 2.7 channels per cm3 of cortical grey matter. The number of contacts per brain region averaged 47 (range 6–178). We found significant differences in the spectral density distributions across the different brain lobes, with beta activity in the frontal lobe (20–24 Hz), a clear alpha peak in the occipital lobe (9.25–10.25 Hz), intermediate alpha (8.25–9.25 Hz) and beta (17–20 Hz) frequencies in the parietal lobe, and lower alpha (7.75–8.25 Hz) and delta (0.75–2.25 Hz) peaks in the temporal lobe. Some cortical regions showed a specific electrophysiological signature: peaks present in 460% of channels were found in the precentral gyrus (lateral: peak frequency range, 20–24 Hz; mesial: 24–30 Hz), opercular part of the inferior frontal gyrus (20–24 Hz), cuneus (7.75–8.75 Hz), and hippocampus (0.75–1.25 Hz). Eight per cent of all analysed channels had more than one spectral peak; these channels were mostly recording from sensory and motor regions. Alpha activity was not present throughout the occipital lobe, and some cortical regions showed peaks in delta activity during wakefulness. This is the first atlas of normal intracranial EEG activity; it includes dense coverage of all cortical regions in a common stereotactic space, enabling direct comparisons of EEG across subjects. This atlas provides a normative baseline against which clinical EEGs and experimental results can be compared. It is provided as an open web resource (https://mni-open-ieegatlas.research.mcgill.ca). 1 Montreal Neurological Institute and Hospital, McGill University, Montreal, Quebec, Canada 2 Department of Medicine and Center for Neuroscience Studies, Queen’s University, Kingston, Ontario, Canada 3 Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA 4 Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University Brno, Czech Republic 5 Department of Neurology, Grenoble-Alpes University Hospital and Grenoble-Alpes University, Grenoble, France 6 Centre hospitalier de l’Universite´ de Montre´al - Hoˆpital Notre-Dame, Montre´al, Que´bec, Canada doi:10.1093/brain/awy035 BRAIN 2018: 141; 1130–1144 | 1130 Received August 24, 2017. Revised November 30, 2017. Accepted January 1, 2018. Advance Access publication March 1, 2018 ß The Author(s) (2018). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: journals.permissions@oup.com Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 120 Correspondence to: Birgit Frauscher, MD Montreal Neurological Institute and Hospital, McGill University, 3801 University Street, Montreal H3A 2B4, Quebec, Canada E-mail: birgit.frauscher@mcgill.ca Keywords: stereo-encephalography; physiology; spectral analysis; human; brain Introduction The scalp EEG during wakefulness in healthy individuals is well defined (Chang et al., 2011), but the knowledge accumulated on normal physiological intracranial EEG activity outside of the context of intracranial event-related potentials, task-related responses or intracranial stimulation (Jerbi et al., 2009; David et al., 2013; Matsumoto et al., 2017) is surprisingly scarce. This contrasts with the vast literature on epileptic activity in intracranial EEG (for a review see Frauscher and Dubeau, 2018). The knowledge of physiological intracranial EEG activity is based mainly on studies performed in the pre-MRI era using visual analysis of the recorded signals (Gastaut 1949; Jasper and Penfield, 1949; Petersen et al., 1953; Sem-Jacobsen et al., 1953, 1955, 1956; Chatrian et al., 1960 a, b; Ribstein, 1960; Perez-Borja et al., 1962; Graf et al., 1984). Evidence suggests that normal alpha, beta and gamma rhythms, and slow wave activity in the theta and delta range are observed in the intracranial EEG following a spatially distributed pattern. For instance, alpha rhythm was described in the occipital lobe, the parietal lobe just posterior to the postcentral gyrus, and the posterior temporal neocortex (Gastaut, 1949; Jasper and Penfield, 1949; Sem-Jacobsen et al., 1953, 1956; Chatrian et al., 1960a; Perez-Borja et al., 1962), and non-reactive alpha activity was found in the frontal lateral neocortex (Jasper and Penfield, 1949). Beta rhythm is most frequently seen over the anterior head regions. In particular, well sustained beta frequencies of $25 Hz are described in the precentral and postcentral gyri with lower frequencies present in the postcentral region (Jasper and Penfield, 1949). Other normal activities such as mu rhythm (Graf et al., 1984) and lambda waves (Chatrian et al., 1960a, b; Perez-Borja et al., 1962; Chatrian, 1976) have also been described. Location-related differences in the occurrence of these rhythms and patterns remain purely descriptive without attempts to determine the exact localization of the positions of the electrode contacts in a common space. Furthermore, quantitative studies of normal activity during resting wakefulness are either lacking or are restricted to certain brain areas, often with small patient numbers. The best evidence exists for the hippocampus where an activity in the theta delta range has been described in quiet wakefulness (Brazier, 1968; Huh et al., 1990; Meador et al., 1991; Nishida et al., 2004; Moroni et al., 2012). In an attempt to create an atlas of normal human brain activity during wakefulness, Kahane (1993) described the quantitative EEG distribution across several cortical areas using intracranial EEG performed in epilepsy patients during presurgical epilepsy work-up. Coordinates of the individual electrode positions were given in Talairach space. In this work, the author already highlighted the methodological difficulties to determine normality in an epileptic brain. Patients with refractory focal epilepsies are the only human subjects where extensive intracranial cortical EEG studies are undertaken, and thus allow access not only to pathological but also to normal brain neurophysiology. We have a poor knowledge of the normative electrophysiological data of brain activity. This is explained by the relatively rare placement of electrodes in healthy brain tissue and the challenge in identifying healthy brain regions, and by the difficulty of standardization of the electrode placement compared to scalp EEG, resulting in problems performing interindividual comparisons of EEG patterns (each patient has at most a few contacts in healthy brain, with great variability in electrode location from patient to patient). Moreover, even in large tertiary epilepsy centres, only a small proportion of patients ($15–25 patients) are explored each year with intracranial electrodes. The current work is a multicentre study aiming to provide an atlas of normal intracranial EEG activity recorded with both stereo-EEG electrodes and cortical grids/strips. These data are provided as an open web resource (https://mni-open-ieegatlas.research.mcgill.ca) developed under the LORIS framework (Das et al., 2016). Localization of the contacts in a common stereotactic space allows the accumulation and superposition of results from a large number of subjects and a direct comparison of EEG activity across subjects. The results of power spectra analysis during wakefulness are presented in this paper. Materials and methods Selection of intracranial EEG recordings Charts and recordings of patients who underwent intracranial EEG investigation as part of their clinical evaluation for epilepsy surgery at the Montreal Neurological Institute and Hospital (MNI), Centre Hospitalier de l’Universite´ de Montre´al (CHUM), and Grenoble-Alpes University Hospital (CHUGA) were screened, starting with the most recent patients at time of data collection (September 2015 for MNI and CHUM, April 2016 for CHUGA), and moving consecutively Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1131 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 121 backward to January 2010 or earlier, to identify cases with 540 recordings fulfilling selection criteria. Recordings of patients included for this study had to fulfil the following four inclusion criteria: (i) presence of at least one channel with normal activity. Such channels are not common (on average 11% of channels per patient as shown in one of our previous studies) (Frauscher et al., 2015). A channel with normal activity is defined as a channel localized in normal tissue as assessed by MRI, is located outside the seizure onset zone, does not show at any time of the circadian cycle interictal epileptic discharges (according to the clinical report of the complete implantation and to a careful investigation of one night of sleep by a board-certified electrophysiologist), and shows the absence of overt slow-wave anomaly; (ii) presence of peri-implantation imaging (CT or MRI) for exact localization of individual electrode contacts (contacts located in the white matter were excluded); (iii) availability of a controlled intracranial EEG recording obtained after a minimum of 72 h after insertion of stereo-EEG electrodes or 1 week after placement of subdural grids or strips (medications are usually not yet lowered), and at least 12 h after a generalized tonic-clonic seizure, 6 h in case of focal clinical seizures, or 2 h in case of purely electrographic seizures, and not after electrical stimulation, as done in our previous work (Frauscher et al., 2015); and (iv) sampling frequency of a minimum of 200 Hz. This minimum sampling frequency was chosen to include as many patients as possible for the analysis of frequencies in the classical Berger frequency bands (0.3–70 Hz). The data collected in the three centres complemented each other well, as different implantation strategies are used (stereoEEG electrodes versus subdural grids/strips, and frame-assisted versus frameless electrode placement), and provide a broad coverage of different cortical regions. Ethical approval was granted at the MNI as lead ethics organization (REB vote: MUHC-15-950). Co-registration and anatomical localization of electrodes and electrode contacts Registration to stereotaxic space and anatomical localization of electrodes was performed using minctools (www.bic.mni.mcgill. ca/ServicesSoftware/Services SoftwareMincToolKit) and the IBIS framework (Drouin et al., 2017). Peri-implantation CT/MRI images showing electrode positions were linearly registered to pre-implantation MRI images (preMRI) of each patient. In turn, preMRIs were non-linearly registered to the ICBM152 2009c non-linear symmetric brain model (Mazziotta et al., 2001; Fonov et al., 2011). The combined transformation allowed estimation of the electrode positions in a common stereotaxic space in order to visualize the standardized atlas of normal EEG activity. Anatomical structures were fully automatically segmented. We used an atlas created with an unbiased method (Fonov et al., 2011) from manually segmented data of 20 normal subjects (Landman and Warfield, 2012). It consists of 132 grey matter labels (66 per hemisphere after excluding basal ganglia, cerebellum, midbrain and brainstem). We chose this atlas because it was detailed and contains deep as well as surface cortical regions. The preMRI was non-linearly registered to the atlas’ averaged MRI and then the inverse transformation was applied to warp the labels back to the patient’s space. In case of lesional epilepsy, segmentations were checked visually by an epileptologist in order to ensure that selected electrode contacts are indeed outside of lesional tissue. Patients with large cortical malformations such as extended polymicrogyria, hemimegalencephaly, agenesis of corpus callosum, or extended encephalomalacic lesions were excluded. Each bipolar channel was represented by a volume made of a cylinder of length equal to the distance between its two electrode contacts, 10 mm diameter and one half sphere at each end. The anatomical localization of each channel was estimated by assessing in which segmented grey matter region laid the majority of its volume. The findings were grouped according to the topographical localization of the channels and plotted onto the standard ICBM152 template. Selection of EEG sections We visually selected 60 s sections (either continuous or consecutive discontinuous 45 s segments after artefact exclusion; 5 s were chosen as a compromise between minimizing the number of segments and obtaining the 60 s sections) of resting wakefulness EEG with eyes closed recorded during standardized conditions. The controlled ‘eyes closed and eyes opened’ recordings were part of the controlled intracranial EEG evaluation at the three participating sites. Analysis of EEG activity of the nonepileptic physiological channels was performed using Matlab (Mathworks). The signals were bandpass filtered at 0.5–80 Hz and downsampled to 200 samples per second if the original sampling rate was higher (original sampling rates were 200, 256, 512, 1000, 1024, and 2000 samples per second). Analysis of the oscillatory component of the signal The spectral density in each channel was estimated with Welch’s method, i.e. averaging the magnitude of the discrete time Fourier transform of 59 overlapping blocks of 2 s duration and 1 s step, weighted by a Hamming window. In each channel the resulting spectral density was normalized to a total power equal to one, making it independent of the EEG signal amplitude. To determine the presence of peaks in the spectrum, we first defined the ‘no-peak set’ as a set of channels with no peak in their spectrum. We defined this set in a data-driven procedure. The normalized spectra were classified into k-groups using the k-means algorithm, with 160 features given by the values at each frequency step (0.5 Hz steps between 0.5 and 80 Hz), Euclidean distance, and 100 repetitions. This separated channels according to their most prominent peaks in their spectrum. By repeating the classification with increasing number of groups k, eventually a group without peaks was found. We determined the presence of this group by requiring that its mean normalized spectrum be lower than the maximum among the other groups. From this group, we took the 50% of the channels closest to its mean to define the no-peak set. Because as the frequency increases, the spectrum becomes more correlated between equally spaced frequency steps, we defined a set of unequal frequency intervals in which to test 1132 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 122 for the presence of peaks in the spectrum. The intervals were selected so that each of them had at least 4% of the power of the average normalized spectrum of all the studied channels. This resulted in 22 frequency intervals (0.5–0.75–1.25–1.75– 2.25–3.25–3.75–4.25–5.25–6.25–6.75–7.75–8.25–9.25–10.25– 11.75–13.25–15.25–17.25–20.25–24.25–31.75–80Hz). The presence of a peak at a given frequency interval in any particular brain region was determined by comparing the distribution of the normalized spectra of all the channels in the region at the desired frequency interval, to the distribution of the no-peak set at the same frequency interval. A one-sided two-sample KolmogorovSmirnov test was used, at 5% significance level with Dunnett’s correction for multiple comparisons (for the number of tested anatomical regions times the number of frequency intervals). The two-sample Kolmogorov-Smirnov test is sensitive to differences in location and shape of the empirical cumulative distribution functions. A second test was performed for brain regions and frequency intervals in which significant differences were found, to determine the percentage of channels in the brain region of interest that had a significantly higher relative power than the no-peak set at the given frequency interval. We performed an uncorrected one-sided Wilcoxon rank-sum test (at 5% significance level), comparing the distribution of the nopeak set to the spectral density of each individual channel. Analysis of the non-oscillatory component of the signal To investigate the non-oscillatory component of the spectrum, we explored the scale-free dynamics of the brain activity by fitting a 1/fb model to the high frequency end of the spectrum. We adapted the methodology used in previous studies (Yamamoto and Hughson, 1993; He et al., 2010). We used non-overlapping Hamming windows of 1 s duration for the computation of the spectrum of the scale-free activity, and fitted the b coefficient in the 10–40 Hz range. We adopted a low end of 10 Hz instead of 1 Hz as in He et al. (2010), because we observed that in most of the channels the scale-free component of the spectrum had a shoulder between 1 and 10 Hz. The upper end of the range (40 Hz) is limited by the sampling rate of our data (the highest frequency not affected by the antialiasing down sampling filters is 80 Hz, but the computation of the scale-free component of the spectrum is correct only up to half this frequency). To determine if some regions had a different b coefficient, we computed the median value of b and performed a permutation test (100 000 permutations of region labels, Bonferroni corrected for 38 regions). For exploring the nested frequencies in the brain activity, we performed a phase-amplitude coupling analysis. We computed the modulation index (Tort et al., 2008) for phases at frequencies between 0.5 and 12 Hz, with 0.5 Hz steps. We explored the coupling to amplitude in 17 frequency bands (4–5–6–7–8– 10–12–14–16–20–24–30–36–42–48, 52–58, 62–70–80 Hz). For computation of the modulation index we used eight phase bins. The Hilbert transform was implemented between 0.5 and 80 Hz, with an FIR filter of order 200. The bandpass filters were elliptic IIR filters of order 6 applied in two passes to achieve zero-phase. To determine the statistical significance of the observed modulation index, a set of 1000 surrogate amplitude signals were analysed for each channel, by partitioning the original time series in eight sections of random length and shuffling them. The mean and standard deviation (SD) of these surrogate data were used to determine the z-score of the modulation index. The proportion of channels with significant modulation index at different phase-frequency bins was computed, and a permutation test was performed to determine regional differences (100 000 permutations of region labels, Bonferroni corrected for 24 Â 17 phase-frequency bins). Results Here we report on the cortical coverage for this project, the different electrode types used, as well as the automatic classification of the no-peak set, which is used subsequently to test for the presence of peaks in the spectrum of channels from different brain regions. We then describe the lobar differences in spectral density distribution, highlight the intracranial EEG signature of some cortical brain regions, stress unexpected findings, and analyse the non-oscillatory component of the signal. Demographical information of the study sample The study sample consisted of 106 patients (54 males; CHUGA: 49 patients, MNI: 40, CHUM: 17) with therapy-refractory focal epilepsy who underwent invasive intracranial EEG investigation during presurgical epilepsy work-up. The mean age was 33.1 Æ 10.8 years. Eightynine patients (84%) were investigated with stereo-EEG electrodes, and 17 (16%) with grids/strips. Of the patients who were explored with grids/strips, all but two had additional stereo-EEG electrodes for exploring deep structures such as the insula or the mesiotemporal lobe. Number and density of cortical channels with normal EEG activity A total of 1785 intracranial EEG channels (left hemisphere, 1066; right hemisphere, 719) recording presumably normal physiological brain activity were identified. The positions of the investigated channels after co-registration in stereotaxic space are given in Fig. 1. There were on average 0.9 channels per cm2 of cortical surface, and 2.7 channels per cm3 of cortical grey matter volume. Restricting the channels to have an equal number of channels per region in both hemispheres (695 channels), we found no differences in the distribution of normalized spectral density between the channels from homologous regions of both hemispheres (two-sided two-sample Kolmogorov-Smirnov test with Dunnett’s correction for 22 comparisons for the tested frequency intervals). Therefore, we grouped left and right hemisphere channels together. The figures show all channel positions flipped to one of the hemispheres of the symmetric atlas. Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1133 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 123 The spectral density is similar in stereo-EEG electrodes and cortical grids/strips Eighty-five per cent of the 1785 channels (1520) were from stereo-EEG electrodes, and 15% (265) from cortical grids and strips (Fig. 1). To compare the spatial coverage achieved with intracerebral electrodes and with cortical grids and strips we classified the cortical surface according to curvature, and compared regions with convex curvature (likely gyri) to regions of concave curvature (likely sulci). We found that grid channels were 5.3-times more frequent in the former, and stereo-EEG electrode channels 1.6-times more frequent in the latter (180 versus 34 and 412 versus 673 in convex versus concave regions, respectively; the remaining channels were in regions with no clear dominant curvature). This indicates that channels from cortical grids/ strips are typically placed atop of gyri, and that stereo-EEG electrodes record not only from the sulci, but also from a significant fraction of the gyri. The normalized spectral density of the neocortical channels recorded with stereo-EEG (commercial DIXI electrodes, home-made MNI electrodes, or Ad-Tech electrodes) or cortical grids/strips (Ad-Tech electrodes) is similar across electrode types (Fig. 2A). Interestingly, there is a $2-fold increase in absolute amplitude (root mean square) in channels from cortical grids/ strips compared to stereo-EEG electrodes (Fig. 2B). The difference is maintained when only neocortical channels are analysed (i.e. excluding mesiotemporal lobe structures). This difference does not affect subsequent results, as normalized spectra are used. Automatic classification of channels and no-peak set The procedure to separate a set of channels with no peaks in the spectrum from sets of channels with various peaks resulted in eight channel groups (no peak, one group and various peaks, seven groups). The median normalized spectral density of the groups can be observed in Fig. 3A, and the groups with spectral peaks correspond to peaks in the low delta band, high delta band, theta band, alpha band, low beta band, high beta band, and gamma band. Channels with delta or gamma activity usually do not show sustained rhythms, but simply an increase in the low or high frequency activity. The spectral density of what we define as ‘the no-peak set’, derived from the group of channels with no peak in their spectrum, is shown in Fig. 3B. This no-peak set is used in subsequent sections to test for the presence of peaks in the spectrum of channels from different brain regions. Figure 3C and D shows the distribution of the groups in the inflated cortex. Note that even though the frequency resolution of this unsupervised classification is poor, it shows a frequency increase from the posterior to the anterior cortical regions. Lobar frequency differences in intracranial EEG Figure 4 shows the difference in the spectral density of the standard EEG frequencies across the different brain lobes (758 frontal channels, 415 temporal, 355 parietal, and 105 occipital). The occipital lobe shows, as expected, a clear peak in alpha activity (significant in the 7.75–10.25 Hz, in Figure 1 Localization of the 1785 EEG channels with normal physiological activity analysed for this study. The 1520 channels from stereo-EEG electrodes are visualized in blue, and the 265 channels from cortical grids and strips are in yellow. Note that for the ‘inflated’ brain display at the bottom, the electrodes are projected on the cortical surface. 1134 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 124 Figure 2 Comparison of different electrode types (stereo-EEG electrode types: DIXI CHUGA, DIXI MNI, MNI MNI, AdTech CHUM, and grids: CHUM). (A) Median spectral density for each electrode type. (B) Signal amplitude (and number of channels) according to electrode type and centre. Note the $100% increase in absolute amplitude (root mean square, RMS) in channels from cortical grids/strips compared to stereo-EEG electrodes. CHUGA = Centre hospitalier de l’universite´ de Grenoble-Alpes; CHUM = Centre hospitalier de l’universite´ de Montre´al; MNI = Montreal Neurological Institute. Figure 3 Unsupervised classification of channels and no-peak set. (A) Median normalized spectral density of the eight different groups obtained by the classification. Channels with delta or gamma activity usually do not show sustained rhythms (no peaks), but simply an increase in the low or high frequency activity. (B) The spectral density of the no-peak set, derived from the group of channels with no peak in their spectrum. (C and D) Distribution of the groups in the inflated cortex (lateral and mesial views). Even though the frequency resolution of this unsupervised classification is poor, the figure gives a rough global idea of the distribution of rhythms in the brain, and shows a posterior to anterior gradient of increasing frequencies. Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1135 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 125 Figure 4 Lobar differences in EEG frequences. (A) Spectra of the different brain lobes in semi-logarithmic graph. The red line corresponds to the median spectral density of all the channels in the region. The 25 and 75 percentiles are indicated by the pink shaded region. The broken red lines show the upper and lower bounds of the spectral distribution at every frequency. The thin black line shows the median spectrum of the nopeak set used to determine the presence of peaks. Vertical black lines separate the common clinical frequency bands indicated by Greek letters. The coloured horizontal segments in the upper part of each graph indicate the presence of peaks. If the segment is present, it indicates that the distribution of the channel spectral densities is significantly higher than the distribution of the no-peak set. The colour of the line indicates the percentage of channels that have a significant deviation compared to the no-peak set at each frequency. (B) Illustration of differences not clearly visible in A. Each panel shows in black the distribution of the energy in the no-peak set for the given frequency interval, and the red line shows the distribution in different brain regions (frontal lobe and temporal lobe). The broken vertical lines indicate the median value, and the coloured area under the curve indicates the percentage of channels that has significantly higher power than the no-peak set in the corresponding frequency interval. The figure illustrates situations in which the test can find differences, not always obvious looking only at the median (in A). Left: Different position (mean, median) in the gamma band of the frontal lobe. Middle: Equal median but different dispersion (SD) in the delta band of the temporal lobe. Right: Equal median but asymmetric distribution (skewness) in the alpha band of the temporal lobe. 1136 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 126 47% of the channels at $10 Hz), whereas faster activity peaks in the beta and gamma band were recorded in the frontal lobe (16.75–80 Hz, in 48% of the channels at $24 Hz). The parietal lobe shows peaks at intermediate frequencies in the alpha (8.75–9.75Hz, in 21% of the channels) as well as in the beta band (16.75–31.75Hz, in 24% of the channels at $20 and 31.5Hz). The temporal lobe shows a borderline alpha peak (7.25–9.25 Hz, in 30% of the channels at $8 Hz), and some channels show a peak in delta activity (0.75–2.25Hz, in 24% of the channels at $1 Hz). Figure 4B shows the difference in the distribution for some of these cases, in which the difference is not clearly appreciated by looking at the median spectral density in Fig. 4A. Some cortical brain regions have a specific EEG signature We analysed the spectrum of channels in different anatomical brain regions. Some neighbouring anatomical regions of the MICCAI atlas were joined in order to attain at least five channels per analysed region. We formed eight regions joining channels from 20 regions (the entorhinal cortex had no electrodes): (i) superior and middle occipital gyri; (ii) inferior occipital gyrus and occipital pole; (iii) lingual gyrus and occipital-fusiform gyrus; (iv) postcentral gyrus and its medial segment; (v) gyrus rectus and anterior, medial, and posterior orbital gyri; (vi) superior frontal gyrus and frontal pole; (vii) temporal pole and planum tempolare; and (viii) fusiform and parahippocampal gyri. We analysed 38 cortical regions (median 40.5 channels per region, range 6–178 channels). Thus, the total number of analysed regions used in the multiple comparison correction was 42, the 38 cortical regions, plus the four lobes presented in the previous subsection. Power spectra of all investigated cortical regions are provided in the Supplementary material. Brain regions such as the posterior insula, the anterior cingulate, and others did not show deviations from the no-peak set. In contrast, other brain regions demonstrated a specific signature: significant peaks present in more than 60% of the channels within a given region were found in the precentral gyrus (lateral segment: 64% of 123 channels at $20–24Hz; medial segment: 72% of 18 channels at $24–31.5Hz), the opercular part of the inferior frontal gyrus (72% of 39 channels at $20–24Hz), the cuneus (68% of 19 channels at $8Hz), and the hippocampus (72% of 36 channels at $1Hz) (Fig. 5). Some cortical areas have more than one spectral peak within the same channel From all analysed channels, 141 (7.9%) had more than one peak in the spectrum (comparing each channel to the no-peak set, without correcting for multiple comparisons). The percentage of channels with two or more peaks was significantly higher than the average of 7.9% in eight brain regions (cumulative binomial distribution with parameters theta = 0.079 and k = number of channels in the region, uncorrected). Most of these regions are sensory/motor regions: medial segment of the precentral gyrus (5/18 or 28% of channels), mid-cingulate region (8/40 or 20%), supplementary motor cortex (11/47 or 23%) and the cortex anterior to the supplementary motor cortex (3/16 or 19%), transverse temporal gyrus (3/14 or 21%), inferior occipital gyrus (4/20 or 20%), precentral gyrus (23/123 or 19%), and postcentral gyrus (11/65 or 17%). In one-third of these channels (48/141) the second peak is at twice the frequency of the first (second harmonic), and in 8% (11 channels) there was a significant peak at the third harmonic. Half of the channels with second harmonics (24/48) correspond to the precentral gyrus, its medial segment, or the postcentral gyrus. Nine of the 11 channels with significant third harmonic belong to these regions. The statistical test to determine if there are two significant peaks at different frequencies requires a multiplecomparison correction for a high number of 231 frequency pairs; therefore the results are not significant for the available number of channels. Unexpected findings In this section, we present some highlights of the findings for a few brain regions. For findings from all 38 brain regions, see the Supplementary material as well as the raw data, which are made available through our webpage (https://mni-open-ieegatlas.research.mcgill.ca). The anterior insula cortex shows a beta peak The insula cortex did not show a significant difference compared to the no-peak set. When subdividing the insula into an anterior and posterior portion, we found a peak in beta activity for the anterior insula in 34% of 71 channels ($20 Hz). There was no significant difference in the spectral power density between the posterior insula and the no-peak set (Supplementary material). The middle cingulate gyrus shows a beta peak The middle cingulate gyrus showed a beta peak (40% of 40 channels $24 Hz, range: 19.75–24.25 Hz), but there was no difference of the power spectral density compared to the no-peak set for the anterior and posterior cingulate gyrus (Supplementary material). Generation of alpha activity in occipital, parietal and temporal lobes Alpha activity was present in a large portion, but not all regions of the occipital lobe (Fig. 6). The cuneus was the region with the highest number of channels showing a clear alpha peak (68% of 19 channels $8Hz, range: 7.75–9.75Hz). This was even more expressed than in the calcarine cortex (58% of 12 channels $9Hz, not significantly different than the nopeak set, presumably because of the relatively low number of channels in the region). There was no significant alpha peak in the inferior occipital gyrus. Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1137 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 127 Apart from the occipital lobe, alpha activity was present, albeit in a lesser degree, in the superior parietal lobule (38% of 53 channels $9 Hz, range: 8.25–9.25 Hz) and in some temporal regions, but at a lower frequency and at the limit of the traditional theta and alpha bands. These findings were significant for the superior temporal gyrus (35% of 79 channels $8 Hz, range: 6.75–9.25 Hz) and the fusiform and parahippocampal gyri (49% of 45 channels $8 Hz, range: 6.75–9.25 Hz) (Fig. 6). No significant peak in alpha activity was seen in the middle and inferior temporal gyrus, the temporal pole and planum temporale, the transverse temporal gyrus and the mesial temporal lobe structures, or in the angular gyrus, supramarginal gyrus, the pars opercularis, precuneus and posterior cingulate gyrus. Some regions show significant peaks in delta activity during wakefulness The most expressed peak in the delta range was found for the hippocampus (72% of 38 channels $1 Hz, range: 0.5–3.25 Hz). Other regions with a delta peak were the inferior occipital gyrus and occipital pole (48% of 23 channels $1.5 Hz, range: 1.25–3.25 Hz), the angular gyrus (38% of 53 channels $1 Hz, range: 0.5–1.25 Hz), the medial frontal cortex (42% of 19 channels $1 Hz, range: 0.5–1.25 Hz), the gyrus rectus/orbital cortex (42% of 45 channels $1 Hz, range: 0.5–1.75 Hz), and the middle temporal gyrus (31% of 129 channels $1 Hz, range: 0.5–1.75 Hz) (Fig. 7). Analysis of the non-oscillatory component of the signal We explored the scale-free brain activity by fitting a 1/fb model to the high frequency end of the spectrum (10–40 Hz). We observed a median value of the b coefficient of 2.29 (range 0.68–4.26; mean 2.35; SD 0.52). We found different median b coefficients at a 5% significance level for the following regions: the b coefficient was higher (steeper decrease of the spectrum for increasing frequency) in the fusiform and parahippocampal gyri (45 channels, median b = 2.84, corrected P = 0.0004) and in the inferior occipital gyrus and occipital pole (23 channels, median b = 2.78, corrected P = 0.024), whereas it was lower in the middle frontal gyrus (178 channels, median b = 2.11, corrected P = 0.001). Figure 8A shows the distribution of the b coefficient in the different brain regions. Regarding the nested-frequencies analysis, we found a low proportion of channels with significant phase-frequency coupling, likely because of the limited duration of the recordings. Figure 8B shows the proportion of channels with statistical significance at 5% without correcting for multiple comparisons (330 phase-amplitude bins), and Fig. 8C corrected for false discovery rate of 5%. In both cases the highest proportion is found between low delta (0.5 Hz) and low alpha band (8–10 Hz). But the proportion of channels in which this coupling was significant is low after correcting for multiple comparisons (28 channels, 1.5% of the 1785 available channels). We found no significant differences between regions. Discussion This is the first atlas of normal intracranial EEG of the adult human brain. Data from physiological channels of many patients with epilepsy were collected to overcome the limited sampling of the brain with intracranial electrodes and the unusual recording from presumably normal brain regions. Furthermore, to superimpose the Figure 5 Spectral density plots of cortical regions showing a significant spectral density peak in `60% of investigated channels. 1138 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 128 results from all subjects on one brain, intracranial electrode positions were standardized in a common 3D environment using a stereotactic template providing the exact coordinates of each recording contact. This atlas of normal brain activity will aid in the better differentiation between physiological and pathological brain activity for clinical work and research in humans. In this paper, results of the quantitative analysis obtained from a large population of human subjects are illustrated. This atlas will be an open resource available for consultation on the web developed under the LORIS framework (Das et al., 2016). Selection of channels from presumably normal brain regions We are aware that selecting ‘true’ normal healthy cortex represents an inaccurate task, since we essentially analyse EEGs from the brains of epileptic patients. In most of these patients, however, some electrodes are placed in non-epileptogenic zones devoid of structural or physiological anomalies. These electrodes are necessary for comparison of cortical physiology and help to define the limits of the eventual surgical resection. Some presumably normal superficial neocortical regions are also recorded as a result of the need to reach a deep structure with a multi-contact electrode. Selecting the most normal brain regions in these patients is as close as we can ever get to a ‘true’ atlas of the normal brain. To ensure as much as possible the selection of channels with normal physiological EEG activity, we followed a strict protocol that involved the consensus of two epileptologists for selection of all the imaging and neurophysiological data for each subject (see ‘Materials and methods’ section). Even if a few channels from pathological regions might have slipped through our careful screening process, it is unlikely to have occurred for many, and the large number of channels in each region makes it likely that our average results are representative of the healthy brain. Intracranial EEG data from different brain regions This atlas extends our knowledge on cortical neurophysiology in humans by providing quantitative and accurate Figure 6 Presence of peaks in alpha activity across the brain. (A) Power spectral density plots of brain regions in the occipital lobe. (B) Power spectral density plots of brain regions outside the occipital lobe with significant alpha activity. Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1139 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 129 Figure 7 Power spectral density plots of brain regions with significant delta activity. Figure 8 Analysis of the non-oscillatory component of the signal. (A) Distribution of the b coefficient of the 1/fb model of the scale free component of the spectrum (10–40 Hz), i.e. the spectrum after removing peaks associated to oscillatory activity. The asterisks indicate regions with coefficient significantly different from the rest. (B) Percentage of channels showing statistically significant phase-amplitude coupling at different frequencies (uncorrected). (C) Percentage of channels showing statistically significant phase-amplitude coupling at different frequencies (corrected for false discovery rate of 5%). 1140 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 130 EEG data for 38 different brain regions. In each region, an average of 2.7 channels per cm3 of cortical grey matter was used for EEG recording. The best coverage was obtained in the areas most frequently explored during epilepsy surgery evaluation, the fronto-temporo-parietal regions. Less frequently explored areas such as the occipital cortex were, however, sufficiently covered. We were also able to identify channels with physiological activity in the mesiotemporal lobe structures, an area often explored, but usually showing epileptic activity. Superimposition of all channels in a common stereotactic 3D environment The lack of standardization in electrodes insertion and placement results in problems performing inter individual comparisons of EEG patterns and, hence, obtaining normative electrophysiological data. To overcome this problem, we used a common 3D environment (IBIS) for semi-automatic co-registration and superimposition of the results from all electrodes from all subjects on one brain template (Mazziotta et al., 2001). To avoid potential bias by a human scorer in the assignment of electrode contacts to anatomical regions, anatomical structures were automatically segmented using an atlas created with an unbiased method (Fonov et al., 2011). This atlas provided detailed volumetric tissue classification of cortical and deep structures (e.g. hippocampus, amygdala). Full brain volume coverage with 66 labels per hemisphere allowed proper assignment of electrode contacts to the anatomical region from which they most likely record. Comparison of stereo-EEG electrodes and cortical grids or strips Compared to scalp EEG, intracranial recordings are not standardized and may show interindividual variations. Stereo-EEG electrodes are in direct contact with the neuronal generators, with a spatial organization that is highly variable and does not follow a standardized electrode positioning like for the 10-20 or 10-10 International EEG System (Jasper, 1958; Nuwer et al., 1998). In contrast, grids/strips have the same distance to the cortical surface and are therefore in the same position with respect to the cellular organization of the cortex. Our study demonstrated that there was no difference in the spectral power distribution between EEG recordings performed with stereo-EEG and cortical grids/strips. The absolute amplitude was two times higher using cortical grids/strips as compared to stereo-EEG. This might be because of the location and orientation of the electrodes with respect to the cortical sources. In case of grids/strips the electrode contacts are always localized atop of the cortex, whereas there are different orientations in stereo-EEG regarding the cellular organization of the cortex. Region-specific EEG signatures as assessed with spectral power analysis After discussing methodological issues we now discuss the main findings. The present work aimed at identifying rhythms characteristic for the different brain regions. To do so we calculated spectral density plots for the different regions investigated. Significances in spectral frequencies were calculated by performing a comparison with a group of automatically selected channels exhibiting no spectral peaks (‘no-peak’ channels). This is a first step in analysing the available data and identifying region-specific characteristics to improve our knowledge on the cortical structural-functional interrelations. Indeed, different methods could have been used. The availability of the present data will allow others to apply different methods and provide further characterization of the human normal iEEG. Distribution of rhythmic activity across different brain regions Current evidence suggests that rhythms are important for the communication between different brain regions (Steriade, 2006; Buzsaki and Schomburg, 2015). To study such rhythms and interactions systematically, it is important to have knowledge of the basic rhythms of each brain region at rest, which is provided by the current atlas. A recent study using MEG in 22 healthy subjects investigated whether rhythmic brain activity is characteristic for different cortical areas. The authors found that individual human brain areas can be identified from their characteristic spectral activation patterns. Moreover, clustering of brain areas according to similarity of spectral profiles reveals known brain networks (Keitel and Gross, 2016). Data of the present atlas could be used to validate these results, and extend them because of the difficulty of MEG to assess deep sources. Another recent study examined brain rhythms of the sensorimotor region in 20 healthy subjects using 64-channel scalp EEG. This study is interesting, although limited to the central region only, as it is a first step towards the identification of cortical areas based on neuronal dynamics rather than on cytoarchitectural features (Cottone et al., 2017). An extension of this work could be to apply this methodology to the intracranial atlas. Focal generation of alpha activity Early work in the canine model demonstrated that the alpha rhythm is generated in the cerebral cortex (Lopes da Silva and Storm Van Leeuwen, 1979). These data obtained by invasive EEG are complemented by more recent non-invasive studies using source analysis showing that the alpha rhythm is due to more than one generator in the posterior cortex. Moreover, a thalamic modulating influence has been suggested (Tyvaert et al., 2008; Chang et al., 2011). Findings from electrocorticography and intracerebral depth EEG studies described alpha activity in the Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1141 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 131 occipital lobe, the parietal lobe just posterior to the postcentral gyrus, and in the posterior temporal neocortex (Frauscher and Dubeau, 2018). Our present work confirmed in part these findings. Interestingly and in contrast to what we initially suspected, a clear peak in alpha activity was not found in all occipital lobe structures; it was most prominent in the mesio-occipital lobe and in particular in the cuneus. Alpha activity was further seen in the superior parietal lobe, as well as the temporal lobe. In the latter, alpha activity was of a lower frequency compared to the occipital lobe. Moreover, there was a peak in alpha activity in the postcentral gyrus. We did not find alpha activity in the frontal lobe, as previously described (Jasper and Penfield, 1949). Peak in beta activity in motor function-related cortical areas In the motor system, oscillations in the beta frequency band have been suggested to play a role in sensorimotor integration (Baker, 2007). Of note, a vast majority of motor function-related eloquent neocortical areas such as the precentral gyrus, the supplementary motor cortex, the mid-cingulate, the anterior insula, the pars triangularis of the inferior frontal gyrus, and the operculum showed a significant peak in the beta frequency band during resting state. This beta peak was most evident in the precentral gyrus, as suggested by previous work during acute electrocorticography (Jasper and Penfield, 1949). In line with our findings, studies examining the execution of voluntary movements showed, in primary sensorimotor areas, a desynchronization of beta activity with the motor response, followed by the appearance of an event-related synchronization in the gamma range between 40 and 60 Hz, and reappearance of beta activity (Sem-Jacobsen et al., 1956; Pfurtscheller and Lopes da Silva, 1999; Szurhaj and Derambure, 2006). Presence of frequency peaks in the different bands depends on the brain region A new finding of this study is that there is more than one peak in certain brain regions, even within the same EEG channel. One example is the precentral gyrus. All channels in this cortical region showed more than one peak in the delta, theta and high beta range. Previous work using acute electrocorticography showed a fast beta rhythm of 25 Hz similar to that in our study in the precentral area (Jasper and Penfield, 1949), but did not refer to underlying slower frequency bands. We further confirmed peaks in alpha and beta activity in the postcentral gyrus. This finding is in line with the notion that there is less beta activity in the postcentral gyrus when performing a direct comparison to the precentral gyrus (Jasper and Penfield, 1949). It is noteworthy that there are certain brain regions, particularly in the temporal lobe or in deep structures such as the anterior and posterior cingulate gyrus as well as the posterior insula, which did not show any difference (statistically significant or perceptible visually) with the set of channels with no identifiable spectral peaks. For other regions, such as the calcarine cortex, there was a clear peak in certain frequency bands, which, however, was not significant, most likely because of the relatively small number of investigated channels. Presence of focal cortical peaks in the delta band in the awake brain Slower frequencies in the theta delta range were found in frontobasal, as well as frontomesial and temporomesial structures. Delta activity in the ventral medial frontal brain portions was already suggested by Sem-Jacobsen et al. (1955) and for the orbitofrontal cortex by Nishida et al. (2004). Slow wave activity in the hippocampus, consisting of irregular slow waves in the delta and theta frequency range was also already reported in the past (Jasper and Penfield, 1949; Brazier, 1968). Interestingly, previous studies showed that the EEG activity in the hippocampus is task-dependent (Bo´dizs et al., 2001; Cantero et al., 2003; Clemens et al., 2009; Lega et al., 2012; Moroni et al., 2012; Watrous et al., 2013; Billeke et al., 2017). Findings of the present study are in line with the described ‘large irregular activity’ characterized by hippocampal sharp waves occurring during immobility. In this context, it is noteworthy to stress that in humans, this slow wave activity is in the delta range; in animals it was shown to be in the theta range (Bo´dizs et al., 2001; Clemens et al., 2009; Watrous et al., 2013). Analysis of the non-oscillatory component of the signal The classical frequency bands in EEG are described based on the various oscillatory rhythms that appear in the EEG. Apart from this oscillatory component, the EEG has a significant, if not major, component of non-oscillatory or scale free activity. We analysed both components and observed differences in the non-oscillatory EEG signal in different brain regions, with higher values in the fusiform and parahippocampal gyri, as well as the inferior occipital gyrus and occipital pole. All these regions are involved in visual processing. Interestingly, this is in keeping with a study analysing scale free activity with functional MRI that showed increased values for the visual cortex (He et al., 2010). We further analysed our data for nested frequencies. We found the strongest phase amplitude coupling between the phase of very low frequency activity at 0.5 Hz and the low alpha band amplitude. The low proportion of channels with significant coupling compared e.g. to He et al. (2010), might be explained by the short length of the recordings in the awake/eyes closed condition. 1142 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 132 Conclusion and future prospects This is the first atlas of normal intracranial EEG activity in a common stereotactic space that allows the accumulation of small amounts of data from each of a large group of subjects, to obtain an atlas covering well most cortical regions. It will aid the neurophysiologist to understand the EEG frequency distributions across the different cortical regions of the human brain better. This atlas provides a normative baseline against which clinical EEGs and experimental results can be compared. It will be an open resource available for augmentation and consultation on the web (https://mni-open-ieegatlas.research.mcgill.ca). We are currently preparing an atlas on sleep, as sleep data are available from most of the subjects and we will be able to compare wake and sleep across the different investigated regions. Also, we still have some regions that are not fully covered, but we intend to allow the atlas to grow by inclusion of data from the original centres as well as from other qualified centres, and to continue the current project in a prospective way. This will also allow to collect longer segments of controlled awake/eyes closed condition for different types of analysis, such as that of nested frequencies. Acknowledgements The authors wish to express their gratitude to Louis Collins, PhD, Vladimir Fonov, PhD, as well as Alan Evans, PhD and the LORIS team from the McConnell Brain Imaging Centre of McGill University, as well as the staff and technicians at the EEG Department at the Montreal Neurological Institute and Hospital, Lorraine Allard, Nicole Drouin, Chantal Lessard and Linda Me´nard, the staff and technicians at the Neurophysiopathology Laboratory of Grenoble-Alpes University Hospital, Dr. Anne-Sophie Job-Chapron, Patricia Boschetti, and Marie-Pierre Noto, and the staff and technicians at the Centre Hospitalier de l’Universite´ de Montre´al. Funding This work was supported by the Savoy Epilepsy Foundation (project grant to B.F. and post-doctoral fellowship to R.Z.), the Botterell Powell’s Foundation (grant to B.F.), and the Canadian Institute of Health Research (grant FDN-143208 to J.G.). Conflicts of interest The authors have no potential conflict of interest with the present study. Outside of the submitted work, B.F. received speaker/advisory board fees sponsored by UCB. J.G. and N.v.E. have received fees for consultancy from Precisis Inc. P.K. received speaker/advisory board fees from UCB and Eisai. D.K.N., D.H., F.D., J.H., L.M., A.O., and R.Z. have nothing to declare. Supplementary material Supplementary material is available at Brain online. References Baker SN. Oscillatory interactions between sensorimotor cortex and the periphery. Curr Opin Neurobiol 2007; 17: 649–55. Billeke P, Ossandon T, Stockle M, Perrone-Bertolotti M, Kahane P, Lachaux JP, et al. Brain state-dependent recruitment of high-frequency oscillations in the human hippocampus. Cortex 2017; 94: 87–99. Bo´dizs R, Ka´ntor S, Szabo´ G, Szuˆcs A, Ero˜ss L, Hala´sz P. Rhythmic hippocampal slow oscillation characterizes REM sleep in humans. Hippocampus 2001; 11: 747–53. Brazier MA. Studies of the EEG activity of limbic structures in man. Electroencephalogr Clin Neurophysiol 1968; 25: 309–18. Buzsaki G, Schomburg EW. What does gamma coherence tell us about inter-regional neural communication. Nat Neurosci 2015; 18: 484–9. Cantero JL, Atienza M, Stickgold R, Kahana MJ, Madsen JR, Kocsis B. Sleep-dependent theta oscillations in the human hippocampus and neocortex. J Neurosci 2003; 23: 10897–903. Chang BS, Schomer DL, Niedermeyer E. Normal EEG and sleep: adults and elderly. In: Schomer DL, Lopes da Silva FH, editors. Niedermeyer’s encephalography. 6th edn. Philadelphia, PA: Lippincott Williams & Wilkins; 2011. p. 183–215. Chatrian GE. The lambda waves. In: Remond A, editor. Handbook of electroencephalography and clinical neurophysiology. vol. 6A. Amsterdam: Elsiever, 1976. p. 123–49. Chatrian GE, Bickford RG, Uihlein A. Depth electrographic study of a fast rhythm evoked from the human calcarine region by steady illumination. EEG Clin Neurophysiol 1960a; 12: 167–76. Chatrian GE, Bickford RG, Petersen MC, Dodge HW, Lazarte JA, Holman CB. Lambda waves with depth electrodes in humans. In: Ramey ER, O’Doherty DS, editors. Electrical studies on the unanesthesized brain. New York: Hoeber PB; 1960b. p. 291–310. Clemens Z, Weiss B, Szucs A, Eross L, Rasonyi G, Halasz P. Phase coupling between rhythmic slow activity and gamma characterizes mesiotemporal rapid-eye-movement sleep in humans. Neurosci 2009; 163: 388–96. Cottone C, Porcaro C, Cancelli A, Olejarczyk E, Salustri C, Tecchio F. Neuronal electrical ongoing activity as a signature of cortical areas. Brain Struct Funct 2017; 222: 2115–26. Das S, Glatard T, MachIntyre LC, Madjar C, Rogers C, Rousseau ME, et al.. The MNI data-sharing and processing ecosystem. Neuroimage 2016; 124: 1188–95. David O, Job AS, De Palma L, Hoffmann D, Minotti L, Kahane P. Probabilistic functional tractography of the human cortex. Neuroimage 2013; 80: 307–17. Drouin S, Kochanowska A, Kersten-Oertel M, Gerard IJ, Zelmann R, De Nigris D et al. IBIS: an OR ready open-source platform for image-guided neurosurgery. Int J Comput Assist Radiol Surg 2017; 12: 363–78. Fonov V, Evans AC, Botteron K, Almli CR, McKinstry RC, Collins DL, et al. Unbiased average age-appropriate atlases for pediatric studies. Neuroimage 2011; 54: 313–27. Physiological awake intracranial EEG BRAIN 2018: 141; 1130–1144 | 1143 Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 133 Frauscher B, Dubeau F. Physiological activity and artifacts in the human epileptic brain studied with intracerebral depth EEG. In: Lhatoo S, Kahane P, Lu¨ders H, editors. Invasive studies of the human epileptic brain: principles and practice of invasive brain recordings and stimulation in epilepsy. Oxford, UK: Oxford University Press; 2018, in press. Frauscher B, von Ellenrieder N, Dubeau F, Gotman J. Scalp spindles are associated with widespread intracranial activity with unexpectedly low synchrony. Neuroimage 2015; 105: 1–12. Gastaut H. Enregistrement sous-cortical de l’activiteelectriquespontanee et provoquee du lobe occipital humain. EEG Clin Neurophysiol 1949; 1: 2005–21. Graf M, Niedermeyer E, Schiemann J, Uematsu S, Long DM. Electrocorticography: Information derived from intraoperative recordings during seizure surgery. Clin Electroencephalogr 1984; 15: 83–91. He BJ, Zempel JM, Snyder AZ, Raichel ME. The temporal structures and functional significance of scale-free brain activity. Neuron 2010; 66: 353–69. Huh K, Meador KJ, Lee GP, Loring DW, Murro AM, King DW, et al. Human hippocampal EEG: effects of behavioral activation. Neurology 1990; 40: 1177–81. Jasper HH. The ten–twenty electrode system of the International Federation. Electroencephalogr Clin Neurophysiol 1958; 10: 367–80. Jasper H, Penfield W. Electrocorticograms in man: effect of voluntary movement upon the electrical activity of the precentral gyrus. Arch Psychiatr Z Neurol 1949; 183: 163–74. Jerbi K, Ossando´n T, Hamame´ CM, Senova S, Dalal SS, Jung J, et al. Task-related gamma-band dynamics from an intracerebral perspective: review and implications for surface EEG and MEG. Hum Brain Mapp 2009; 30: 1758–71. Kahane P. Activite´s e´lectriques intra-ce´re´brales de repos dites normales chez l’homme e´pileptique: essai de caracte´risation et proble`mes me´thodologiques. PhD thesis. Universite´ J. Fourier, 1993. Keitel A, Gross J. Individual human brain areas can be identified from their characteristic spectral activation fingerprints. PLoS Biol 2016; 14: e1002498. Landman BA, Warfield SK, editors. MICCAI 2012 workshop on multi-atlas labeling. Create Space Independent Publishing Platform; 2012. Lega AJ, Jacobs J, Kahana M. Human hippocampal theta oscillations and the formation of episodic memories. Hippocampus 2012; 22: 748–61. Lopes da Silva FH, Storm van Leeuwen W. The cortical source of the alpha rhythm. Neurosci Lett 1977; 6: 237–41. Matsumoto R, Kunieda T, Nair D. Single pulse electrical stimulation to probe functional and pathological connectivity in epilepsy. Seizure 2017; 44: 27–36. Mazziotta J, Toga A, Evans A, Fox P, Lancaster J, Zilles K, et al. A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM). Philos Trans R Soc Lond B Biol Sci 2001; 356: 1293–322. Meador KJ, Thompson JL, Loring DW, Murro AM, King DW, Gallagher BB, et al. Behavioral state-specific changes in human hippocampal theta activity. Neurology 1991; 41: 869–72. Moroni F, Nobili L, De Carli F. Slow EEG rhythms and inter-hemispheric synchronization across sleep and wakefulness in the human hippocampus. Neuroimage 2012; 60: 497–504. Nishida M, Hirai N, Miwakeichi F, Maehara T, Kawai K, Shimizu H, et al. Theta oscillation in the human anterior cingulate cortex during all-night sleep: an electrocorticographic study. Neurosci Res 2004; 50: 331–41. Nuwer MR, Comi G, Emerson R, Fuglsang-Frederiksen A, Gue´rit JM, Hinrichs H, et al. IFCN standards for digital recording of clinical EEG. International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol 1998; 106: 259–61. Perez-Borja C, Chatrian GE, Tyce FA, Rivers MH. Electrographic patterns of the occipital lobe in man: a topographic study based on use of implanted electrodes. Electroencephalogr Clin Neurophysiol 1962; 14: 171–82. Petersen MC, Bickford G, Sem-Jacobsen CW, Dodge HW Jr. The depth electrogram in schizophrenic patients. Proc Staff Meet Mayo Clin 1953; 28: 170–5. Pfurtscheller G, Lopes da Silva FH. Event-related EEG/MEG synchronization and desynchronization: basic principles. Clin Neurophysiol 1999; 110: 1842–57. Ribstein M. Exploration of the human brain by depth electrodes. [in Freanch]. Electroencephalogr Clin Neurophysiol Suppl 1960; Suppl 16: 1–129. Sem-Jacobsen CW, Bickford RG, Petersen MC, Dodge HW. Depth distribution of normal electroencephalographic rhythms. Proc Staff Meet Mayo Clin 1953; 28: 156–61. Sem-Jacobsen CW, Petersen MC, Dodge HW, Layarte JA, Holman CB. Electroencephalographic rhythms from the depths of the frontal lobe in 60 psychotic patients. EEG Clin Neurophysiol 1955; 7: 193–210. Sem-Jacobsen CW, Petersen MC, Dodge HW, Layarte JA, Holman CB. Electroencephalographic rhythms from the depths of the parietal, occipital and temporal lobes in man. EEG Clin Neurophysiol 1956; 8: 263–78. Steriade M. Grouping of brain rhythms in corticothalamic systems. Neuroscience 2006; 137: 1087–106. Szurhaj W, Derambure P. Intracerebral study of gamma oscillations in the human sensorimotor cortex. Prog Brain Res 2006; 159: 297–310. Tort AB, Kramer MA, Thorn C, Gibson DJ, Kubota Y, Graybiel AM, et al. Dynamic cross-frequency couplings of local field potential oscillations in rat striatum and hippocampus during performance of a T-maze task. Proc Natl Acad Sci USA 2008; 105: 20517–22. Tyvaert L, Levan P, Grova C, Dubeau F, Gotman J. Effects of fluctuating physiological rhythms during prolonged EEG-fMRI studies. Clin Neurophysiol 2008; 119: 2762–74. Watrous AJ, Lee DJ, Izadi DJ, Gurkoff GG, Shahlaie K, Ekstrom AD. A comparative study of human and rat hippocampal low frequency oscillations during spatial navigation. Hippocampus 2013; 23: 656–61. Yamamoto Y, Hughson RL. Extracting fractal components from time series. Physica D 1993; 68: 250–64. 1144 | BRAIN 2018: 141; 1130–1144 B. Frauscher et al. Downloadedfromhttps://academic.oup.com/brain/article-abstract/141/4/1130/4915909bygueston29April2020 134 Annex 4: Dolezalova I, Brazdil M, Hermanova M, Horakova I, Rektor I, Kuba R. Intracranial seizure onset patterns in unilateral temporal lobe epilepsy and their relationship to other variables. Clinical Neurophysiology. 2013;124(6):1079-1088. 135 Intracranial EEG seizure onset patterns in unilateral temporal lobe epilepsy and their relationship to other variables Irena Dolezˇalová a,b,⇑ , Milan Brázdil a,b , Markéta Hermanová b,c , Iva Horáková c , Ivan Rektor a,b , Robert Kuba a,b a Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic b Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic c First Department of Pathological Anatomy, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic a r t i c l e i n f o Article history: Accepted 28 December 2012 Available online 13 February 2013 Keywords: Temporal lobe epilepsy Invasive EEG Outcome Histopathology Localization Frequency Predictive factors h i g h l i g h t s  Very early seizure onset frequencies in invasive EEG in patient with temporal lobe epilepsy could be predictive factor for their outcome.  Higher frequencies (P8 Hz) could indicate better surgical outcome.  Slower frequencies (<8 Hz) could imply worse surgical outcome. a b s t r a c t Objective: We performed a retrospective study to determine the different types of seizure onset patterns (SOP) in invasive EEG (IEEG) in patients with temporal lobe epilepsy (TLE). Methods: We analyzed a group of 51 patients (158 seizures) with TLE who underwent IEEG. We analyzed the dominant frequency during the first 3 s after the onset of ictal activity. The cut-off value for distinguishing between fast and slow frequencies was 8 Hz. We defined three types of SOPs: (1) fast ictal activity (FIA) – frequency P8 Hz; (2) slow ictal activity (SIA) – frequency <8 Hz; and (3) attenuation of background activity (AT) – no clear-cut rhythmic activity during the first 3 s associated with changes of IEEG signal (increase of frequency, decrease of amplitude). We tried to find the relationship between different SOP types and surgery outcome, histopathological findings, and SOZ localization. Results: The most frequent SOP was FIA, which was present in 67% of patients. More patients with FIA were classified postoperatively as Engel I than those with SIA and AT (85% vs. 31% vs. 0) (P < 0.001). There were no statistically significant differences in the type of SOP, in the histopathological findings, or in the SOZ localization. Conclusion: In patients with refractory TLE, seizure onset frequencies P8 Hz during the first 3 s of ictal activity are associated with a better surgical outcome than frequencies <8 Hz. Significance: Our study suggests that very early seizure onset frequencies in IEEG in patients with TLE could be the independent predictive factor for their outcome, regardless of the localization and etiology. Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. 1. Introduction Temporal lobe epilepsy (TLE) is the most common form of drug resistant partial epilepsy (Williamson et al., 1998). Epilepsy surgery is a well-established and evidence-based method for treatment in TLE patients (Wiebe et al., 2001). In the context of epilepsy surgery, the precise identification of the epileptogenic zone, which is defined as the zone responsible for seizure generation and which has to be removed completely for seizure freedom, is crucial (Bancaud et al., 1965; Lüders and Awad, 1992; Rosenow and Lüders, 2001). There is currently no diagnostic modality that can be used to measure the entire epileptogenic zone directly. We can deduce the extent of the epileptogenic zone only indirectly from the seizure onset zone (SOZ) (Rosenow and Lüders, 2001). In cases where the SOZ cannot be properly identified on the basis of non-invasive investigations, invasive electroencephalography (IEEG) monitoring is a well-established method for its localization (Wyler et al., 1984, 1993; Devinsky et al., 1989; Spencer, 1989; Sperling and O’Connor, 1989; Spencer et al., 1990, 1993; So, 1992, 1995; Arroyo et al., 1993; Spencer and Lamoureux, 1996). 1388-2457/$36.00 Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinph.2012.12.046 ⇑ Corresponding author at: Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Pekarˇská 53, 656 91 Brno, Czech Republic. Tel.: +420 543 182 645; fax: +420 543 182 624. E-mail address: irena.dolezalova@fnusa.cz (I. Dolezˇalová). Clinical Neurophysiology 124 (2013) 1079–1088 Contents lists available at SciVerse ScienceDirect Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph 136 The major limitation of IEEG is the recording of cerebral activity from an extremely confined region; the SOZ can be missed or falsely localized to the regions with propagated electrographic seizure activity (Spencer et al., 1982, 1993; Gloor, 1984; So, 1992, 1995; Arroyo et al., 1993; Spencer and Lamoureux, 1996). It has been suggested that the different morphologies of intracranial EEG seizure onsets have different degrees of localizing value (Lieb et al., 1981; Engel et al., 1989; Engel, 1990; Faught et al., 1992; Weinand et al., 1992; Spencer, 1994; Park et al., 1996). Published studies with subdural (Weinand et al., 1992) and epidural (Faught et al., 1992) electrodes in patients with TLE showed that fast activity at seizure onset can be associated with seizure freedom after surgery. In some studies, slower ictal onset frequencies, mainly from the theta range, in intracranial electrodes in patients with TLE can suggest a ‘‘propagated’’ electrographic pattern (Schiller et al., 1998). Engel et al. (1989) separated patients with TLE into two groups according to their ictal onset patterns: hypersynchronous patterns (HYP) and low-voltage fast patterns (LVF), which are described in 70–80% of patients (Engel, 1990; Velasco et al., 2000; Ogren et al., 2009). Many investigators focus on the qualities of these two patterns, mainly in terms of the predominant localization and histopathological findings (Lieb et al., 1981; Townsend and Engel, 1991; Spanedda et al., 1997; Spencer et al., 1992a,b; Velasco et al., 2000; Schiller et al., 1998; Ogren et al., 2009). Patients who cannot be classified according to these criteria are not included in most studies of seizure classification (Bragin et al., 2005), resulting in the loss of important information. In the present study, patients with medically refractory TLE were categorized into three groups according to the seizure onset frequencies recorded in IEEG. These three groups were then correlated according to surgical outcomes, SOZ localizations, and histopathological findings. We hypothesized that this approach would reveal variations between frequency groups and patient characteristics. 2. Methods 2.1. Patient selection In this retrospective study, we included all patients with refractory TLE who underwent epilepsy surgery at the Brno Epilepsy Center of Masaryk University Hospital between 1996 and 2011 and who had undergone IEEG prior to the surgery. Other inclusion criteria were: (1) IEEG showing that SOZ is restricted to unilateral temporal lobe structures (we excluded patients with bitemporal independent SOZs and patients with contemporary involvement of temporal and extratemporal lobe structures); (2) surgery performed according to non-invasive data and IEEG results; and (3) at least 1 year of follow-up after the surgery. We included a total of 51 patients who fulfilled these criteria. All patients gave their informed consent. The study was approved by the ethics committee of St. Anne’s University Hospital. 2.2. Presurgical evaluation All 51 patients underwent a comprehensive evaluation, including detailed history and neurological examination, neuropsychological testing, magnetic resonance imaging (MRI), and both scalp and invasive video-EEG monitoring. Fluorodeoxyglucose-positron emission tomography (FDG-PET) and/or interictal single photon emission tomography (SPECT) was performed in all patients; other functional neuroimaging techniques such as ictal SPECT and subtraction ictal SPECT co-registered to MRI (SISCOM) were performed only in patients with incongruent preoperative data. In total, 37 patients had MRI abnormalities prior to surgery. The remaining 14 patients showed no clear MRI abnormalities. 2.3. Invasive EEG procedure All patients underwent the IEEG procedure. Of the 51 patients, 47 were investigated with chronically stereotactically implanted intracerebral electrodes. The number of electrodes per patient ranged from 3 to 14 (with an average of 7.9 ± 2.2). In 4 patients, we used a combination of 1 or 2 stereotactically implanted intracerebral electrodes and subdural strip electrodes. In three cases 8-contact strips were used, and in 1 case a combination of 6-contact and 8-contact strips was used. In 20 out of 51 patients (40%), the exploration was limited to temporal lobe structures; in the other 31 patients (60%), extratemporal structures were explored as well. The anatomical targeting of intracerebral electrodes and the position of subdural strips were established in each patient according to available non-invasive information and hypotheses about the localization of SOZ. Intracerebral 5-, 10-, and 15-contact platinum semiflexible Microdeep electrodes DIXI or ALCIS (with an electrode diameter of 0.8 mm, a contact length of 2 mm, and a 1.5 mm distance between contacts) and 6- and 8-contact platinum subdural strip electrodes (Radionics) were used. The invasive video-EEG recording was performed with the 64-channel Brain Quick system (Micromed), and the 128-channel Alien Deymed system was used for intracranial video-EEG recording. Monopolar recordings (with a reference electrode on the processus mastoideus) and special bipolar montages were used to evaluate the EEG activity. EEG was amplified with a bandwidth of 0.4–100 Hz at a sampling rate of 128 Hz. 2.4. Invasive EEG analysis The character and localization of seizure onset (SO) were analyzed independently by two of the authors (ID & RK) and disagreement was resolved by consensus after review. SO was assessed visually and defined as a sustained rhythmic change in the IEEG accompanied by subsequent clinically typical sei- zureactivity,notexplainedbylevelofarousalandclearlydistinguished from background IEEG and interictal activity (Spencer et al., 1992b). We examined only seizures with clinical accompaniment, because the significance of pure electrographic seizures is not clearly determined (Sperling and O’Connor, 1990). We divided the SOs according to their localization, as (1) mesial (amygdala, hippocampus, parahippocampal gyrus), (2) temporopolar, and (3) laterobasal (laterobasal cortex dorsally from temporal pole). The temporal pole delineation defined by Chabardès et al. (2002) was applied. The posterior border of the temporal pole region was taken as an oblique line joining the rostral ends of the superior and inferior temporal sulci projected onto another line passing through the anterior edge of the temporofrontal transitional fold. On the basal side, the border passed through the rostral end of the inferior temporal sulcus and terminated at the rhinal sulcus in the mesial part of the hemisphere (Chabardès et al., 2002). We analyzed the frequency of SO during the first 3 s of ictal activity as defined above. The dominant frequency of each SO was determined based on spectral analysis, using the fast Fourier transform algorithm, and was defined as a frequency with its maximal peak in the given spectrum. The dominant frequency was measured in the contact where the ictal activity first appeared. If the simultaneous involvement of two or more contacts was present, only the highest dominant frequency was used for analysis. We conducted the frequency analysis according to the following principles:  Some types of initial EEG changes, i.e., brief bursts of spikes and attenuation of the background activity lasting less than 3 s, were not considered part of seizure onset (Lee et al., 2000). If they were the initial EEG changes, we used the frequency of subsequent ictal activity. 1080 I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 137  If there was a rhythmic pattern in some electrodes and attenuation of background activity in others, we classified this SO according to the frequency of the rhythmic activity, following a similar principle used by Lee et al. (2000).  If the first change was an attenuation lasting more than 3 s and no rhythmic activity was present simultaneously, we defined it as a unique pattern. The patients were divided into two groups according to the dominant SO frequency. The cut-off value to distinguish between fast and slow frequencies was 8 Hz, which was established as the lower borderline of the alpha frequency range. When 2 or 3 SOPs were present, the predominant pattern was used for the analysis; Lee et al. applied the same concept (Lee et al., 2000). This predominant pattern was defined as the most common pattern (more than half of all seizures) in an individual. If we were not able to determine the predominant pattern in a patient, this patient was excluded from further analysis. Based on these criteria, we defined three types of seizure onset patterns (SOP): Fast ictal activity (FIA) – frequency Pthan 8 Hz (Fig. 1); slow ictal activity (SIA) – frequency 15 Hz). (2) Similarly, SIA was subdivided into 2 bands with a cut-off value of 2 Hz, i.e. frequencies 62 Hz (SIA 6 2 Hz) and frequencies >2 Hz (SIA > 2 Hz). This separation was based on the works of Engel et al. (1989), who designated ictal onset activity of frequencies 62 Hz as a HYP pattern. 2.5. Surgical procedure The extent of the resection was based on the results of a presurgical evaluation including non-invasive and invasive data. Engel’s Fig. 1. Fast ictal activity (FIA). (A) Localization of intracerebral electrodes: electrodes on the left side: A0 – mesial, amygdala, lateral middle temporal gyrus; B0 – mesial anterior hippocampus, lateral middle temporal gyrus; C0 – mesial posterior hippocampus, lateral middle temporal gyrus; D0 – middle temporal gyrus (posterior part); T0 – superior temporal gyrus; Tp0 – temporal pole; I0 – frontal operculum; O0 – orbitofrontal cortex. (B) Ictal findings: the arrow indicates the development of FIA in B0 1–4 (hippocampus on the left side). I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 1081 138 classification (Engel et al., 1987) was used to evaluate surgical effectiveness. 2.6. Neuropathological examination Evaluable formalin-fixed paraffin-embedded tissues of temporal lobe resection specimens were available from all patients. The paraffin-embedded tissue specimens, slides, and histopathology reports were retrieved from the files of the First Department of Pathological Anatomy of St. Anne’s University Hospital and independently re-evaluated by two histopathologists exprerienced in the evaluation of specimens obtained from epilepsy surgery patients (MH & IH), and discrepancies were resolved by consensus. All examined resected tissues were identically treated, fixed in a Fig. 2. Slow ictal activity (SIA). (A) Localization of intracerebral electrodes: electrodes on the left side: A0 – mesial amygdala, lateral middle temporal gyrus; B0 – mesial anterior hippocampus, lateral middle temporal gyrus; C0 – mesial posterior hippocampus, lateral middle temporal gyrus; D0 – middle temporal gyrus; P0 – temporal pole, T0 – superior temporal gyrus (posterior part); Z0 – superior temporal gyrus (anterior part); Po0 – precuneus; Fo0 – frontal operculum; O0 – orbitofrontal cortex. (B) Ictal findings: the arrow indicates the development of SIA in Z0 1–4 (left superior temporal gyrus). Fig. 3. Attenuation of background activity (AT). (A) Localization of intracerebral electrodes: electrodes on the left side: A0 – mesial amygdala, lateral middle temporal gyrus; B0 – mesial anterior hippocampus, lateral middle temporal gyrus; C0 – mesial posterior hippocampus, lateral middle temporal gyrus; T0 – superior temporal gyrus; Tp0 – temporal pole; O0 – orbitofrontal cortex; G0 – mesial anterior cingulate gyrus, lateral middle frontal gyrus; P0 – mesial medial frontal gyrus, lateral superior frontal gyrus; Fo0 – frontal operculum. Electrodes on the right side: B – mesial anterior hippocampus, lateral middle temporal gyrus; C – mesial posterior hippocampus, lateral middle temporal gyrus. (B) Ictal findings: AT appears initially in both left-sided and in right-sided electrodes (frame); this is followed, in approximately 5 s, by rhythmic ictal activity. 1082 I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 139 10% neutral buffered formalin, grossly inspected, measured, and cut so as to obtain representative tissue slices perpendicular to the cortical surface. Representative tissue slices were routinely processed and paraffin embedded. Five micrometers thick tissue sections were stained by hematoxylin and eosin, evaluated under light microscope, and reported. NeuN immunohistochemistry (using mouse monoclonal anti-NeuN antibody, dilution 1:100, clone A-60, Millipore), was performed on preselected tissue sections if there was an inconclusive picture in hematoxylin and eosin. Focal cortical dysplasias were classified according the classification system reported by Blümcke et al. (2011). 2.7. Statistical analysis The Kruskal–Wallis test was used to test possible relationships between individual SOPs and patient demographic data, i.e. age at epilepsy onset, epilepsy duration, and age at time of surgery. The Wilcoxon–Mann–Whitney test was used to compare average frequencies between groups of patients with different Engel outcomes. The chi-square test was used to analyze the relationship between individual SOPs and surgical outcome, SOZ localization, and histopathological findings. The chi-square test was also used to analyze the relationship between surgical outcome and SOZ localization and histopathological findings. P-value <0.05 was considered to be statistically significant. 3. Results 3.1. Demographic data The study included 51 patients: 20 females and 31 males. Patient age at the time of surgery ranged from 13 to 54 years (with an average of 31.5 ± 6.6 years). The age at epilepsy onset ranged from 0 to 42 years (with an average of 13.4 ± 7.4 years), and epilepsy duration ranged from 4 to 36 years (with an average of 18.3 ± 7.9 years). All data concerning the whole series of our patients are summarized in Table 1. 3.2. Surgery and outcome Surgery was performed on the right side in 26 out of 51 patients (51%) and on the left side in 25 (49%). Standard anteromesial temporal lobe resection (AMTR) was done in 36 patients (70%), tailored lesionectomy in 12 patients (24%), AMTR plus tailored topectomy of the lateral temporal cortex in 2 patients (4%), and temporal pole resection sparing the hippocampus in 1 patient (2%). Using Engel’s classification, 33 (65%) patients were classified as Engel I, 5 (10%) as Engel II, 5 (10%) as Engel III, and 8 (15%) as Engel IV. Patients were divided into two groups according to outcome: Engel I (excellent outcome) and Engel II–IV. The duration of follow-up ranged from 1 to 15 years (with an average of 8.3 ± 4.6 years). 3.3. Ictal onset frequency and patterns We recorded 158 seizures in 51 patients. In all patients, the first detectable changes were present in intracranial electrodes. We excluded 6 seizures from the analysis (5 for artifacts preventing proper interpretation of invasive EEG, 1 for damage of electrodes). The remaining 152 seizures were distributed from 1 to 6 seizures per patient (with average of 3.0 ± 1.1). We found that 43 (84%) of 51 patients had an exclusive SOP (SIA, FIA, or AT). The other 8 patients were categorized according to their predominant SOP. FIA was present in 34 (67%) patients, SIA in 13 (25%), and AT in 4 (8%) (Fig. 4). There were no statistically significant differences in patients with different SOPs in terms of age at epilepsy onset, epilepsy duration, or age at time of surgery (Table 2). 3.4. Relationship of SOPs to outcome We found a borderline statistically significant difference in average frequencies between patients with Engel I outcome and patients with Engel II–IV outcomes. The average frequency was 13.3 ± 6.0 Hz in patients with Engel I and 9.2 ± 7.9 Hz in patients classified as Engel II–IV (P = 0.081). We found that 29 (85%) of the 34 patients with FIA had an Engel I outcome, in comparison with 4 (31%) of the 13 with SIA and 0 (0%) of the 4 AT (P < 0.001) (Table 3). When the FIA patients were subdivided (cut-off value 15 Hz) into two subgroups, 18 patients (53%) had FIA 8–15 Hz and 16 (47%) had FIA > 15 Hz (Table 4). The patients with FIA 8–15 Hz had a better prognosis: 17 (94%) had an Engel I outcome; 12 patients (75%) with FIA > 15 Hz had an Engel I outcome. The difference between these two groups was not statistically significant (P = 0.164). When we subdivided patients with SIA into two subgroups (cutoff value 2 Hz), we saw that 7 (54%) patients had SIA 6 2 Hz and 6 (46%) patients had SIA > 2 Hz (Table 4). SIA 6 2 Hz was associated with a better outcome than SIA > 2 Hz: 4 (57%) patients with SIA 6 2 Hz had an excellent outcome (Engel I), in comparison with 0 (0%) with SIA > 2 Hz, this difference was borderline statistically significant (P = 0.070). 3.5. Localization of SOZ, histopathological examination and SOP SO was characterized as mesial in 38 patients (75%), as temporopolar in 7 (13%), and as laterobasal in 6 (12%). There were no statistical differences in SOZ localization and surgical outcome or SOP (Tables 5 and 6). In 31 out of 51 patients (60%), extratemporal electrodes were used, which enabled us to study the speed of extratemporal propagation. Three patients with FIA exhibited fast involvement (within 2 s) of the frontal lobe structures. In Table 1, they are numbered as 28, 35, and 48. No patient with SIA showed such fast involvement of the extratemporal structures. In all of the patients with AT, extratemporal electrodes were used. Three out of 4 patients exhibited early widely-distributed and long-lasting attenuation; in the remaining patient with AT, the attenuation was limited to the temporal pole area. After that, rhythmic ictal activity developed; this activity was initially limited to temporal regions and subsequently spread to the extratemporal regions. A histopathological examination was performed in all 51 patients. The most common pathology was hippocampal sclerosis (HS), which was found in 20 patients (39%); both HS and focal cortical dysplasia (FCD) were found in 2 patients (4%). For further analysis, we grouped these 22 (43%) patients into one group. In 6 patients (12%), isolated FCD was present; all these cases were classified as FCD Ia. Other types of lesion were found in 12 cases (23%): non-specific gliosis in 5, dysembryoplastic neuroepithelial tumor (DNET) in 4, other benign tumors in 2, and a vascular malformation was revealed in 1. In the remaining 11 (22%) patients, the histopathological specimen analysis was negative. There were no statistical differences in histopathological findings and surgical outcome or SOP (Tables 5 and 6), but 10 patients out of 11 (91%) with negative histopathology exhibited FIA. 4. Discussion Our study shows that patients with refractory TLE caused by different types of lesions exhibit a high grade of SOP uniformity. I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 1083 140 As defined in our study, 84% of all patients had exclusively only one type of SOP. The literature data show a higher diversity of SOPs in each individual patient. For example, Park et al. (1996) found 49% uniformity of SOP, and Lee et al. (2000) showed 70% uniformity of SOP in an individual patient. This diverse degree of uniformity could be conditioned by the number of seizures per patient (3.0 in our study vs. 4.6 in Park et al. and 6.6 in Lee et al.). We did not find any relationship between the type of SOP and the demographic characteristics of our patients, i.e. age at epilepsy onset, epilepsy duration, and age at time of surgery. Differences in seizure onset frequencies regarding demographic data have not been described in literature. The most frequent type of SOP in our patients was FIA, which was present in 67% of patients, followed by SIA (in 25%) and AT (in 8%). This distribution is indirectly comparable to other studies in TLE (Park et al., 1996; Schiller et al., 1998). Table 1 Main clinical and EEG features in 51 evaluated patients. Patient No. Number of seizures AF (Hz) Side R/L Localization of SOZ M/TP/LB SOP MRI P/N Histopathology HS/FCD/NG/OTH Engel outcome 1 2 8 R M FIA N NG I 2 4 NA R M AT P HS IV 3 4 12 L M FIA P HS I 4 2 10 L TP FIA P OTH I 5 3 2 L M SIA P FCD I 6 4 13 L M FIA P HS I 7 2 14 R M FIA P HS I 8 5 6 L LB SIA P OTH IV 9 3 2 R M SIA P OTH I 10 5 NA L TP AT P FCD IV 11 4 13 L M FIA P HS I 12 1 18 L M FIA N NG I 13 1 10 L M FIA P OTH I 14 3 5 L LB SIA N FCD IV 15 2 13 L LB FIA P OTH I 16 6 13 R M FIA P HS I 17 3 18 R M FIA P OTH I 18 4 5 L M SIA P HS I 19 4 20 L M FIA P HS II 20 3 16 R M FIA N FCD I 21 7 6 L M SIA P OTH IV 22 2 9 L M FIA P HS I 23 1 14 L M FIA P HS I 24 3 20 R LB FIA P OTH I 25 2 20 R M FIA N NG I 26 1 NA R M AT P HS II 27 1 25 L M FIA P HS I 28 2 23 R TP FIA N NG I 29 3 16 R M FIA P HS I 30 3 17 R TP FIA N FCD III 31 4 2 L M SIA P OTH I 32 2 20 R M FIA P HS I 33 2 2 R M SIA P HS I 34 1 4 R M SIA P HS III 35 3 19 R TP FIA N NG I 36 5 20 R M FIA P HS I 37 2 13 L M FIA N HS I 38 7 NA L TP AT N NG IV 39 2 24 R M FIA N NG III 40 3 10 R M FIA P NG I 41 1 4 R M SIA P OTH IV 42 3 2 R M SIA P HS II 43 3 14 L TP FIA P HS II 44 1 22 L M FIA P HS II 45 3 12 R LB FIA N NG I 46 4 11 R LB FIA N NG I 47 4 9 L M FIA P OTH I 48 2 13 R M FIA P OTH I 49 2 18 R M FIA N NG III 50 6 1 L M FIA P FCD IV 51 2 2 L M FIA P HS III AF, average frequency; R, right; L, left; SOZ, seizure onset zone; M, medial; TP, temporopolar; LB, laterobasal; SOP, seizure onset pattern; N, negative; P, positive; HS, hippocampal sclerosis; FCD, focal cortical dysplasia; NG, negative finding; OTH, other type of lesion; NA, not applicable. FIA 67% SIA 25% AT 8% Fig. 4. The percentage representation of individual seizure-onset patters (SOPs). 1084 I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 141 In a study by Park et al. (1996) characterized SOs of patients with refractory non-lesional TLE into three patterns (rhythmic activity, repetitive epileptiform discharges, and attenuation of background activity), or their combination. In Park et al. (1996), 49% of the patients had a uniform type of SOP: rhythmic activity was present in 38% of cases and repetitive epileptiform discharges in 11%. Among seizures with rhythmic activity, 83% occurred in the alpha or beta range; the remaining 17% were in the theta or delta range. This result correlates with the representation of FIA and SIA in our study. In a study by Schiller et al. (1998), SOPs were distributed in mesial TLE (MTLE) and in neocortical TLE (NTLE) as follows: early beta buzz (rhythmic discharges in beta range) in 48% of MTLE and in 74% of NTLE, rhythmic sharp activity in the alpha–theta range in 28% of MTLE and in 15% of NTLE, rhythmic sharp wave activity in the delta range in 20% of MTLE and in 11% NTLE, and irregular or semi-rhythmic spikes and sharp waves in 4% of MTLE; this pattern was not found in NTLE. The first two patterns (i.e. early beta buzz and rhythmic sharp activity in the alpha theta range) reflect FIA in our study; the second two patterns (i.e. rhythmic sharp wave activity in the delta range and irregular/semi-rhythmic spikes and sharp waves) reflect SIA. Most previous studies analyzing SOPs in TLE divide them into HYP and LVF; these two ictal onsets patterns are defined by their frequency and amplitude and are present in 70–80% of patients (Engel et al., 1989; Engel, 1990; Velasco et al., 2000; Ogren et al., 2009). The remaining 20–30% of patients cannot be classified according to those criteria; for this reason we used different SOP criteria in our study. FIA at least partially corresponds with LVF, which is defined as high-frequency (>10 Hz), low amplitude discharges (Spencer, 1998; Velasco et al., 2000). HYP was defined as a preictal phenomenon of low frequency (62 Hz) long-lasting (P5 s) spiking and was described as the most frequently encountered pattern in MTLE, in approximately 50% cases (Engel et al., 1989; Velasco et al., 2000). In our study, HYP can correlate with SIA 6 2 Hz, but this pattern was present in 18% of our patients. This discrepancy in percent representation of SIA 6 2 Hz can be explained by certain characteristics Table 2 Relationship of SOP to patient demographic data. SOP Mean age at epilepsy onset Mean duration of epilepsy Mean age at time of surgery FIA (N = 34) 13.8 ± 7.9 17.5 ± 7.5 30.1 ± 6.7 SIA (N = 13) 13.6 ± 7.5 20.2 ± 8.5 33.8 ± 6.3 AT (N = 4) 10.5 ± 3.5 17.8 ± 7.6 28.3 ± 7.3 Statistics P = 0.837 P = 0.719 P = 0.432 SOP, seizure onset pattern; FIA, fast ictal activity; SIA, slow ictal activity; AT, attenuation of background activity. Table 3 Relationship of SOP to surgical outcome. SOP Outcome Engel I Engel II–IV FIA (N = 34) 29 (85%) 5 (15%) SIA (N = 13) 4 (31%) 9 (69%) AT (N = 4) 0 (0%) 4 (100%) Statistics P < 0.001 SOP, seizure onset pattern; FIA, fast ictal activity; AT, attenuation of background activity. Table 4 More detailed division of FIA and SIA and their relations to surgical outcome. SOP Outcome Engel I Engel II–IV FIA 8–15 Hz (N = 18) 17 (94%) 1 (6%) FIA > 15 Hz (N = 16) 12 (75%) 4 (25%) Statistics P = 0.164 SIA 6 2 Hz (N = 7) 4 (57%) 3 (43%) SIA > 2 Hz (N = 6) 0 (0%) 6 (100%) Statistics P = 0.070 SOP, seizure onset pattern; FIA, fast ictal activity; SIA, slow ictal activity. Table 5 Relationship of surgical outcome to localization of SOZ and to histopathological findings. Outcome Localization of SOZ Histopathology Mesial Temporopolar Laterobasal HS FCD Negative Other Engel I (N = 33) 26 (79%) 3 (9%) 4 (12%) 14 (43%) 2 (6%) 8 (24%) 9 (27%) Engel II–IV (N = 18) 12 (67%) 4 (22%) 2 (11%) 8 (44%) 4 (22%) 3 (17%) 3 (17%) Statistics P = 0.878 P = 0.362 SOZ, seizure onset zone; HS, hippocampal sclerosis; FCD, focal cortical dysplasia. Table 6 Relationship of SOP to localization of SOZ and to histopathological findings. SOP Localization of SOZ Histopathology Mesial Temporopolar Laterobasal HS FCD Negative Other FIA (N = 34) 25 (74%) 5 (15%) 4 (12%) 15 (44%) 2 (6%) 10 (30%) 7 (20%) SIA (N = 13) 11 (85%) 0 (0%) 2 (15%) 5 (39%) 3 (23%) 0 (0%) 5 (38%) AT (N = 4) 2 (50%) 2 (50%) 0 (0%) 2 (50%) 1 (25%) 1 (25%) 0 (0%) Statistics P = 0.120 P = 0.158 SOP, seizure onset pattern; SOZ, seizure onset zone; FIA, fast ictal activity; SIA, slow ictal activity; AT, attenuation of background activity; HS, hippocampal sclerosis; FCD, focal cortical dysplasia. I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 1085 142 of HYP and by the IEEG indication criteria of individual centers. First, seizures beginning with HYP tend to remain confined to the epileptogenic region, failing to propagate outside the ipsilateral hippocampus, or doing so only after an extended period of seconds or even several minutes, and after transitioning to another type of ictal discharge (Lieb et al., 1987; Engel, 1996; Bragin et al., 1999). Second, HYP is present in ‘‘advanced’’ grades of HS (Bratz, 1899; Townsend and Engel, 1991; Spencer et al., 1992a; Velasco et al., 2000). Third, only unilateral hippocampal impairment is found on MRI in HYP patients (Ogren et al., 2009). On the basis of these HYP characteristics, we can speculate that our patient’s noninvasive data was sufficiently conclusive for us to perform surgery without an IEEG, whereas in other centers, IEEG would be required in these cases. We found a borderline significant difference in average seizure onset frequency between patients classified as Engel I and Engel II– IV (13.3 Hz vs. 9.2 Hz). This is because patients with seizure onset frequency between 8 and 15 Hz tend to have the best prognosis; frequencies from both higher and lower ranges were associated with worse surgical results. FIA, especially at frequencies between 8 and 15 Hz, was generally associated with an excellent outcome. 85% of patients with FIA (95% with FIA = 8–15 Hz, 75% with FIA > 15 Hz) were classified as Engel I. This observation is in concordance with previously published studies with subdural (Weinand et al., 1992) and epidural (Faught et al., 1992) electrodes, which showed that seizure onset frequencies higher than 8 Hz are associated with good surgical outcomes. We see a contradiction between the excellent outcome of our patients with FIA and the results of previously published studies by Lin et al. (2007) and by Ogren et al. (2009). In many patients with TLE who exhibit LVF, Lin et al. found significant damage of neocortical areas together with hippocampal atrophy. Ogren et al. (2009) observed significant bilateral hippocampal atrophy on MRI in LVF patients. Ogren et al. suggested two possible explanations for these observations. The first explanation is based on the high tendency of LVF to quickly propagate to the frontal lobes and the contralateral temporal lobe (Lieb et al., 1981; Spanedda et al., 1997; Velasco et al., 2000). Damage of neocortical areas and contralateral hippocampal atrophy could result from long-term exposure to propagated ictal activity. Alternatively, these findings could reflect the presence of undetected secondary epileptogenic regions. Considering our results, we are more inclined to the first suggestion. From another point of view, this study included only patients who had surgery after IEEG. Patients with extensive damage of both hippocampi, with bitemporal epilepsy, or with non-conclusive IEEG data are less likely to have surgical treatment. In our study, patients with SIA had worse outcomes: only 30% were classified as Engel I. This difference cannot be explained by the involvement of extratemporal structures. Firstly, patients with contemporary involvement of both temporal and extratemporal structures were not included in our study. Secondly, the very fast involvement of extratemporal structures in the ictal process was present only in patients with FIA, and all these patients were classified as Engel I. Patients with SIA 6 2 Hz tended to have better outcomes than patients with SIA > 2 Hz. More than 50% of patients with SIA 6 2 Hz had an excellent outcome; this pattern corresponds with HYP as defined above. None of our 6 patients with SIA > 2 Hz had an excellent outcome. This pattern is consistent with rhythmic theta–delta activity, which was described by Schiller et al. (1998) as a unique pattern of electrographic seizure propagation and indicates that the epileptogenic zone remains uncovered by the intracranial electrodes. Only a limited number of studies have emphasized the AT pattern and its significance (Wieser and Siegel, 1991; Gotman, 1999; Alarcon et al., 1995; Wendling et al., 2003). Electrodecremental events involve high-frequency activity between 20 and 80 Hz. In our study, these events were widely distributed and long-lasting, subsequent clear-cut rhythmic activity evolved initially in temporal regions and then later in extratemporal regions as well. We have two possible explanations for the association between AT and worse outcome (no patient was classified as Engel I; moreover, 3 out of 4 AT patients were classified as Engel IV), although these explanations cannot be generalized because of the small sample size. The first explanation is that AT could reflect extratemporal seizure origin with preferential propagation of ictal activity to the temporal lobe structures. The second explanation is based on the work of Alarcon et al. (1995). He supposed that electrodecremental events are not a part of the ictal process itself, but reflect generalized cerebral changes that allow particularly susceptible regions to develop paroxysmal behavior that gradually recruits neighboring regions. Alarcon et al. postulated that partial seizures can begin in regions that are prone to generate paroxysmal ictal activity in response to previously described generalized electrical changes. We did not find any statistically significant relations between SOPs and localization of SOZ or histopathological findings. Most (more than 70%) SOPs were distributed in the mesial temporal regions. In the literature, the differences between LVF and HYP onset patterns in SOZ locations are delineated. LVF is found more often in patients with extrahippocampal ictal onset (Engel, 1996, 2001). Bragin et al. (2005) found that the peak amplitudes of LVF onsets are located in the lateral temporal lobe indicating neocortical ictal onsets. In our group of patients with laterobasal SOZ, FIA was present in almost 70%, but this result may be influenced by the small sample size. HYP has been reported to be the most common pattern in patients with MTLE (Engel et al., 1989; Velasco et al., 2000). We did not observe SIA 6 2 Hz, which correlates with this pattern so frequently (in 18% of patients with MTLE). In our study, 44% of patients with FIA and 40% of patients with SIA had HS. Several histopathological and MRI studies have focused on detailed HS characterization and on the delineation of HS differences by use of HYP and LVF onset patterns. According to histopathological studies, hippocampal cell loss is typically found to be greater in HYP patients than in LVF patients, but this observation was not confirmed by MRI studies. In HYP patients, the cell loss is predominantly localized in areas CA1, CA3, CA4, dentate gyrus, and prosubiculum of hippocampus, sparing area CA2, which is the picture of ‘‘classical’’ HS (Bratz, 1899; Townsend and Engel, 1991; Spencer et al., 1992a,b; Velasco et al., 2000; Ogren et al., 2009). In contrast, Ogren et al. (2009) found extensive damage of area CA2 in patients with LVF. Velasco et al. (2000) showed greater or equivalent cell loss in CA2 compared to all hippocampal subfields, except CA1, in patients with LVF. This atypical distribution of damage in the LVF group appears more diffuse than ‘‘classical’’ HS and may indicate a more diffuse temporal lobe disorder distinct from classical HS (Ogren et al., 2009). We are not able to determine retrospectively a sufficiently detailed and precise histopathological data to define the differences between FIA and SIA in distribution of hippocampal atrophy, which is one of the limitations of our study. The association of FIA and negative histopathological finding is interesting and quite challenging to interpret. Negative histopathological findings are common, appearing in more than 50% of samples in a group of MRI negative, PET positive patients, which is a group of patients with good surgical outcome (Carne et al., 2004; Immonen et al., 2010). In our group, 8 out of 11 patients (73%) with negative histopathological findings had an excellent outcome; probably some could have been categorized as MRI negative, PET positive, but PET was performed only in patients after 2005. In this context, we can hypothesize that the association of FIA and negative histopathology could reflect favorable prognosis for these patients. 1086 I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 143 In our study, we did not find any relations between surgical outcome and SOZ localization or histopathological findings, but this result could be influenced by indication criteria for IEEG and patient selection. We explored patients with HS only if there were discrepancies in their clinical and/or non-invasive data. On the other hand, almost all patients with negative MRI findings underwent IEEG in the past, although they would now be categorized as MRI-negative, PET-positive TLE and would have surgery on the basis of their non-invasive data (Kuba et al., 2011; LoPinto-Khoury et al., 2011). We see some technical deficiencies in our study. The first is the 128 Hz sampling rate of our EEG monitoring system; this sampling rate does not allow us to study high frequency oscillations (HFOs), i.e. frequencies between 80 and 500 Hz. Both ictal and interictal HFOs are specific to areas of seizure generation, which is supported by the correlation between the surgical removal of tissue-generating HFOs and a good surgical outcome (Akiyama et al., 2011; Jacobs et al., 2009, 2010; Nariai et al., 2011; Ochi et al., 2007; Wu et al., 2010). The increase of HFOs was found during the seconds preceding the SO, during intrinsic SO, and during the propagation of ictal activity; these preictal and ictal HFOs are thought to be more specific for the SOZ than interictal HFOs (Allen et al., 1992; Fisher et al., 1992; Jirsch et al., 2006; Khosravani et al., 2009; Nariai et al., 2011; Zijlmans et al., 2011). The second technical limitation is that we did not provide a relationship between ictal activity and interictal activity, used as a resting state; this process would enable the analysis of higher frequencies than low beta activity. On the other hand, this method is not common in the daily praxis of epilepsy centers, and the main purpose of this work was to provide a helpful patient evaluation tool for clinicians, both epileptologists and neurophysiologists. It is a challenge to perform an analogous study with more detailed processing of ictal, preictal, and postictal data, and with a higher number of patients; such a study could provide a useful standard to the decision-making process about surgery. Our data suggests that SOP in IEEG could be an independent predictive factor for surgical outcome in patients with TLE. Higher frequencies, as a first change in IEEG, indicate better surgical outcomes, and we could hypothesize that they are associated with initial electrographic changes of seizure onsets. Slower frequencies, especially between 3 and 8 Hz, imply worse surgical outcome and could be linked with the propagation of seizure activity. Acknowledgments We thank Anne Johnson for grammatical assistance. This work was supported by the ‘‘CEITEC – Central European Institute of Technology’’ Project (CZ.1.05/1.1.00/02.0068) from the European Regional Development Fund. The study was supported by Grant GACR P304/11/1318. References Akiyama T, McCoy B, Go CY, Ochi A, Elliott IM, Akiyama M, et al. Focal resection of fast ripples on extraoperative intracranial EEG improves seizure outcome in pediatric epilepsy. Clin Neurophysiol 2011;116:1528–67. Alarcon G, Binnie CD, Elwes RD, Polkey CE. Power spectrum and intracranial EEG patterns at seizure onset in partial epilepsy. Electroencephalogr Clin Neurophysiol 1995;94:326–37. Allen PJ, Fish DR, Smith SJ. Very high-frequency rhythmic activity during SEEG suppression in frontal lobe epilepsy. Electroencephalogr Clin Neurophysiol 1992;82:155–9. Arroyo S, Lesser RP, Awad IA, Goldring S, Sutherling WW, Resnick TJ. Subdural and epidural grids and strips. In: Engel Jr J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 377–86. Bancaud J, Talairach J, Bonis A, Schaub C, Szikla G, Morel P, et al. La stéréoencéophalographie dans l´épilepsie: informations neurophysiopathologiques apportées par l´investigation fonctionnelle stéréotaxique. Paris: Masson; 1965. Blümcke I, Thom M, Aronica E, Armstrong DD, Vinters HV, Palmini A. The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 2011;52:158–74. Bragin A, Engel Jr J, Wilson CL, Vizentin E, Mathern GW. Electrophysiologic analysis of a chronic seizure model after unilateral hippocampal KA injection. Epilepsia 1999;40:1210–21. Bragin A, Wilson CL, Fields I, Fried I, Engel Jr J. Analysis of seizure onset on the basis of wideband EEG recordings. Epilepsia 2005;46(Suppl. 5):59–63. Bratz E. Ammonshornbefunde bei epileptikern. Arch Psychiatr Nervenkr 1899;31:820–35. Carne RP, O’Brien TJ, Kilpatrick CJ, MacGregor LR, Hicks RJ, Murphy MA. MRInegative PET positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain 2004;127:2276–85. Chabardès S, Kahane P, Minotti L, Hoffmann D, Benabid AL. Anatomy of the temporal pole region. Epileptic Disord 2002;4(Suppl. 1):9–15. Devinsky O, Sato S, Kufta CV, Ito B, Rose DF, Theodore WH, et al. Electroencephalographic studies of simple partial seizures with subdural electrode recordings. Neurology 1989;39:233–527. Engel Jr J. The Hans Berger lecture. Functional explorations of the human epileptic brain and their therapeutic implications. Electroencephalogr Clin Neurophysiol 1990;76:296–316. Engel Jr J. Introduction to temporal lobe epilepsy. Epilepsy Res 1996;26:141–50. Engel Jr J. Mesial temporal lobe epilepsy: what have we learned? Neuroscientist 2001;7:340–52. Engel Jr J, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel Jr J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 553–71. Engel Jr J, Babb TL, Crandall PH. Surgical treatment of epilepsy: opportunities for research into basic mechanisms of human brain function. Acta Neurochir Suppl (Wien) 1989;46:3–8. Faught E, Kuzniecky RI, Hurst DC. Ictal EEG wave forms from epidural electrodes predictive of seizure control after temporal lobectomy. Electroencephalogr Clin Neurophysiol 1992;83:229–35. Fisher RS, Webber WR, Lesser RP, Arroyo S, Uematsu S. High-frequency EEG activity at the start of seizure. J Clin Neurophysiol 1992;1992:441–8. Gloor P. Electroencephalography and the role of intracerebral depth electrode recordings in the selection of patients for surgical treatment of epilepsy. In: Porter RJ, Mattson RH, Ward AA, Dam M, editors. Advances in epileptology, the XVth epilepsy international symposium. New York: Raven Press; 1984. p. 433–7. Gotman J. Automatic detection of seizures and spikes. J Clin Neurophysiol 1999;16:130–40. Immonen A, Jutila L, Muraja-Murro A, Mervaala A, Aikia M, Lamusuo S, et al. Longterm epilepsy surgery outcomes in patients with MRI-negative temporal lobe epilepsy. Epilepsia 2010;51:2260–9. Jacobs J, LeVan P, Chatillon CE, Olivier A, Dubeau F, Gotman J. High frequency oscillations in intracranial EEGs mark epileptogenicity rather than lesion type. Brain 2009;132:1022–37. Jacobs J, Zijlmans M, Zelmann R, Chatillon CE, Hall J, Olivier A, et al. High-frequency electroencephalographic oscillations correlate with outcome of epilepsy surgery. Ann Neurol 2010;67:209–20. Jirsch JD, Urrestarazu E, LeVan P, Olivier A, Dubeau F, Gotman J. High-frequency oscillation during human focal seizures. Brain 2006;129:1593–608. Khosravani H, Mehrotra N, Rigby M, Hader WJ, Pinnegar CR, Pillay N. Spatial localization and time-dependent changes of electrographic high frequency oscillations in human temporal lobe epilepsy. Epilepsia 2009;50:605–16. Kuba R, Tyrlíková I, Chrastina J, Slaná B, Pazˇourková M, Hemza J, et al. ‘‘MRI-negative PET-positive’’ temporal lobe epilepsy: invasive EEG findings, histopathology, and postoperative outcomes. Epilepsy Behav 2011;22:537–41. Lee SA, Spencer DD, Spencer SS. Intracranial EEG seizure-onset patterns in neocortical epilepsy. Epilepsia 2000;41:297–307. Lieb JP, Engel Jr J, Brown WJ, Gevins AS, Crandall PH. Neuropathological findings following temporal lobectomy related to surface and deep EEG patterns. Epilepsia 1981;22:539–49. Lieb JP, Hoque K, Skomer CE, Song XW. Inter-hemispheric propagation of human mesial temporal lobe seizures: a coherence/phase analysis. Electroencephalogr Clin Neuropsysiol 1987;67:101–19. Lin JJ, Salamon N, Lee AD, Dutton RA, Geaga JA, Hayashi KM, et al. Reduced neocortical thickness and complexity mapped in mesial temporal lobe epilepsy with hippocampal sclerosis. Cereb Cortex 2007;17:2007–18. LoPinto-Khoury C, Sperling MR, Skidmore C, Nei M, Evans J, Sharan A, et al. Surgical outcome in PET-positive, MRI-negative patients with temporal lobe epilepsy. Epilepsia 2011;53:342–8. Lüders HO, Awad I. Conceptual considerations. In: Lüders HO, editor. Epilepsy surgery. New York: Raven Press; 1992. p. 51–62. Nariai H, Nagasawa T, Juhasz C, Sood S, Chugani HT, Asano E. Statistical mapping of ictal high-frequency oscillations in epileptic spasm. Epilepsia 2011;52:63–74. Ochi A, Otsubo H, Donner EJ, Elliott I, Iwata R, Funaki T, et al. Dynamic changes of ictal high-frequency oscillations in neocortical epilepsy: using multiple band frequency analysis. Epilepsia 2007;48:286–96. Ogren JA, Bragin A, Wilson CL, Hoftman GD, Lin JJ, Dutton RA, et al. Threedimensional hippocampal atrophy maps distinguish two common temporal lobe seizure-onset patterns. Epilepsia 2009;50:1361–70. Park YD, Murro AM, King DW, Gallagher BB, Smith JR, Yaghmai F. The significance of ictal depth EEG patterns in patients with temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol 1996;99:412–5. I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 1087 144 Rosenow F, Lüders H. Presurgical evaluation of epilepsy. Brain 2001;124:1683–700. Schiller Y, Cascino GD, Busacker NE, Sharbrough FW. Characterization and comparison of local onset and remote propagated electrographic seizure recorded with intracranial electrodes. Epilepsia 1998;39:380–8. So NK. Depth electrode studies in mesiotemporal epilepsy. In: Lüders H, editor. Epilepsy surgery. New York: Raven Press; 1992. p. 371–84. So NK. The clinical neurophysiology of epilepsy surgery. In: Hopkins A, Shorvon S, Cascino G, editors. Epilepsy. London: Chapman & Hall Medical; 1995. p. 283–308. Spanedda F, Cendes F, Gotman J. Relations between EEG seizure, morphology, interhemispheric spread, and mesial temporal atrophy in bitemporal epilepsy. Epilepsia 1997;38:1300–14. Spencer SS. Controversies in epilepsy: depth versus subdural electrode studies for unlocalized epilepsy. J Epilepsy 1989;39:527–33. Spencer SS. Substrates of localization-related epilepsies: biologic implication of localizing finding in humans. Epilepsia 1998;39:114–23. Spencer SS, Lamoureux D. Invasive electroencephalography evaluation for epilepsy surgery. In: Shorvon S, Dreifuss F, Fish D, Thomas D, editors. The treatment of epilepsy. Oxford: Blackwell Science; 1996. p. 562–88. Spencer SS, Spencer DD. Entorhinal-hippocampal interactions in medial temporal lobe epilepsy. Epilepsia 1994;35:721–7. Spencer SS, Spencer DD, Williamson PD, Mattson RH. The localizing value of depth electroencephalography in 32 patients with refractory epilepsy. Ann Neurol 1982;12:248–53. Spencer SS, Williamson PD, Spencer DD, Mattson R. Combined depth and subdural electrodes investigation in uncontrolled epilepsy. Neurology 1990;40:74–9. Spencer SS, Kim J, Spencer DD. Ictal spikes: a marker of specific hippocampal cell loss. Electroencephalogr Clin Neurophysiol 1992a;83:104–11. Spencer SS, Guimaraes P, Katz A, Kim J, Spencer DD. Morphological patterns of seizures recorded intracranially. Epilepsia 1992b;33:537–45. Spencer SS, So NK, Engel Jr J, Williamson PD, Levesque MF, Spencer DD. Depth electrodes. In: Engel Jr J, editor. Surgical treatment of epilepsies. New York: Raven Press; 1993. p. 359–76. Sperling MR, O’Connor MJ. Comparison of depth and subdural electrodes in recording temporal lobe seizures. Neurology 1989;39:1497–504. Sperling MR, O’Connor MJ. Auras and subclinical seizures: characteristics and prognostic significance. Ann Neurol 1990;28:320–8. Townsend JB, Engel Jr J. Clinicopathological correlations of low voltage fast and high amplitude spikes and wave mesial temporal stereoncephalographic ictal onsets. Epilepsia 1991;32:21. Velasco AL, Wilson CL, Babb TL, Engel Jr J. Functional and anatomic correlates of two frequently observed temporal lobe seizure-onset patterns. Neural Plast 2000;7:49–63. Weinand ME, Wyler AR, Richey ET, Phillips BB, Somes GW. Long-term ictal monitoring with subdural strip electrodes: prognostic factors for selecting temporal lobectomy candidates. J Neurosurg 1992;77:20–8. Wendling F, Bartolomei F, Bellanger JJ, Bourien J, Chauvel P. Epileptic fast intracerebral EEG activity: evidence for spatial decorrelation of seizure onset. Brain 2003;126:1449–59. Wiebe S, Blume WT, Girvin JP, Eliasziw M. Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group. A randomized, controlled trial of surgery for temporal-lobe epilepsy. New Engl J Med 2001;345:311–8. Wieser HG, Siegel AM. Analysis of foramen ovale electrode-recorded seizures and correlation with outcome following amygdalohippocampectomy. Epilepsia 1991;32:838–50. Williamson P, Engel Jr J, Munari C. Anatomic classification of localization-related epilepsies. In: Engel Jr J, editor. Epilepsy: a comprehensive textbook. New York: Lippincott-Raven; 1998. p. 2465–26477. Wu JY, Sankar R, Lerner JT, Matsumoto JH, Vinters HV, Mathern GW. Removing interictal fast ripples on electrocorticography linked with seizure freedom in children. Neurology 2010;75:1686–94. Wyler AR, Ojemann GA, Lettich E, Ward Jr AA. Subdural strip electrodes for localizing epileptogenic foci. J Neurosurg 1984;60:1195–200. Wyler AR, Wilkus RJ, Blume WT. Strip electrodes. In: Engel Jr J, editor. Surgical treatment of epilepsies. New York: Raven Press; 1993. p. 387–97. Zijlmans M, Jacobs J, Kahn YU, Zelmann R, Dubeau F, Gotman J. Ictal and interictal high frequency oscillations in patients with focal epilepsy. Clin Neurophysiol 2011;122:664–71. 1088 I. Dolezˇalová et al. / Clinical Neurophysiology 124 (2013) 1079–1088 145 Annex 5: Pail M, Rehulka P, Cimbalnik J, Dolezalova I, Chrastina J, Brazdil M. Frequencyindependent characteristics of high-frequency oscillations in epileptic and non-epileptic regions. Clinical Neurophysiology. 2017;128(1):106-114. 146 Frequency-independent characteristics of high-frequency oscillations in epileptic and non-epileptic regions Martin Pail a,⇑ , Pavel Rˇehulka a , Jan Cimbálník b , Irena Dolezˇalová a , Jan Chrastina c , Milan Brázdil a,d a Brno Epilepsy Center, Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic b International Clinical Research Center, St. Anne’s University Hospital, Brno, Czech Republic c Brno Epilepsy Center, Department of Neurosurgery, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic d Behavioral and Social Neuroscience Research Group, CEITEC – Central European Institute of Technology, Masaryk University, Brno, Czech Republic a r t i c l e i n f o Article history: Accepted 15 October 2016 Available online 29 October 2016 Keywords: High frequency oscillations Ripples Fast ripples Temporal lobe epilepsy Extratemporal lobe epilepsy Seizure onset zone Epileptogenic zone Irritative zone h i g h l i g h t s  The highest rate of fast ripples is in the seizure onset zone (SOZ).  A worse prognosis (no seizure freedom after surgery) is associated with higher amplitudes of fast ripples outside the SOZ.  In the SOZ, high frequency oscillations are more frequent, shorter, and have higher amplitudes. a b s t r a c t Objective: The purpose of the presented study is to determine whether there are frequency-independent high-frequency oscillation (HFO) parameters which may differ in epileptic and non-epileptic regions. Methods: We studied 31 consecutive patients with medically intractable focal (temporal and extratemporal) epilepsies who were examined by either intracerebral or subdural electrodes. Automated detection was used to detect HFO. The characteristics (rate, amplitude, and duration) of HFO were statistically compared within three groups: the seizure onset zone (SOZ), the irritative zone (IZ), and areas outside the IZ and SOZ (nonSOZ/nonIZ). Results: In all patients, fast ripples (FR) and ripples (R) were significantly more frequent and shorter in the SOZ than in the nonSOZ/nonIZ region. In the group of patients with favorable surgical outcomes, the relative amplitude of FR was higher in the SOZ than in the IZ and nonIZ/nonSOZ regions; in patients with poor outcomes, the results were reversed. The relative amplitude of R was significantly higher in the SOZ, with no difference between patients with poor and favorable surgical outcomes. Conclusions: FR are more frequent, shorter, and have higher relative amplitudes in the SOZ area than in other regions. The study suggests a worse prognosis in patients with higher amplitudes of FR outside the SOZ. Significance: Various HFO parameters, especially of FR, differ in epileptic and non-epileptic regions. The amplitude and duration may be as important as the frequency band and rate of HFO in marking the seizure onset region or the epileptogenic area and may provide additional information on epileptogenicity. Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. 1. Introduction Over the past few years, there has been growing interest in the analysis of interictal high frequency oscillations (HFO), primarily with the goal of understanding their value for identifying the epileptogenic zone and their correlation with epileptogenicity. HFO promise to be more specific than interictal spikes for epileptogenic brain tissue and even more specific than the seizure-onset area (Jacobs et al., 2008). HFO are short-lasting field potentials, which arise as a result of the synchronization of neuronal populations. HFO have been identified and defined in terms of frequency: ripples (80–250 Hz), fast ripples (250–600 Hz) (reviewed in Bragin et al., 2010; Engel et al., http://dx.doi.org/10.1016/j.clinph.2016.10.011 1388-2457/Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. ⇑ Corresponding author at: Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Pekarˇská 53, 656 91 Brno, Czech Republic. Fax: +420 543 182 624. E-mail address: martin.pail@fnusa.cz (M. Pail). Clinical Neurophysiology 128 (2017) 106–114 Contents lists available at ScienceDirect Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph 147 2009), and very high frequency oscillations (1000–2500 Hz) (Usui et al., 2015). HFO have been widely studied in animals and humans, in mesial temporal and neocortical structures, under physiological and pathological conditions, using microelectrodes or commercial macroelectrodes, and during interictal and ictal periods (Bragin et al., 1999a,b; Staba et al., 2002; Worrell et al., 2004, 2008; Urrestarazu et al., 2007; Jacobs et al., 2008, 2009; Bagshaw et al., 2009; Brázdil et al., 2010; Crépon et al., 2010). However, the reliability of HFO as a biomarker of epileptogenicity and the seizure-onset zone remains uncertain (Haegelen et al., 2013; Jobst, 2013; Wang et al., 2013). Ripples, observed as a physiological finding in the hippocampus and parahippocampal structures, are thought to be functionally involved in memory consolidation (Buzsáki et al., 1992; Axmacher et al., 2010; Lachaux et al., 2012). The spontaneous occurrence of ripples in humans is also believed to be physiological in the primary visual cortex (Nagasawa et al., 2012; Wang et al., 2013) and in the primary motor cortex (Wang et al., 2013). The presumption of the exclusively physiological nature of ripples was, however, impugned by evidence of HFO in ripple ranges recorded in the dentate gyrus after epileptogenic insult in an animal model of kainate-induced status epilepticus (Bragin et al., 1999b, 2004). Conversely, fast ripples were repeatedly reported as a biomarker of epileptogenesis and epileptogenicity, both in animal models and in human epilepsy (Bragin et al., 1999a,b; Staba et al., 2002). It is important that HFO in the fast ripple range (at about 600 Hz) can also be considered physiological, as they were previously evoked during stimulation of the somatosensory cortex (Curio et al., 1997). Thus the HFO frequency range, in general, is not sufficient to differentiate physiological and pathological HFO (Bragin, 2007). On the other hand, there is evidence of favorable epilepsy surgery outcome after the removal of tissue generating interictal HFO, especially fast ripples (Jacobs et al., 2010; Wu et al., 2010; Akiyama et al., 2011). Several authors have tried to distinguish between pathological and physiological HFO on the basis of a specific regional distribution in respective mesial temporal structures (Jiruska and Bragin, 2011); some have investigated the difference between taskinduced and spontaneous HFO (Nagasawa et al., 2012; Matsumoto et al., 2013; Brázdil et al., 2015); others have studied the association of HFO with epileptiform discharges (Crépon et al., 2010; Urrestarazu et al., 2007; Wang et al., 2013). Interictal HFO (both ripples and fast ripples) rates were proven significantly higher within the seizure onset zone (SOZ) than outside it (Jacobs et al., 2009). The purpose of the present study is to identify whether there are any other frequency-independent HFO parameters that potentially differ in areas within the SOZ, within the irritative zone (IZ), and in areas outside the IZ/SOZ. 2. Methods 2.1. Subjects Our sample comprised 31 patients (19 females; 12 males) ranging in age from 13 to 56 years (mean age 33.4 years, SD = 10.5), all with medically intractable focal epilepsies (Table 1). All the subjects fulfilled the diagnostic criteria for either medically intractable temporal lobe epilepsy (TLE) – 22 subjects or extratemporal lobe epilepsy (ETLE) – 9 subjects. The diagnosis was made according the ILAE criteria (Commission on Classification and Terminology of ILAE, 1989). The main demographic and clinical characteristics of the included subjects are shown in Table 1. 2.1.1. Presurgical evaluation All 31 patients underwent a comprehensive presurgical evaluation, including a detailed history and neurological examination, magnetic resonance imaging (MRI), neuropsychological testing, and scalp and invasive video-EEG monitoring (Table 1). Most of the subjects had not previously undergone intracranial surgery. One subject underwent resection of venous malformation within the left P-O region before SEEG and re-operation; in one subject a resection of the pole of the left temporal lobe had been performed, and in one subject a limited left AMTR was performed before SEEG monitoring. Prior to invasive EEG, two subjects had a vagus nerve stimulation system implanted, with unfavorable seizure frequency outcome. The duration of clinical monitoring and the location and number of implanted electrodes were determined in accordance with clinical considerations. 2.1.2. Surgery and outcome measure Most of the patients (28) underwent surgical intervention (implantation of VNS was performed in 7 patients and brain resection in 21 patients; details are shown in Table 1). The follow-up interval after epilepsy surgery was at least 12 months. After surgical resection, 8 patients were rated as Engel IA, one patient was Engel IIA, and 11 patients were Engel III or IV; the Engel rating is unknown for one patient (due the loss to follow-up care). This study was approved by the St. Anne’s University Hospital Research Ethics Committee and the Ethics Committee of Masaryk University. All patients signed an informed consent form. 2.2. SEEG Depth electrodes (mostly SEEG; grids and strips were used in two patients) were implanted to localize the seizure origin prior to surgical treatment. The sites of electrode placements were individualized according to seizure semiology, clinical history, noninvasive EEG investigations, and neuroimaging results. Standard intracerebral 5-, 10-, and 15-contact Micro Deep semi-flexible multicontact platinum electrodes (ALCIS) were used with a diameter of 0.8 mm, a contact length of 2 mm, an inter-contact distance of 1.5 mm, and a contact surface area of 5 mm2 . Their position within the brain was afterwards verified by MRI with electrodes in situ (see Table 1). In two patients, platinum subdural strip and grid electrodes (Radionics) were used. Thirty minutes of artifactfree continuous interictal SEEG data (recorded during wakefulness) was analyzed for each subject. This period is sufficient based on the results of previously published papers (Jacobs et al., 2008; Zelmann et al., 2009; Andrade-Valença et al., 2012). The EEGs were acquired at 25 kHz sampling frequency and subsequently low-pass filtered and downsampled to 5 kHz. High harmonics produced by the system (artificial harmonics) are accounted for during EEG acquisition. A reference average montage was used for the analysis. 2.2.1. Labeling of analyzed contacts The contacts in each subject were divided into three groups for further HFO analysis. The distribution was done visually (by coauthors M.P. and I.D.) by a standard analysis of ictal and interictal SEEG recordings. The first group was labelled SOZ contacts: the channels that revealed the first ictal activity. The second group was labelled IZ contacts: the channels where interictal epileptiform discharges were detected, but no seizure onset was detected. The remaining non-spiking contacts were labelled nonSOZ/nonIZ. Only contacts localized in gray matter were included in the study. 2.2.2. Automated detection of HFO in resting awake SEEG The algorithm for HFO detection is explained in more detail in the Supplementum. HFO were detected by a custom Matlab detection algorithm. The raw signal (Supplementary Fig. S1) was divided M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 107 148 Table1 Demographicandclinicaldataofthepatients. Subject gender Ageat seizure onset Ageat invasive EEG Seizure frequency/monthly Febrile seizures/precipitating events MRI(signsof)Brainlobesexploredand numberofimplanted electrodes RegionSOZ (epileptogeniczone) Intervention/histopathologyPostsurgical outcomes(Engel) (years) 1(F)1620CPS/4plus0NormalT(3),T0 (2)HbilaterallyVNS– 2(F)1957CPS/2plus,sGTCS/60Postischemiclesions withinleftT-OandH F0 (1),T0 (5)LeftHAMTR/gliosis,hemosiderin withinTpole IA(3.5) 3(F)434CPS/500HypotrophicrightHT0 (3),T(3)RightHAMTR/hippocampalsclerosis gradeIII IA(3.5) 4(F)619CPS/4PerinatalhypoxiaGliosisandcortical abnormalitiesleftPO O0 (2),T0 (2),P0 (1),O(2),F(2)LeftTPOregionCortectomy/negatIV(3) 5(F)234SPS,CPS/4PerinatalasphyxiaPolymicrogyrialeftF, postoperativechangesof leftPO F0 (8),P0 (2),T0 (1)LeftPlobe,pericentral area VNS– 6(F)1633CPS/5PerinatalhypoxiaBilat.HST0 (4),T(4)Hbilaterally(mainly rightside) RightAMTR/negatIII(2.5) 7(M)0.521CPS/30,MS/3000AbnormalityofleftHF0 (9),P0 (1),T0 (1),F(2)LeftSMA(myoclonic seizures),leftFpole (CPS) ResectionofleftFpole/negatIV(2.5) 8(F)9.513CPS/30PerinatalasphyxiaMalrotationofleftHP0 (5),T0 (5),F0 (1),O(1),P(1)Undetected–– 9(M)927CPS/30plusPerinatalasphyxiaBilat.abnormal gyrificationandgliotic changesofPO O(2),T(3),P(3),O0 (3),T0 (1), P0 (3) Olobebilaterally (mainlyrightside) PartialresectionofrightOlobe/ ulegyria,FCDIIIA IIIA(2) 10(F)930CPS/3PerinatalasphyxiaNormalT(3),T0 (8),O0 (2)LeftlateralTlobe,GTSIncompleteresectionofleftGTS/ negat IVA(2) 11(F)1726CPS/12MeningoencephalitisNormalT(1),T0 (8)LateralmesialTlobeLeftAMTR/FCDIBIA(2) 12(F)2856CPS/80Righthippocampal sclerosis T(2),T0 (3)RightHRightAMTR/negatIIIA(2) 13(M)140CPS/2plus0HypotrophicleftHT0 (6),P0 (2),O0 (2)LeftHLeftAMTR/negatIA(2) 14(F)1127sGTCS/30NormalF(4),F0 (8)LeftGFS,GFM,GFMedPartialcortectomyofleftFlobeIIIA(2) 15(M)1926CPS/40FCDleftTand hyperintensityofH T0 (7),F0 (1),T(1)Leftanterotemporal, rightmesialtemporal VNS– 16(F)1034SPS,CPS/10plusPerinatalasphyxiaNormalF0 (5),P0 (2),F(6),P(2)Rightdorsolateral parasagittal premotoricarea PartialcortectomyofrightFlobe, partialcallosotomy/FCD1C Notavailable 17(M)2738CPS/25plus0NormalT(3),I0 (2),T0 (6),P0 (1),O0 (1), F0 (1) MesialTregion bilaterally VNS– 18(F)648CPS/60plusFebrileseizuresPostoperativegliosisright P P(6),T(3),O(1)RightTPcortex,right hippocampus VNS– 19(M)3341CPS/300Focalhyperintensityright basalT T(4),T0 (3)RightHRightAMTR/FCDIIIb gangliogliom IA(1.5) 20(M)1125CPS/12plus0NormalTPOgridsRightlateralTlobeCorticalresectionofrightTaTO region/negat IIA(2) 21(F)1740CPS/200NormalF0 (1),O0 (2),T0 (7),T(2)Undetected–– 22(F)1740CPS/200NormalFTPO0 gridsLeftlaterobasal posteriorTlobe Cortectomyofleftlaterobasal posteriorTlobe/negat IIIA(1.5) 23(M)829CPS/20plus0NormalT0 (7),F0 (2),T(2)BitemporalLeftAMTR/negatIVA(1) 24(F)233CPS/6plusMeningoencephalitisPostencephaliticchanges leftT T(2),T0 (6),F0 (3)LeftHLeftAMTR/hippocampalsclerosis, postmeningoencephalitic changes IA(1) 25(M)1240CPS/40plusTraumaPosttraumaticchangesof leftTandGFI F0 (6),T0 (5)LeftTpole,lateralT, lateralprefrontalarea Resectionoftemporalpole, lesionectomyF left/posttraumaticchanges IIIA(1) 26(F)233sGTCS/100Postoperativechanges rightPO T(3),P(2),O(2)RightGTS,LPI–– 27(F)222CPS/40Postoperativechanges rightPO P(2),F(2),T(3)Undetected(right posteriorquadrant) VNS– 108 M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 149 into sliding statistical windows (10 s) and filtered in a series of overlapping, logarithmically spaced frequency bands. Each band was processed as follows: The power envelope was computed using the Hilbert transform and normalized by Eq. (1) to stress the signal peaks (Supplementary Fig. S2). xnormed ¼ x À P2=3 P1=3 À P2=3 ð1Þ To overcome the effects of Gibb’s phenomenon, a ‘‘frequency stability” metric was computed. The band passed filtered signal (narrow band) was transformed to a cosine representation of its phase. The same transformation was applied to a band passed filtered signal with the same high cut-off frequency but a four times lower low cut-off frequency (broad band) (Supplementary Fig. S3a, b). A sliding window with the width of four cycles of the narrow band low cut-off frequency was applied to the narrow band signal and the root mean square (RMS) was calculated, thus generating the ‘‘signal”. Similarly, the RMS was calculated on the signal created by subtracting the narrow band signal from the broad band signal, generating ‘‘noise”. The frequency stability was calculated as a ‘‘signal-to-noise” ratio. The produced signal was normalized Eq. (1). The dot product of the amplitude and frequency stability metric was calculated. Negative values were set to 0. Putative HFO detections were obtained by thresholding the normalized Eq. (1) dot product of the power envelopes and frequency stability with 0.1 (Supplementary Fig. S4). To increase algorithm specificity, amplitude, and frequency stability, the dot product and duration minimum/maximum thresholds were applied on putative HFO detections. The metric threshold values were obtained from cumulative distribution functions fitted to the HFO metric distributions previously marked by expert reviewers. The parameters of the fitted gamma functions were specific for each band. The parameters (Supplementary Table S1) and examples of true positive detection and false positive detection (Supplementary Fig. S5) are included in the Supplementum. 2.3. Statistical analysis The rates of HFO per contacts within the three groups (SOZ, IZ, and nonSOZ/nonIZ) were statistically compared (SOZ Â IZ; SOZ Â nonSOZ/nonIZ; IZ Â nonSOZ/nonIZ). This statistical analysis was performed using an independent two sample t-test separately for ripple range and fast ripple range. We performed statistical analyses on the whole dataset and separately for patients with favorable postoperative outcomes (Engel I or II – 9 patients) and the other patients with ‘‘poor outcomes” (postsurgical outcomes of Engel III or IV and patients with presumed poor outcomes due to more than one detected SOZ according to SEEG). Furthermore, we performed a statistical analysis comparing the duration and relative amplitude of HFO (separately for R and FR ranges) for contacts in the areas, as defined above. This analysis was performed in all patients and subsequently in patients with favorable or poor surgical outcomes. For this analysis, we again used an independent two sample t-test. Bonferroni’s correction for multiple comparisons was applied where needed. 3. Results The statistical analysis and complete results of HFO detection (rates, duration, and amplitudes) are shown in Table 2 and Supplementary Table S2, respectively. Table1(continued) Subject gender Ageat seizure onset Ageat invasive EEG Seizure frequency/monthly Febrile seizures/precipitating events MRI(signsof)Brainlobesexploredand numberofimplanted electrodes RegionSOZ (epileptogeniczone) Intervention/histopathologyPostsurgical outcomes(Engel) (years) 28(M)2135CPS/5CommotiocerebriBilat.hypotrophic hippocampi T(6),T0 (2)Hbilaterally(mainly rightside) VNS– 29(M)3137CPS/4Prolongedfebrile seizures NormalT0 (8),F0 (2),T(2)Hbilaterally(mainly leftside) LeftAMTR/negatIA(1) 30(F)927CPS/5MeningoencephalitisLeftHST0 (8),T(2),P0 (1),I0 (1),F0 (1)LeftHLeftAMTR/FCDIIIAIIIA(1) 31(M)251CPS/3plusFebrileseizuresRightHatrophy,slight changesofdensity T(5),F(3),T0 (2)RightHRightAMTR/negatIA(1) CPS–complexpartialseizure;MS–myoclonicseizures;plus–sporadicictalgeneralization;SOZ–seizureonsetzone;SMA–supplementarymotorarea;LPI–lobulusparietalisinferior;GTS–gyrustemporalissuperior;GFS– gyrusfrontalissuperior;GFM–gyrusfrontalismedius;GFMed–gyrusfrontalismedius;H–hippocampus;HS–hippocampalsclerosis;DNET–dysembryoplasticneuroepithelialtumor;FCD–focalcorticaldysplasia;AMTR– anteromedialtemporalresection;E–extratemporal;T–temporal;R–right;L–left;FCD–focalcorticaldysplasia;F–frontal;P–parietal;T–temporal;O–occipital,I–insular;0 -left. M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 109 150 3.1. Rates As expected, the HFO rate per contact in all patients was statistically higher in the SOZ than in the nonSOZ/nonIZ regions (Fig. 1) in the fast ripple frequency range (p = 0.018) and ripple range (p = 0.038). Specifically, HFO in the ripple range were identified. The mean number of HFO per contact (per 30 min) was 109.13 within the SOZ (SD = 79.19), 96.03 in the IZ (SD = 82.24) and 68.65 for nonSOZ/nonIZ regions (SD = 49.24). The differences between groups of contacts were significant only between SOZ and nonSOZ/nonIZ regions. In the fast ripple range, the mean HFO count per contact (per 30 min) were 50.02 in the SOZ (SD = 50.75), 31.56 in the IZ (SD = 34.89) and 23.23 in nonSOZ/nonIZ (SD = 21.84). The only significant result was in the comparison of SOZ and nonSOZ/nonIZ regions; see above. No statistically significant results were seen in comparison of regions separately in groups of patients with favorable or poor outcomes. 3.2. Duration 3.2.1. Ripples The mean duration of detected HFO was 54.40 ms in the SOZ (SD = 25.58), 56.31 ms in the IZ (SD = 27.52) and 56.12 ms in nonSOZ/nonIZ regions (SD = 29.16). Statistical analysis showed significantly shorter durations of HFO in the SOZ than in either the IZ or nonSOZ/nonIZ regions (p < 0.001); the difference between the IZ and nonSOZ/nonIZ regions was not significant (p = 0.361). 3.2.2. Fast ripples Similarly, automated detection in the fast ripple range detected the mean durations of HFO of 27.43 ms in the SOZ (SD = 16.04), 29.13 ms in the IZ (SD = 18.30) and 30.41 ms, in nonSOZ/nonIZ regions (SD = 20.94); the shortest durations of FR were in the Table 2 The results of statistical analysis (two sample t-test, p values) comparing the rates, duration, and relative amplitudes of HFO (all over separately for ripples [first column] and fast ripples range [second column]) per contacts in the redefined areas (SOZ, IZ, nonSOZ/nonIZ); also in this analysis particularly in all patients and in patients with favorable or poor outcomes. Rates nonSOZ/nonIZ IZ All Favorable outcome Poor outcome All Favorable outcome Poor outcome SOZ 0.038 0.018 0.417 0.143 0.102 0.076 1.078 0.228 0.102 0.092 0.974 0.407 IZ 0.245 0.543 1.243 1.136 0.464 0.491 x x x x x x nonSOZ/nonIZ x x x x x x DuraƟon nonSOZ/nonIZ IZ All Favorable outcome Poor outcome All Favorable outcome Poor outcome SOZ <<0.001 <<0.001 <<0.001 <<0.001 0.877 <<0.001 <<0.001 <<0.001 <<0.001 <0.001 <<0.001 <<0.001 IZ 0.361 <<0.001 <<0.001 0.460 <<0.001 0.027 x x x x x x nonSOZ/nonIZ x x x x x x Amplitude nonSOZ/nonIZ IZ All Favorable outcome Poor outcome All Favorable outcome Poor outcome SOZ <<0.001 0.430 <<0.001 <<0.001 <<0.001 <<0.001 <<0.001 0.309 <<0.001 0.014 0.030 <<0.001 IZ <<0.001 0.013 <<0.001 <<0.001 <<0.001 1.136 x x x x x x nonSOZ/nonIZ x x x x x x Fig. 1. Counts of fast ripples per 30 min in particular areas (SOZ, IZ, and nonSOZ/ nonIZ) in all patients. 110 M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 151 SOZ, HFO durations were longer in nonSOZ/nonIZ regions than in the IZ (p < 0.001) and in nonSOZ/nonIZ regions than in the SOZ (p < 0.001). The difference between the IZ and SOZ was also significant (p < 0.001). In the group of patients with favorable outcomes, shorter R and FR durations were seen in the SOZ (p < 0.001) than in nonSOZ/ nonIZ and IZ regions (see Fig. 2). In the group of patients with poor outcomes, the longest duration of R was in the IZ (p < 0.001); the difference between nonSOZ/ nonIZ Â SOZ was not significant (p = 0.877). In both subgroups of patients, the shortest FR duration was seen in the SOZ (p < 0.001). 3.3. Amplitudes 3.3.1. Ripples Automated detection in the ripple range detected the relative amplitudes of HFO in the SOZ of 77.47 (SD = 164.35), in the IZ of 66.14 (SD = 231.74) and in nonSOZ/nonIZ regions of 56.30 (SD = 151.74). Statistical analysis showed significantly higher amplitudes of HFO in the SOZ than in either the IZ or nonSOZ/nonIZ regions (p < 0.001); the difference between the IZ and nonSOZ/ nonIZ regions was also significant (p < 0.001). 3.3.2. Fast ripples Automated detection in the fast ripple range detected relative amplitudes of HFO in the SOZ of 76.04 (SD = 164.95), in the IZ of 80.93 (SD = 458.49) and in nonSOZ/nonIZ regions of 73.79 (SD = 241.17). The differences between SOZ and IZ or nonSOZ/ nonIZ were not significant. Nevertheless, in the group of patients with favorable outcomes, the relative amplitude of HFO (both R and FR) was higher in the SOZ than in the IZ and nonIZ/nonSOZ region (p < 0.001) (see Figs. 3 and 4). In the group of patients with poor outcomes, the highest amplitude of R was seen in the SOZ (p < 0.001) (versus nonSOZ/ IZ); p = 0.030 (IZ) and the lowest in nonIZ/nonSOZ region (p < 0.001). The relative amplitude of FR was lower in the SOZ than in either the IZ or nonSOZ/nonIZ regions (p < 0.001) (see Fig. 5). 4. Discussion Interictal HFO analyses in patients with epilepsy have been reported useful for SOZ identification (Urrestarazu et al., 2007; Fig. 2. Durations (ms) of fast ripples in particular areas (SOZ, IZ, and nonSOZ/nonIZ) in patients with favorable outcomes. Fig. 3. Relative amplitudes of fast ripples in particular areas (SOZ, IZ, and nonSOZ/ nonIZ) in patients with favorable outcomes. Fig. 4. Relative amplitudes of fast ripples in particular areas (SOZ, IZ, and nonSOZ/ nonIZ) in patients with favorable outcomes. Bar graphs represent the population mean. Ticks represent a 95% confidence interval of the mean calculation. Fig. 5. Relative amplitudes of fast ripples in particular areas (SOZ, IZ, and nonSOZ/ nonIZ) in patients with poor outcomes. Bar graphs represent the population mean. Ticks represent a 95% confidence interval of the mean calculation. M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 111 152 Jacobs et al., 2008, 2009). In the present study, patients with both temporal and extratemporal epilepsy were examined to ascertain facts about HFO characteristics in areas within and outside the SOZ and the IZ. For these purposes, we used the concept of an epileptogenic zone (Lüders and Awad, 1992; Rosenow and Lüders, 2001), and used the term ‘‘irritative zone” as it has been used elsewhere (Blanco et al., 2011) to investigate HFO characteristics in more detail, to observe the area between the SOZ and nonSOZ, and to eliminate potential overlap of interictal active (‘‘spiking”) channels within SOZ and non-SOZ. Only patients who underwent surgical resections and had good postsurgical outcomes reflect the correct determination of the seizure onset zone. For that reason, we divided patients into two subgroups: favorable outcome and others (‘‘poor outcome”). In our study, HFO were detected in all (SOZ, IZ, and nonSOZ/ nonIZ) areas, with a higher absolute HFO rate for events in the R range than in the FR range. These findings are not surprising; the occurrence of high frequency activity unassociated with the SOZ is well documented in the hippocampi (Staba et al., 2002; Axmacher et al., 2010) and in the primary motor, somatosensory, and visual cortices (Curio, 2000; Nagasawa et al., 2012; Matsumoto et al., 2013; Wang et al., 2013). The results indicate that the detection algorithm, in addition to detecting pathological HFO, may also detect HFO that might be physiological, might be a fragment or propagation of pathological HFO arising from elsewhere (Crépon et al., 2010), or might be false positive detections. Higher HFO rates in the R range than in the FR range may be explained by an association between HFO and interictal spiking (Urrestarazu et al., 2007; Jacobs et al., 2008; Wang et al., 2013). There is evidence that larger networks are involved in R than in FR generation in SOZ areas (Chrobak and Buzsáki, 1996; Bragin et al., 2002) and that FR generators are spatially localized to a region of less than 1 mm3 (Bragin et al., 2002). Furthermore there is evidence of HFO under detection, especially in higher frequency bands when using macroelectrodes (as in our cases), i.e. in the FR range, which results in a lower mean frequency of detected HFO (Worrell et al., 2008). According to previously published data, which is further corroborated by our study, the FR and R rate are significantly higher in the SOZ than in nonSOZ/nonIZ regions (Urrestarazu et al., 2007; Jacobs et al., 2008, 2009; Andrade-Valença et al., 2012). Interestingly, in patients with favorable outcomes, a prominent (though not significant) difference between rates of FR and R in the SOZ and the IZ can be observed, which might represent a more focal generator (epileptogenic tissue) of these pathological HFO; this difference was less seen in other patients. As in a study by Jacobs et al. (2008), the distinction between pathological and normal areas was worse for R than for FR. Nevertheless, the explanation for the high rates of R and FR in nonSOZ/nonIZ regions is unclear. It cannot be definitely demonstrated whether all of these marked events were actually pathological, physiological, or the propagation of pathological HFO arising from elsewhere, as was mentioned above. Spontaneous HFO of a physiological nature are difficult to distinguish from epileptogenic ones, particularly during wakefulness (Jacobs et al., 2008; Curio, 2000) since the frequency and amplitude measures alone cannot be used for this purpose (Engel et al., 2009; Nagasawa et al., 2012; Matsumoto et al., 2013). Finally, the overlap (in both amplitude and duration) between the pathological and physiological ripples is extensive (Alkawadri et al., 2014). The main reason for focusing on the IZ is evidence of HFO linking with interictal epileptiform discharges in both mesiotemporal and neocortical epilepsy (Urrestarazu et al., 2007; Jacobs et al., 2008; Wang et al., 2013). The vast majority of interictal HFO (up to 73% of R and 92% of FR) is associated with interictal spikes or sharp waves (Urrestarazu et al., 2007; Jacobs et al., 2008). A comparison of HFO associated with interictal epileptiform discharges and unassociated HFO revealed no differences in terms of the duration (Urrestarazu et al., 2007), or the associated HFO had longer durations than the unassociated HFO with spikes (Jacobs et al., 2008). Wang and colleagues (2013) showed HFO detected in the SOZ area were of shorter duration than those not correlated to the SOZ area. As in the study by Wang et al. (2013), our data suggest longer durations of both R and FR in the IZ or nonSOZ/nonIZ than in the SOZ. These findings might be also a consequence of analyzing awake recordings with less expressed interictal discharges, and so more HFO detected this phenomenon and with shorter duration. This finding might be also explained by the work of Nagasawa (2012), who revealed that the duration of spontaneous HFO in the ripple range (namely from the occipital cortex) of a physiological nature were significantly longer than that of epileptogenic ripple HFO. Similar results were presented by Alkawadri et al. (2014). These observations are still consistent with the hypothesis that longer durations of HFO may represent longer excitatory neural processing (Niessing et al., 2005; Nishida et al., 2008; Koch et al., 2009; Manning et al., 2009; Nagasawa et al., 2012). In other studies, however, the duration of pathological (in the SOZ) and physiological (nonSOZ region) FR was not diverse (Nagasawa et al., 2012; Alkawadri et al., 2014), or the longer duration of both R (Brázdil et al., 2015) and FR was revealed within the SOZ (Jacobs et al., 2008; Matsumoto et al., 2013). The discrepancies among studies in the duration of R and FR in various regions might be explained by different proportions of included focal epilepsies (hippocampus vs. neocortex). There have been noteworthy results regarding relative HFO amplitude in epileptic and non-epileptic regions in particular subgroups of patients. Interestingly, in the group of patients with favorable outcomes, the relative HFO amplitude (especially FR) was higher in the SOZ than in other regions; in other patients with ‘‘poor outcomes”, the results of FR analysis were reversed. These results for both R and FR might contribute to neurosurgical resection planning, showing possible worse prognosis in patients with higher amplitudes especially of FR outside the SOZ. This indicates that it is possible that the true SOZ was not adequately detected. Another reason may be a more dispersed or multifocal epileptogenic zone/SOZ, a pattern which is too diffuse to permit a successful resective strategy, and so usually results in VNS implantation. Yet another reason for poor outcome may be that the result of the epileptogenic zone may include the actual epileptogenic zone (generating seizures before surgery) as well as a potential epileptogenic zone which is an area of the cortex that may generate seizures after the presurgical SOZ has been resected (Rosenow and Lüders, 2001). Seizures originating from areas not covered by electrodes but propagating to the actual electrode positions might lead to misinterpretation (Zijlmans et al., 2012). HFO are more stable and more expressed, and the likelihood for artifact contamination is lower, during sleep. However, a review of recently presented data indicates that the effect of sleep on HFO expression differs among regions (Dümpelmann et al., 2015) and so this phenomenon might influence the observation. Some patients also experience postoperative nausea and general discomfort so they rarely reach deep stages of sleep. The differences in HFO rates between the SOZ and other remote areas were disclosed in wakefulness periods as significant (Bagshaw et al., 2009). Based on this data we decided to analyze awake recordings. Our findings emphasize the importance of the careful interpretation of HFO, especially in cases with extensive spatial sampling or when there is an overlap between the epileptic and physiologic areas (Alkawadri et al., 2014). Based on our data, it is useful to include both ripples and fast ripples in the evaluation of the potential epileptogenic region (Zijlmans et al., 2012). Currently, there are no established criteria for distinguishing physiological from pathological HFO and SOZ areas (Engel et al., 2009; Jacobs et al., 112 M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 153 2016). Nevertheless, it seems that FR generated in the SOZ are more frequent, shorter, and have higher relative amplitudes than in other regions. There is, however, no clear cut-off value for these characteristics which can separate the SOZ, the IZ, and other regions. 5. Conclusion HFO parameters (rate, amplitude, and duration) differ in epileptic and non-epileptic regions. We suggest that amplitude and duration may be as important as frequency band and rate of HFO in marking the seizure onset region or epileptogenic area and may provide additional information on epileptogenicity. To conclude, FR are more frequent, shorter, and have higher relative amplitudes in the SOZ area than in other regions. The study suggests a possible worse prognosis in patients with higher amplitudes of FR outside the SOZ. Acknowledgments The study was supported by the ‘‘CEITEC – Central European Institute of Technology” project (CZ.1.05/1.1.00/02.0068) from the European Regional Development Fund and by MŠMT CˇR Research Program no. MSM0021622404. The technical part of the study was supported by the GACR project P103/11/0933 and the Application Laboratories of Advanced Microtechnologies and Nanotechnologies (CZ.1.05/2.1.00/01.0017), co-funded by the Operational Programme ‘‘Research and Development for Innovations”, the European Regional Development Fund, and the state budget. Supported by European Regional Development Fund - Project FNUSA-ICRC (No. CZ.1.05/1.1.00/02.0123) and Ministry of Education, Youth and Sports of the Czech Republic project no. LH15047 (KONTAKT II). Conflict of interest: None of the authors has any conflict of interest to disclose. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.clinph.2016.10. 011. References Akiyama T, McCoy B, Go CY, Ochi A, Elliott IM, Akiyama M, et al. Focal resection of fast ripples on extraoperative intracranial EEG improves seizure outcome in pediatric epilepsy. Epilepsia 2011;52:1802–11. Alkawadri R, Gaspard N, Goncharova II, Spencer DD, Gerrard JL, Zaveri H, et al. The spatial and signal characteristics of physiologic high frequency oscillations. Epilepsia 2014;55:1986–95. Andrade-Valença L, Mari F, Jacobs J, Zijlmans M, Olivier A, Gotman J, et al. Interictal high frequency oscillations (HFOs) in patients with focal epilepsy and normal MRI. Clin Neurophysiol 2012;123:100–5. Axmacher N, Cohen MX, Fell J, Haupt S, Dümpelmann M, Elger CE, et al. Intracranial EEG correlates of expectancy and memory formation in the human hippocampus and nucleus accumbens. Neuron 2010;65:541–9. Bagshaw AP, Jacobs J, LeVan P, Dubeau F, Gotman J. Effect of sleep stage on interictal high-frequency oscillations recorded from depth macroelectrodes in patients with focal epilepsy. Epilepsia 2009;50:617–28. Blanco JA, Stead M, Krieger A, Stacey W, Maus D, Marsh E, et al. Data mining neocortical high-frequency oscillations in epilepsy and controls. Brain 2011;134:2948–59. Bragin A, Wilson CL, Engel Jr J. Voltage depth profiles of high-frequency oscillations after kainic acid-induced status epilepticus. Epilepsia 2007;48(Suppl 5):35–40. Erratum in: Epilepsia 2007; 48:2379. Bragin A, Engel Jr J, Wilson CL, Fried I, Buzsáki G. High-frequency oscillations in human brain. Hippocampus 1999;9:137–42. Bragin A, Engel Jr J, Wilson CL, Fried I, Mathern GW. Hippocampal and entorhinal cortex high-frequency oscillations (100–500 Hz) in human epileptic brain and in kainic acid-treated rats with chronic seizures. Epilepsia 1999;40:127–37. Bragin A, Mody I, Wilson CL, Engel Jr J. Local generation of fast ripples in epileptic brain. J Neurosci 2002;22:2012–21. Bragin A, Wilson CL, Almajano J, Mody I, Engel Jr J. High-frequency oscillations after status epilepticus: epileptogenesis and seizure genesis. Epilepsia 2004;45:1017–23. Bragin A, Engel Jr J, Staba RJ. High-frequency oscillations in epileptic brain. Curr Opin Neurol 2010;23:151–6. Brázdil M, Halámek J, Jurák P, Daniel P, Kuba R, Chrastina J, et al. Interictal highfrequency oscillations indicate seizure onset zone in patients with focal cortical dysplasia. Epilepsy Res 2010;90:28–32. Brázdil M, Cimbálník J, Roman R, Shaw DJ, Stead MM, Daniel P, et al. Impact of cognitive stimulation on ripples within human epileptic and non-epileptic hippocampus. BMC Neurosci 2015;16:47. Buzsáki G, Horváth Z, Urioste R, Hetke J, Wise K. High frequency network oscillation in the hippocampus. Science 1992;256(5059):1025–7. Chrobak JJ, Buzsáki G. High-frequency oscillations in the output networks of the hippocampal–entorhinal axis of the freely behaving rat. J Neurosci 1996;16:3056–66. Crépon B, Navarro V, Hasboun D, Clemenceau S, Martinerie J, Baulac M, et al. Mapping interictal oscillations greater than 200 Hz recorded with intracranial macroelectrodes in human epilepsy. Brain 2010;133:33–45. Curio G. Linking 600 Hz ‘‘spikelike” EEG/MEG wavelets (‘‘1-bursts”) to cellular substrates: concepts and caveats. J Clin Neurophysiol 2000;17:377–96. Curio G, Mackert BM, Burghoff M, Neumann J, Nolte G, Scherg M, et al. Somatotopic source arrangement of 600 Hz oscillatory magnetic fields at the human primary somatosensory hand cortex. Neurosci Lett 1997;234:131–4. Dümpelmann M, Jacobs J, Schulze-Bonhage A. Temporal and spatial characteristics of high frequency oscillations as a new biomarker in epilepsy. Epilepsia 2015;56:197–206. Engel Jr J, Bragin A, Staba R, Mody I. High-frequency oscillations: what is normal and what is not? Epilepsia 2009;50:598–604. Haegelen C, Perucca P, Châtillon C-E, Andrade-Valença L, Zelmann R, Jacobs J, et al. High-frequency oscillations, extent of surgical resection, and surgical outcome in drug-resistant focal epilepsy. Epilepsia 2013;54:848–57. Jacobs J, LeVan P, Chander R, Hall J, Dubeau F, Gotman J. Interictal high-frequency oscillations (80–500 Hz) are an indicator of seizure onset areas independent of spikes in the human epileptic brain. Epilepsia 2008;49:1893–907. Jacobs J, LeVan P, Châtillon C-E, Olivier A, Dubeau F, Gotman J. High frequency oscillations in intracranial EEGs mark epileptogenicity rather than lesion type. Brain 2009;132:1022–37. Jacobs J, Zijlmans M, Zelmann R, Châtillon C-E, Hall J, Olivier A, et al. High-frequency electroencephalographic oscillations correlate with outcome of epilepsy surgery. Ann Neurol 2010;67:209–20. Jacobs J, Vogt C, LeVan P, Zelmann R, Gotman J, Kobayashi K. The identification of distinct high-frequency oscillations during spikes delineates the seizure onset zone better than high-frequency spectral power changes. Clin Neurophysiol 2016;127(1):129–42. Jiruska P, Bragin A. High-frequency activity in experimental and clinical epileptic foci. Epilepsy Res 2011;97:300–7. Jobst BC. Are HFOs still UFOs? The known and unknown about high frequency oscillations in epilepsy surgery. Epilepsy Curr. 2013;13:273–5. Koch SP, Werner P, Steinbrink J, Fries P, Obrig H. Stimulus-induced and statedependent sustained gamma activity is tightly coupled to the hemodynamic response in humans. J Neurosci 2009;29:13962–70. Lachaux JP, Axmacher N, Mormann F, Halgren E, Crone NE. High-frequency neural activity and human cognition: past, present and possible future of intracranial EEG research. Prog Neurobiol 2012;98:279–301. Lüders HO, Awad I. Conceptual considerations. In: Lüders HO, editor. Epilepsy surgery. New York: Raven Press; 1992. p. 51–62. Manning JR, Jacobs J, Fried I, Kahana MJ. Broadband shifts in local field potential power spectra are correlated with single-neuron spiking in humans. J Neurosci 2009;29:13613–20. Matsumoto A, Brinkmann BH, Stead SM, Matsumoto J, Kucewicz MT, Marsh WR, et al. Pathological and physiological high-frequency oscillations in focal human epilepsy. J Neurophysiol 2013;110:1958–64. Nagasawa T, Juhász C, Rothermel R, Hoechstetter K, Sood S, Asano E. Spontaneous and visually driven high-frequency oscillations in the occipital cortex: intracranial recording in epileptic patients. Hum Brain Mapp 2012;33: 569–83. Niessing J, Ebisch B, Schmidt KE, Niessing M, Singer W, Galuske RA. Hemodynamic signals correlate tightly with synchronized gamma oscillations. Science 2005;309(5736):948–51. Nishida M, Juhász C, Sood S, Chugani HT, Asano E. Cortical glucose metabolism positively correlates with gamma-oscillations in nonlesional focal epilepsy. Neuroimage 2008;42:1275–84. Rosenow F, Lüders H. Presurgical evaluation of epilepsy. Brain 2001;124:1683–700. Staba RJ, Wilson CL, Bragin A, Fried I, Engel Jr J. Quantitative analysis of highfrequency oscillations (80–500 Hz) recorded in human epileptic hippocampus and entorhinal cortex. J Neurophysiol 2002;88:1743–52. Urrestarazu E, Chander R, Dubeau F, Gotman J. Interictal high-frequency oscillations (100–500 Hz) in the intracerebral EEG of epileptic patients. Brain 2007;130:2354–66. Usui N, Terada K, Baba K, Matsuda K, Usui K, Tottori T, et al. Significance of veryhigh-frequency oscillations (over 1000 Hz) in epilepsy. Ann Neurol 2015;78:295–302. Wang S, Wang IZ, Bulacio JC, Mosher JC, Gonzalez-Martinez J, Alexopoulos AV, et al. Ripple classification helps to localize the seizure-onset zone in neocortical epilepsy. Epilepsia 2013;54:370–6. Worrell GA, Parish L, Cranstoun SD, Jonas R, Baltuch G, Litt B. High-frequency oscillations and seizure generation in neocortical epilepsy. Brain 2004;127:1496–550. M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 113 154 Worrell GA, Gardner AB, Stead SM, Hu S, Goerss S, Cascino GJ, et al. High-frequency oscillations in human temporal lobe: simultaneous microwire and clinical macroelectrode recordings. Brain 2008;131:928–37. Wu JY, Sankar R, Lerner JT, Matsumoto JH, Vinters HV, Mathern GW. Removing interictal fast ripples on electrocorticography linked with seizure freedom in children. Neurology 2010;75:1686–94. Zelmann R, Zijlmans M, Jacobs J, Châtillon C-E, Gotman J. Improving the identification of high frequency oscillations. Clin Neurophysiol 2009;120:1457–64. Zijlmans M, Jiruska P, Zelmann R, Leijten FS, Jefferys JG, Gotman J. High-frequency oscillations as a new biomarker in epilepsy. Ann Neurol 2012;71:169–78. 114 M. Pail et al. / Clinical Neurophysiology 128 (2017) 106–114 155 Annex 6: Dolezalova I, Brazdil M, Rektro I, Tyrlikova I, Kuba R. Syncope with atypical trunk convulsions in a patient with malignant arrhythmia. Epileptic Disorders. 2013;15(2):171-174. 156 doi:10.1684/epd.2013.0564 Epileptic Disord, Vol. 15, No. 2, June 2013 171 Correspondence: Irena Doleˇzalová Brno Epilepsy Center, First Department of Neurology, St Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Pekaˇrská 53, 656 91 Brno, Czech Republic Clinical commentary with video sequences Epileptic Disord 2013; 15 (2): 171-4 Syncope with atypical trunk convulsions in a patient with malignant arrhythmia Irena Doleˇzalová 1,2 , Milan Brázdil 1,2 , Ivan Rektor 1,2 , Ivana Tyrlíková 1 , Robert Kuba 1,2 1 Brno Epilepsy Center, First Department of Neurology, St Anne’s University Hospital and Faculty of Medicine 2 Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic Received October 9, 2012; Accepted February 17, 2013 ABSTRACT – Syncope is a condition often misdiagnosed as epilepsy. Syncope caused by cardiac disturbance is a life-threatening condition and accurate diagnosis is crucial for patient outcome. We present a case study of a 71-year-old woman who was referred to our epilepsy centre with a diagnosis of refractory epilepsy. We diagnosed convulsive syncope caused by malignant cardiac arrhythmia based on the presence of cardiac asystole lasting for 20-30 seconds, which was caused by sick sinus syndrome combined with third-degree atrioventricular block. The most prominent feature of this syncope was atypical trunk (abdominal or thoracoabdominal) convulsions, which were accompanied by other motor signs (head and eye deviation and brief jerks of the extremities). In the periods between attacks, all investigations, including standard 12-lead ECG and 24-hour ECG monitoring, were normal. This case study highlights the challenge in differential diagnosis of sudden loss of consciousness. [Published with video sequences] Key words: syncope, epilepsy, arrhythmia, asystole, trunk convulsion, sick sinus syndrome, central pattern generators The misdiagnosis of epilepsy is a relatively common occurrence. Of patients diagnosed with epilepsy who were referred to epilepsy centres, 20-30% were found to be misdiagnosed (Benbadis, 2009). The most common conditions misdiagnosed as epilepsy were psychogenic non-epileptic attacks, followed by syncope (Benbadis, 2009). Five basic mechanisms to describe the genesis of syncope have been described: syncope may be neurally mediated, sometimes referred to as a reflex (vasovagal syncope, carotid sinus syncope, and situation syncope), caused by orthostatic hypotension, cardiac arrhythmias, structural cardiac or pulmonary disease, or mediated by the central nervous system (ictal bradycardic syncope) (Crompton and Berkovic, 2009). Cardiac arrhythmias, in particular, can cause life-threatening events. We present a case study of a patient who was referred to our epilepsy 157 172 Epileptic Disord, Vol. 15, No. 2, June 2013 I. Doleˇzalová, et al. centre with a diagnosis of intractable epilepsy and later diagnosed with syncope and atypical trunk convulsions induced by malignant arrhythmia. Case study We present the case of a 71-year-old female who was treated for coronary artery disease, arterial hypertension, and type 2 diabetes mellitus. Her family and social history were insignificant. The first episode of unconsciousness occurred at the age of 70 without any triggering factors. Witnesses reported a sudden loss of consciousness, accompanied by brief jerks of the extremities and contractions of the abdominal wall. The patient was admitted to a local hospital, where physical examination and other investigations (basic laboratory examination, 12-lead ECG, 24-hour ECG monitoring, EEG, and brain CT) were performed. The laboratory examination and ECG monitoring were described to be normal. On EEG, there was a greater representation of theta waves in both temporal regions, but this was interpreted to be normal based on the age of the patient. Based on CT, there was only mild cortical atrophy with moderate ischaemic/degenerative changes of white matter. The frequency of seizures was approximately one seizure per month. After the second attack, antiepileptic medication with lamotrigine (LTG) was started, but was ineffective. Carbamazepine (CBZ) and levetiracetam (LEV) were then introduced without any significant influence on seizure frequency. One year later, the frequency of seizures suddenly increased and started to appear daily. For this reason, the patient was hospitalised in a local hospital and then referred to our epilepsy centre with a diagnosis of intractable epilepsy. She was admitted to our video-EEG unit in the late afternoon for urgent care and during the first night, three habitual paroxysmal Fp1- F7 F7 - T3 T3 - T5 T5 - O1 Fp2 - F8 F8 - T4 T4 - T6 T6 - O2 F3 - C3 C3 - P3 Fz - Cz Cz - pz F4 - C4 C4 - P4 Fp1- O1 Fp2 - O2 EKG - Ref 1s Figure 1. EEG during heart asystole. The heart asystole is followed by the deceleration and subsequent attenuation of background activity. The EEG is contaminated with muscular artefacts. events were recorded. For video-EEG monitoring, standard scalp electrode positions (10-20 system), onechannel ECG, and simultaneous video recordings were performed. The attacks always started with heart asystole, followed by diffuse slowing-down and attenuation of EEG, accompanied by muscle and movement artefacts (figure 1). Within approximately 10 seconds, the patient lost consciousness. Head and eye deviation appeared, followed by brief jerks of the upper and lower extremities, and subsequently distinct abdominal or thoracoabdominal convulsions developed. These convulsions lasted for 3-5 seconds. After spontaneous restitution of heart activity, the patient almost immediately regained consciousness without any signs of confusion, but nausea and occasional vomiting were present at the end (video sequence 1). We recorded one seizure in which only abdominal/thoracoabdominal contractions were present (video sequence 2). The duration of asystole ranged from 20 to 30 seconds. The patient was immediately admitted to our cardiology department. Based on the 12-channel ECG results, she was diagnosed with sick sinus syndrome, probably combined with third-degree atrioventricular block, and was urgently implanted with a Biotronic Talos DR pacemaker. After discharge from the hospital, the antiepileptic medication was withdrawn. The patient has now been free of syncopal attacks without any antiepileptic drugs for almost two years. Discussion This case study highlights the challenge in differential diagnosis of a loss of consciousness. An incorrect diagnosis may have severe consequences; the patient may not be treated appropriately, moreover, some antiepileptic drugs can be potentially 158 Epileptic Disord, Vol. 15, No. 2, June 2013 173 Syncope with convulsions and malignant arrhythmia harmful to patients with cardiac disease. The use of sodium channel blockers (for example, phenytoin and carbamazepine) in patients with cardiac-related syncope may worsen undetected heart disease through their negative chronotropic and dromotropic effects (Kenneback et al., 1991). The percentage of syncopes with motor manifestations is reported to vary; some convulsions were reported in 12% of blood donors who experienced vasovagal syncope, compared with 45% of patients with malignant ventricular arrhythmias and implanted defibrillators (Lin et al., 1982; Aminoff et al., 1988). In a study by Lempert et al. (1994), almost 90% of healthy volunteers, in whom syncope was provoked artificially by hyperventilation, squatting, and Valsalva manoeuvre, presented with syncope with some type of convulsion. However, we presume that this is an over-representation, influenced by the study design. The pattern of movement usually consists of multifocal arrhythmic jerks of the extremities, generalised myoclonus, and some additional movements such as head turns, oral automatisms, and righting movements (Crompton and Berkovic, 2009). In our patient, the most prominent motor characteristics were atypical trunk convulsions (we were unable to distinguish reliably between abdominal and thoracoabdominal contractions based on video recordings), a syncopal feature which has probably not been previously reported. Other similar reports include those of Ambrosetto et al. (2009) and Gasparini et al. (2011) who reported case studies of syncope with bipedal activity and gestural automatisms. The underlying pathological mechanism is believed to be the release, most likely disinhibition, of central pattern generators by anoxic cortical inactivation. Central pattern generators located at the subcortical level (mainly the brainstem and spinal cord) are responsible for rhythmic behaviour and are “normally” under neocortical control (Tassinari et al., 2009). In the literature, the characteristics of patient history and description of attacks, which enable physicians to differentiate between syncope and epileptic seizures, as well as identify different aetiologies of syncope, are summarised (Sheldon et al., 2002; Crompton and Berkovic, 2009). We would like to point out that history or physical signs of cardiac disease should lead to a high suspicion of cardiac aetiology (Alboni et al., 2001). In our patient, a history of coronary artery disease was known. It is important to stress that establishing a correct diagnosis can be complicated by the possible coexistence of syncope and seizure within one attack (Crompton and Berkovic, 2009). Attacks in which an initial syncopal event triggers an epileptic seizure are rare and occur mainly in children (Horrocks et al., 2005). Conversely, focal-onset seizures can cause bradycardia-induced syncope as a result of the disruption of cardiac autonomic neural discharges (Crompton and Berkovic, 2009). This case study is an illustrative example of the importance of differential diagnosis in patients with loss of consciousness. The differential diagnosis is crucial mainly in cases with atypical course progression or clinical manifestation and in elderly patients with somatic comorbidities. In our patient, establishing a correct diagnosis and the subsequent implantation of a cardiac pacemaker were probably lifesaving. Acknowledgments and disclosures. We thank Anne Johnson for grammatical assistance in writing the manuscript. This work was supported by the CEITEC-Central European Institute of Technology project (CZ.1.05/1.1.00/02.0068) from the European Regional Development Fund. None of the authors has any conflict of interests to disclose. Legends for videosequences Video sequence 1 00:00 development of asystole 00:10 head and eye deviation to the left, followed by brief myoclonus of right-sided extremities 00:14 trunk convulsions 00:24 heart rate reappears spontaneously 00:29 patient regains consciousness and is oriented Video sequence 2 00:00 development of asystole 00:10 trunk convulsions 00:20 heart rate reappears spontaneously 00:23 patient regains consciousness and is oriented Key words for video research on www.epilepticdisorders.com Syndrome: non epileptic paroxysmal disorder Etiology: syncope (cardiac) Phenomenology: head deviation; motor seizure (complex); nonepileptic paroxysmal event Localization: not applicable 159 174 Epileptic Disord, Vol. 15, No. 2, June 2013 I. Doleˇzalová, et al. References Alboni P, Brignole M, Menozzi C, et al. Diagnostic value of history in patients with syncope with or without heart disease. J Am Coll Cardiol 2001; 37: 1921-8. Ambrosetto G, Montagna P, Vetrugno R, Cortelli P. Paroxysmal bipedal activity during syncope related to carotid body tumor. Epilepsy Behav 2009; 15: 388-90. Aminoff MJ, Scheinman MM, Griffin JC, Herre JM. Electrocerebral accompaniments of syncope associated with malignant ventricular arrhythmias. Ann Intern Med 1988; 108: 791-6. Benbadis S. The differential diagnosis of epilepsy: a critical review. Epilepsy Behav 2009; 15: 15-21. Crompton DE, Berkovic SF. The borderland of epilepsy: clinical and molecular features of phenomena that mimic epileptic seizures. Lancet Neurol 2009; 8: 370-81. Gasparini S, Ferlazzo E, Cianci V, et al. Gestural automatisms during syncope related to cervical malignancy. Epilepsy Behav 2011; 20: 566-8. Horrocks A, Nechay A, Stephenson JBP, Zuberi SM. Anoxicepileptic seizures: observational study of epileptic seizures induced by syncopes. Arch Dis Child 2005; 90: 1283-7. Kenneback G, Bergfeldt L, Vallin H, Tomson T, Edhag O. Electrophysiologic effects and clinical hazards of carbamazepine treatment for neurologic disorders in patients with abnormalities of the cardiac conduction system. Am Heart J 1991; 121: 1421-9. Lempert T, Bauer M, Schmidt D. Syncope-a videometric analysis of 56 episodes of transient cerebral hypoxia. Ann Neurol 1994; 36: 233-7. Lin JTL, Ziegler DK, Lay CW, Bayer W. Convulsive syncope in blood donors. Ann Neurol 1982; 11: 525-8. Sheldon R, Rose S, Ritchie D, et al. Historical criteria that distinguish syncope from seizures. J Am Coll Cardiol 2002; 40: 142-8. Tassinari CA, Cantalupo G, Högl B, et al. Neuroethological approachtofrontolimbicepilepticseizuresandparasomnias: the same central pattern generators for the same behaviours. Rev Neurol 2009; 165: 762-8. 160 Annex 7: Dolezalova I, Brazdil M, Kahane P. Temporal lobe epilepsy? Things are not always what they seem. Epileptic Disorders. 2017;19(1):59-66. 161 doi:10.1684/epd.2017.0897 Epileptic Disord, Vol. 19, No. 1, March 2017 59 VIDEO ONLINE Correspondence: Irena Doleˇzalová Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Pekarska 53, Brno 65691 Czech Republic Electroclinical reasoning report Epileptic Disord 2017; 19 (1): 59-66 Temporal lobe epilepsy? Things are not always what they seem Irena Doleˇzalová 1 , Milan Brázdil 1,2 , Philippe Kahane 3 1 Brno Epilepsy Center, First Department of Neurology, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University 2 Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic 3 Clinique de Neurologie, University Hospital of Grenoble, France Received August 21, 2016; Accepted December 29, 2016 ABSTRACT – Temporal lobe epilepsy is the most frequent form of drugresistant epilepsy referred to epilepsy surgery centres. The vast majority of lesional cases can be operated on without invasive investigation which is often not the case for non-lesional cases. Invasive investigation in non-lesional cases, however, may lead to unexpected results, as illustrated in the following case report. [Published with video sequence on www.epilepticdisorders.com] Key words: temporal lobe epilepsy, invasive EEG, pseudotemporal epilepsy, posterior cingulate Clinical data A 33-year-old left-handed female patient started to have seizures at the age of 14. She was born from a normal uncomplicated pregnancy, and was treated for asthma in childhood. Familial history was irrelevant with the exception of a paternal cousin who suffered from moderate intellectual disability associated with epilepsy. The patient had a secondary school education and was working episodically as a civil servant. Her neurological examination was normal, except for a discrete right inferior facial palsy. The first episode was described as an abrupt loss of contact accompanied by abnormal verbal behaviour (she repeated the same questions several times), followed by headache. A diagnosis of epilepsy was established and the patient was subsequently treated with valproic acid, which failed to control the seizures. The seizures persisted on a weekly to monthly basis despite different AEDs used in monotherapy or in combination (carbamazepine, clobazam, lamotrigine, and levetiracetam). Aura was described as an epigastric constriction, sometimes associated with gustatory hallucinations. A seizure was observed by her neurologist who described a behavioural change (the patient seemed “confused”), a preserved ability to denominate objects, right-handed nose rubbing, and amnesia (with no recollection of denominating objects). 162 60 Epileptic Disord, Vol. 19, No. 1, March 2017 I. Doleˇzalová, et al. Figure 1. Preoperative MRI showing a left-sided thalamic infarction. Hippocampal sclerosis or other possible epileptogenic lesions were not identified. EEG showed infrequent right fronto-temporal sharp waves and left frontal spikes during drowsiness. MRI showed the cavity of a left thalamic infarct, without any other abnormalities (figure 1). Comments The overall clinical presentation appears to be similar to that of temporal lobe epilepsy (TLE), which can be considered “cryptogenic” despite the presence of a left thalamic infarct. The inconsistent gustatory aura, however, could suggest an early perisylvian involvement during the seizures (Hausser-Hauw and Bancaud, 1987). The neurological examination (right inferior facial palsy) indicates the involvement of the left hemisphere (Robillard et al., 1983), while ictal semiology (right-handed nose rubbing) could suggest right hemispheric involvement (Catenoix et al., 2004). Preservation of speech suggests the involvement of the non-dominant hemisphere for language (Gillig et al., 1988). Non-invasive investigations During video-EEG monitoring, seizures were characterized by an unpleasant epigastric constriction, inconsistent preservation of speech, loss of contact, a change in facial expression, oroalimentary automatisms (mastication), upper limb automatisms (mainly left), and nose wiping or face rubbing (without clear lateralization) (video sequence 1). Seizures lasted approximately one minute and the patient sometimes remained confused during the postictal state, without language deficit. Interictal EEG activity mainly showed, during sleep, interictal epileptiform discharges (IEDs) over the left fronto-temporal region (phase reversal at Fb1, T3; figure 2A) and, much less frequently, over the right frontal region (phase reversal at F4; figure 2B) or the left frontal region (phase reversal at F3; figure 2C). Seizures slightly differed in their EEG pattern, but typically indicated seizure onset over the left anterior temporal lobe, with secondary left hemispheric involvement and contralateral spread (figure 3). Neuropsychological examination showed visual and verbal encoding difficulties; executive functions and attention were described as normal. Functional MRI supported bilateral speech representation. 18FDG-PET showed bitemporal, but left-predominant, hypometabolism (figure 4). Comments Based on interictal EEG abnormalities, neuropsychological findings, and PET results, the patient could appear to have bitemporal epilepsy. However, ictal EEG consistently indicated the left hemisphere at seizure onset, and seizures systematically propagated over the contralateral hemisphere, which could explain bilateral dysfunctions. The association of left-sided TLE with seizures arising from the dominant hemisphere and preserved ictal speech would be rare. In these cases, an extratemporal lobe seizure origin with temporal propagation of ictal activity seems to be more probable (Kaiboriboon et al., 2006). However, the structural or functional damage of a dominant hemisphere by epileptic activity can cause neuronal reorganization and redistribution of “critical” functions, as has been reported for speech and handedness which were shown to shift to the other hemisphere (Orsini and Satz, 1986; Adcock et al., 2003). This could be supported by left-handedness and bilateral speech representation on functional MRI. However, the distribution of IEDs over both frontal regions raised a red flag which forced us to also consider extra-TLE with propagation of ictal activity to the temporal region. Invasive EEG (iEEG) was considered as mandatory to answer the following question: does this patient have TLE, temporal “plus” epilepsy (Barba et al., 2007), or extra-TLE? 163 Epileptic Disord, Vol. 19, No. 1, March 2017 61 Temporal lobe epilepsy? f7 fb1 fb1 tp1 tp1 t5 fp1 f7 f7 t3 t3 t5 t5 o1 fp1 f3 f3 c3 c3 p3 p3 o1 fz cz cz pz f8 fb2 fb2 tp2 tp2 t6 fp2 f8 f8 t4 t4 t6 t6 o2 fp2 f4 f4 c4 c4 p4 1 sec p4 o2 ecg2 ecg1 f7 fb1 fb1 tp1 tp1 t5 fp1 f7 f7 t3 t3 t5 t5 o1 fp1 f3 f3 c3 c3 p3 p3 o1 fz cz cz pz f8 fb2 fb2 tp2 tp2 t6 fp2 f8 f8 t4 t4 t6 t6 o2 fp2 f4 f4 c4 c4 p4 1 sec p4 o2 ecg2 ecg1 A B Figure 2. Distribution of interictal epileptiform discharges (IEDs) on scalp EEG. (A) The majority of IEDs were present in the left-sided fronto-temporal area (phase reversal at electrodes Fb1, T3). (B) There were also less frequent IEDs over the right-sided frontal area (phase reversal at electrodes F4). 164 62 Epileptic Disord, Vol. 19, No. 1, March 2017 I. Doleˇzalová, et al. f7 fb1 fb1 tp1 tp1 t5 fp1 f7 f7 t3 t3 t5 t5 o1 fp1 f3 f3 c3 c3 p3 p3 o1 fz cz cz pz f8 fb2 fb2 tp2 tp2 t6 fp2 f8 f8 t4 t4 t6 t6 o2 fp2 f4 f4 c4 c4 p4 p4 o2 ecg2 ecg1 1 sec C Figure 2. (Continued ). (C) Rarely, IEDs could be found over the left frontal regions (phase reversal at electrode F3). f7 fb1 fb1 tp1 tp1 t5 fp1 f7 f7 t3 t3 t5 t5 o1 fp1 f3 f3 c3 c3 p3 p3 o1 fz cz cz pz f8 fb2 fb2 tp2 tp2 t6 fp2 f8 f8 t4 t4 t6 t6 o2 fp2 f4 f4 c4 c4 p4 p4 o2 ecg1 ecg2 1 sec Figure 3. Ictal scalp EEG showing the typical pattern for the patient. The seizures started with a short flattening over the left anterior temporal lobe, followed by irregular theta waves in the same area. After, clear-cut rhythmic activity appeared; this rhythmic activity started on the left side, but rapidly spread over the right hemisphere. The left-sided ictal activity predominated throughout the entire seizure. Invasive investigation A large temporo-perisylvian stereoelectroencephalography (sEEG) study was performed that targeted the mesial and lateral temporal lobe structures, as well as the extra-temporal lobe regions that can trigger temporal lobe symptoms (e.g. the orbito-frontal cortex, insular cortex, and posterior cingulate cortex) (figure 5). IEDs either involved mesiotemporal structures or, independently, the posterior cingulate region (figure 6A) . All seizures, however, started within the posterior cingulate before becoming symptomatic when they involved the mesiotemporal lobe structures (figure 6B and 7). Comments This patient is a typical example of what has been described as pseudo-temporal epilepsy, in which seizures start in extra-temporal lobe structures and later spread over the temporal lobe region, therefore 165 Epileptic Disord, Vol. 19, No. 1, March 2017 63 Temporal lobe epilepsy? Figure 4. Preoperative positron emission tomography (PET) showing bilateral temporal hypometabolism, which was slightly more pronounced on the left side. mimicking temporal semiology (Andermann, 2003). Extra-temporal lobe regions include the orbito-frontal cortex (Shihabuddin et al., 2001), the insula (Isnard et al., 2000), and the parieto-occipital regions (Williamson et al., 1992; Palmini et al., 1993; Liava et al., 2014; Francione et al., 2015), especially the posterior cingulate (Devinsky et al., 1995; Alkawadri et al., 2013; Enatsu et al., 2014). Alkawadri et al. (2013) reported a group of 14 patients with cingulate epilepsy; within this group there were also four patients with posterior cingulate epilepsy and their seizures resembled temporal lobe seizures in all patients (Alkawadri et al., 2013). Enatsu et al. (2014) analysed a case series of seven patients with posterior cingulate epilepsy, four of whom exhibited temporal lobe seizures and the rest had motor seizures (including bilateral tonic seizures and hypermotor seizures). The differential diagnosis between temporal and pseudo-temporal epilepsy remains a diagnostic challenge, especially in non-lesional cases. The distribution of IEDs can be very helpful in terms of searching for a solution. IEDs in TLE often remain limited to the temporal lobe region; in other types of epilepsies they spread over different regions (Remi et al., 2011). However, IEDs in posterior cingulate epilepsy have a large tendency to propagate into the temporal region, as was demonstrated by Alkawadri et al. (2013). These authors reported “typical” temporal IEDs in three out of four patients (in two patients over anterior temporal areas and in one patient over posterior temporal areas); in contrast, frontal IEDs were found only in one patient (Alkawadri et al., 2013). Also, Elwan et al. (2013) did not findanystatisticalsignificantdifferencesbetweentemporal and pseudo-TLE when analysing the distribution of IEDs and seizure-onset pattern on scalp EEG (Elwan et al., 2013). To sum up, IEDs often also propagate over temporal areas in extra-temporal epilepsies. However, the presence of IEDs over extra-temporal areas is very often associated with an extra-temporal seizure origin despite typical temporal seizure semiology. Action taken Three surgical strategies could be considered: – perform a standard temporal lobe resection, i.e. a resection of the symptomatogenic zone; this X’ Y’ Z’ V’ S’ R’ U’ C’ F’ E’ B’ Et’ A’ T’ I’ O’ Figure 5. Electrode positions for invasive EEG monitoring. The positionsofelectrodesinthelefttemporallobearethefollowing: A’: amygdala; B’: anterior hippocampus; C’: posterior hippocampus; I’: temporal pole; Et’: entorhinal cortex; T’, U’: temporal operculum; E’, F’: temporobasal region. The positions of elec- trodesintheleftextra-temporallobestructuresarethefollowing: O’: fronto-orbital cortex; R’: frontal operculum; S’: parietal operculum; V’: cingulum (posterior part); X’: anterior insula (dotted line); Y’: posterior insula (dotted line); Z’ (dotted line). 166 64 Epileptic Disord, Vol. 19, No. 1, March 2017 I. Doleˇzalová, et al. A B r’7 r’8 s’7 s’8 v’1 v’2 v’14 v’15 x’1 x’2 x’7 x’8 y’1 y’2 y’8 y’9 t’1 t’2 t’7 t’8 u’1 u’2 u’7 u’8 o’2 o’3 o’5 o’6 o’10 o’11 i’3 i’4 a’1 a’2 a’10 a’11 et’1 et’2 et’5 et’6 et’10 et’11 b’1 b’2 b’10 b’11 c’1 c’2 c’11 c’12 e’1 e’2 e’11 e’12 f’1 f’2 f’9 f’10 G2 G2 Fz cz G2 G2 1 sec r’7 r’8 s’7 s’8 v’1 v’2 v’14 v’15 x’1 x’2 x’7 x’8 y’1 y’2 y’8 y’9 t’1 t’2 t’7 t’8 u’1 u’2 u’7 u’8 o’2 o’3 o’5 o’6 o’10 o’11 i’3 i’4 a’1 a’2 a’10 a’11 et’1 et’2 et’5 et’6 et’10 et’11 b’1 b’2 b’10 b’11 c’1 c’2 c’11 c’12 e’1 e’2 e’11 e’12 f’1 f’2 f’9 f’10 G2 G2 Fz cz G2 G2 ecg2 ecg1 1sec Figure 6. The results of invasive EEG. (A) Distribution of interictal epileptiform discharges (IEDs) on invasive EEG. IEDs dominated in the area of left-sided mesiotemporal structures (left amygdala: electrode Aˇı1-2; left entorhinal cortex: electrode Etˇı1-2; and left hippocampus: electrodes Bˇı1-2 and Cˇı1-2), but there were also IEDs and bursts of rapid activity (blue arrow) in the area of the left posterior cingulum (electrode Vˇı1-2). (B) Ictal findings on invasive EEG. The seizures started with low-voltage fast (LVF) activity in the left side of the posterior cingulum (first arrow); LVF activity propagated to the left-sided mesiotemporal structures, where the symptomatogenic zone was localized (second arrow). strategy should be considered as only palliative and could postoperatively aggravate the memory deficit; – perform a new iEEG study in order to better delineate the amount of the mesial parietal cortex that should be resected; this is a possible option, although the benefit/risk ratio cannot be clearly evaluated; – perform a very focal resection of the posterior cingulate cortex surrounding the cingulate electrode. The latter strategy was finally chosen, as the shape and signal of that part of the posterior cingulate retrospectively appeared “abnormal” (figure 8). The histopathological specimen, however, proved negative.Anotherpossibilitywouldbetoperformthermocoagulation in the area of the posterior cingulum. The effectiveness of this procedure is not very high; only 7% of patients remain seizure-free one year after surgery (Bourdillon et al., 2016). 167 Epileptic Disord, Vol. 19, No. 1, March 2017 65 Temporal lobe epilepsy? Figure 7. More detailed view of the position of electrode Vˇı. The low-voltage fast activity started in the mesial contacts of this electrode, encompassing the left side of the posterior cingulum, as seen in this image. Follow-up The patient is completely seizure-free three years after surgery (Class 1; according to International League Against Epilepsy Classification [Wieser et al., 2001]). AEDs have been substantially reduced and at present, the patient is only being given 800 mg of carbamazepine. Interictal EEG shows infrequent right temporal sharp waves. Conclusion This case report illustrates that atypical presentation of non-lesional TLE epilepsy must be referred for iEEG evaluation. Invasive evaluation must be designed to allow a distinction between mesial and neocortical temporal lobe onset, as well as consideration for possible extra-temporal onset, as in the present case. Supplementary data. Summary didactic slides are available on the www.epilepticdisorders.com website. Disclosures. None of the authors have any conflict of interest to declare. Legend for video sequence A typical patient seizure. The seizure semiology appears late (approximately 30 seconds from the first change in EEG). In this first seizure, limb automatisms of the right arm are observed and speech is preserved during the seizure. In the postictal state, the patient was amnesic for the seizure. Key words for video research on www.epilepticdisorders.com Phenomenology: aura (abdominal), automatisms Localisation: cingulate gyrus, temporal (left) Epilepsy syndrome: focal non-idiopathic parietal Aetiology: not applicable Figure 8. The second preoperative MRI. We retrospectively identified a discrete change on MRI (yellow circle). The position of seizure onsets is marked by red dots. References Adcock JE, Wise RG, Oxbury JM, Oxbury SM, Matthews PM. Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy. Neuroimage 2003; 18(2): 423-38. Alkawadri R, So NK, Van Ness PC, Alexopoulos AV. Cingulate epilepsy: report of 3 electroclinical subtypes with surgical outcomes. JAMA Neurol 2013; 70(8): 995-1002. Andermann F. Pseudotemporal vs neocortical temporal epilepsy: things aren’t always where they seem to be. Neurology 2003; 61(6): 732-3. Barba C, Barbati G, Minotti L, Hoffmann D, Kahane P. Ictal clinical and scalp-EEG findings differentiating temporal lobe epilepsies from temporal ‘plus’ epilepsies. Brain 2007; 130(7): 1957-67. Bourdillon P, Isnard J, Catenoix H, et al. Stereo electroencephalography-guided radiofrequency thermocoagulation (SEEG-guided RF-TC) in drug-resistant focal epilepsy: results from a 10-year experience. Epilepsia 2016. Ahead of print. Catenoix H, Guenot M, Isnard J, Fischer C, Mauguiere F, Ryvlin P. Intracranial EEG study of seizure-associated nose wiping. Neurology 2004; 63(6): 1127-9. Devinsky O, Morrell MJ, Vogt BA. Contributions of anterior cingulate cortex to behaviour. Brain 1995; 118(1): 279-306. Elwan SA, So NK, Enatsu R, Bingaman WE. Pseudotemporal ictalpatternscomparedwithmesialandneocorticaltemporal ictal patterns. J Clin Neurophysiol 2013; 30(3): 238-46. Enatsu R, Bulacio J, Nair DR, et al. Posterior cingulate epilepsy: clinical and neurophysiological analysis. J Neurol Neurosurg Psychiatry 2014; 85(1): 44-50. Francione S, Liava A, Mai R, et al. Drug-resistant parietal epilepsy: polymorphic ictal semiology does not preclude good post-surgical outcome. Epileptic Disord 2015; 17(1): 32-46. Gillig P, Sackellares JC, Greenberg HS. Right hemisphere partial complex seizures: mania, hallucinations, and speech disturbances during ictal events. Epilepsia 1988; 29(1): 26-9. Hausser-Hauw C, Bancaud J. Gustatory hallucinations in epileptic seizures. Electrophysiological, clinical and anatomical correlates. Brain 1987; 110(2): 339-59. 168 66 Epileptic Disord, Vol. 19, No. 1, March 2017 I. Doleˇzalová, et al. Isnard J, Guenot M, Ostrowsky K, Sindou M, Mauguiere F. The role of the insular cortex in temporal lobe epilepsy. Ann Neurol 2000; 48(4): 614-23. Kaiboriboon K, Parent JM, Barbaro NM, Walker JA, Garcia PA.Speechpreservationduringlanguage-dominant,lefttemporal lobe seizures: report of a rare, potentially misleading finding. Epilepsia 2006; 47(8): 1343-6. Liava A, Mai R, Tassi L, et al. Paediatric epilepsy surgery in the posterior cortex: a study of 62 cases. Epileptic Disord 2014; 16(2): 141-64. Orsini DL, Satz P. A syndrome of pathological lefthandedness. Correlates of early left hemisphere injury. Arch Neurol 1986; 43(4): 333-7. Palmini A, Andermann F, Dubeau F, et al. Occipitotemporal epilepsies: evaluation of selected patients requiring depth electrodes studies and rationale for surgical approaches. Epilepsia 1993; 34(1): 84-96. TEST YOURSELF EDUCATION (1) In pseudotemporal epilepsy, A. do seizures always spread to mesio-temporal lobe structures? B. is seizure onset localized to the extra-temporal lobe structures? C. is the seizure semiology similar to that of temporal lobe seizures? D. is an extra-temporal lobe lesion always present on MRI? (2) Which type of extra-temporal epilepsy can manifest as pseudotemporal epilepsy (more than one answer may be correct)? A. Occipital B. Parietal C. Frontal D. Insular (3) Which of the following examinations is the most relevant in the diagnosis of non-lesional pseudo-temporal epilepsy? A. Video-EEG monitoring B. 18-FDG PET C. Neuropsychological evaluation D. Invasive EEG Note: Reading the manuscript provides an answer to all questions. Correct answers may be accessed on the website, www.epilepticdisorders.com, under the section “The EpiCentre”. Remi J, Vollmar C, de Marinis A, Heinlin J, Peraud A, Noachtar S. Congruence and discrepancy of interictal and ictal EEG with MRI lesions in focal epilepsies. Neurology 2011; 77(14): 1383-90. Robillard A, Saint-Hilaire JM, Mercier M, Bouvier G. The lateralizing and localizing value of adversion in epileptic seizures. Neurology 1983; 33(9): 1241-2. Shihabuddin B, Abou-Khalil B, Delbeke D, Fakhoury T. Orbito-frontal epilepsy masquerading as temporal lobe epilepsy-a case report. Seizure 2001; 10(2): 134-8. Wieser HG, Blume WT, Fish G, et al. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001; 42(2): 282-6. Williamson PD, Boon PA, Thadani VM, et al. Parietal lobe epilepsy: diagnostic considerations and results of surgery. Ann Neurol 1992; 31(2): 193-201. 169 Annex 8: Dolezalova I, Schachter S, Chrastina J, Hemza J, Hermanova M, Rektor I, Pazourkova M, Brazdil M. Atypical handedness in mesial temporal lobe epilepsy. Epilepsy & Behavior. 2017;72:78-81. 170 Atypical handedness in mesial temporal lobe epilepsy Irena Doležalová a , Steven Schachter b , Jan Chrastina c , Jan Hemza c , Markéta Hermanová d , Ivan Rektor a,e , Marta Pažourková f , Milan Brázdil a,e, ⁎ a Brno Epilepsy Center, First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic b Harvard Medical School, Boston, MA, USA c Brno Epilepsy Center, Department of Neurosurgery, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic d Brno Epilepsy Center, First Department of Pathological Anatomy, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic e Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic f Brno Epilepsy Center, Department of Radiology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic a b s t r a c ta r t i c l e i n f o Article history: Received 6 November 2016 Revised 2 January 2017 Accepted 26 January 2017 Available online 7 June 2017 Objective: The main aim of our study was to investigate the handedness of patients with mesial temporal lobe epilepsy (MTLE). We also sought to identify clinical variables that correlated with left-handedness in this population. Methods: Handedness (laterality quotient) was assessed in 73 consecutive patients with MTLE associated with unilateral hippocampal sclerosis (HS) using the Edinburgh Handedness Inventory. Associations between rightand left-handedness and clinical variables were investigated. Results: We found that 54 (74.0%) patients were right-handed, and 19 (26%) patients were left-handed. There were 15 (36.6%) left-handed patients with left-sided seizure onset compared to 4 (12.5%) left-handed patients with right-sided seizure onset (p = 0.030). Among patients with left-sided MTLE, age at epilepsy onset was significantly correlated with handedness (8 years of age [median; min-max 0.5–17] in left-handers versus 15 years of age [median; min-max 3–30] in right-handers (p b 0.001). Conclusions: Left-sided MTLE is associated with atypical handedness, especially when seizure onset occurs during an active period of brain development, suggesting a bi-hemispheric neuroplastic process for establishing motor dominance in patients with early-onset left-sided MTLE. © 2017 Elsevier Inc. All rights reserved. Keywords: Mesial temporal lobe epilepsy Handedness Left-handed Right-handed Age at epilepsy onset Atypical dominance 1. Introduction The left and right hemispheres differ in their functional organization and specialization. The left hemisphere is typically responsible for processes involved in language and verbal memory. Left-sided language dominance is present in nearly all right-handers as well as the majority of left-handers. Damage to the left hemisphere in utero or during early childhood is often associated with network reorganization, resulting in atypical lateralization of hemispheric functions such as language, verbal memory, and motor dominance (handedness) [1]. For example, several authors have demonstrated atypical lateralization of language and memory functions in patients with left-sided mesial temporal lobe epilepsy (L-MTLE) [2–7]. Based on this work, it appears that the crucial variable determining the likelihood of cerebral reorganization is the age at epilepsy onset; i.e., the lower the age of epilepsy onset, the greater the probability of atypical lateralization of function [6]. In addition to atypical hemispheric dominance for language and verbal memory associated with L-MTLE, there is a correlation between leftsided cerebral damage or seizure onset and atypical handedness [8–14]. For example, in 1959, Penfield and Roberts reported that 27% of 246 patients with epilepsy and focal left-sided brain injury were left-handed compared to only 8% of 276 patients with epilepsy and focal rightsided brain injury [10]. What remains unclear is the extent to which frontal lobe pathology, with or without frontal lobe seizure onset, is a pre-requisite for atypical handedness in patients with epilepsy, given the localization of motor control in the frontal lobe. In light of the growing evidence that MTLE is a brain network disease, with widespread extra-temporal anatomic and functional alterations [15], we hypothesized that MTLE would be associated with atypical handedness. We further speculated that atypical handedness would be associated with L-MTLE, reflecting the increased likelihood of atypical handedness lateralization associated with left-sided lesions, and early age of seizure onset, given the known correlation between young age and atypical hemispheric function [1,16]. The main aim of our study was therefore to assess handedness in patients with MTLE. We also sought to identify clinical variables that correlated with left-handedness in this population. Epilepsy & Behavior 72 (2017) 78–81 ⁎ Corresponding author at: Brno Epilepsy Center, First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Pekařská 53, 656 91 Brno, Czech Republic. E-mail address: mbrazd@muni.cz (M. Brázdil). http://dx.doi.org/10.1016/j.yebeh.2017.01.034 1525-5050/© 2017 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh 171 2. Methods We retrospectively analyzed handedness in 73 consecutive adult patients (41 females, 32 males) with drug-refractory MTLE associated with unilateral hippocampal sclerosis (HS). Handedness was assessed with the Edinburgh Handedness Inventory [17]. A laterality quotient (LQ) which ranged from −100 to 100 was calculated for each patient as described previously [17]. Patients with LQ from 100 to 0 were defined as right-handers. Patients with LQ from 0 to −100 were defined as left-handers. All patients underwent investigation for epilepsy surgery, including brain MRI, neuropsychological tests, and interictal/ictal semi-invasive video-EEG monitoring with sphenoidal electrodes, including ictal semiology analysis. Positron emission tomography (PET), interictal/ictal single photon emission tomography (SPECT), subtraction ictal SPECT co-registered to MRI (SISCOM) and invasive EEG (SEEG) were performed only in a subset of patients based on their clinical data and expert consensus. The inclusion criteria differ between surgically and medically treated patients. In medical group, only patients with unilateral temporal seizure onset and anatomically concordant HS on MRI were included. The HS was radiologically defined as unilateral hippocampal atrophy and increased signal in the hippocampus [18]. We excluded patients with suspicion of dual pathology, including bilateral HS, and bilateral seizure onset. In surgical group, we included only patients who underwent anteromesial temporal lobe resection, the presence of HS was confirmed histopathologically and was classified as having Engel 1a outcome at one year postsurgical follow-up [19]. Demographic data, including age at epilepsy onset, duration of epilepsy, performance of surgery, number of seizures per month, occurrence of generalized tonic-clonic seizures (GTCS), and type or age of initial precipitating injury were obtained by medical chart review. The statistical analysis utilized Fisher's exact test or Mann-Whitney test according to their condition of validity. For all tests, a p value b0.05 was considered statistically significant. All patients gave their informed consent prior to participation in the study and study approval was granted by the ethical committee at St. Anne's University Hospital. 3. Results Among 73 enrolled patients with MTLE (46 [63%] patients underwent surgery, 27 [37%] were treated medically), HS was lateralized to the right in 32 (43.8%) patients and to the left in 41 (56.2%) patients. We did not find any statistically significant differences in demographics between patients with R-MTLE and L-MTLE (Table 1). Overall, 54 (74.0%) patients were right-handed and 19 (26%) patients were left-handed. The proportion of left-handers was statistically significantly higher in patients with L-MTLE than in patients with R-MTLE: 15 (36.6%) patients with L-MTLE were left-handed compared with 4 (12.5%) patients with R-MTLE (p = 0.030, Fig. 1). We then assessed for clinical variables that correlated with handedness in patients with L-MTLE, and, separately, in patients with R-MTLE (Table 2). This revealed that in patients with L-MTLE MTLE, age of epilepsy onset correlated with handedness (Fig. 2). Left-handed patients with L-MTLE had an earlier age of epilepsy onset (median 8 years of age, min-max 0.5–17) compared with right-handed patients with L-MTLE (median 15 years of age, min-max 3–30; p b 0.001). No other clinical variables, including the presence of initial precipitating injury, correlated with handedness in patients with L-MTLE, and there were no clinical variables associated with handedness in the patients with R-MTLE. Table 1 The comparison of demographic data between patients with right- and left-sided MTLE. Right-sided MTLE (n = 32) Left-sided MTLE (n = 41) p-Value Age at epilepsy onset Median (min-max) 13.5 (0.5–41) 13 (0.5–30) 1.000 Duration of epilepsy Median (min-max) 15 (5–20) 15 (6–26) 1.000 Surgery Y/N n(%) 30 (93.8)/2(6.2) 38 (92.7)/3 (7.3) 1.000 Number of seizures per month ≥5 seizures/b5 seizures n(%) 26 (81.3)/6 (18.7) 34 (82.9)/17 (17.1) 1.000 GTCS Y/N n(%) 14 (43.8)/18 (56.3) 16 (39.0)/25 (61.0) 0.811 Initial precipitating injury n (%) 9 (28.1) 19 (46.3) 0.147 GTCS – generalized tonic-clonic seizures, MTLE – mesial temporal lobe epilepsy, N – no, Y – yes. Fig. 1. Difference in left-handers representation between patients with left- and rightsided MTLE MTLE – mesial temporal lobe epilepsy. Table 2 Comparison of demographic data between left- and right-handers, analysis done separately for left-sided MTLE and right-sided MTLE. Left- sided MTLE Left-handers (n = 15) Right-handers (n = 26) p-Value Age at epilepsy onset Median (min-max) 8 (0.5–17) 15 (3–30) b0.001 Duration of epilepsy Median (min-max) 16 (6–26) 16 (8–25) 0.103 Surgery Y/N n(%) 13 (86.7)/2 (13.3) 25 (96.2)/1 (3.8) 0.543 Number of seizures per month ≥5 seizures/b5 seizures n(%) 11 (73.3)/4 (26.7) 23 (88.5)/3 (11.5) 0.390 GTCS Y/N n(%) 9 (34.6)/17 (65.4) 7 (46.7)/8 (53.3) 0.517 Initial precipitating injury n (%) 7 (46.6) 12 (46.2) 1.000 Right-sided MTLE Left-handers (n = 4) Right-handers (n = 28) p-Value Age at epilepsy onset Median (min-max) 19 (1.5–30) 13.5 (0.5–41) 1.000 Duration of epilepsy Median (min-max) 15 (10–20) 14.5 (5–25) 0.129 Surgery Y/N n(%) 3 (75)/1 (25) 27 (96.4)/1 (3.6) 0.238 Number of seizures per month ≥ 5 seizures/b 5 seizures n(%) 4 (100)/0 (0) 22 (78.6)/6 (21.4) 0.566 GTCS Y/N n(%) 2 (50)/2 (50) 12 (42.9)/16 (57.1) 1.000 Initial precipitating injury n (%) 9 (32.1) 0 (0) 0.303 GTCS – generalized tonic-clonic seizures, MTLE – mesial temporal lobe epilepsy, N – no, Y – yes. 79I. Doležalová et al. / Epilepsy & Behavior 72 (2017) 78–81 172 4. Discussion There is compelling evidence in the literature supporting altered lateralization of language and verbal memory functions in patients with MTLE. Approximately 20–30% of patients with L-MTLE exhibit atypical (bilateral or right-sided) language dominance. In contrast, atypical language dominance is rare in patients with R-MTLE [6,7]. Similar reorganization of verbal memory lateralization has been reported. For example, Richardson et al. proved atypical activation of right hippocampus in leftsided MTLE [2]. Powell et al. confirmed these results, and moreover, demonstrated that reorganization of memory was not sufficient to support normal memory functioning [3]. An analysis of verbal memory processing (encoding and retrieval) showed different patterns of activation among healthy controls, patients with L-MTLE and patients with R-MTLE [4]. For verbal memory processes, there was greater activation of frontal, temporal, and parietal lobes in L-MTLE patients than in healthy controls. Less information is available for handedness patterns in patients with MTLE. We published a case report of a 19-year-old female patient with drug-refractory MTLE [20] who had unequivocal shift in handedness correlated in time with epilepsy onset. We hypothesized that this handedness shift was caused by seizures. In the present study, we extend the earlier findings by showing a 3-fold elevation of lefthandedness in L-MTLE compared to R-MTLE and, further, a correlation with age of epilepsy onset. The similar increase of left-handedness was present in the study of Sveller et al. (2006) who analyzed the group of 74 patients with left-sided epilepsy and 70 control subjects. There were 18% left-handers in patients' group in comparison with 4% left-handers in control group [21]. The higher incidence of atypical handedness (left-handedness and ambidexterity) in association with epilepsy was also present in study of Slezicki et al. (2009). Handedness was not associated with sex, age, seizure type, age at epilepsy onset, type or side of EEG or brain imaging abnormalities, family history of seizures, drug-resistance, or history of epilepsy surgery [22]. This study has several limitations. First, we included both operated and unoperated patients. In operated patients, the side of epileptogenic zone was confirmed by complete seizure freedom after surgery. In unoperated patients, while we required a high level of consistency in pre-surgical data, we lack definitive confirmation of the seizure focus. Second, we utilized the Laterality Quotient as originally described by Oldfield [17], though another scoring method, the Laterality Score, is more sensitive to atypical or pathological left-handedness. Third, handedness in family members was not fully evaluated in our study due to incomplete data in some patients. Finally, hand preference, especially for writing, can vary by cultural background [23]. However, all participants in our study were from the Czech or Slovak Republics, with a left-handedness rate among the general population of 10% (National Health Institute official report), similar to the proportion of lefthanded patients with R-MTLE in our study (12%). This suggests that the elevated rate of left-handedness in patients with L-MTLE represents so-called “pathological” left-handedness. Interestingly, the only clinical variable that correlated with lefthandedness in patients with L-MTLE was young age at epilepsy onset. The median age at epilepsy onset was 8 years in left-handers compared with a median age of 15 years in right-handers. This suggests that the reorganization of motor dominance is age-dependent, as has also been demonstrated for language dominance [6]. Our results are further supported by other studies focusing on structural and functional brain changes associated with long-term MTLE [16,24–26]. For example, Janszky et al. showed that interictal epileptiform discharges activated brain plasticity and subsequent hemispheric reorganization for language functions from left to right [26]. We presume a similar phenomenon is responsible for reorganization of motor dominance, especially given the evidence of altered connectivity between temporal, frontal, and parietal lobes in MTLE [16], but this requires further study. In addition, it would be valuable to utilize functional MRI in this population for confirmation of hemispheric dominance and to correlate handedness with lateralization of language function and verbal memory. In conclusion, L-MTLE is associated with atypical handedness, especially when seizure onset occurs during an active period of brain development, suggesting a bi-hemispheric neuroplastic process for establishing motor dominance in patients with early-onset left-sided MTLE. Conflict of interest The authors have no conflicts of interest to disclose. References [1] Renteria ME. Cerebral asymmetry: a quantitative, multifactorial, and plastic brain phenotype. Twin Res Hum Genet 2012;15:401-13. [2] Richardson MP, Strange BA, Duncan JS, Dolan RJ. Preserved verbal memory function in left medial temporal pathology involves reorganisation of function to right medial temporal lobe. Neuroimage 2003;20(Suppl. 1):S112-9. [3] Powell HW, Richardson MP, Symms MR, Boulby PA, Thompson PJ, Duncan JS, et al. Reorganization of verbal and nonverbal memory in temporal lobe epilepsy due to unilateral hippocampal sclerosis. Epilepsia 2007;48:1512-25. [4] Alessio A, Pereira FR, Sercheli MS, Rondina JM, Ozelo HB, Bilevicius E, et al. Brain plasticity for verbal and visual memories in patients with mesial temporal lobe epilepsy and hippocampal sclerosis: an fMRI study. Hum Brain Mapp 2013;34:186-99. [5] Brazdil M, Chlebus P, Mikl M, Pazourkova M, Krupa P, Rektor I. Reorganization of language-related neuronal networks in patients with left temporal lobe epilepsy — an fMRI study. Eur J Neurol 2005;12:268-75. [6] Brazdil M, Zakopcan J, Kuba R, Fanfrdlova Z, Rektor I. Atypical hemispheric language dominance in left temporal lobe epilepsy as a result of the reorganization of language functions. Epilepsy Behav 2003;4:414-9. [7] Adcock JE, Wise RG, Oxbury JM, Oxbury SM, Matthews PM. Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy. Neuroimage 2003;18:423-38. [8] Dellatolas G, Luciani S, Castresana A, Remy C, Jallon P, Laplane D, et al. Pathological left-handedness. Left-handedness correlatives in adult epileptics. Brain 1993; 116(Pt 6):1565-74. [9] Orsini DL, Satz P. A syndrome of pathological left-handedness. Correlates of early left hemisphere injury. Arch Neurol 1986;43:333-7. [10] Penfield W, Roberts L. Speech and brain mechanisms. Princeton: Princeton University Press; 1959. [11] Rey M, Dellatolas G, Bancaud J, Talairach J. Hemispheric lateralization of motor and speech functions after early brain lesion: study of 73 epileptic patients with intracarotid amytal test. Neuropsychologia 1988;26:167-72. [12] Satz P, Strauss E, Wada J, Orsini DL. Some correlates of intra- and interhemispheric speech organization after left focal brain injury. Neuropsychologia 1988;26:345-50. [13] Strauss E, Wada J. Lateral preferences and cerebral speech dominance. Cortex 1983; 19:165-77. Fig. 2. Difference in age at epilepsy onset between left-handers and right-handers within the left-sided MTLE. 80 I. Doležalová et al. / Epilepsy & Behavior 72 (2017) 78–81 173 [14] Woods RP, Dodrill CB, Ojemann GA. Brain injury, handedness, and speech lateralization in a series of amobarbital studies. Ann Neurol 1988;23:510-8. [15] Englot DJ, Konrad PE, Morgan VL. Regional and global connectivity disturbances in focal epilepsy, related neurocognitive sequelae, and potential mechanistic underpinnings. Epilepsia 2016;57:1546–57. [16] Liao W, Zhang Z, Pan Z, Mantini D, Ding J, Duan X, et al. Altered functional connectivity and small-world in mesial temporal lobe epilepsy. PLoS One 2010;5:e8525. [17] Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. [18] Bronen RA, Cheung G, Charles JT, et al. Imaging findings in hippocampal sclerosis: correlation with pathology. AJNR Am J Neuroradiol 1991;12:933–40. [19] Engel J. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE task force on classification and terminology. Epilepsia 2001;42:796–803. [20] Chlebus P, Brazdil M, Hlustik P, Mikl M, Pazourkova M, Krupa P. Handedness shift as a consequence of motor cortex reorganization after early functional impairment in left temporal lobe epilepsy—an fMRI case report. Neurocase 2004;10:326-9. [21] Sveller C, Hons BN, Briellmann RS, Saling MM, Lillywhite L, Abbott DF, et al. Relationship between language lateralization and handedness in left-hemispheric partial epilepsy. Neurology 2006;67:1813–7. [22] Slezicki KI, Cho YW, Yi SD, Brock MS, Pfeiffer MH, McVearry KM, et al. Incidence of atypical handedness in epilepsy and its association with clinical factors. Epilepsy Behav 2009;16:330–4. [23] Kim H, Yi S, Son EI, Kim J. Evidence for the pathological right-handedness hypothesis. Neuropsychology 2001;15:510-5. [24] Brazdil M, Marecek R, Fojtikova D, Mikl M, Kuba R, Krupa P, et al. Correlation study of optimized voxel-based morphometry and (1)H MRS in patients with mesial temporal lobe epilepsy and hippocampal sclerosis. Hum Brain Mapp 2009;30:1226-35. [25] Pail M, Brazdil M, Marecek R, Mikl M. 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 2010;51:511-8. [26] Janszky J, Mertens M, Janszky I, Ebner A, Woermann FG. Left-sided interictal epileptic activity induces shift of language lateralization in temporal lobe epilepsy: an fMRI study. Epilepsia 2006;47:921-7. 81I. Doležalová et al. / Epilepsy & Behavior 72 (2017) 78–81 174 Annex 9: Kojan M, Dolezalova I, Koritakova E, Marecek R, Rehak Z, Hermanova M, Brazdil M, Rektor I. Predictive value of preoperative statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy with hippocampal sclerosis. Epilepsy & Behavior. 2018;79:46-52. 175 Predictive value of preoperative statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy with hippocampal sclerosis Martin Kojan a,b , Irena Doležalová a,b , Eva Koriťáková c , Radek Mareček a,b , Zdeněk Řehák d , Markéta Hermanová e , Milan Brázdil a,b , Ivan Rektor a,b, ⁎ a Brno Epilepsy Center, First Department of Neurology, St. Anne's University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic b CEITEC – Central European Institute of Technology, Neuroscience Centre, Masaryk University, Brno, Czech Republic c Institute of Biostatistics and Analyses, Faculty of Medicine, Masaryk University, Brno, Czech Republic d Department of Nuclear Medicine, PET Centre, RECAMO, Masaryk Memorial Cancer Institute (MMCI), Brno, Czech Republic e First Department of Pathological Anatomy, St. Anne's University Hospital and Medical Faculty of Masaryk University, Brno, Czech Republic a b s t r a c ta r t i c l e i n f o Article history: Received 4 July 2017 Revised 9 November 2017 Accepted 11 November 2017 Available online 13 December 2017 Objective: This study was designed to use statistical parametric mapping of interictal positron-emission tomography using [18F]Fluorodeoxyglucose (FDG-PET) to compare the brain metabolisms of patients with mesial temporal lobe epilepsy (MTLE)/hippocampal sclerosis and controls. Another aim of this study was to analyze the potential differences among patients in terms of epilepsy duration, side of hippocampal sclerosis, histopathological findings, insult in their history, and postoperative outcomes. Methods: We analyzed FDG-PET scans from 49 patients with MTLE/hippocampal sclerosis and 24 control subjects. We analyzed the differences in regional glucose metabolism between the patients and the control group and within the patient group using multiple variables. Results: We observed widespread hypometabolism in the patient group in comparison with the control group in temporal and extratemporal areas on the epileptogenic side (ES). On the nonepileptogenic side (NES), we observed the most hypometabolism in the thalamus and the anterior and middle cingulate gyrus. In the group of patients with more severe hippocampal sclerosis, we observed statistically significant hypometabolism in the insula on the ES. In patients with poor postoperative outcomes, we found statistically significant hypometabolism in the insula on the ES and the temporal pole (TP) on the NES. Patients with any insult in their history showed hypermetabolism in the TP on both sides. Conclusion: Our study showed that there are widespread changes in metabolism in patients with MTLE in comparison to controls, either inside or outside the temporal lobe. There are significant differences among these patients in terms of postoperative outcomes, degree of hippocampal sclerosis, and insults in their history. © 2017 Elsevier Inc. All rights reserved. 1. Introduction The value of Positron-emission tomography (PET) using [18F]fluorodeoxyglucose (FDG-PET) in the presurgical evaluation of patients with mesial temporal lobe epilepsy (MTLE) associated with hippocampal sclerosis (HS) [MTLE/HS] is well established [1,2]. In some earlier papers, the visual analysis of PET was found to be reliable both in lateralizing the seizure onset zone and in predicting excellent postoperative outcomes in temporal lobe epilepsy (TLE) [3,4]. However, visual analysis is associated with large variability among investigators. For that reason, methods of quantitative PET analysis have been developed. The first approach is based on glucose metabolic rate calculations in predefined regions of interest (ROIs). These values are compared with the rates from the contralateral side homologous ROI or with the normalized metabolic values from a control group. The definition of the ROI, and its size, shape, and position varies among studies, making data comparison very difficult [2]. Voxel-based approaches such as statistical parametric mapping (SPM) [5] and the asymmetry index [6] do Epilepsy & Behavior 79 (2018) 46–52 Abbreviations: SPM, statistical parametric mapping; ES, epileptogenic side; NES, nonepileptogenic side; FDR, false discovery rate; MTLE, mesial temporal lobe epilepsy; TLE, temporal lobe epilepsy; HS, hippocampal sclerosis; TP, temporal pole; FCD, focal cortical dysplasia; ROI, region of interest. ⁎ Corresponding author at: Brno Epilepsy Center, Department of Neurology, St. Anne's University Hospital, Pekařská 53, Brno 65691, Czech Republic. E-mail address: ivan.rektor@fnusa.cz (I. Rektor). https://doi.org/10.1016/j.yebeh.2017.11.014 1525-5050/© 2017 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh 176 not depend on the selection of the ROI. The SPM method assesses the null hypothesis at each voxel with univariate statistics and constructs an image from its results. Statistical parametric mapping does not require a priori hypotheses about the location and extent of effects. We conducted this study with interictal FDG-PET using SPM analysis in order to compare regional glucose between patients with MTLE/HS and a group of control subjects and to analyze the potential differences among patients in terms of epilepsy duration, side of HS, histopathological findings, insult in their history, and postoperative outcomes. 2. Methods 2.1. Case selection, demographics, and history data We retrospectively reviewed all of the patients with histopathologically proven HS who underwent anterior-medial temporal lobe resection at the Brno Epilepsy Center between 2005 and 2011. The review included patients with a well-documented postoperative outcome for at least 3 years after the surgery and comprehensive results of the histopathological investigation. Patients with other pathologies including dual pathology on magnetic resonance imaging (MRI) and/or without a histopathological investigation were excluded from the study unless there was a finding of focal cortical dysplasia (FCD) of the pole of the resected temporal lobe, which was associated with HS. Clinical data included personal disease history, age at epilepsy onset, epilepsy duration, age at the time of the evaluation, and the postoperative outcome according to the International League Against Epilepsy (ILAE) classification [7]. 2.2. Presurgical evaluation The patients were epilepsy surgery candidates who underwent a comprehensive presurgical evaluation at the Brno Epilepsy Center. Magnetic resonance imaging scans were obtained using the Siemens 1.5 T MRI scanner. Video-EEG monitoring was performed on the 64-channel and 128-channel Alien Deymed systems. All patients had a neuropsychological evaluation targeting memory functions, and patients with language-dominant TLE underwent Wada testing in order to predict the postoperative memory outcome. If the noninvasive evaluations provided discordant data concerning the potential epileptogenic zone, invasive Electroencephalography (EEG) was performed. In all patients, functional neuroimaging techniques (FDG-PET) were performed; and in most patients, interictal/ictal Single-photon emission computed tomography (SPECT) were also performed. Standard anterior-medial temporal lobe resection was based on the results of a presurgical evaluation. 2.3. Histopathology Standard histopathological examination of hippocampal and temporal pole (TP) resection specimens was performed on formalin-fixed paraffin-embedded tissues. For the grade of HS, the grading system by Wyler was used [8]. For FCD in the TP, the classification system reported by Palmini et al. [9] was used. 2.4. Patient population, surgery, outcome, and histopathology We included 49 patients (27 female, 22 male) in the final analysis. The age of the patients at the time of the preoperative investigation ranged from 16 to 59 years with a median of 40 years. The active epilepsy duration ranged from 4 to 58 years with a median of 25 years. Of the 49 patients, 21 (42.9%) had some type of insult in their history (12 patients had encephalitis/meningoencephalitis, and 9 patients had febrile seizures). Of the 49 patients, 27 (55.1%) had leftsided (language-dominant) MTLE/HS, and 22 patients (44.9%) had right-sided (language-nondominant) MTLE/HS. The time after surgery ranged from 2 to 7 years with a median of 4.5 years. At the last follow-up visit, according to the ILAE outcome classification [7], 31 out of 49 patients (63.3%) were classified as Outcome Group 1 (completely seizure-free since the surgery), 7 (14.3%) as Outcome Group 2 (only auras since the surgery), 6 (12.2%) as Outcome Group 3 (one to three seizure days per year), 3 (6.1%) as Outcome Group 4 (from 4 seizure days per year to 50% seizure reduction from baseline seizure days), and 2 (4.1%) as Outcome Group 5 (less than 50% seizure reduction from baseline seizure days). Precise grading of HS (Wyler grading system) [8] was available for 34 patients. Of those 34 patients, 6 (17.6%) were classified as Wyler I/II (low-grade HS) and 28 (82.4%) as Wyler III/IV (high-grade HS). Patients classified as low-grade HS did not differ significantly from the patients classified as high-grade HS in terms of postoperative outcome (ILAE I + II in 55.5% of patients with low-grade HS and 67.6% in high-grade HS respectively; Fisher's exact test, p = 1.0). Histolopathological evaluations revealed FCD of type I a/b in the TP in 13 patients (26.5%) (type Ia in 8 patients and type Ib in 5 patients). All patients signed informed consent forms. The study was approved by the local ethics committee. 2.5. Control group For the control group, we selected 24 (13 female, 11 male) patients with FDG-PET images acquired within oncological screenings. The age of the control subjects at the time of imaging ranged from 16 to 54 years with a median of 34.5 years. The control subjects were without any findings on their MRI scans and without any neurological or psychiatric diagnoses. No disorder in the central nervous system was found, and no drugs influencing brain metabolism were taken by these control subjects. Their FDG-PET findings were assessed as normal. 2.6. FDG-PET image acquisition [18F]Fluorodeoxyglucose PET scans were available for SPM analysis in all 49 (27 female, 22 male) patients and in the 24 (13 female, 11 male) control subjects. Both patients and controls underwent the same procedure. Patients were imaged as outpatients in the interictal state. The PET images for both groups were acquired using a Siemens ECAT ACCEL PET scanner (three detection rings with lutetium orthosilicate type crystals and 16.2 cm axial field of view (FOV); Erlangen, Germany) in 3-D mode using the “Brain” protocol. The intrinsic spatial resolution of the scanner was 6.3 mm at full width at half maximum (FWHM) 1 cm from the center of the FOV, and 6.7 mm at FWHM 10 cm from the center of the FOV. Subjects prepared by fasting for 6 h before the scan and resting in a quiet, darkened room for 50– 60 min after FDG administration. The dose of FDG administered was 200 MBq ± 15% per subject with no weight differentiation. The emission acquisition time in 3-D mode was 10 min. Forty-seven tomographic slices with a 3-mm slice thickness were reconstructed with a 128 × 128 iteration matrix with 6 iterations and 16 subsets, and a 6 mm FWHM Gaussian filter was applied. 2.7. Image preprocessing for SPM analysis Spatial preprocessing and statistical analysis were performed using SPM8 (Wellcome Department of Cognitive Neurology, Institute of Neurology, University College London, U.K.) and MATLAB version 2011b (MathWorks Inc., Natick, MA, USA). All FDG-PET images, both patient and control scans, were spatially normalized into the standardized stereotactic Montreal Neurological Institute space using our in-house FDG-PET template. This template was created following instructions introduced by Soma et al. [10]. A three-dimensional isotropic Gaussian kernel with 8 mm FWHM was used for smoothing all spatially normalized images. To remove the effect of global metabolism, we used global normalization with a proportional scaling method [11]. 47M. Kojan et al. / Epilepsy & Behavior 79 (2018) 46–52 177 2.8. Computing of statistic parametric maps To lateralize the epileptogenic side (ES) to the left, we flipped 22 FDG-PET preprocessed images of right patients with MTLE (8 M, 14 F) in the direction of the x-axis. For the group analysis, we used the Mann–Whitney U test and created parametric maps (PMs) using a voxelwise approach. We applied a false discovery rate (FDR) control for the results. In all analyses, FDR q = 0.05 was used as a level of statistical significance. These PMs were used in further analyses. 2.9. Comparison between different brain regions To calculate the amount and the ratio of statistically significant voxels in different brain regions, an experienced neurologist selected 11 separate ROIs. These ROIs are presented in Table 1. For each ROI for the ES and the nonepileptogenic side (NES), we created a mask using the WFU PickAtlas software toolbox (Functional MRI Laboratory, Wake Forest University School of Medicine, U.S.) [12] using the included automated anatomical labeling (AAL) atlas [13]. These masks were applied on the computed PMs of the whole brain to allow comparison between ROIs with different sizes. The ratio between the size of the ROI in voxels and the number of statistically significant voxels from the PMs matching this ROI was computed for all ROIs. 3. Results 3.1. SPM-PET analysis, comparison of patients with controls By analyzing statistical parametric maps created using the Mann– Whitney U test, we found widespread statistically significant lower signal values (hypometabolism) in the patient group than in the control group, either on the ES or the NES. On the ES, the most pronounced hypometabolism was detected in the TP, parahippocampal gyrus, and amygdalo-hippocampal complex. Of the extratemporal sites on the ES, we observed the most pronounced hypometabolism in the orbitofrontal cortex and insula, followed by the thalamus and posterior cingulate gyrus. The most hypometabolism of the NES was observed in the thalamus and anterior cingulate gyrus and middle cingulate gyrus. All of our results are presented in Fig. 1; the ratios of significantly hypometabolic voxels on predefined ROIs on the ES and the NES are shown in Table 1. No significant differences between patients and controls were revealed in the TP, parahippocampal gyrus, or amygdalo-hippocampal complex on the NES (Fig. 1, Table 1). Table 1 Ratio of voxels in ROIs with statistically significant hypometabolism in the patient group relative to the control group, FDR adjusted p-value b 0.05. ROI ES NES ROI size Significant Significant Voxels Voxels % Voxels % Amygdala 192 28 14.58 0 0.00 Hippocampus 878 319 36.33 0 0.00 PHG 962 669 69.54 0 0.00 AHC 1578 744 47.15 0 0.00 TP 2563 1571 61.30 0 0.00 Insula 1134 632 55.73 54 4.76 OF 1175 861 73.28 90 7.66 GCA 795 35 4.40 113 14.21 GCM 1422 251 17.65 115 8.09 GCP 345 136 39.42 20 5.80 GCA + GCM 2296 300 13.07 248 10.80 Thalamus 972 307 31.58 114 11.73 ES – epileptogenic side, NES – nonepileptogenic side, PHG – parahippocampal gyrus, AHC – amygdalo-hippocampal complex, TP – temporal pole, OF – orbitofrontal cortex, GCA –anterior cingulate gyrus, GCM – middle cingulate gyrus, GCP – posterior cingulate gyrus. Fig. 1. Statistical parametric mapping of patients (N = 49) versus controls (N = 23) coregistered to T1 MRI. Epileptogenic side on the left showing large hypometabolism in the temoporal and extratemporal regions. False discovery rate adjusted p-value b 0.05. 48 M. Kojan et al. / Epilepsy & Behavior 79 (2018) 46–52 178 3.2. SPM-PET analysis, comparison of patient groups We did not find any significant differences between the right and left-sided patients with MTLE/HS after FDR correction. We did not find any significant differences between patients with and without FCD in the TP, neither on the ES nor the NES. These groups differ in metabolism only in the posterior cingulate gyrus on the NES. In the group of patients with more severe HS (Wyler III/IV) (N = 28), we observed statistically significant hypometabolism in the insula on the ES in comparison with the patients classified as Wyler I/II (N = 6) (Fig. 2). In patients with poor postoperative outcomes (ILAE III–V), we found statistically significant hypometabolism in the insula on the ES and on the NES in the TP and middle and posterior neocortex in comparison to the patients with better postoperative outcome (ILAE I/II) (Fig. 3). The comparison of the patients with and without insult (encephalitis/ meningoencephalitis and/or febrile seizures) in the history showed significant hypermetabolism in the TP either on the ES or the NES and in the middle temporal gyrus on the NES in patients with these insults in their history. The finding was more pronounced on the NES (Fig. 4). Patients with longer epilepsy duration (N10 years) did not demonstrate any significant differences from patients with shorter duration (≤10 years). 4. Discussion 4.1. Comparison of patients with MTLE/HS to controls Our study on patients with MTLE/HS clearly revealed very widespread hypometabolism of temporal regions (TP, parahippocampal gyrus, and amygdalo-hippocampal complex), extratemporal regions (orbitofrontal cortex, insula, posterior and anterior cingulate gyrus), and the thalamus. Glucose hypometabolism typically involved the temporal lobe ipsilateral to the HS, which is in agreement with previous studies using both visual and SPM analysis [14,15]. In the further text, we compare our results to results of another studies focusing on analysis of MTLE associated with HS proved by MRI or by histopathological finding. Some other studies have shown that hypometabolism often involves the extratemporal structures, mostly the ipsilateral frontal lobe and insula [16–18]. In addition to these finding, we observed hypometabolism on the side contralateral to HS in extratemporal regions, namely the anterior and middle cingulate gyrus. Another interesting finding in our study is the hypometabolism of subcortical areas. We did not observe any hypometabolism in the putamen, pallidum, or caudate nucleus, neither on the ES nor the NES. Nevertheless, we found the hypometabolism of the thalamus bilaterally to be more pronounced on the ES. Van Bogaert et al. [2] demonstrated hypometabolism in the ipsilateral and contralateral thalamus and Fig. 2. Patients with severe hippocampal sclerosis Wyler grades 3 and 4 (N = 28) showing relative hypometabolism in the insula on the epileptogenic side as compared with the Wyler grades 1 and 2 group (N = 6). Coregistered to T1 MRI. False discovery rate adjusted p-value b 0.05. Fig. 3. ILAE classification 3 and 4 groups (N = 11) showing hypometabolism in the insula on the epileptogenic side and in the temporal lobe on the nonepileptogenic side relative to the ILAE 1 and 2 groups (N = 38) Coregistered to T1 MRI. False discovery rate adjusted p-value b 0.05. 49M. Kojan et al. / Epilepsy & Behavior 79 (2018) 46–52 179 caudate nucleus. Thalamic hypometabolism ipsilateral to the seizure side in patients with TLE was frequently reported in earlier studies using either visual or quantitative analysis [14,19,20]. Newberg et al., using SPM analysis, showed that hypometabolism, mostly of the contralateral thalamus, but also of the ipsilateral (to the epileptogenic side) thalamus, might be a predictor of poorer postoperative outcomes in patients with refractory TLE [21]. Thalamic hypometabolism probably correlates with the more widespread dysfunction of the temporal and extratemporal part of the limbic system that is observed in patients with refractory TLE. For future studies, it would be useful to determine which specific part of the thalamus shows the hypometabolism in these patients with TLE, i.e., for example in the anterior thalamic nuclei, which are part of the limbic system. 4.2. Comparison in relation to postoperative outcome Our results clearly showed that patients with worse long-term postoperative outcomes had significant hypometabolism in the insula ipsilateral to the surgery and in the TP contralateral to the surgery, i.e., contralateral to HS. This finding might indicate a more pronounced bitemporal hypometabolism or a different pattern of hypometabolism outside the temporal lobe on the ES in comparison to patients with excellent postoperative outcome. We did not find statistically significant hypometabolism in the temporal lobe on the NES when comparing all patients and controls. Takahashi et al. [6] used a voxel-based comparison of FDG-PET to and controls and showed that significant hypometabolism in seizure-free patients preoperatively was restricted to the ipsilateral temporal tip and was not present in a postoperative nonseizure free group. Moreover, they showed that hypometabolism in the ipsilateral hippocampal, frontal, and thalamic areas were larger in the postoperative seizure-free group than in the nonseizure-free group. On the other hand, in the contralateral frontal and thalamic areas, the extents of hypometabolism were smaller in the seizure-free group than in the nonseizure-free group [6]. Studies focused on bilateral temporal hypometabolism showed relatively wide variability in terms of incidence, varying from 10 to 30% [22, 23]. Previous studies using semiquantitative analysis showed that bilateral temporal hypometabolism is usually associated with bilateral temporal epilepsy, bilateral MRI findings, worse cognitive function, and bilateral independent seizure onset zones at the scalp EEG [22,24]. Another study revealed that bilateral temporal hypometabolism on FDGPET in patients with MTLE is present in patients who had electrographic features, suggesting the early contralateral spread of ictal activity [17]. All of these variables might have a bad prognostic predictive value for seizure outcome in patients with TLE [23,25,26]. One recent study [27] showed that the duration from the last seizure could be an important factor that needs to be considered when analyzing the bitemporal involvement on interictal FDG-PET. Their multivariate analysis showed that bilateral temporal hypometabolism is significantly more often present when PET is performed within 2 days after the last seizure. Our study suggests that the finding of contralateral temporal hypometabolism on preoperative FDG-PET in a highly select group of patients with MTLE/HS is a negative prognostic factor for the postoperative outcome. Concerning the finding of insular hypometabolism ipsilateral to the side of HS in patients with worse postoperative prognosis revealed by our study, we have to mention particularly the possible existence of “temporal-plus” (T+) epilepsies. This concept has been defined and analyzed in detail by some authors [28,29]. Temporal-plus epilepsies are very specific forms of seizures of multilobar origin which are characterized by the involvement of a complex epileptogenic network including the temporal lobe and the connected structures, mainly the orbitofrontal cortex, the insula, the frontal and parietal operculum, and the temporo-parieto-occipital junction. In general, patients with unifocal TLE are hardly distinguishable from the patients with T+ epilepsy on the basis of general clinical features or MRI data. Furthermore, the presence of HS, which is known as one of the best prognostic factors for successful temporal lobe surgery [30,31], did not distinguish unifocal TLE from patients with T+ epilepsy. Only a minority of the patients in our series were investigated with invasive EEG, so we can only speculate that some of them might have T+ epilepsy, with the second focus in the insular cortex, which can very often be a part of the epileptogenic network in patients with MTLE/HS [28,29]. In conclusion, our results may indicate that hypometabolism in the insular cortex on preoperative FDG-PET using SPM analysis is a predictor of worse postoperative outcomes in patients with MTLE/HS and is an indication for invasive EEG in those patients. 4.3. Glucose metabolism in relation to histopathological findings We demonstrated statistically significant hypometabolism in the insula on the ES in the group of patients with more severe HS (Wyler III/IV) in comparison with those with Wyler I/II classification. These groups did not differ in terms of postoperative outcome. This outcome Fig. 4. Hypermetabolism on the nonepileptogenic side within the group of patients with encephalitis/meningoencephalitis and/or febrile seizures in the history (N = 21) compared with the group without any insult (N = 27). Corregistered to T1 MRI. False discovery rate adjusted p-value b 0.05. 50 M. Kojan et al. / Epilepsy & Behavior 79 (2018) 46–52 180 was even better in the group of patients with HS III/IV. Foldvary et al. [32] evaluated the neuronal density in mesiotemporal regions including the hippocampal subfield and correlated it with hypometabolism in the temporal lobe using a visual range scale. The neuronal density in CA1, CA4, subiculum, and dental granular cell layer did not correlate with the severity of the hypometabolism [32]. These data are supported by other studies evaluating the relationship between hippocampal cell density, hippocampal volume, and the degree of hypometabolism [33, 34]. On the other hand, Dlugos et al. [35] revealed that hilar cell densities correlated positively and significantly with hypometabolism in the bilateral thalamus, putamen, and globus pallidus, and in the ipsilateral caudate. Dentate granule cell densities correlated positively and significantly with hypometabolism in the bilateral thalamus and putamen. There was no significant correlation between cell densities and glucose metabolism in any cortical region, including the hippocampus [35]. Although these data are controversial, it might be postulated that hippocampal cell loss results in decreased efferent synaptic activity to the various cortical and subcortical regions resulting in subsequent hypometabolism. We did not find significant differences between patients with or without FCD in the TP in any temporal lobe (both on the ES and the NES). These groups differ in metabolism only in the posterior cingulate gyrus on the NES. An isolated study analyzed this particular issue, demonstrating a different pattern of hypometabolism in patients with isolated HS and HS + FCD. The most prominent hypometabolism in HS patients was in the anterior and mesial parts of the temporal lobe; in patients with HS + FCD, it was in the lateral temporal lobe [36]. Out data are only marginally comparable to those data, because we used SPM analysis and the other study used an asymmetry index based on a comparison of the contralateral and ipsilateral side ROI counts, and because no patient in our series had FCD type II, whereas they were included in the other study. 4.4. Study limitations While our method is based on statistical comparisons between groups, we have to rule out all outliers in the groups that may influence the results. Subjects with major brain malformations may not be correctly spatially normalized with the template image and the corresponding areas in analysis are not perfectly aligned, which may lead to incorrect results. In cases when a subject is missing a certain volume of brain tissue due to injury, disease, or surgery, the results of the analysis are focused in those missing volumes. Changes in metabolism in the remaining tissue thus cannot be revealed using this method. All subjects with such variance from the template image must be excluded from this type of analysis. Based on these criteria, two patients were removed from this study in order to not contaminate the findings. We also suppose that this method could be beneficial in patient groups other than HS, i.e., in groups of nonlesional cases or of complicated cases with discordant presurgical data, but questions about the advantages of this method cannot be answered properly based on this study, because only patients with HS were included. 5. Conclusions Our study shows that there are widespread significant changes in metabolism in patients with MTLE/HS in comparison to controls, either inside or outside the temporal lobe and mostly ipsilateral to the side of HS. These differences are most pronounced on the TP, the mesial part of the temporal lobe, and the orbitofrontal cortex ipsilateral to the side of HS. There are significant differences among these patients in terms of the postoperative outcome, degree of HS, and presence of insults in the patient history. Acknowledgments The authors would like to thank prof. Robert Kuba, who was the mentor of the first author until his premature death. This research was carried out under the project CEITEC 2020 (LQ1601) with financial support from the Ministry of Education, Youth and Sports of the Czech Republic under the National Sustainability Programme II; by the project MEYS-NPS I-LO1413 and by the Ministry of Health, Czech Republic - conceptual development of research organization (MMCI, 00209805). We thank Anne Johnson for grammatical assistance. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.yebeh.2017.11.014. References [1] Spencer SS. The relative contributions of MRI, SPECT, and PET imaging in epilepsy. Epilepsia 1994;35:S72–89. [2] Van Bogaert P, Massager N, Tugendhaft P, Wikler D, Damhaut P, Levivier M, et al. Statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy. Neuroimage 2000;12:129–38. [3] Delbeke D, Lawrence SK, AbouKhalil BW, Blumenkopf B, Kessler RM. Postsurgical outcome of patients with uncontrolled complex partial seizures and temporal lobe hypometabolism on (18)FDG-positron emission tomography. Invest Radiol 1996; 31:261–6. [4] Radtke RA, Hanson MW, Hoffman JM, Crain BJ, Walczak TS, Lewis DV, et al. Temporal-lobe hypometabolism on PET - predictor of seizure control after temporal lobectomy. Neurology 1993;43:1088–92. [5] Signorini M, Paulesu E, Friston K, Perani D, Colleluori A, Lucignani G, et al. Rapid assessment of regional cerebral metabolic abnormalities in single subjects with quantitative and nonquantitative F-18 FDG PET: a clinical validation of statistical parametric mapping. Neuroimage 1999;9:63–80. [6] Takahashi M, Soma T, Kawai K, Koyama K, Ohtomo K, Momose T. Voxel-based comparison of preoperative FDG-PET between mesial temporal lobe epilepsy patients with and without postoperative seizure-free outcomes. Ann Nucl Med 2012;26: 698–706. [7] Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross JH, Boas WV, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia 2010;51: 676–85. [8] Wyler AR, Dohan FC, Schweitzer JB, Berry AD. A grading system for mesial temporal pathologz (hippocampal sclerosis) from anterior temporal lobectomy. J Epilepsy 1992;5:220–5. [9] Palmini A, Najm I, Avanzini G, Babb T, Guerrini R, Foldvary-Schaefer N, et al. Terminology and classification of the cortical dysplasias. Neurology 2004;62:S2–8. [10] Soma T, Momose T, Takahashi M, Koyama K, Kawai K, Murase K, et al. Usefulness of extent analysis for statistical parametric mapping with asymmetry index using inter-ictal FGD-PET in mesial temporal lobe epilepsy. Ann Nucl Med 2012;26: 319–26. [11] Fox PT, Mintun MA, Reiman EM, Raichle ME. Enhanced detection of focal brain responses using intersubject averaging and change-distribution analysis of subtracted PET images. J Cereb Blood Flow Metab 1988;8:642–53. [12] Maldjian JA, Laurienti PJ, Kraft RA, Burdette JH. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. Neuroimage 2003;19:1233–9. [13] Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard O, Delcroix N, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 2002;15: 273–89. [14] Henry TR, Mazziotta JC, Engel J. Interictal metabolic anatomy of mesial temporallobe epilepsy. Arch Neurol 1993;50:582–9. [15] Kim YK, Lee DS, Lee SK, Kim SK, Chung CK, Chang KH, et al. Differential features of metabolic abnormalities between medial and lateral temporal lobe epilepsy: quantitative analysis of F-18-FDG PET using SPM. J Nucl Med 2003;44:1006–12. [16] Bouilleret V, Dupont S, Spelle L, Baulac M, Samson Y, Semah F. Insular cortex involvement in mesiotemporal lobe epilepsy: a positron emission tomography study. Ann Neurol 2002;51:202–8. [17] Chassoux F, Semah F, Bouilleret V, Landre E, Devaux B, Turak B, et al. Metabolic changes and electro-clinical patterns in mesio-temporal lobe epilepsy: a correlative study. Brain 2004;127:164–74. [18] Wong CH, Bleasel A, Wen L, Eberl S, Byth K, Fulham M, et al. The topography and significance of extratemporal hypometabolism in refractory mesial temporal lobe epilepsy examined by FDG-PET. Epilepsia 2010;51:1365–73. [19] Choi JY, Kim SJ, Hong SB, Seo DW, Hong SC, Kim BT, et al. Extratemporal hypometabolism on FDG PET in temporal lobe epilepsy as a predictor of seizure outcome after temporal lobectomy. Eur J Nucl Med Mol Imaging 2003;30:581–7. 51M. Kojan et al. / Epilepsy & Behavior 79 (2018) 46–52 181 [20] Khan N, Leenders KL, Hajek M, Maguire P, Missimer J, Wieser HG. Thalamic glucose metabolism in temporal lobe epilepsy measured with 18F-FDG positron emission tomography (PET). Epilepsy Res 1997;28:233–43. [21] Newberg AB, Alavi A, Berlin J, Mozley PD, O'Connor M, Sperling M. Ipsilateral and contralateral thalamic hypometabolism as a predictor of outcome after temporal lobectomy for seizures. J Nucl Med 2000;41:1964–8. [22] Kim MA, Heo K, Choo MK, Cho JH, Park SC, Lee JD, et al. Relationship between bilateral temporal hypometabolism and EEG findings for mesial temporal lobe epilepsy: analysis of F-18-FDG PET using SPM. Seizure Eur J Epilepsy 2006;15:56–63. [23] Steinhoff BJ, So NK, Lim SH, Luders HO. Ictal scalp EEG in temporal-lobe epilepsy with unitemporal versus bitemporal interictal epileptiform discharges. Neurology 1995;45:889–96. [24] Koutroumanidis M, Hennessy MJ, Seed PT, Elwes RDC, Jarosz J, Morris RG, et al. Significance of interictal bilateral temporal hypometabolism in temporal lobe epilepsy. Neurology 2000;54:1811–21. [25] Hirsch LJ, Spencer SS, Williamson PD, Spencer DD, Mattson RH. Comparison of bitemporal and unitemporal epilepsy defined by depth electroencephalogarohy. Ann Neurol 1991;30:340–6. [26] Holmes MD, Dodrill CB, Ojemann GA, Wilensky AJ, Ojemann LM. Outcome following surgery in patients with bitemporal interictal epileptiform patterns. Neurology 1997;48:1037–40. [27] Tepmongkol S, Srikijvilaikul T, Vasavid P. Factors affecting bilateral temporal lobe hypometabolism on 18F-FDG PET brain scan in unilateral medial temporal lobe epilepsy. Epilepsy Behav 2013;29:386–9. [28] Barba C, Barbati G, Minotti L, Hoffmann D, Kahane P. Ictal clinical and scalp-EEG findings differentiating temporal lobe epilepsies from temporal ‘plus’ epilepsies. Brain 2007;130:1957–67. [29] Ryvlin P, Kahane P. The hidden causes of surgery-resistant temporal lobe epilepsy: extratemporal or temporal plus? Curr Opin Neurol 2005;18:125–7. [30] Berkovic SF, McIntosh A, Howell RA, Mitchell A, Sheffield LJ, Hopper JL. Familial temporal lobe epilepsy: a common disorder identified in twins. Ann Neurol 1996;40: 227–35. [31] Garcia PA, Laxer KD, Barbaro NM, Dillon WP. Prognostic value of qualitative Magnetic-Resonance-Imaging hippocampal abnormalities in patients undergoing lobectomy for medically refractory seizures. Epilepsia 1994;35:520–4. [32] Foldvary N, Lee N, Hanson MW, Coleman RE, Hulette CM, Friedman AH, et al. Correlation of hippocampal neuronal density and FDG-PET in mesial temporal lobe epilepsy. Epilepsia 1999;40:26–9. [33] Obrien TJ, Newton MR, Cook MJ, Berlangieri SU, Kilpatrick C, Morris K, et al. Hippocampal atrophy is not a major determinant of regional hypometabolism in temporal lobe epilepsy. Epilepsia 1997;38:74–80. [34] Henry TR, Babb TL, Engel J, Mazziotta JC, Phelps ME, Crandall PH. Hippocampal neuronal loss and regional hypometabolism in temporal-lobe epilepsy. Ann Neurol 1994;36:925–7. [35] Dlugos DJ, Jaggi J, O'Connor WM, Ding XS, Reivich M, O'Connor MJ, et al. Hippocampal cell density and subcortical metabolism in temporal lobe epilepsy. Epilepsia 1999;40:408–13. [36] Diehl B, LaPresto E, Najm I, Raja S, Rona S, Babb T, et al. Neocortical temporal FDGPET hypometabolism correlates with temporal lobe atrophy in hippocampal sclerosis associated with microscopic cortical dysplasia. Epilepsia 2003;44:559–64. 52 M. Kojan et al. / Epilepsy & Behavior 79 (2018) 46–52 182 Annex 10: Dolezalova I, Brazdil M, Chrastina J, Hemza J, Hermanova M, Janousova E, Pazourkova M, Kuba R. Differences between mesial and neocortical magnetic-resonanceimaging-negative temporal lobe epilepsy. Epilepsy & Behavior. 2016;62:21-26. 183 Differences between mesial and neocortical magnetic-resonance-imaging-negative temporal lobe epilepsy Irena Doležalová a, ⁎, Milan Brázdil a,b , Jan Chrastina c , Jan Hemza c , Markéta Hermanová d , Eva Janoušová e , Marta Pažourková f , Robert Kuba a,b a Brno Epilepsy Center, First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic b Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic c Brno Epilepsy Center, Department of Neurosurgery, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic d First Department of Pathological Anatomy, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic e Institute of Biostatistics and Analyses, Faculty of Medicine, Masaryk University, Brno, Czech Republic f Department of Radiology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic a b s t r a c ta r t i c l e i n f o Article history: Received 3 November 2015 Revised 8 April 2016 Accepted 11 April 2016 Available online 2 June 2016 Objective: The aim of this study was to assess clinical and electrophysiological differences within a group of patients with magnetic-resonance-imaging-negative temporal lobe epilepsy (MRI-negative TLE) according to seizure onset zone (SOZ) localization in invasive EEG (IEEG). Methods: According to SOZ localization in IEEG, 20 patients with MRI-negative TLE were divided into either having mesial SOZ–mesial MRI-negative TLE or neocortical SOZ–neocortical MRI-negative TLE. We evaluated for differences between these groups in demographic data, localization of interictal epileptiform discharges (IEDs), and the ictal onset pattern in semiinvasive EEG and in ictal semiology. Results: Thirteen of the 20 patients (65%) had mesial MRI-negative TLE and 7 of the 20 patients (35%) had neocortical MRI-negative TLE. The differences between mesial MRI-negative TLE and neocortical MRI-negative TLE were identified in the distribution of IEDs and in the ictal onset pattern in semiinvasive EEG. The patients with neocortical MRI-negative TLE tended to have more IEDs localized outside the anterotemporal region (p = 0.031) and more seizures without clear lateralization of ictal activity (p = 0.044). No other differences regarding demographic data, seizure semiology, surgical outcome, or histopathological findings were found. Conclusions: According to the localization of the SOZ, MRI-negative TLE had two subgroups: mesial MRI-negative TLE and neocortical MRI-negative TLE. The groups could be partially distinguished by an analysis of their noninvasive data (distribution of IEDs and lateralization of ictal activity). This differentiation might have an impact on the surgical approach. © 2016 Elsevier Inc. All rights reserved. Keywords: MRI-negative temporal lobe epilepsy Mesial MRI-negative TLE Neocortical MRI-negative TLE Clinical semiology Invasive EEG Semiinvasive EEG Positron emission tomography 1. Introduction Patients with temporal lobe epilepsy (TLE) are the most common candidates for epilepsy surgery. There is a high probability that a patient with a lesional (MRI-positive) case will become seizure-free after surgery; the rates reach almost 80% in patients with hippocampal atrophy and concordant EEG findings [1]. In contrast, seizure-free rates are traditionally reported much lower in nonlesional or MRI-negative TLE. According to a meta-analysis, these rates may be 2.7 times lower than in lesional ones. For this reason, these patients are not considered to be good surgical candidates and are often excluded from surgical programs [2–7]. Studies focusing on MRI-negative TLE have been published in the past few years in which the surgical outcome seemed to be only slightly worse or even comparable to the lesional cases [8–15]. This is valid especially for patients with fluorodeoxyglucose positron emission tomography (FDG-PET) hypometabolism localized to one temporal lobe. In agreement with other noninvasive data, this subgroup is sometimes referred to as MRI-negative, PET-positive TLE [8–10,12–15]. The key point of surgery in MRI-negative TLE is to distinguish whether mesial or neocortical temporal lobe structures are primarily involved in seizure generation and, subsequently, to indicate the approach for appropriately tailored surgery. If this process is not performed correctly, the surgery does not result in seizure cessation and sometimes can even lead to serious memory decline, resulting in a patient with persistent seizures despite surgery and disabling memory problems [16]. The precise demarcation of cerebral tissue resection could be based on the findings of perioperative electrocorticography or chronic invasive EEG (IEEG), but the significance of perioperative electrocorticography remains controversial. There is still a very limited Epilepsy & Behavior 61 (2016) 21–26 ⁎ Corresponding author at: Brno Epilepsy Center, First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Pekařská 53, 656 91 Brno, Czech Republic. Tel.: +420 543 182 645; fax: +420 543 182 624. E-mail address: irena.dolezalova@fnusa.cz (I. Doležalová). http://dx.doi.org/10.1016/j.yebeh.2016.04.027 1525-5050/© 2016 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh 184 number of IEEG studies focusing on patients with MRI-negative TLE [14, 15,17–20]. In this study, we divided patients with MRI-negative TLE according to the localization of their SOZ in IEEG into mesial MRI-negative TLE and neocortical MRI-negative TLE, and we tried to differentiate between these two types on the basis of their noninvasive data analysis. 2. Methods 2.1. Patient selection This retrospective study included a total of 20 patients with MRInegative TLE who underwent epilepsy surgery for refractory epilepsy at the Brno Epilepsy Center of Masaryk University Hospital between 2003 and 2013. All patients underwent the IEEG procedure prior to surgery. Of the 20 patients, 18 (90%) were examined with chronically stereotactically implanted intracerebral electrodes. In one patient, we used a combination of stereotactically implanted intracerebral electrodes and subdural strip electrodes. In one other patient, the combination of one subdural strip and grid electrodes and intracerebral electrodes was used. According to the localization of the SOZ in IEEG, patients were divided into mesial MRI-negative TLE (i.e., SOZ localized to the amygdala, hippocampus, and parahippocampal gyrus) or neocortical MRInegative TLE. The study aimed to find differences between mesial and neocortical MRI-negative TLE based on the analysis of noninvasive data and semiinvasive EEG with the use of sphenoidal electrodes. The patients were compared on the following variables: demographic data, FDG-PET findings, interictal semiinvasive EEG, ictal semiinvasive EEG, and seizure semiology. Finally, patients with mesial MRI-negative TLE and patients with neocortical MRI-negative TLE were compared with regard to the results of surgical treatment and histopathological findings. The study was approved by the ethics committee of St. Anne's University Hospital. 2.2. Demographic data Patient demographic data were obtained from patient charts. The following demographic data were analyzed: potential causes for the development of epilepsy, age at epilepsy onset, duration of epilepsy, frequency of complex partial seizures (CPSs) per month during the 6 months before surgery, and the presence/absence of a generalized tonic–clonic seizure (GTCS) in the 6 months before surgery. 2.3. FDG-PET scans The FDG-PET images were acquired using a Siemens ECAT ACCEL PET scanner (three detection rings with lutetium-orthosilicate-type crystals and 16.2-cm axial FOV) (Erlangen, Germany) in 3-D mode using Brain protocol. The intrinsic spatial resolution of the scanner was 6.3 mm at full width at half maximum (FWHM) 1 cm from the center of the FOV and 6.7 mm at full width at half maximum 10 cm from the center of the FOV. The subjects fasted for 6 h before the scan and rested in a quiet, darkened room for 50–60 min after FDG administration. The dose of FDG administered was 200 MBq ± 15% per subject with no weight differentiation. The emission acquisition time in 3-D mode was 10 min. Forty-seven tomographic slices with a 3-mm slice thickness were reconstructed with a 128 × 128 iteration matrix with 6 iterations and 16 subsets and a 6-mm FWHM Gaussian filter applied. The FDG-PET findings were analyzed visually by authors (I.D. and R.K.) experienced in reviewing FDG-PET scans of patients with epilepsy. According to the localization of FDG-PET hypometabolism, patients were categorized to have the following: 1. hypometabolism on the operated side (Fig. 1); 2. hypometabolism on either side or hypometabolism contralateral to the operated side. Patients with hypometabolism on the operated side were subdivided into having hypometabolism in the anterior and medial parts of the temporal lobe or in the lateral part of the temporal lobe. 2.4. Semiinvasive EEG Semiinvasive EEGs were recorded based on the international 10–20 system with additional electrodes (anterotemporal T1, T2; sphenoidal Sp1, Sp2; and supraorbital SO1, SO2). The semiinvasive EEG was analyzed by two authors (I.D. and R.K.); discrepancies were resolved by consensus. 2.4.1. Semiinvasive EEG — interictal analysis Five-minute samples from every hour of interictal EEG during the first 24 h were analyzed for the presence and distribution of IEDs. According to IED distribution, patients were classified into the following: 1. unilateral IEDs (i.e., ≥90% of IEDs lateralized to the operated side) versus bilateral IEDs (i.e., b90% of IEDs lateralized to the operated side); 2. strictly anterotemporal IEDs versus also having extraanterotemporal IEDs on the operated side. Anterotemporal IEDs were described by phase reversal at Sp1/Sp2, T1/T2, or F7/F8. 2.4.2. Semiinvasive EEG — ictal analysis According to the type of seizure onset pattern in semiinvasive EEG, the patients were classified as having all seizures correctly regionalized/ lateralized over the operated temporal lobe versus having at least one seizure nonlateralized or falsely regionalized/lateralized. The definition of regionalized/lateralized seizure onset was published by Steinhoff et al. [21] and Risinger et al. [22]. 2.5. Analysis of ictal semiology The patient records were reviewed in order to analyze the type of aura. The following ictal signs were clinically evaluated during seizures: early oroalimentary automatisms (considered only if present within 20 s from the beginning of a seizure), early nonversive head turning, lateralized ictal immobility of the upper limb, ictal dystonia, rhythmic ictal nonclonic hand (RINCH) motions, periictal nose wiping, periictal face rubbing, periictal bed leaving, and periictal vegetative symptoms (retching with/without vomiting, coughing, urinary urge, water drinking; all were evaluated during the seizure and/or within 2 min after seizure Fig. 1. FDG-PET with hypometabolism on operated side in patient with MRI-negative TLE. 22 I. Doležalová et al. / Epilepsy & Behavior 61 (2016) 21–26 185 termination). Periictal nose wiping, periictal face rubbing, and periictal bed leaving were evaluated only in CPSs; other signs were evaluated in both CPSs and GTCSs. An ictal sign was counted as present in a patient if it was found in at least one of that patient's seizures. 2.6. Surgery and surgical outcome, neuropathological examination The extent of the resection was based on the results of a presurgical evaluation. Surgical effectiveness was categorized at year 1 after surgery according to Engel classification as Engel I (no seizures, with or without nondisabling auras) or as Engel II–IV [23]. Focal cortical dysplasias (FCDs) were classified according to the classification system reported by Blümcke et al. [24]. 2.7. Statistics The differences in demographic data, patient characteristics, semiinvasive interictal/ictal EEG analysis, and clinical semiology between mesial MRI-negative TLE and neocortical MRI-negative TLE were calculated using the Mann–Whitney test or Fisher's exact test according to their condition of validity. For all tests, a p-value b 0.05 was considered statistically significant. 3. Results 3.1. The localization of the SOZ in IEEG A total of 64 seizures were recorded by IEEG in 20 patients with MRInegative TLE; the number of seizures per patient ranged from 1 to 5 (with an average of 3.3 ± 1.2). According to SOZ localization in IEEG, 13 of the 20 patients (65%) had mesial MRI-negative TLE, and 7 of the 20 patients (35%) had neocortical MRI-negative TLE. In patients with mesial MRI-negative TLE, 3 (15%) of the 13 showed alternating involvement of the temporal pole and mesial temporal lobe structures, with a majority of seizures originating in the mesial temporal lobe structures. 3.2. Demographic data The age at epilepsy onset ranged from 0 to 36 years (with an average of 15.6 ± 9.7 years); the duration of epilepsy ranged from 3 to 41 years (with an average of 17.2 ± 11.1 years). The data concerning the main clinical features of the patients are summarized in Table 1. The SOZ was localized on the left side in 11 of the 20 patients (55%) and on the right side in 9 of the 20 (45%). No significant differences between mesial and neocortical MRI-negative TLE were found in age at epilepsy onset, duration of epilepsy, or side of the SOZ (Table 2). Potential causes for the development of epilepsy were present in 9 of the 20 patients (45%). There were no statistically significant differences between patients with mesial and neocortical MRI-negative TLE in the absence/presence of potential causes for the development of epilepsy (Table 2). The monthly CPS frequency during the 6 months before surgery ranged from 3 to 15 (with an average of 8.6 ± 3.7); GTCS during the 6 months preceding surgery were present in 13 of the 20 patients (65%). There were no statistically significant differences between mesial and neocortical MRI-negative TLE in the monthly CPS frequency or in the absence/presence of GTCS (Table 2). 3.3. FDG-PET scans Of the 20 patients, 12 (60%) had FDG-PET hypometabolism on the operated side. The other 8 (40%) patients had no ipsilateral FDG-PET hypometabolism or contralateral hypometabolism. There were no statistically significant differences with respect to the localization of FDG-PET hypometabolism (hypometabolism on operated side vs. no hypometabolism or contralateral hypometabolism) between mesial and neocortical MRI-negative TLE (p = 0.642, Table 3). When analyzing the localization of hypometabolism on the operated side, the hypometabolism was present in the anterior and medial parts of the temporal lobe in 5 (39%) of the 13 patients with mesial MRI-negative TLE and in 3 (43%) of the 7 patients with neocortical MRI-negative TLE (p = 1.000). Table 1 Main clinical features of 20 evaluated patients with MRI-negative TLE. Patient no. Localization of the SOZ (M/N) Age at epilepsy onset (years) Duration of epilepsy (years) Frequency of CPSs per month Occurrence of GTCS (+/−) Potential cause Side of operation (R/L) Type of operation Outcome (Engel I–IV) Histopathological finding (FCD/MCD/negative) 1 Mes 16 11 3 − ME L AMTR I Negative 2 Neo 11 14 10 + Negative R Tailored cortectomy I Negative 3 Mes⁎ 23 18 10 + Negative L AMTR II FCD Ia 4 Mes 14 9 20 + Negative L AMTR IV FCD Ib 5 Mes 36 11 8 + Negative L AMTR I Negative 6 Mes 6 31 5 − Negative R AMTR III Negative 7 Neo 11 14 15 − Negative L Tailored cortectomy IV Negative 8 Mes 17 30 18 + PT R AMTR I FCD Ia 9 Neo 17 10 12 − ME L AMTR I FCD Ia 10 Mes⁎ 1 40 5 − ME R AMTR IV Negative 11 Mes 28 5 5 + Trauma R AMTR I Negative 12 Neo 10 24 9 + FS R AMTR III FCD Ia 13 Mes 0 41 3 + Negative L AMTR I Negative 14 Neo 9 20 3 + Negative L Tailored cortectomy III Negative 15 Mes⁎ 3 23 10 + Negative L AMTR I Negative 16 Neo 16 3 5 − Negative L Resection of pole sparing hippocampus IV Negative 17 Mes 25 10 8 + PT R AMTR III Negative 18 Neo 29 4 10 + Trauma R AMTR + cortectomy I Negative 19 Mes 23 12 5 − Negative R AMTR III Negative 20 Mes 17 14 8 + PT L AMTR I Negative AMTR — anteromesial temporal lobe resection, CPS — complex partial seizure, FCD — focal cortical dysplasia, FS — febrile seizure, GTCS — generalized tonic-clonic seizure, L — left, Mes — mesial, ME — meningitis/encephalitis, neo — neocortical, PT — perinatal trauma, R — right, SOZ — seizure onset zone. ⁎ Seizure onset with alternating involvement of temporal pole and mesial temporal lobe structures. 23I. Doležalová et al. / Epilepsy & Behavior 61 (2016) 21–26 186 3.4. Semiinvasive EEG 3.4.1. Semiinvasive EEG — interictal analysis When analyzing interictal semiinvasive EEG, 9 of the 20 patients (45%) exhibited predominantly unilateral IEDs; 11 of the 20 patients (55%) exhibited bilateral IEDs. There was no significant difference in IED distribution (unilateral vs. bilateral IEDs) between patients with mesial MRI-negative TLE and patients with neocortical MRI-negative TLE (Table 3). The IEDs in an extraanterotemporal region were present in 5 of the 20 patients (25%): in 1 of the 13 patients (7.7%) with mesial MRI-negative TLE and in 4 of the 7 patients (57%) with neocortical MRI-negative TLE. This difference reached statistical significance (p = 0.031) (Table 3). 3.4.2. Semiinvasive EEG — ictal analysis When analyzing semiinvasive ictal EEG, 7 of the 20 patients (35%) had all of their seizures correctly regionalized/lateralized; the other 13 of the 20 patients (65%) had at least one seizure either nonlateralized or falsely regionalized/lateralized. Seven of the 13 patients (54%) had all of their seizures correctly regionalized/lateralized in mesial MRI-negative TLE; this was true for none of the 7 patients (0%) with neocortical MRI-negative TLE. This difference was statistically significant (p = 0.044, Table 3). 3.5. Ictal semiology Nineteen of the 20 patients (95%) exhibited an aura. Six (30%) had epigastric aura; another type of aura was present in the other 13 patients (65%) (Table 4). We did not find any significant differences between mesial MRI-negative TLE and neocortical MRI-negative TLE in the type of aura (epigastric vs. other). A total number of 144 seizures (109 CPSs, 35 GTCSs; 80 seizures were recorded in semiinvasive EEG, the other 64 in IEEG) were available for clinical semiology evaluation; the number of evaluated seizures per patient ranged from 5 to 12 (with an average of 7.2 ± 1.8). The number of evaluated CPSs per patient ranged from 0 to 11 (with an average of 5.5 ± 2.3); the number of evaluated GTCSs per patient ranged from 0 to 5 (with an average of 1.85 ± 1.5). We did not perform a statistical analysis of the semiology between patients with mesial MRI-negative TLE and patients with neocortical MRI-negative TLE because of the small number of patients, but it seemed that there were some signs more often associated with mesial seizure origin (early oroalimentary automatisms, early noninvasive head turning, RINCH) or lateral seizure origin (periictal face rubbing, Table 4). 3.6. Surgery and surgical outcome Standard anteromesial temporal lobe resection (AMTR) was performed in 14 patients (70%), tailored cortectomy of the lateral temporal cortex in 3 (15%), AMTR plus tailored cortectomy of the lateral temporal cortex in 2 patients (10%), and resection of the temporal pole sparing the hippocampus in 1 patient (5%). When analyzing surgical outcome at year 1 after surgery, 12 of the 20 patients (60%) were characterized as Engel I and the other 8 (40%) as Engel II–IV (1 patient Engel II, 4 patients Engel III, 3 patients Engel IV). There were no significant differences in surgical outcome at year 1 after surgery between mesial MRI-negative TLE and neocortical MRInegative TLE, but the tendency of mesial MRI-negative TLE to result in better surgical outcome was present (Table 5). Table 2 The differences in demographic data between mesial and neocortical MRI-negative TLE. Mesial (n = 13) Neocortical (n = 7) p-Value Age at epilepsy onset (mean ± SD, min–max) 16.1 ± 11.1, 0–36 14.7 ± 7.0, 9–29 0.721 Duration of epilepsy (mean ± SD, min–max) 19.6 ± 12.2, 5–41 12.7 ± 7.8, 3–24 0.322 CPSs per month (mean ± SD, min–max) 8.3 ± 3–20 9.1 ± 3–15 0.423 Presence of GTCS, n (%) 9 (69) 4 (57) 0.651 Presence of potential insults, n (%) 6 (46) 3 (43) 1.000 – Encephalitis/meningitis, n (%) 2 (15) 1 (14) – Head injury, n (%) 1 (8) 1 (14) – Febrile seizures, n (%) 0 (0) 1 (14) – Perinatal trauma, n (%) 3 (23) 0 (0) Side of the SOZ — right, n (%) 6 (46) 3 (43) 1.000 CPS — complex partial seizure, GTCS — generalized tonic-clonic seizure, SOZ — seizure onset zone. Table 3 The differences in FDG-PET findings and semiinvasive EEG between mesial and neocortical MRI-negative TLE. Mesial (n = 13) Neocortical (n = 7) p-Value FDG-PET findings Clearcut hypometabolism, n (%) 7 (54) 5 (71) 0.642 Interictal analysis Unilateral IEDs, n (%) 5 (38) 4 (57) 0.642 Bilateral IEDs, n (%) 8 (62) 3 (43) Extraanterotemporal IEDs, n (%) 1 (8) 4 (57) 0.031 Ictal analysis All seizures correctly regionalized/lateralized, n (%) 7 (54) 0 (0) 0.044 At least 1 seizure nonlateralized or falsely regionalized/lateralized, n (%) 6 (46) 7 (100) IEDs — interictal epileptiform discharges. Table 4 The differences in clinical semiology between mesial and neocortical MRI-negative TLE. Mesial (n = 13) Neocortical (n = 7) p-Value Type of aura Epigastric aura, n (%) 4 (31) 2 (29) 1.000 Other type, n (%) 9 (69) 5 (71) Clinical semiology Early oroalimentary automatisms, n (%) 7 (54) 2 (29) Early nonversive head turning, n (%) 10 (77) 3 (43) Immobility of the upper limb, n (%) 6 (46) 2 (28) Ictal dystonia, n (%) 3 (23) 2 (29) RINCH, n (%) 4 (31) 0 (0) Periictal nose wiping, n (%) 9 (69) 5 (71) Periictal face rubbing, n (%) 6 (46) 6 (86) Periictal bed leaving, n (%) 3 (23) 2 (29) Retching, n (%) 2 (15) 0 (0) Coughing, n (%) 6 (46) 2 (29) Urinary urge, n (%) 2 (15) 1 (14) Water drinking, n (%) 2 (15) 0 (0) RINCH — rhythmic ictal nonclonic hand motions. Table 5 The differences in histopathological findings and in surgical outcome at year 1 after surgery defined on the basis of Engel classification between mesial and neocortical MRInegative TLE. Mesial (n = 13) Neocortical (n = 7) p-Value Outcome Engel I, n (%) 9 (69) 3 (42) 0.356 Engel II–IV, n (%) 4 (31) 4 (57) Histopathological findings Negative, n (%) 10 (77) 5 (71) 1.000 FCD, n (%) 3 (23) 2 (29) FCD — focal cortical dysplasia. 24 I. Doležalová et al. / Epilepsy & Behavior 61 (2016) 21–26 187 3.7. Histopathological findings The histopathological findings were negative in 15 of the 20 patients (75%); FCD was present in 5 patients (25%) (FCD Ia was present in 4 of the 5 patients (80%) with FCD; FCD Ib was present in 1 of the 5 patients (20%) with FCD). There were no statistically significant differences in histopathological findings between mesial MRI-negative TLE and neocortical MRI-negative TLE (Table 5). 4. Discussion Patients with MRI-negative TLE represent a distinct subgroup of the patients with nonlesional epilepsy — they differ from the others because of a relatively good surgical outcome, and therefore, they come to the forefront of many centers for epilepsy surgery [8,12–14,25,26]. On the basis of MRI and FDG-PET, Carne et al. described a group of patients with MRI-negative, PET-positive TLE. This patient type was subsequently characterized in detail by other authors [8–10,12,13]. For patients with MRI-negative, PET-positive TLE, FDG-PET hypometabolism is typical, which correlates with interictal and ictal electrophysiological findings, and these patients had excellent surgical outcomes. In our study of MRI-negative TLE, 60% of the patients had characteristic FDG-PET hypometabolism that corresponded to the side of operation; the remaining 40% of the patients had either no findings on FDG-PET or the hypometabolism was lateralized to the nonoperated side. This is in agreement with the study by Lee et al. in which a correlation of hypometabolism on PET and side of resection was described in 59% of patients [14]. Patients with MRI-negative TLE can be classified according to the localization of their SOZ as mesial MRI-negative TLE or neocortical MRInegative TLE. In our study, 65% of the patients were classified as having mesial MRI-negative TLE, and 35% of the patients were classified as having neocortical MRI-negative TLE. Luther et al. stated that 25% of patients have neocortical MRI-negative TLE [17]. In a study by Lee et al., a group of 32 patients with MRI-negative TLE was analyzed, with 25% of the patients having “pure” neocortical seizure onset, 38% of the patients having “pure” mesial seizure onset, and the rest of the patients with independent or simultaneous mesial and neocortical seizure onset [14]. In our analysis, we tried to find differences between mesial and neocortical MRI-negative TLE based on noninvasive and semiinvasive data analysis. These differences might be helpful in surgery planning, as they could distinguish between patients who could proceed directly to surgery and patients in whom IEEG or perioperative electrocorticography is necessary. In our study, we found differences based on semiinvasive EEG analysis, namely different distribution of IEDs and different ictal onset pattern. In the literature, there are also references about different ictal semiology between patients with mesial MRI-negative TLE and patients with neocortical MRI-negative TLE [14]. In our study, extraanterotemporal IEDs were found in approximately 25% of the patients. We found a statistically significantly higher incidence of extraanterotemporal IEDs in patients with neocortical MRI-negative TLE than in patients with mesial MRI-negative TLE (57% vs. 8%). In the past, the predominance of lateral neocortical IEDs was reported to be associated with lesional neocortical TLE [27]. When analyzing ictal onset patterns in semiinvasive EEG, we found that only 35% of the patients had all of their seizures correctly lateralized; the remaining 65% of the patients had at least one seizure nonlateralized or falsely regionalized/lateralized. We found that patients with mesial MRI-negative TLE had all of their seizures correctly lateralized more often than patients with neocortical MRI-negative TLE (54% vs. 0%). This difference is also supported by literature focusing on lesional TLE. In a study by O'Brien et al., almost 80% of patients with neocortical TLE had bilateral ictal activity at onsets, compared to almost 50% of patients with mesial TLE [28]. According to the results, we believe that the lateralization of ictal activity at the beginning of seizures and the distribution of IEDs could be an indicator of SOZ localization and a red flag for surgery, but the possibility of unequivocal localization (mesial vs. neocortical) seems to be limited. We did not find any significant differences in demographic data (age at epilepsy onset, duration of epilepsy, frequency of CPSs, the presence/ absence of GTCS, the presence of potential insults for the development of epilepsy, or side of the SOZ) between patients with mesial MRI-negative TLE and patients with neocortical MRI-negative TLE. No differences in a group of patients with lesional TLE between mesial and neocortical TLE were reported with respect to seizure frequency, proportion with GTCS, or epilepsy history [27,28]. The only characteristic of lesional mesial TLE is the occurrence of febrile seizures by medical history, which is conditioned by the fact that the febrile seizure is a known risk factor for hippocampal sclerosis development [29,30]. Similarly to other authors, we found only a very low number of patients with FS (approximately 5% of patients exhibited FS in both studies) [8]. We did not prove any significant differences in the representation of bilateral IEDs between mesial and neocortical MRI-negative TLE. The same results were reported by other authors analyzing interictal EEG differences between mesial and neocortical MRI-negative and lesional TLE [14,28,31]. In the literature, some ictal semiology features are associated with mesial seizure onset (i.e., epigastric aura, ipsilateral limb automatism, contralateral dystonic posturing, and oroalimentary automatisms) and others with neocortical seizure onset (i.e., psychic, auditory, and visual aura and early aphasia) [29,32–36]. A study by Lee et al. focusing on differences between mesial and neocortical MRI-negative TLE reported a difference in clinical semiology that reached statistical significance [14]. All of the patients with exclusively neocortical temporal seizure onset had typical neocortical temporal semiology in comparison to patients with mesial and independent or simultaneous mesial and neocortical seizure onset. We observed similar tendencies in our study. On the other hand, there are also studies which did not find any statistically significant difference in semiology between mesial and neocortical TLE [28]. As mentioned at the beginning of this section, MRI-negative TLE is characterized by a better surgical outcome than other nonlesional types of epilepsy, especially in cases of MRI-negative, PET-positive TLE. LoPinto-Khoury et al. reported that 76% of patients with MRI-negative, PET-positive TLE are classified as Engel I in year 2 and 75% in year 5 after surgery [12]. In a study by Carne et al., the surgical outcome in patients with MRI-negative, PET-positive TLE was even slightly better than in patients with hippocampal sclerosis [8]. In a study published by Kuba et al. analyzing patients with MRI-negative, PET-positive TLE who had surgery in our department, 71% of the patients were classified as Engel I at the final follow-up [13]. In our present study, 60% of patients were postoperatively classified as Engel I in year 1 after surgery. Our results are in agreement with two studies focusing on MRI-negative TLE: Lee et al. (who reported Engel class I in 65% of patients) and Burkholder et al. (who reported Engel class I in 55% of patients) [14,15]. When looking at differences in surgical outcome between mesial and neocortical MRI-negative TLE, we found that patients with mesial MRI-negative TLE tended to have better surgical prognosis than patients with neocortical MRI-negative TLE (69% seizure-free vs. 42%), but these results did not reach statistical significance. Similar results were published by Lee et al.: seizure freedom was attained in 70% of patients with mesial temporal onset and in 33% of patients with neocortical seizure onset [14]. In our study, 75% of the patients had negative histopathological findings, and the remaining 25% of the patients exhibited signs of FCD. The Negative histopathology seems to be common in MRInegative TLE [8,11]. Carne et al. found FCD in 10% of the patients and HS in 10% of the patients [8]. Similar results were published by Immonem et al., who found negative histopathological findings in 68% of patients with nonlesional TLE; there were HS, signs of gliosis in the hippocampus/amygdala, oligodendroglioma, and dysembryoplastic neuroepithelial tumor (DNET) in the remaining 32% of the patients in 25I. Doležalová et al. / Epilepsy & Behavior 61 (2016) 21–26 188 their study [11]. We did not prove any statistically significant differences in pathology between patients with mesial and neocortical MRInegative TLE. The biggest limitation of our study is the small sample size of 20 patients with MRI-negative TLE. In such a small group, it is probable that we missed some of the differences. We believe that there are distinctions between mesial MRI-negative TLE and neocortical MRI-negative TLE with respect to seizure semiology and to surgical outcome, but we were not able to prove them statistically. The other problems of our study are its retrospective design and the “purity” of the group with mesial MRI-negative TLE, in which 15% of the patients had seizure onset alternating between the area of the temporal pole and mesial temporal lobe structures. 5. Conclusions In conclusion, approximately two-thirds of patients with MRInegative TLE have the seizure onset localized to the mesial temporal lobe structures, and one-third have the seizure onset localized to the neocortical areas. Noninvasive data, specifically the distribution of IEDs and the lateralization of ictal activity, contain some features which help distinguish between mesial and neocortical MRI-negative TLE. Patients with mesial MRI-negative TLE tend to have IEDs more restricted to the anterotemporal area and to have all of their seizures clearly lateralized into one temporal lobe. Despite these findings, it seems that only IEEG is able to provide adequate information about precise SOZ localization, which is essential for surgery. Only a correctly tailored resection can result in seizure freedom in a given patient with a minimal risk of a cognitive decline [14]. If we manage to localize the SOZ correctly in patients with MRI-negative TLE, almost 60% of patients can become seizure-free after the surgery, which is higher than in other types of nonlesional epilepsies and almost comparable to lesional ones. Acknowledgments We would like to thank Anne Johnson for grammatical assistance. This work was supported by the project “CEITEC — Central European Institute of Technology” (CZ.1.05/1.1.00/02.0068) from the European Regional Development Fund. References [1] Cascino GD. Surgical treatment for epilepsy. Epilepsy Res 2004;60:179–86. [2] Berkovic SF, McIntosh AM, Kalnins RM, Jackson GD, Fabinyi GCA, Brazenor GA, et al. Preoperative MRI predicts outcome of temporal lobectomy — an actuarial analysis. Neurology 1995;45:1358–63. [3] Holmes MD, Born DE, Kutsy RL, Wilensky AJ, Ojemann GA, Ojemann LM. Outcome after surgery in patients with refractory temporal lobe epilepsy and normal MRI. Seizure 2000;9:407–11. [4] Chapman K, Wyllie E, Najm I, Ruggieri P, Bingaman W, Luders J, et al. Seizure outcome after epilepsy surgery in patients with normal preoperative MRI. J Neurol Neurosurg Psychiatry 2005;76:710–3. [5] Tatum WO, Benbadis SR, Hussain A, Al-Saadi S, Kaminski B, Heriaud LS, et al. Ictal EEG remains the prominent predictor of seizure-free outcome after temporal lobectomy in epileptic patients with normal brain MRI. Seizure 2008;17:631–6. [6] Bell ML, Rao S, So EL, Trenerry M, Kazemi N, Stead SM, et al. Epilepsy surgery outcomes in temporal lobe epilepsy with a normal MRI. Epilepsia 2009;50:2053–60. [7] Tellez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 2005;128:1188–98. [8] Carne RP, O'Brien TJ, Kilpatrick CJ, MacGregor LR, Hicks RJ, Murphy MA, et al. MRI-negative PET-positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain 2004;127:2276–85. [9] Carne RP, Cook MJ, MacGregor LR, Kilpatrick CJ, Hicks RJ, O'Brien TJ. “Magnetic resonance imaging negative positron emission tomography positive” temporal lobe epilepsy: FDG-PET pattern differs from mesial temporal lobe epilepsy. Mol Imaging Biol 2007;9:32–42. [10] Carne RP, O'Brien TJ, Kilpatrick CJ, MacGregor LR, Litewka L, Hicks RJ, et al. ‘MRInegative PET-positive’ temporal lobe epilepsy (TLE) and mesial TLE differ with quantitative MRI and PET: a case control study. BMC Neurol 2007;7:16. [11] Immonen A, Jutila L, Muraja-Murro A, Mervaala E, Äikiä M, Lamusuo S, et al. Longterm epilepsy surgery outcomes in patients with MRI-negative temporal lobe epilepsy. Epilepsia 2010;51:2260–9. [12] LoPinto-Khoury C, Sperling MR, Skidmore C, Nei M, Evans J, Sharan A, et al. Surgical outcome in PET-positive, MRI-negative patients with temporal lobe epilepsy. Epilepsia 2012;53:342–8. [13] Kuba R, Tyrlikova I, Chrastina J, Slana B, Pazourkova M, Hemza J, et al. “MRI-negative PET-positive” temporal lobe epilepsy: invasive EEG findings, histopathology, and postoperative outcomes. Epilepsy Behav 2011;22:537–41. [14] Lee RW, Hoogs MM, Burkholder DB, Trenerry MR, Drazkowski JF, Shih JJ, et al. Outcome of intracranial electroencephalography monitoring and surgery in magnetic resonance imaging-negative temporal lobe epilepsy. Epilepsy Res 2014;108: 937–44. [15] Burkholder DB, Sulc V, Hoffman EM, Cascino GD, Britton JW, So EL, et al. Interictal scalp electroencephalography and intraoperative electrocorticography in magnetic resonance imaging-negative temporal lobe epilepsy surgery. JAMA Neurol 2014; 71:702–9. [16] Helmstaedter C, Kurthen M, Lux S, Reuber M, Elger CE. Chronic epilepsy and cognition: a longitudinal study in temporal lobe epilepsy. Ann Neurol 2003;54:425–32. [17] Luther N, Rubens E, Sethi N, Kandula P, Labar DR, Harden C, et al. The value of intraoperative electrocorticography in surgical decision making for temporal lobe epilepsy with normal MRI. Epilepsia 2011;52:941–8. [18] Schwartz TH, Bazil CW, Walczak TS, Chan S, Pedley TA, Goodman RR. The predictive value of intraoperative electrocorticography in resections for limbic epilepsy associated with mesial temporal sclerosis. Neurosurgery 1997;40:302–11. [19] San-juan D, Tapia CA, Gonzalez-Aragon MF, Martinez Mayorga A, Staba RJ, AlonsoVanegas M. The prognostic role of electrocorticography in tailored temporal lobe surgery. Seizure 2011;20:564–9. [20] Wray CD, McDaniel SS, Saneto RP, Novotny Jr EJ, Ojemann JG. Is postresective intraoperative electrocorticography predictive of seizure outcomes in children? J Neurosurg Pediatr 2012;9:546–51. [21] Steinhoff BJ, So NK, Lim S, Lüders HO. Ictal scalp EEG in temporal lobe epilepsy with unitemporal versus bitemporal interictal epileptiform discharges. Neurology 1995; 45:889–96. [22] Risinger MW, Engel Jr J, Van Ness PC, Henry TR, Crandall PH. Ictal localization of temporal lobe seizures with scalp/sphenoidal recordings. Neurology 1989;39: 1288–93. [23] Engel J, Ness PV, Rasmussen T, Ojemann L. Outcome with respect to epileptic seizures. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. [24] Blümcke I, Thom M, Aronica E, Armstrong DD, Vinters HV, Palmini A, et al. The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc task force of the ILAE Diagnostic Methods Commission. Epilepsia 2011;52:158–74. [25] Tellez-Zenteno JF, Hernandez-Ronquillo LH, Moien-Afshari F, Wiebe S. Surgical outcomes in lesional and non-lesional epilepsy: a systematic review and meta-analysis. Epilepsy Res 2010;89:310–8. [26] Spencer S, Huh L. Outcomes of epilepsy surgery in adults and children. Lancet Neurol 2008;7:525–37. [27] Pfänder M, Arnold S, Henkel A, Weil S, Werhahn KJ, Eisensehr I, et al. Clinical features and EEG findings differentiating mesial from neocortical temporal lobe epilepsy. Epileptic Disord 2002;4:189–95. [28] O'Brien TJ, Newton MR, Cook MJ, Berlangieri SU, Kilpatrick C, Morris K, et al. Hippocampal atrophy is not a major determinant of regional hypometabolism in temporal lobe epilepsy. Epilepsia 1997;38:74–80. [29] French JA, Williamson PD, Thadani VM, Darcey TM, Mattson RH, Spencer SS, et al. Characteristics of medial temporal lobe epilepsy: I. Results of history and physical examination. Ann Neurol 1993;34:774–80. [30] Harvey AS, Grattan-Smith JD, Desmond PM, Chow CW, Berkovic SF. Febrile seizures and hippocampal sclerosis: frequent and related findings in intractable temporal lobe epilepsy of childhood. Pediatr Neurol 1995;12:201–6. [31] Burgerman RS, Sperling MR, French JA, Saykin AJ, O'Connor MJ. Comparison of mesial versus neocortical onset temporal lobe seizures: neurodiagnostic findings and surgical outcome. Epilepsia 1995;36:662–70. [32] Dupont S, Semah F, Boon P, Saint-Hilaire JM, Adam C, Broglin D, et al. Association of ipsilateral motor automatisms and contralateral dystonic posturing: a clinical feature differentiating medial from neocortical temporal lobe epilepsy. Arch Neurol 1999;56:927–32. [33] Villanueva V, Serratosa JM. Temporal lobe epilepsy: clinical semiology and age at onset. Epileptic Disord 2005;7:83–90. [34] Tatum WO. Mesial temporal lobe epilepsy. J Clin Neurophysiol 2012;29:356–65. [35] Bercovici E, Kumar BS, Mirsattari SM. Neocortical temporal lobe epilepsy. Epilepsy Res Treat 2012;2012:1031–60. [36] Kennedy JD, Schuele SU. Neocortical temporal lobe epilepsy. J Clin Neurophysiol 2012;29:366–70. 26 I. Doležalová et al. / Epilepsy & Behavior 61 (2016) 21–26 189 Annex 11: Rehulka P, Dolezalova I, Janousova E, Tomasek M, Marusic P, Brazdil M, Kuba R. Ictal and postictal semiology in patients with bilateral temporal lobe epilepsy. Epilepsy & Behavior. 2014;41:40-46. 190 Ictal and postictal semiology in patients with bilateral temporal lobe epilepsy Pavel Řehulka a,b, ⁎, Irena Doležalová a,b , Eva Janoušová c , Martin Tomášek d , Petr Marusič d , Milan Brázdil a,b , Robert Kuba a,b a Brno Epilepsy Center, First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic b Behavioural and Social Neuroscience Research Group, Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic c Institute of Biostatistics and Analyses, Masaryk University, Brno, Czech Republic d Department of Neurology, Charles University in Prague, Second Faculty of Medicine, Motol University Hospital, Prague, Czech Republic a b s t r a c ta r t i c l e i n f o Article history: Received 11 July 2014 Revised 7 September 2014 Accepted 10 September 2014 Available online 1 October 2014 Keywords: Epilepsy Bitemporal Bilateral temporal lobe epilepsy Invasive EEG Semiology Postictal unresponsiveness Bilateral temporal lobe epilepsy is characterized by evidence of seizure onset independently in both temporal lobes. The main aim of the present study was to determine whether patients with evidence of independent bilateral temporal lobe epilepsy (biTLE) can be identified noninvasively on the basis of seizure semiology analysis. Thirteen patients with biTLE, as defined by invasive EEG, were matched with 13 patients with unilateral temporal lobe epilepsy (uniTLE). In all 26 patients, the frequency of predefined clusters of ictal and periictal signs were evaluated: ictal motor signs (IMSs), periictal motor signs (PIMSs), periictal vegetative signs (PIVSs), the frequency of early oroalimentary automatisms (EOAs), and the duration of postictal unresponsiveness (PU). Some other noninvasive and clinical data were also evaluated. A lower frequency of IMSs was noted in the group with biTLE (patients = 46.2%, seizures = 20.7%) than in the group with uniTLE (patients = 92.3%, seizures = 61.0%) (p = 0.030; p b 0.001, respectively). The individual IMS average per seizure was significantly lower in the group with biTLE (0.14; range = 0–1.0) than in the group with uniTLE (0.80; range = 0–2.6) (p = 0.003). Postictal unresponsiveness was longer than 5 min in more patients (75.0%) and seizures (42.9%) in the group with biTLE than in the group with uniTLE (patients = 30.8%, seizures = 18.6%) (p = 0.047; p = 0.002). The frequency of EOAs, PIMSs, PIVSs, and other clinical data did not differ significantly. There is a lower frequency of ictal motor signs and longer duration of postictal unresponsiveness in patients with biTLE. © 2014 Elsevier Inc. All rights reserved. 1. Introduction Bilateral temporal lobe epilepsy or also known as bitemporal epilepsy is often vaguely characterized by the existence of independent seizureonset zones in both temporal lobes. Bitemporal epilepsy is not clearly defined and is usually suspected when independent bilateral temporal seizures are recorded in scalp EEG. Bitemporal epilepsy has been defined by depth electrodes, as clear clinical and scalp EEG differences that could noninvasively distinguish bitemporal epilepsy from unilateral temporal lobe epilepsy have not been established [1]. Patients with bitemporal epilepsy are generally considered to be poorer surgery candidates than patients with unilateral temporal lobe epilepsy [2–5]. Seizure semiology is an important part of the presurgical assessment of epilepsy surgery candidates. To the best of our knowledge, the seizure semiology in bitemporal epilepsy and that in unilateral temporal lobe epilepsy have not been compared in detail. The main goal of this study was to reveal potential differences between the patients with bitemporal epilepsy and those with unilateral temporal lobe epilepsy in terms of history data, semi-invasive EEG findings, and seizure semiology. 2. Methods 2.1. Group definition We reviewed all of the patients with temporal lobe epilepsy (TLE) who underwent invasive video-EEG at one of two epilepsy centers in Czech Republic: the Brno Epilepsy Center at St. Anne's University Hospital between 1999 and 2012 and the Epilepsy Center Motol at the University Hospital Motol in Prague between 2006 and 2012. We defined the following criteria for identifying the patients in the group with bitemporal epilepsy (biTLE): independent bitemporal seizure origin, defined on the basis of invasive EEG as (1) spontaneous clinical seizures arising independently from both temporal lobes (electrographic Epilepsy & Behavior 41 (2014) 40–46 ⁎ Corresponding author at: Brno Epilepsy Center First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Pekařská 53, 656 91 Brno, Czech Republic. Tel.: +420 543 182 645; fax: +420 543 182 624. E-mail address: rehulka.pavel@fnusa.cz (P. Řehulka). http://dx.doi.org/10.1016/j.yebeh.2014.09.033 1525-5050/© 2014 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh 191 seizures were not included in this analysis because their clinical significance is not yet clearly determined [6]) and/or (2) habitual complex partial seizures (CPSs) elicited by the electrical stimulation of the temporal lobe contralateral to the spontaneous seizures. For comparison purposes, we formed a control group of patients with unilateral temporal lobe epilepsy (the group with uniTLE). The subjects in this group were patients with uniTLE who were completely seizure-free for at least two years after epilepsy surgery. Patients from the group with uniTLE were matched with patients from the group with biTLE in terms of age at the onset of epilepsy, age at evaluation, duration of epilepsy, and gender. We selected a group of 26 patients who fulfilled these matching criteria (13 in the group with biTLE and 13 in the group with uniTLE). The study was approved by the Ethics Committee of St. Anne's University Hospital. 2.2. Presurgical evaluation All 26 patients underwent a comprehensive presurgical evaluation, including detailed history and neurological examination, neuropsychological testing, magnetic resonance imaging (MRI), and scalp video-EEG monitoring. Bilateral carotid sodium amobarbital/methohexital testing was performed in 11 patients of the group with biTLE and 12 patients of the group with uniTLE; unilateral testing was available in one patient of the group with biTLE. Interictal and/or ictal single-photon emission computed tomography (SPECT) was performed in 10 patients of the group with biTLE and in seven patients in the group with uniTLE. Fluorodeoxyglucose positron emission tomography (FDG-PET) was performed in 12 patients from the group with biTLE and in seven patients from the group with uniTLE. 2.3. Scalp EEG, semi-invasive EEG, and invasive EEG procedures In all 13 patients in the group with biTLE, invasive video-EEG monitoring was a part of the presurgical evaluation. In the patients from the group with uniTLE, depth EEG was performed in four patients because their noninvasive data were insufficient to proceed directly to surgery. Evidence indicates that all 26 patients had mesial temporal lobe epilepsy: in all of the patients in the group with biTLE and in four of the patients from the group with uniTLE who underwent invasive EEG (patients 14, 15, 18, and 23), the seizure-onset zone (SOZ) was found within the hippocampus, amygdala, or temporal pole (i.e., anteromesio-temporal onset) (see Table 1); in the remaining nine patients from the group with uniTLE (patients 16, 17, 19, 20, 21, 22, 24, 25, and 26) the lesion localization, long-term video-EEG monitoring with scalp/sphenoidal electrodes, and FDG-PET findings led us to consider them to be patients with mesial temporal lobe epilepsy. Scalp/sphenoidal EEG was performed using the international 10–20 electrode placement system. Multicontact depth electrodes inserted orthogonally or diagonally into both amygdalohippocampal complexes were used in all patients of the group with biTLE. A combination of two stereotactically implanted depth electrodes and subdural strip electrodes was used in two patients (patients 12 and 13). Only preoperative EEG data were used for further analysis in all of the patients. 2.4. Surgery and outcome measure Eight patients from the group with biTLE and all of the patients from the group with uniTLE underwent resective surgery. We recorded the Table 1 Demographic and clinical characteristics, histopathology, and outcome in the group with biTLE (patients: 1–13) and in the group with uniTLE (patients: 14–26). M — male; F — female; TBI — traumatic brain injury; PI — perinatal insult; FSs — febrile seizures; M/E — meningitis/encephalitis; LD — language dominance according to the Wada test; L — left; R — right; ⁎ — only unilateral testing performed; SOZ — localization of seizure-onset zone proven by invasive EEG (if performed); AHC — amygdalohippocampal complex; Tpol — temporal pole; HS — hippocampal sclerosis; DNET — dysembryoplastic neuroepithelial tumor; TL — temporal lobe; MCD — malformation of cortical development; HA — hippocampal atrophy; MAng — meningioangiomatosis; HIMAL — hippocampal malrotation; FCD — focal cortical dysplasia; VNS — vagus nerve stimulation; AMTR — anteromedial temporal lobe resection; LE — lesionectomy. Patient Sex/age at evaluation (years) Insult LD Side-SOZ identified by invasive EEG MRI finding Side/surgery Histopathology Follow-up surgery (years) Follow-up VNS (years) Outcome (Engel) Outcome (Mc Hugh) Bitemporal group 1 M/33 TBI L L-AHC; R-AHC Normal – – – 14 – I 2 M/41 PI, FSs L L-AHC; R-AHC HS L/AMTR HS grade IV 3 2 III A V 3 F/29 – L R-AHC; L-AHC Tumor R/AMTR DNET 3 – I B – 4 F/25 – L R-AHC; L-AHC AHC hyperintensity R/AMTR Gliosis 10 – I A – 5 M/23 – L L-Tpol, AHC; R- AHC TL hypotrophy L/AMTR FCD IA 2 1 IV V 6 F/19 – L L-Tpol, AHC; R- AHC Normal L/temporal pole resection Normal 2 1 IV III 7 F/41 M/E L R-AHC; L-AHC HS R/AMTR HS grade III 2 – II A – 8 F/33 – L⁎ R-AHC; L-AHC MCD – – – 2 – V 9 M/47 PI – L-AHC; R-AHC HA – – – 2 – II 10 F/51 – L R-AHC; L-AHC TL lesion R/AMTR MAng 1 – III A – 11 F/29 PI L R-AHC; L-AHC HIMAL – – – 4 – V 12 M/41 – L L-Tpol, AHC; R-AHC Suspected FCD L/AMTR FCD IA 5 – IV – 13 M/43 PI, M/E L L-Tpol, AHC; R-AHC HS – – – – – – Unitemporal group 14 M/33 – L R-AHC HA R/AMTR Normal 9 – I A – 15 M/29 PI, FSs – – HS R/AMTR HS grade uncertain 10 – I A – 16 M/41 M/E, FSs L L-AHC HS L/AMTR HS grade uncertain 9 – I A – 17 M/21 PI L – HA L/AMTR Normal 3 – I A – 18 F/41 – L L-AHC Normal L/AMTR FCD IA 3 – I A – 19 M/40 – L – Cavernoma R/extended LE Cavernoma 9 – I A – 20 F/36 PI, FSs L – HS R/AMTR HS grade uncertain 8 – I A – 21 M/31 – L – HS L/AMTR HS grade III 7 – I A – 22 F/24 – L – Tumor R/extended LE DNET 9 – I A – 23 F/47 – L L-AHC Normal L/AMTR Normal 2 – I A – 24 M/33 – L – Normal R/AMTR Normal 5 – I A – 25 M/43 – L – Normal R/AMTR FCD IA 3 – I A – 26 M/36 M/E L – HA L/AMTR Normal 4 – I A – 41P. Řehulka et al. / Epilepsy & Behavior 41 (2014) 40–46 192 seizure outcome at the last follow-up visit for all of the patients and classified them based on Engel's classification [7]. Individual features, including side of surgery, follow-up duration, and outcome at the most recent follow-up visit, are presented in Table 1. 2.5. VNS procedure and outcome measure Five patients from the group with biTLE received vagus nerve stimulation (VNS). The implantation procedure was performed in a standardized manner and under general anesthesia. We recorded the seizure reduction at the most recent follow-up visit for all of the patients and classified them based on the McHugh classification for measuring the efficacy of VNS [8]. Individual follow-up duration and outcome at the most recent follow-up visit are mentioned in Table 1. 2.6. Histopathological examination A histopathological examination was performed on patients who underwent resective surgery. If focal cortical dysplasia was present in the temporopolar region, the recent ILAE classification system [9] was used; proven cases of hippocampal sclerosis were classified according to the classification of Wyler [10]. 2.7. Clinical variables Anamnestic data were collected using a retrospective review of the medical records, including the presence or absence of insults prior to the development of the epilepsy such as traumatic brain injury, history of perinatal insult, history of febrile seizures, history of central nervous system infection (i.e., meningitis or encephalitis), anamnestic experience of aura, the presence of b1 or ≥1 secondarily generalized tonic– clonic seizure(s) (GTCS(s)) per year (at time of study), and total seizure frequency per month (TSF) calculated as the mean frequency during the six months prior to the preoperative evaluation. 2.8. Semi-invasive ictal EEG parameters Ictal EEGs were analyzed independently by two authors of this study (PR and ID); disagreements, if present, were resolved by consensus. The following ictal parameters were evaluated: • Seizure onset (SO) — defined as the time when the first unequivocal change in patient behavior or in EEG appeared, depending on whether EEG or behavioral change developed first; • Presence of early rhythmic theta/alpha activity (ERTA) — considered as a rhythmic 4- to 8-Hz activity present in scalp/sphenoidal electrodes Sp1, Sp2, T1, and T2 within 20 s from the SO; • Time to the development of rhythmic theta/alpha activity (TRTA) — defined as the time from the SO to the development of rhythmic 4to 8-Hz activity in scalp/sphenoidal electrodes Sp1, Sp2, T1, and T2; • Bitemporal propagation time (BPT) — defined as the time from the SO to the propagation of rhythmic ictal activity to at least one of the contralateral electrodes Sp1, T1, T3 or Sp2, T2, T4; • Ictal activity duration (IAD) — defined as the time from the SO to the end of the ictal activity (EIA), i.e., to the replacement of rhythmic ictal activity with irregular slowing or diffuse attenuation without any rhythmic ictal activity within all scalp/sphenoidal electrodes. 2.9. Seizure semiology Scalp, semi-invasive, and invasive ictal video-EEG recordings were assessed, and the following ictal and periictal signs were evaluated: • Early oroalimentary automatisms (EOAs) — defined as repetitive movements of the mouth, such as mastication, swallowing, lip smacking, or blowing, considered only if present within 20 s from the SO • Ictal motor signs (IMSs) occurring during the ictal period of CPSs, which included the following: – Early nonversive head turning — defined as a natural and unforced lateral movement of the head within first 30 s after the SO [11] – Lateralized ictal immobility of the upper limb — defined as a condition lasting more than 10 s in which one of the upper extremities performs no movement and rests in a natural position next to the body without visually detected dystonic or tonic posturing, while the second upper limb demonstrates movement [12] – Ictal dystonia — defined as an unnatural posture of one upper extremity associated with the rotatory component [13] – Rhythmic ictal nonclonic hand (RINCH) motions — defined as unilateral, rhythmic, nonclonic, nontremor motions of an upper extremity or opening–closing motions of the hand lasting more than 5 s [14] • Periictal vegetative symptoms (PIVSs), which included retching with/ without vomiting, cough, urinary urge, and water drinking; all during a seizure and/or within 2 min after seizure termination, as defined by EIA [15] • Periictal motor signs (PIMSs), which included the following: – Periictal nose wiping or face rubbing — defined as the movement of one upper extremity directed to the face, rubbing of the nose during the seizure or within 2 min after seizure termination, as defined by EIA [16] – Periictal bed leaving — defined as lateralized leaving from bed during seizures or within 3 min after seizure termination, as defined by EIA [17] EOAs, IMSs, and PIVSs were analyzed both in CPSs and GTCSs; PIMSs only in CPSs • Postictal unresponsiveness (PU), which was analyzed only in CPSs and defined as the time from the end of EEG ictal activity to the recovery of normal contact, i.e., contact without impaired speech and without confusion. We used this definition because, usually, it is not possible to distinguish unequivocally between postictal dysphasia and postictal confusion in all seizures. The duration of PU was categorized according to its duration as short (≤3 min), medium (3–5 min), and long (N 5 min). 2.10. Statistical analysis The distribution of clinical variables was compared between the group with biTLE and that with uniTLE using the nonparametric Mann–Whitney test or Fisher's exact test according to their condition of validity. For all tests, p value b 0.05 was considered statistically significant. For TRTA, BPT, IAD, and IMSs, we calculated the individual average for each patient, and then we calculated the median of individual values for the group with biTLE and that with uniTLE because the individual values in both groups did not have a normal distribution. Table 2 Clinical characteristics of the group with biTLE and that with uniTLE; GTCSs — generalized tonic–clonic seizures; TSF — total seizure frequency per month; NS — statistically nonsignificant. biTLE uniTLE p-Value Number of patients 13 13 Perinatal insult 4 (30.8%) 3 (23.1%) NSa Febrile seizures 1 (7.7%) 3 (23.1%) NSa CNS infection 2 (15.4%) 2 (15.4%) NSa GTCS b 1 per year 10 (76.9%) 10 (76.9%) NSa Experience of aura 11 (84.6%) 11 (84.6%) NSa TSF in median [min–max] 7 [1–25] 7 [1–12] NSb a Fisher's exact test. b Mann–Whitney test. 42 P. Řehulka et al. / Epilepsy & Behavior 41 (2014) 40–46 193 3. Results 3.1. Demographic data The study included a total of 26 patients: 13 fulfilled the criteria for the group with biTLE and 13 fulfilled the criteria for the group with uniTLE. In the group with biTLE, there were 7 females and 6 males, and the age at the evaluation ranged from 19 to 51 years (with an average of 34 ± 9.8 years). In the group with uniTLE, there were 4 females and 9 males, and the age at the evaluation ranged from 21 to 47 years (with an average of 35 ± 8.2 years). The data regarding patient characteristics are summarized in Table 1. 3.1.1. Presurgical evaluation — language dominance In 24 patients, left-sided language dominance was proven according to the Wada test (Table 1). In patients 9 and 15, the Wada test was not performed, but other clinical data and neuropsychological results indicated left-sided dominance. 3.1.2. Surgery, histopathological findings, and outcome Of the 13 patients in the group with biTLE, eight (61.5%) underwent resective surgery, four (30.8%) received primarily VNS, and one (7.7%) refused VNS implantation. Three patients received VNS after unsuccessful resective surgery. AMTR was performed in seven (87.5%) out of the eight patients who underwent resective surgery; the resection of temporal pole was performed in one (12.5%) patient. In the group with uniTLE, all of the patients underwent resective surgery; AMTR was performed in 11 (84.6%) out of 13 patients, and extended lesionectomy was performed in the remaining two (15.4%) patients. Histopathological findings were available in eight (61.5%) patients in the group with biTLE and in all 13 patients in the group with uniTLE (Table 1). Regarding the outcome in the eight patients from the group with biTLE who underwent resective surgery, two (25%) patients were classed as Engel I (IA and IB), one (12.5%) patient was classed as Engel II (IIA), two (25%) patients were classed as Engel III (both IIIA), and three (37.5%) patients were classed as Engel IV. Seizure outcome in seven patients with biTLE after VNS implantation were classified, using the modified McHugh classification, as the following: I (one patient; 14.3%), II (one patient; 14.3%), III (one patient; 14.3%), and V (four patients; 57.1%). In the group with uniTLE, all of the patients were completely seizure-free after surgery, as defined in the inclusion criteria. 3.2. Differences between the group with biTLE and the group with uniTLE 3.2.1. Anamnestic data Anamnestic data describing the group with biTLE and the group with uniTLE are summarized in Table 2. No statistically significant differences were found between the group with biTLE and that with uniTLE. 3.2.2. Semi-invasive ictal EEG parameters A total of 93 seizures were recorded in 26 patients during semiinvasive EEG. A total of 35 seizures were analyzed in the group with biTLE (range = 2–6 per patient; average of 2.69 ± 1.18) and 58 seizures in the group with uniTLE (range = 2–9 per patient; average of 4.46 ± 2.30). Technically acceptable recordings of all evaluated ictal EEG parameters were available in 33 (94.3%) out of 35 seizures in the group with biTLE and in 52 (89.7%) out of 58 seizures in the group with uniTLE. We did not find any significant differences in terms of predefined semiinvasive ictal EEG parameters between the group with biTLE and the group with uniTLE (Table 3). 3.2.3. Seizure semiology A total of 164 seizures recorded during both semi-invasive EEG and invasive EEG monitoring were included in the seizure semiology analysis. In the group with biTLE, a total of 86 seizures were analyzed (range = 4–13 per patient; average of 6.62 ± 2.66), including 67 CPSs. Similarly, in the group with uniTLE, a total of 78 seizures were analyzed (range = 3–9 per patient; average of 6.00 ± 1.87), including 61 CPSs. At least one IMS was observed in 18 (20.7%) out of 87 seizures in the group with biTLE and in 47 (61.0%) out of 77 seizures in the group with uniTLE (p b 0.001). The individual IMS average per seizure was significantly lower in the group with biTLE, with a median of 0.14 (range = 0–1.0) than in the group with uniTLE, with a median of 0.80 (range = 0–2.6) (p = 0.003) (Table 4). The distribution of each particular lateralizing sign in both groups is shown in Table 5. Table 5 also provides information about the percentage of lateralizing signs that were concordant with SOZ (in the group with uniTLE) or with the predominant SOZ (in the biTLE group). Postictal unresponsiveness shorter than 3 min was observed in 18 (28.6%) out of 63 seizures in the group with biTLE and in 34 (57.6%) out of 59 seizures in the group with uniTLE. In contrast, PU was longer than 5 min in 27 (42.9%) out of 63 seizures in the group with biTLE and in 11 (18.6%) out of 59 seizures in the group with uniTLE. This different distribution of PU among seizures of both groups reached statistical significance (p = 0.002) (Table 6). In the group with biTLE, the invasive part of the evaluation revealed six patients with a predominant SOZ in the left (i.e., language-dominant) hemisphere, six patients with a predominant SOZ in the right (i.e., language nondominant) hemisphere, and one patient with a 1:1 ratio of seizures originating from both hemispheres. In the group with uniTLE, there were seven patients with a SOZ in the left (i.e., language-dominant) hemisphere and six patients with a SOZ in the right (i.e., language nondominant) hemisphere. In the group with biTLE, we analyzed a total of 63 seizures, from which 29 were registered in patients with a predominant SOZ in the left (i.e., languagedominant) side (46.0%) and 34 seizures in patients with a predominant SOZ in the right (i.e., language nondominant) side. In the group with uniTLE, we analyzed a total of 59 seizures, from which 35 were in patients with a predominant SOZ in the left (i.e., language-dominant) side (59.3%) and 24 seizures in patients with a predominant SOZ in the right (i.e., language nondominant) side. Other features of ictal and periictal semiology did not differ significantly between the two groups (Table 4). Table 3 Characteristics of biTLE and uniTLE semi-invasive ictal EEG data. ERTA — presence of early rhythmic theta/alpha activity; TRTA — time to rhythmic theta/alpha activity; BPT — bitemporal propagation time; IAD — ictal activity duration; NS — statistically nonsignificant. biTLE uniTLE p-Value ERTA/number of analyzed seizures 31/34 (91.2%) 52/53 (98.1%) NSa Individual TRTA average [min–max] (seconds) 8.0 [2.0–60.5] 4.3 [1.0–10.5] NSb Individual BPT average [min–max] (seconds) 13.5 [1.0–64.5] 10.5 [3.5–30.0] NSb Individual IAD average [min–max] (seconds) 80.0 [49.0–332.5] 74.5 [26.0–143.0] NSb a Fisher's exact test. b Mann–Whitney test. 43P. Řehulka et al. / Epilepsy & Behavior 41 (2014) 40–46 194 4. Discussion 4.1. Definition of bitemporal epilepsy Although the existence of independent bitemporal seizure onsets has been clearly demonstrated by invasive EEG recordings, the term “bitemporal epilepsy” differs substantially among published studies and is used in various clinical situations. Bitemporal epilepsy was defined in previous invasive studies as intractable TLE with “at least one independent seizure arising from each temporal lobe” [1,3–5], “predominance of seizure origin b80% in one from both temporal lobes” [19], or was defined by various combinations of criteria (scalp and sphenoidal EEG, MRI findings, and neuropsychological results) independently on invasive EEG results [20]. In other studies, the term “bitemporal epilepsy” was interpreted as clinical seizures arising independently from both temporal lobes in scalp EEG recordings [21] or bilateral independent seizure origin in scalp or invasive EEG recordings [22,23]. As a rule, patients with only interictal bitemporal epileptiform abnormalities were not declared as patients with bitemporal epilepsy [19,24–29]. Our definition of a group with bitemporal epilepsy is similar to that of Hirsch et al. [1], with the exception of the inclusion of patients with habitual complex partial seizures elicited by electrical stimulation. 4.2. Anamnestic data We did not observe any significant difference between the two groups with respect to epilepsy risk factors. In agreement with a previous study by Hirsch et al. [1], we did not notice any statistical difference between the group with biTLE and the group with uniTLE in terms of TSF, GTCS frequency, or anamnestic experience of aura. We did not evaluate other history data such as age at epilepsy onset, duration of epilepsy, or histopathological findings. These parameters do not differ statistically between bitemporal and unitemporal patients [1], although one study showed an earlier age at epilepsy onset in bitemporal patients [18]. 4.3. Semi-invasive EEG ictal parameters We did not find any significant differences between the group with biTLE and that with uniTLE in terms of predefined EEG variables. The high incidence of ERTA in both the group with biTLE and the group with uniTLE is a typical ictal EEG finding in patients with TLE [30]. Parameters describing time course of seizures (BPT and IAD) were not found to play any role in biTLE identification because both groups varied extremely, both interindividually and intraindividually. This supports the previously observed vast diversity in propagation models [31–35]. 4.4. Ictal and postictal semiology We systematically assessed the semiology of biTLE in comparison with uniTLE. Hirsch et al. [3] did not find significant differences in bitemporal epilepsy when analyzing the clinical uniformity of seizures originating in each temporal lobe. In contrast to some previous comprehensive semiologic studies in TLE [36,37], we analyzed the frequency of separate symptoms (EOAs and PU) as well as the frequency of predefined symptom clusters (IMSs, PIVSs, and PIMSs). The first main difference between the group with biTLE and the group with uniTLE was the lower frequency of IMSs in the course of the seizures in patients from the group with biTLE. The second significant difference was longer PU in the group with biTLE than in the group with uniTLE. One possible reason for the lack of IMSs in the majority of the seizures in the group with biTLE may be a more rapid contralateral seizure spread in biTLE than in uniTLE, which does not permit the evolution of typically lateralized TLE semiology. Although the short contralateral spread may be due to the presence of a more seizure-prone contralateral hippocampus [31], there may also be evidence of specific independent or connected networks in both temporal foci. Our results from scalp and sphenoidal EEG did not directly support these hypotheses. The hypotheses should be tested with a depth electrode study with a detailed analysis of the ictal involvement of the main structures of both temporal lobes, which was impossible in our study. To the best of our knowledge, the term “postictal unresponsiveness” has not been used and evaluated in CPSs in TLE to clinically describe the altered behavioral state in the meaning of postictal dysphasia and/or confusion. Privitera et al. [38] introduced the Cincinnati method of postictal language assesment, in which postictal confusion does not interfere or rarely interferes with postictal language testing [39]. The Table 4 Comparison of ictal and periictal semiology between the group with biTLE and that with uniTLE. EOAs — early oroalimentary automatisms; IMSs — ictal motor signs; PIVSs — periictal vegetative symptoms; PIMSs — periictal motor signs; NS — statistically nonsignificant. biTLE uniTLE p-Value Number of patients/seizures 13/87 13/77 EOAs in ≥1 seizure 9 (69.2%) 7 (53.8%) NSa EOAs in seizures 25 (28.7%) 18 (23.4%) NSa IMSs in ≥1 seizure 6 (46.2%) 12 (92.3%) p = 0.030a ≥1 IMS in seizures 18 (20.7%) 47 (61.0%) p b 0.001a Individual IMS average per seizure [min–max] 0.14 [0–1.0] 0.80 [0–2.6] p = 0.003b PIVSs in ≥1 seizure 4 (30.8%) 7 (53.8%) NSa ≥1 PIVS in seizures 13 (14.9%) 14 (18.2%) NSa PIMSs in ≥1 seizure 11 (91.7%)c 11 (84.6%) NSa ≥1 PIMS in seizures 40 (62.5%) 34 (57.6%) NSa a Fisher's exact test. b Mann–Whitney test. c Seizures of patient no. 9 were not analyzed [all registered seizures were secondarily generalized]. Table 5 The distribution of the occurrence among the four ictal motor signs by group and lateralizing value with respect to the predominant side of seizure-onset zone. IMSs — ictal motor signs; SOZ — localization of seizure-onset zone; ENHT — early nonversive head turning; ID — ictal dystonia; LII — lateralized ictal immobility of the upper limb; RINCH — rhythmic ictal nonclonic hand motions. ENHT was considered concordant if it occurred ipsilateral to the predominant SOZ; the other three IMSs (ID, LII, and RINCH) were considered concordant if they occurred contralateral to the predominant SOZ. biTLE uniTLE Total number of seizures where IMSs are present IMSs concordant with the predominant side of SOZ Total number of seizures where IMSs are present IMSs concordant with the side of SOZ ENHT 5 4/5 (80%) 31 27/31 (87.1%) ID 1 1/1 (100%) 16 16/16 (100%) LII 18 15/18 (83.3%) 30 28/30 (93.3%) RINCH 1 0/1 (0%) 4 4/4 (100%) Table 6 Postictal responsiveness in the group with biTLE and that with uniTLE. CPSs — complex partial seizures; PU — postictal unresponsiveness; NS — statistically nonsignificant. biTLE uniTLE p-Value Number of patients/seizures 12a /63 13/59 PU in all CPSs ≤ 5 min 3 (25.0) 9 (69.2) p = 0.047b PU in all CPSs N 5 min 9 (75.0) 4 (30.8) PU b 3 min 18 (28.6) 34 (57.6) p = 0.002b PU 3–5 min 18 (28.6) 14 (23.7) PU N 5 min 27 (42.9) 11 (18.6) a Seizures of patient no. 9 were not analyzed [all registered seizures were secondarily generalized]. b Fisher's exact test. 44 P. Řehulka et al. / Epilepsy & Behavior 41 (2014) 40–46 195 ability to correctly read a test phrase during the first 60 s after the end of ictal EEG activity lateralizes seizure origin to the nondominant hemisphere whether [38] or not it is propagated to the dominant temporal lobe [40]. When multiple task postictal language testing was used previously, there were no significant differences between left-sided seizures and right-sided seizures in the mean interval to the first correct verbal response [41]. The main limitation of multiple task postictal testing is that there is no single task presentation (and, consequently, no clearcut time of each task presentation). In the present study, we used multiple task testing, but we measured the interval to full recovery of normal contact without impaired speech and without confusion. Late phases of the postictal period were divided into three relatively long periods, with the first cutoff notably long after the sensitive period of the Cincinnati method (i.e., 3 min after the end of the seizure). This difference between the group with biTLE and the group with uniTLE is not caused directly by a different seizure duration (IAD) or by the distribution of the predominant SOZ side among both hemispheres in the group with uniTLE and the group with biTLE. Based on these results, we suppose that the difference in “postictal unresponsiveness” between the two groups was not caused by more seizures arising from the language-dominant side in patients in the group with biTLE, even though invasive recordings of all of the seizures were not used in this part of analysis. Further detailed study is needed. Postictal unresponsiveness longer than 3 min probably indicates temporal lobe seizures with origin or propagation in/to the dominant hemisphere and/or seizures with prolonged postictal confusion. The major proportion of patients and seizures with PU longer than 5 min in the group with biTLE may be explained by hypothetically more profound functional impairment of both temporal lobes in the group with biTLE. One may speculate that the effect of a seizure is more profound in the group with biTLE, because both temporal lobes are seizure-prone. 4.5. Surgery and outcome In our study, 61.5% of the patients in the group with biTLE underwent resective surgery, which is comparable with the literature. In previous series with invasive EEG, the percentage of evaluated bitemporal patients who had surgery was 43–53% [3,4,21]. Literature data concerning the postoperative outcome in patients with bitemporal epilepsy are difficult to interpret because outcome assessment differs among studies. Published data imply that 75–100% seizure reduction (not including auras) was achieved in 50.0–93.3% of operated patients [3,4,18,21], and completely seizure-free outcome was achieved in 12.5–45.5% of operated patients [3,5,18]. Favorable VNS efficacy in the treatment of bitemporal epilepsy was documented in a proportion of patients as comparable with other epilepsy syndromes [22,42]. Our results of resective surgery and VNS efficacy are limited because of the low patient number. One patient reached seizure freedom; VNS was effective (McHugh I or II) in about one-third of our patients. 5. Conclusions Our study analyzes differences in seizure semiology between bitemporal TLE and unilateral TLE. We demonstrated a higher frequency of ictal motor signs in patients in the group with uniTLE and longer postictal unresponsiveness in patients in the group with biTLE. The major limitation of this study was the small sample size. Further evaluation on the basis of invasive EEG is needed to confirm the results of our study. Acknowledgments We are grateful to Radka Kubíková for work on the Wada test data collection. We thank Anne Johnson for grammatical assistance. This study was supported by the “CEITEC — Central European Institute of Technology” project (CZ.1.05/1.1.00/02.0068) from the European Regional Development Fund. Conflict of interest 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] Hirsch LJ, Spencer SS, Williamson PD, Spencer DD, Mattson RH. Comparison of bitemporal and unitemporal epilepsy defined by depth electroencephalography. Ann Neurol 1991;30:340–6. [2] So N, Olivier A, Andermann F, Gloor P, Quesney LF. Results of surgical treatment in patients with bitemporal epileptiform abnormalities. Ann Neurol 1989;25:432–9. [3] Hirsch LJ, Spencer SS, Spencer DD, Williamson PD, Mattson RH. Temporal lobectomy in patients with bitemporal epilepsy defined by depth electroencephalography. Ann Neurol 1991;30:347–56. [4] Sirven JI, Malamut BL, Liporace JD, O'Connor MJ, Sperling MR. Outcome after temporal lobectomy in bilateral temporal lobe epilepsy. Ann Neurol 1997;42:873–8. [5] Boling W, Aghakhani Y, Andermann F, Sziklas V, Olivier A. Surgical treatment of independent bitemporal lobe epilepsy defined by invasive recordings. J Neurol Neurosurg Psychiatry 2009;80:533–8. [6] Sperling MR1, O'Connor MJ. Auras and subclinical seizures: characteristics and prognostic significance. Ann Neurol 1990;28:320–8. [7] Engel Jr J, Van Ness PC, Rasmussen TB. Outcome with respect to epileptic seizures. In: Engel J, editor. Surgical treatment of epilepsies. 2nd ed. New York: Raven Press; 1993. p. 609–21. [8] McHugh JC, Singh HW, Phillips J, Murphy K, Doherty CP, Delanty N. Outcome measurement after vagal nerve stimulation therapy: proposal of a new classification. Epilepsia 2007;48:375–8. [9] Blümcke I, Thom M, Aronica E, Armstrong DD, Vinters HV, Palmini A. The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 2011; 52:158–74. [10] Wyler AR, Dohan FC, Schweitzer JB, Berry III AD. A grading system for mesial temporal pathology (hippocampal sclerosis) from anterior temporal lobectomy. J Epilepsy 1992;5:220–5. [11] Abou-Khalil B, Fakhoury T. Significance of head turn sequences in temporal lobe onset seizures. Epilepsy Res 1996;23:245–50. [12] Kuba R, Krízová J, Brázdil M, Tyrlíková I, Rektor I. Lateralized ictal immobility of the upper limb in patients with temporal lobe epilepsy. Eur J Neurol 2005;12:886–90. [13] Kotagal P, Lüders H, Morris HH, Dinner DS, Wyllie E, Godoy J, et al. Dystonic posturing in complex partial seizures of temporal lobe onset: a new lateralizing sign. Neurology 1989;39:196–201. [14] Lee GR, Arain A, Lim N, Lagrange A, Singh P, Abou-Khalil B. Rhythmic ictal nonclonic hand (RINCH) motions: a distinct contralateral sign in temporal lobe epilepsy. Epilepsia 2006;47:2189–92. [15] Musilová K, Kuba R, Brázdil M, Tyrlíková I, Rektor I. Occurrence and lateralizing value of “rare” peri-ictal vegetative symptoms in temporal lobe epilepsy. Epilepsy Behav 2010;19:372–5. [16] Leutmezer F, Serles W, Lehrner J, Pataraia E, Zeiler K, Baumgartner C. Postictal nose wiping: a lateralizing sign in temporal lobe complex partial seizures. Neurology 1998;51:1175–7. [17] Musilová K, Kuba R, Brázdil M, Tyrlíková I, Rektor I. Peri-ictal bed leaving in temporal lobe epilepsy: incidence and lateralizing value. Epilepsy Behav 2011;21:143–6. [18] Hufnagel A, Elger CE, Pels H, Zentner J, Wolf HK, Schramm J, et al. Prognostic significance of ictal and interictal epileptiform activity in temporal lobe epilepsy. Epilepsia 1994;35:1146–53. [19] So N, Gloor P, Quesney LF, Jones-Gotman M, Olivier A, Andermann F. Depth electrode investigations in patients with bitemporal epileptiform abnormalities. Ann Neurol 1989;25:423–31. [20] Spanedda F, Cendes F, Gotman J. Relations between EEG seizure morphology, interhemispheric spread, and mesial temporal atrophy in bitemporal epilepsy. Epilepsia 1997;38:1300–14. [21] Holmes MD, Miles AN, Dodrill CB, Ojemann GA, Wilensky AJ. Identifying potential surgical candidates in patients with evidence of bitemporal epilepsy. Epilepsia 2003;44:1075–9. [22] Alsaadi TM, Laxer KD, Barbaro NM, Marks Jr WJ, Garcia PA. Vagus nerve stimulation for the treatment of bilateral independent temporal lobe epilepsy. Epilepsia 2001; 42:954–6. [23] Spencer D, Gwinn R, Salinsky M, O'Malley JP. Laterality and temporal distribution of seizures in patients with bitemporal independent seizures during a trial of responsive neurostimulation. Epilepsy Res 2011;93:221–5. [24] Chung MY, Walczak TS, Lewis DV, Dawson DV, Radtke R. Temporal lobectomy and independent bitemporal interictal activity: what degree of lateralization is sufficient? Epilepsia 1991;32:195–201. 45P. Řehulka et al. / Epilepsy & Behavior 41 (2014) 40–46 196 [25] Steinhoff BJ, So NK, Lim S, Lüders HO. Ictal scalp EEG in temporal lobe epilepsy with unitemporal versus bitemporal interictal epileptiform discharges. Neurology 1995; 45:889–96. [26] Benbadis SR, So NK, Antar MA, Barnett GH, Morris HH. The value of PET scan (and MRI and Wada test) in patients with bitemporal epileptiform abnormalities. Arch Neurol 1995;52:1062–8. [27] Holmes MD, Dodrill CB, Ojemann GA, Wilensky AJ, Ojemann LM. Outcome following surgery in patients with bitemporal interictal epileptiform patterns. Neurology 1997;48:1037–40. [28] Serles W, Pataraia E, Bacher J, Olbrich A, Aull S, Lehrner J, et al. Clinical seizure lateralization in mesial temporal lobe epilepsy: differences between patients with unitemporal and bitemporal interictal spikes. Neurology 1998;50:742–7. [29] Ergene E, Shih JJ, Blum DE, So NK. Frequency of bitemporal independent interictal epileptiform discharges in temporal lobe epilepsy. Epilepsia 2000;41:213–8. [30] Williamson PD, French JA, Thadani VM, Kim JH, Novelly RA, Spencer SS, et al. Characteristics of medial temporal lobe epilepsy: II. Interictal and ictal scalp electroencephalography, neuropsychological testing, neuroimaging, surgical results, and pathology. Ann Neurol 1993;34:781–7. [31] Lieb JP, Engel Jr J, Babb TL. Interhemispheric propagation time of human hippocampal seizures. I. Relationship to surgical outcome. Epilepsia 1986;27:286–93. [32] Spencer SS, Williamson PD, Spencer DD, Mattson RH. Human hippocampal seizure spread studied by depth and subdural recording: the hippocampal commissure. Epilepsia 1987;28:479–89. [33] Spencer SS, Spencer DD, Williamson PD, Mattson R. Combined depth and subdural electrode investigation in uncontrolled epilepsy. Neurology 1990;40:74–9. [34] Spencer SS, Marks D, Katz A, Kim J, Spencer DD. Anatomic correlates of interhippocampal seizure propagation time. Epilepsia 1992;33:862–73. [35] Adam C, Saint-Hilaire JM, Richer F. Temporal and spatial characteristics of intracerebral seizure propagation: predictive value in surgery for temporal lobe epilepsy. Epilepsia 1994;35:1065–72. [36] Fakhoury T, Abou-Khalil B, Peguero E. Differentiating clinical features of right and left temporal lobe seizures. Epilepsia 1994;35:1038–44. [37] Kotagal P, Lüders HO, Williams G, Nichols TR, McPherson J. Psychomotor seizures of temporal lobe onset: analysis of symptom clusters and sequences. Epilepsy Res 1995;20:49–67. [38] Privitera MD, Morris GL, Gilliam F. Postictal language assessment and lateralization of complex partial seizures. Ann Neurol 1991;30:391–6. [39] Privitera M, Kim KK. Postictal language function. Epilepsy Behav 2010;19:140–5. [40] Ficker DM, Shukla R, Privitera MD. Postictal language dysfunction in complex partial seizures: effect of contralateral ictal spread. Neurology 2001;56:1590–2. [41] Devinsky O, Kelley K, Yacubian EM, Sato S, Kufta CV, Theodore WH, et al. Postictal behavior. A clinical and subdural electroencephalographic study. Arch Neurol 1994;51:254–9. [42] Kuba R, Brázdil M, Novák Z, Chrastina J, Rektor I. Effect of vagal nerve stimulation on patients with bitemporal epilepsy. Eur J Neurol 2003;10:91–4. 46 P. Řehulka et al. / Epilepsy & Behavior 41 (2014) 40–46 197 Annex 12: Chrastina J, Novak Z, Zeman T, Kocvarova J, Pail M, Dolezalova I, Jarkovsky J, Brazdil M. Single-center long-term results of vagus nerve stimulation for epilepsy: A 10-17 year follow-up study. Seizure – European Journal of Epilepsy. 2018;59:41-47. 198 Single-center long-term results of vagus nerve stimulation for epilepsy: A 10–17 year follow-up study Jan Chrastinaa, *, Zdenek Nováka , Tomáš Zemana , Jitka Ko9cvarováb , Martin Pailb , Irena Doležalováb , Jirí Jarkovskýc , Milan Brázdilb,d a Department of Neurosurgery, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic b Brno Epilepsy Center, Departments of Neurology and Neurosurgery, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic c Institute of Biostatistics and Analyses, Masaryk University Medical Faculty, Brno, Czech Republic d Behavioral and Social Neuroscience Research Group, CEITEC – Central European Institute of Technology, Masaryk University, Brno, Czech Republic A R T I C L E I N F O Article history: Received 27 January 2018 Received in revised form 12 March 2018 Accepted 26 April 2018 Keywords: Epilepsy Vagus nerve stimulation Long term follow up Responder Battery replacement A B S T R A C T Purpose: The paper presents a long-term follow-up study of VNS patients, analyzing seizure outcome, medication changes, and surgical problems. Method: 74 adults with VNS for 10 to 17 years were evaluated yearly as: non-responder – NR (seizure frequency reduction <50%), responder – R (reduction  50% and <90%), and 90% responder – 90R (reduction  90%). Delayed R or 90R ( 4 years after surgery), patients with antiepileptic medication changes and battery or complete system replacement were identified. Statistical analysis of potential outcome predictors (age, seizure duration, MRI, seizure type) was performed. Results: The rates of R and 90R related to the patients with outcome data available for the study years 1, 2, 10, and 17 were for R 38.4%, 51.4%, 63.6%, and 77.8%, and for 90R 1.4%, 5.6%, 15.1%, and 11.1%. The absolute numbers of R and 90R increased until years 2 and 6. Antiepileptic therapy was changed in 62 patients (87.9%). There were 11 delayed R and four delayed 90R, with medication changes in the majority. At least one battery replacement was performed in 51 patients (68.9%), 49 of whom R or 90R. VNS system was completely replaced in 7 patients (9.5%) and explanted in 7 NR (9.5%). No significant predictor of VNS outcome was found. Conclusions: After an initial increase, the rate of R and 90R remains stable in long-term follow-up. The changes of antiepileptic treatment in most patients potentially influence the outcome. Battery replacements or malfunctioning system exchange reflect the patient’s satisfaction and correlate with good outcomes. © 2018 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. 1. Introduction Sincethe first implantation inhumans in 1988 and Foodand Drug Administration (FDA) approval in 1997, vagus nerve stimulation (VNS) has become an accepted palliative treatment modality for patients with drug-resistant epilepsy not suitable for resective surgery. The effect of VNS on seizure frequency and severity was confirmed by randomized controlled trials [1,2,3,4]. However the follow-up period in the first three studies did not exceed 26 weeks [1,2,3]. The last paper, a metanalysis of 74 studies with 3321 enrolled patients, proved a significant reduction of seizures at 3–12 months after surgery (36%) and an increasing effect of VNS on seizure reduction at >1year after surgery [4]. Other studies confirm a cumulative effect of VNS in medium-duration follow-up. For example, the median seizure reduction in 454 patients enrolled in five double-blind US studies improved from 35% to 44% at two years [5]. Similarly, a European study confirmed that treatment duration wassignificantlycorrelatedwiththepercentageofseizurefrequency reduction [6]. With increased experience, studies reporting postVNS outcomes for up to 5 years [7,8] and studies covering follow-up periods exceeding 10–11 years have been published [9,10,11,12]. The effectofVNSonseizurereductionhasbeendiscussedincombination with other aspects of VNS: post-VNS quality of life [12] and surgical problems [9]. Abbreviations: FDA, food and drug administration; VNS, vagus nerve stimulation; ILAE, international league against epilepsy; MRI, magnetic resonance imaging; NR, non-responder; R, responder; 90R, 90% responder; DBS, deep brain stimulation. * Corresponding author at: Department of Neurosurgery, Masaryk University Medical Faculty, St. Anne’s Hospital Brno Pekarská 53, 656 91, Brno, Czech Republic. E-mail addresses: jan.chrastina@fnusa.cz (J. Chrastina), zdenek.novak@fnusa.cz (Z. Novák), tomas.zeman@fnusa.cz (T. Zeman), jitka.krizova@fnusa.cz (J. Ko9cvarová) , martin.pail@fnusa.cz (M. Pail), irena.dolezalova@fnusa.cz (I. Doležalová), jarkovsky@iba.muni.cz (J. Jarkovský), milan.brazdil@fnusa.cz (M. Brázdil). https://doi.org/10.1016/j.seizure.2018.04.022 1059-1311/© 2018 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. Seizure 59 (2018) 41–47 Contents lists available at ScienceDirect Seizure journal homepage: www.elsevier.com/locate/yseiz 199 The first aim of the study is to present the results of VNS systems implanted for  10 years in patients with drug-resistant epilepsy followed in a specialized epilepsy center in a longitudinal time axis. The effect of VNS on seizure reduction is quantified yearly in terms of responder rates for the entire study period from stimulation onset to study completion. There were marked developments in the field of antiepileptic drugs during the study period. In terms of new medications during the study period (January 2000–November 2017), the date of levetiracetam registration was 29 Sept 2000, for zonisamide 10 Mar 2005 and for pregabalin 6 July 2004. The impact of medication changes during the follow-up period on seizure outcome is also studied, particularly in patients with late or delayed VNS response. The satisfaction with VNS treatment can be quantified by sophisticated scales evaluating the different aspects of the quality of life [13], but a simple criterion – the frequency of battery replacement after depletion (indicating the patient’s or the physician’s will to continue VNS therapy in mutual agreement) – can provide a simplified answer. Moreover, surgical problems with the implanted system require revision and replacement and the rate of patients willing to undergo the surgery to continue the VNS therapy during the prolonged follow-up period also provide data about patient satisfaction with the results. Therefore the second aim of the study is the analysis of surgical problems of longterm VNS, with attention to reoperations and system complications in the long-term follow-up period, for which there are only a few comparable papers [10,12]. Finally statistical analysis of potential long- term VNS outcome predictors was performed; unfortunately there are currently no universally accepted outcome predictors for VNS. However, based on previously published papers aiming to find VNS outcome predictors, age, seizure duration, age at seizure onset, MRI findings (diffuse, focal, mesiotemporal sclerosis and negative) and the prevailing seizure type (complex partial seizure – focal aware – ILAE 2017, simple partial seizure – focal impaired awareness – ILAE 2017, and other), were selected as potential predictors for statistical analysis [4,14–16]. 2. Material and methods Adult patients (age  18 years) with VNS systems (Cyberonics, Inc., Houston, Texas, USA) implanted at the author’s department for  10 years ago were retrospectively identified from a prospectively constructed database of the patients who had surgery for drug-resistant epilepsy at the comprehensive Brno Epilepsy Center. Prior to VNS implantation, all patients underwent a detailed preoperative investigation at the center and were ruled out as suitable candidates for resective epilepsy surgery. After standard implantation of the left vagus nerve stimulation system (ZN, JC), all the patients were followed at the First Department of Neurology at regular intervals (2, 4, 6, 12 and 18 months; then yearly). VNS parameter adjustments and modifications of antiepileptic medication were made strictly according to the clinical decision of the epileptologist. Based onpost-VNS seizure reduction, the patients were graded at the follow-up visits as non-responders – NR (seizure frequency reduction<50%), responders–R(reduction  50%and<90%),or90% responders – 90R (reduction  90%). When the VNS system was switched off due to a response that was not clinically relevant, the patient was moved to the “stimulation off” category and remained there until the end of the follow-up period; other patients were categorized as having had system explantation (“explantation”), as lost to follow-upcare (“patientlost”),orashavingdiedforanyreason (“dead”). Missing data for a single follow-up visit moved the patient to the “data missing” category for that particular follow-up point. Patients with late response to VNS (shifting from the NR category to the R category, or from the R to the 90R category at  4 years after surgery), as well as patients with fluctuating or worsening response to VNSduringthe follow-upperiod,were identified.The percentages of R, 90R, and Good outcomes (R + 90R) for each study year were related to all the patients with follow-up results available for the defined study year (excluding patients categorized as “dead”, “data missing”, or “lost”). The medical reports of late responders were checked for medication changes during the year prior to the improved response. The percentage of patients with medication changes (including permanent dosage changes) during the follow-up period was also calculated. Patients with battery replacement during the follow-up period were identified from the database and confirmed from surgical reports. Similarly, patients requiring complete VNS system replacement (including the helical electrode) were identified, and the cause of system failure was determined from the surgical report. The rates of 90R and Good outcomes were selected as the endpoints of the statistical analysis. The time points for analysis were defined at year 10 and final follow up. Absolute and relative frequencies were accepted for the description of endpoints occurrence. Logistic regression was used for the analysis of relation between predictors and endpoints; odds ratios, their 95% confidence intervals and statistical significance were used for the description of this relationship. 3. Results The study included 74 adult patients (33 males, 41 females) who had VNS implanted between January 2000 and November 2007. The mean patient age at implantation was 31.1 years (range 18–59 years; standard deviation 10.65 years). There was a history of previous neurosurgical operations in 20 patients (simple lesionectomies in six patients, failed extratemporal or temporal resections in four patients, stereotactic lesional surgeries in seven patients, stereoelectroencephalography not providing adequate data for resective surgery in three patients). Two patients had brain hypothermia. The first case was a 35-year-old male with seizure onset at the age of 2 years (frontal absences with secondary generalization). Hypothermic treatment was administered at the age of 17 years, but without effect on seizure frequency or severity. MRI showed gliotic changes after ventricular punctures. Because the investigations failed to localize the epileptogenic focus, the patient had VNS system implanted, and was categorized as a Responder. The second patient (a 37-year-old male) was treated at another department for multifocal epileptic seizures by stereotactic surgeries (coagulation of the right Forel field, rostral cingulum, and right thalamic nuclei) and with brain hypothermia. After VNS, he was graded as 90R. Six patients were lost to follow-up before study year 10 (one 90R and five R at the last follow-up). Two patients died before study year 10 (one NR suffered a severe brain injury after epileptic seizure and one NR due to malignant retroperitoneal tumor). The outcome data at the year 10 follow-up visits were available for 66 patients. Follow-up data of patients with VNS implanted for 17 years were available for nine patients (two patients were lost to follow up; one patient died from neurodegenerative disease). During the whole study period, there were 8 patients lost to follow-up care (one 90R and 7 R at the last follow-up). Five patients died during the follow-up period. In three patients, the cause of death was unrelated to epilepsy (retroperitoneal malignancy, urinary bladder cancer, and neurodegenerative disease). In one patient, the death was related to epilepsy: a fatal brain injury caused by a fall during an epileptic seizure. The possibility of SUDEP could not be excluded in one patient who reportedly died from heart failure. 42 J. Chrastina et al. / Seizure 59 (2018) 41–47 200 General overviews of the results are presented in Table 1 (follow-up years 1–9) and Table 2 (follow-up years 10–17) and Graphs 1 and 2. The rates of R, 90R and Good outcomes can be related to the complete group of patients (including patients with no available data because they were lost to follow-up, dead, or had data missing). This calculation underestimates the percentage of good results, because it in fact presumes that all patients with unavailable results had bad outcomes. The second possibility is that the rates of R, 90R, and Good outcomes are related to the patients with available follow-up data. This calculation overestimates the percentage of good results when the data unavailability is caused by the dropout of patients who are not doing well. Seven of the patients who were lost to follow-up were categorized as R and one as 90R before their loss. Of the five patients who died during the study period, two patients were graded before their death as 90R, one as R, and two as NR. Three of the patients categorized as “data missing” at a certain follow-up point were graded as NR at the next follow-up; two of them finally reached R grading. The other two patients categorized as “data missing” at a certain visit were evaluated as R and 90R at the next follow-up. Because good results were documented in the vast majority of the patients who were lost to follow-up care, dead, or had data missing, the rates of R and 90R were related to the group of patients with follow-up data available (excluding “lost”, “dead”, or “data missing”); in author’s opinion, this minimizes the distortion of results. The probable reason that patients with good outcomes dropped out was their choice to continue follow-up care in their native country or closer to their home (our center was the first in the country to start systematic VNS implantation). From the 28 patients evaluated as VNS responders at the oneyear follow-up visit, the number of responders increased to 37 at the two-year follow-up visit and remained stable with minimal fluctuations (36–42) until study year 10. The number of 90R increased until study year 6. This trend is also reflected in the dynamics of Good outcomes (R + 90R): the peak number was reached at study year 6 and then remained stable with minimal fluctuations (49–52). The nearly steady number of R from study year 2 can be explained by the dynamic equilibrium of patient inflow from the NR to R category together with the outflow from R to 90R and by the dropout of Responders. The 90R rate increases from 1.4% at year 1 to 19.1% at study year 6, then slightly decreases due to the dropout of two 90R patients reaching 15.4% at study year 10. The R and Good outcome rates increase from 38.4% and 39.8% at study year 1, to 51.4% and 57.0% for study year 2, and to 63.1% and 78.7% at study year 10. Because the number of study patients decreased during years 10 to 17, only the rates of R, 90R and Good outcomes were analyzed. The 90R rates fluctuate between 11.1% (study year 17; only 9 patients available for analysis) and 20% (study year 15). The R rate remains relatively stable between years 10 and 16 (56.1% to 63.1%) with an R rate of 77.8% at study year 17. The Good outcomes rate varies between 70.8% and 81.8% for study years 10–16 and reaches 88.9% for study year 17. Twelve patients were graded as 90R at their final follow-up visit (including one patient lost to subsequent follow-up care and one patient who died of neurodegenerative disease). A seizure-free period lasting at least one year prior to the final follow-up was achieved in two patients (one patient had a 13 year-seizure-free period after three years evaluated as NR; one patient was 9 years seizure free after a one-year follow-up evaluation as NR). Table 1 VNS outcomes – years 1–9. Study Year 1 2 3 4 5 6 7 8 9 NR 44 30 28 21 16 11 10 9 6 R 28 37 37 36 36 37 38 38 41 90R 1 4 7 10 12 13 11 12 10 Stimulation Off 0 0 1 2 3 3 4 4 6 Explantation 0 1 1 1 3 4 4 4 4 Patient Lost 0 0 0 2 3 5 5 6 6 Dead 0 0 0 0 0 1 1 1 1 Data Missing 1 2 0 2 1 0 1 0 0 Together 74 74 74 74 74 74 74 74 74 Good results (R + 90R) 29 41 44 46 48 50 49 50 51 Bad results (NR+ Stimulation Off + Explantation) 44 31 30 24 22 18 18 17 16 R rate (%) 38.4 51.4 50.0 51.4 51.4 54.4 56.7 56.7 61.2 90R rate (%) 1.4 5.6 9.5 14.3 17.1 19.1 16.4 17.8 14.7 R + 90R rate (%) 39.8 57.0 59.5 65.7 68.5 73.5 73.1 74.5 75.9 Table 2 VNS outcomes – years 10–17. Study year 10 11 12 13 14 15 16 17 NR 3 3 1 1 0 0 0 0 R 42 34 28 23 17 15 10 7 90R 10 10 9 6 4 5 3 1 Stimulation Off 6 5 4 4 3 1 1 0 Explantation 5 5 7 7 5 4 2 1 Patient Lost 6 3 3 3 3 2 2 2 Dead 2 2 2 4 4 4 2 1 Data Missing 0 0 0 0 0 0 0 0 Together 74 62 54 48 36 31 20 12 Good results (R + 90R) 52 44 37 29 21 20 13 8 Bad results (NR+ Stimulation Off + Explantation) 14 13 12 12 8 5 3 1 R rate (%) 63.6 59.6 60.8 56.1 58.6 60.0 63.1 77.8 90R rate (%) 15.1 17.5 18.3 14.6 13.9 20.0 18.7 11.1 R + 90R rate (%) 78.7 77.1 75.5 70.7 72.5 80.0 81.8 88.9 J. Chrastina et al. / Seizure 59 (2018) 41–47 43 201 During the study period, previously switched off VNS systems were explanted in seven NR patients because of further evaluation for resective surgery, deep brain stimulation (DBS), or at the patient’s request. The system was completely removed in six patients. In one patient the battery was removed for cosmetic reasons and the helical electrodes around the nerve were intentionally left at the patient’s request. Antiepileptic therapy remained unchanged during the study period in nine patients (12.1%). 11 patients (14.8%) achieved an R grade not earlier than at the year 4 follow-up visit. During the last year before their favorable response, there were medication changes in nine of them (Table 3); six had increased their medication dose. Four patients improved from R to 90R not earlier than at the year 4 follow-up visit. Three of Graph 1. The dynamics of R and 90R rate in time – study years 1–9. Graph 2. The dynamics of R and 90R rate in time – study years 10–17. 44 J. Chrastina et al. / Seizure 59 (2018) 41–47 202 them had changes in antiepileptic treatment in the previous year (Table 4). Three patients with fluctuating response to VNS were identified. One patient was initially graded as R; at the years 3 and 4 visits, she deteriorated to the NR category, but during all the following visits she was graded as R. Another patient was graded as R during the entire follow-up period except for at the years 5 and 6 follow-up visits, where she was graded as NR. One patient with a fluctuating response deteriorated to NR after the year 5 follow-up visit (R year 1, NR years 2 and 3, R years 4 and 5, NR after year 5). During the study period, one or more battery replacements were performed due to the depleted battery in 51 patients (68.9%). At the time of battery replacement, 49 patients (66.2%) were graded as R or 90R. Battery replacement was also indicated in two NR patients; the seizure reduction rate was 30–40% in both with positive effects from extrastimulation. In 11 patients, the battery was replaced twice; in one patient, it was replaced three times. A complete system replacement was necessary for seven patients (categorized as R before system problems) (9.5%); in one of them, the entire system had to be replaced twice. In all patients, the indication for system replacement was excessive system impedance with decreasing stimulation efficacy. During surgical revision, only fibrosis around the helical electrodes and nerve was found in all patients. There was no case of electrode violation or helical electrode dislocation. Further surgeries in NR patients were performed after VNS system explantation; these included DBS (bilateral anterior thalamic nucleus) in two patients and resective surgeries in two patients (anteromedial temporal resection initially refused by the patient and frontal topectomy after stereoelectroencephalography). In all the patients operated on after VNS explantation, the results of postexplantation MRI did not differ from the preVNS findings. Among the remaining nonresponders who underwent VNS explantation, there was no case with a resectable lesion detected even using high field MRI. The rate of VNS system infection and wound breakdown was 0%. There was one case of seroma and another case of hematoma around the battery, both conservatively treated without further consequences. Among the parameters potentially predicting long term VNS outcome no significant predictor of VNS outcomes as measured by the frequencies of endpoints (90R and Good outcomes) was found using the selected statistical methodology. 4. Discussion The presented study has a long follow-up period, lasting from 10 to 17 years. The results confirm that the benefit to patients with VNS from systematic treatment in a specialized epilepsy center increases over time. When comparing our results with a long-term study by Wasade et al. [12] with 207 patients, the overall results regarding seizure control are comparable (>50% seizure reduction in 68% of patients) with the exception of higher seizure freedom rates in the Wasade study (20%). In the Wasade study, 22 patients were surveyed from a group of 36 patients with VNS durations of 10 to 14 years (13 with favorable outcomes), and 11 patients were surveyed from a group of 24 patients with follow-up care lasting over 15 years (seven favorable outcomes). Our study had a higher rate of patients available for analysis. The steady rise of good outcome rates (R + 90R) after study year 2 together with the percentage of patients with late response to VNS (>4 years postimplantation) support prolonging the waiting period before evaluating VNS as ineffective. The adequate duration of this period is unclear [8]. Salinsky et al. [17] suggested that patient response during the first three months could foretell subsequent treatment effectiveness. However, Schachter et al. [18] encouraged continuing VNS for up to two years before discontinuation due to inefficacy. A long-term increase in VNS responder rates was supported by the extensive multicenter study published by Kuba et al. [19] with responder rates of 44.4% at one year after stimulation was initiated increasing to 58.7% after two years and 64.4% at five years. In a paper by Uthman et al. [20], the decrease of seizure frequency was 26% after one year, 30% after five years, and 52% after 12 years of VNS. Ryzí et al. [21] assessed long-term seizure outcomes in a group of 15 children with the mean seizure reduction of 42.5% at one year, 54.9% at two years, and 58.3% at five years. The responder rates remained stable at study year 2, at 60%, and at study year 5, at 60%. Serdaroglu et al. [11] confirmed that, once achieved, positive VNS results are stable or improve over time. In our study, only one patient deteriorated to NR after five years of follow-up observation with a fluctuating response. Despite Table 3 VNS patients with delayed R rate. Patient number Year of significant response Medication change Type of change 1 4 no change no change 2 8 yes increased lacosamide dose 3 4 yes increased topiramate dose 4 6 yes increased zonisamide dose 5 4 no change no change 6 7 yes increased levetiracetam dose 7 4 yes stopped levetiracetam, started lamotrigine 8 6 yes increased pregabalin dose 9 10 yes clonazepam added 10 4 yes stopped primidone, started topiramate 11 4 yes increased levetiracetam dose Table 4 VNS patients with delayed 90R rate. Patient number Year of 90R Grading before 90R Medication change Type of medication change 1 4 R no change no change 2 4 R yes carbamazepine dose reduction, increase of VNS stimulation 3 4 R yes stopped pregabalin, started clonazepam 4 5 R yes added clonazepam and lamotrigine J. Chrastina et al. / Seizure 59 (2018) 41–47 45 203 the difficult calculations of the R rate and 90R rate considering the drop of patients, the results are comparable with the literature data both for study period years 1–10 and years 11–17. During the study period, antiepileptic treatment was changed in 87.9% and remained unchanged in 12.1% of patients. The high percentage of patients with antiepileptic medication changes was undoubtedly related to the boom in this field. In a paper by Arcand et al. [22], the percentages of VNS patients with medication changes in type or dose during follow-up care were 57% at 6 months, 33% at 12 months, 59% at 24 months, and 81% at 36 months (mainly dose increases). The percentage of responders did not match the increased number of patients with medication changes – 43% at 6 months, 48% at 12 months, 41% at 24 months, and 50% at 36 months. A detailed analysis proved that maximum improvements correlated with the time of changed medication. These results correlate with our findings about medication changes in patients with late VNS response: in the year prior to the delayed VNS response improvement, there was an antiepileptic drug dose increase in 54.5% and a medication change in 27.2% of delayed responders. Medication was changed in 50% of the delayed 90R patients. Not all papers confirm the positive role of medication adjustments in VNS outcome. In a paper by García-Pallero et al. [23], the percentage of responders in the group with no medication changes allowed was 63%; in the comparable group of patients with medication changes permitted, 45.2% were responders. According to the authors, the absence of changes in antiepileptic drugs helps to optimize the stimulation parameters. The most frequent reason for revision surgery of the VNS system in our group was battery replacement due to battery depletion. In a study of 1234 patients, Lam et al. [24] reported the average incidence of revision surgeries within six years of followup care as <1% for electrode revision, <3% for battery revision or removal, 4–10% for battery replacement, and <1% for infection washout, confirming the highest frequency for battery replacements during a shorter follow-up period. In a study by Couch et al. [25] describing 1144 VNS procedures, 46% of patients required at least one battery replacement or revision surgery, mostly for battery depletion (27%), poor seizure control efficacy (9%), lead malfunction (8%), and infection (2%). In our patients, the indications for battery replacement were decreasing or depleted battery capacity. Attention should be paid to the timing of system revision or battery replacement. According to Vonck et al. [26], prereplacement seizure control could not be regained in 2 of 14 patients with replacement postponed for several months. Similarly, Tatum et al. [27] support battery replacement before the end of battery life because of the symptoms preceding the end of battery life, such as seizure and behavioral worsening. In our study, no case was observed of the loss of the pre-replacement seizure control after battery depletion and replacement. The indication for system replacement in our study was increased system impedance with the clinical correlation of seizure worsening; there were no cases of infection or system violation. During surgery, fibrosis around the nerve portion with helical electrodes was found in all patients. No case of electrode displacement was observed. The 9.5% incidence of complete system replacement correlates with literature data reporting the incidence of device malfunction at 4–16.8% of implanted systems [28]. Apart from standard surgical problems of battery or complete system replacement, other aspects of VNS revision surgeries should be discussed. The effect of VNS therapy may be measured by seizure reduction, but also by patient satisfaction and the impact of VNS on the patient’s quality of life. For example, Ryvlin et al. [13] proved that VNS therapy as a treatment adjunct to best medical practices in patients with pharmacoresistant focal seizures was associated with a significant improvement in health-related quality of life compared with best medical practices alone. In addition to sophisticated scales, the willingness of the patient (or the physician) to continue therapy by means of undergoing a relatively simple surgery (battery replacement) or a more difficult intervention, such as complete VNS system replacement, may be considered an indicator of treatment success. Based on this simple criterion, the VNS success rate is 78.5% (battery replacement rate 68.9% and complete system replacement 9.5%). This figure correlates with a study reporting that 80% of the surveyed patients considered the VNS worthwhile [12]. Although the incidence of infection was 0%, the increasing risk of implant complications during the prolonged follow-up treatment cannot be excluded despite the smaller battery and cable size as compared to currently available DBS hardware. This assumption is supported by a study of 247 VNS patients with a mean follow-up period of 12 years, in which the incidence of complications was 8.6%, including postoperative hematoma (1.9%) and infection (2.6%) [10]. 5. Conclusions The paper provides an exceptionally long-term follow-up view of VNS patients, involving seizure outcome and other factors. Longterm follow-up results indicate that vagus nerve stimulation is a safe and effective palliative treatment option for drug-resistant epilepsy and its efficacy does not decrease with time. After an initial steady increase until year 6 the number of patients with good VNS outcomes (R + 90R) remained stable. Changes in antiepileptic treatment took place during the entire follow-up period in 87.9% of patients, with potential impacts on seizure reduction. In most patients with delayed response to VNS the improved seizure control was preceded by a change in antiepileptic treatment during the previous year. Patient’s age, seizure duration, age at seizure onset, MRI findings and the prevailing seizure type are not significant predictor of long term VNS outcomes.The patient’s or treating physician’s satisfaction and willingness to continue VNS treatment is indicated by the rate of depleted battery replacement and complete system replacement and corresponds with the rate of good VNS outcomes. In terms of surgical complications in this long-term follow-up study, VNS proved to be a safe technique with a very low rate of transient treatable problems. Conflict of interest The authors have no conflict of interest to declare. Acknowledgements This research did not receive any specific grant from funding agencies in the public, commercial, or not-for- profit sectors References [1] The Vagus Nerve Stimulation Study Group. A randomized controlled trial of chronic vagus nerve stimulation for treatment of medically intractable seizures. Neurology 1995;45:224–30. [2] Handforth A, DeGiorgio CM, Schachter SC, Uthman BM, Naritoku DK, Tecoma ES, et al. Vagus nerve stimulation therapy for partial onset seizures: a randomized active-control trial. Neurology 1998;51:48–55. [3] Ben-Menachem E, Mañon-Espaillat R, Ristanovic R, Wilder BJ, Stefan H, Mirza W, et al. Vagus nerve stimulation for treatment of partial seizures: 1: a controlled study of effect on seizures. First International Vagus Nerve Study Group. Epilepsia 1994;35:616–26. [4] Englot DJ, Chang EF, Auguste KI. Vagus nerve stimulation for epilepsy: a metaanalysis of efficacy and predictors of response. J Neurosurg 2011;115:1248–55. [5] Morris 3rd GL, Mueller WM. Long-term treatment with vagus nerve stimulation in patients with refractory epilepsy. The Vagus Nerve Stimulation Study Group E01-E05. Neurology 1999;53:1731–5. 46 J. Chrastina et al. / Seizure 59 (2018) 41–47 204 [6] Scherrmann J, Hoppe C, Kral T, Wilder BJ, Stefan H, Mirza W. Vagus nerve stimulation: clinical experience in a large patient series. J Clin Neurophysiol 2001;18:408–14. [7] García-Navarette E, Torres CV, Gallego I, Navas M, Pastor J, Sola RG. Long-term results of vagal nerve stimulation for adults with medication-resistant epilepsy who have been on unchanged antiepileptic medication. Seizure 2013;22:9–13. [8] Spanaki MV, Allen LS, Mueller WM, Morris 3rd GL. Vagus nerve stimulation therapy: 5-year or greater outcome at a university-based epilepsy center. Seizure 2004;13:587–90. [9] Elliott RE, Morsi A, Kalhorn SP, Marcus J, Sellin J, Kang M, et al. Vagus nerve stimulation in 436 consecutive patients with treatment-resistant epilepsy: long-term outcomes and predictors of response. Epilepsy Behav 2011;20:57– 63. [10] Révész D, Rydenhag B, Ben-Menachem E. Complications and safety of vagus nerve stimulation: 25 years of experience at a single center. J Neurosurg Pediatr 2016;18:97–104. [11] Serdaroglu A, Arhan E, Kurt G, Erdem A, Hirfanoglu T, Aydin K, et al. Long term effect of vagus nerve stimulation in pediatric intractable epilepsy: an extended follow-up. Childs Nerv Syst 2016;32:641–6. [12] Wasade VS, Schultz L, Mohanarangan K, Gaddam A, Schwalb JM, SpanakiVarelas M. Long-term seizure and psychosocial outcomes of vagus nerve stimulation for intractable epilepsy. Epilepsy Behav 2015;53:31–6. [13] Ryvlin P, Gilliam FG, Nguyen DK, Colicchio G, Iudice A, Tinuper P, et al. The long-term effect of vagus nerve stimulation on quality of life in patients with pharmacoresistant focal epilepsy: the PuLsE (Open Prospective Randomized Long-term Effectiveness) trial. Epilepsia 2014;55:893–900. [14] Arcos A, Romero L, Gelabert M, Prieto A, Pardo J, Osorio XR, et al. Can we predict the response in the treatment of epilepsy with vagus nerve stimulation. Neurosurg Rev 2014;37:661–8. [15] Chrastina J, Kocvarova J, Novak Z, Dolezalova I, Svoboda M, Brazdil M. Older age and longer epilepsy duration do not predict worse seizure reduction outcome after vagus nerve stimulation. J Neurol Surg A Centr Eur Neurosurg 2018;79:152–8. [16] Janszky J, Hoppe M, Behne F, Tuxhorn I, Pannek HW, Ebner A. Vagus nerve stimulation: predictors of seizure freedom. J Neurol Neurosurg Psychiatry 2005;76:384–9. [17] Salinsky MC, Uthman BM, Ristanovic RK, Wernicke JF, Tarver WB. Vagus nerve stimulation for the treatment of medically intractable seizures: results of a 1year open-extension trial. Vagus Nerve Stimulation Study Group. Arch Neurol 1996;53:1176–80. [18] Schachter SC. Vagus nerve stimulation therapy summary: five years after FDA approval. Neurology 2002;59:S15–20. [19] Kuba R, Brázdil M, Kalina M, Procházka T, Hovorka J, Nežádal T, et al. Vagus nerve stimulation: longitudinal follow-up of patients treated for 5 years. Seizure 2009;18:269–74. [20] Uthman BM, Reichl AM, Dean JC, Eisenschenk S, Gilmore R, Reid S, et al. Effectiveness of vagus nerve stimulation in epilepsy patients: a 12-year observation. Neurology 2004;63:1124–6. [21] Ryzí M, Brázdil M, Novák Z, Chrastina J, Ošlejšková H, Rektor I, et al. Long-term vagus nerve stimulation in children with focal epilepsy. Acta Neurol Scand 2013;127:316–22. [22] Arcand J, Waterhouse K, Hernandez-Ronquillo L, et al. Efficacy of vagal nerve stimulation for drug-resistant epilepsy: is it the stimulation or medication. Can J Neurol Sci 2017;44:532–7. [23] García-Pallero MA, García-Navarrete E, Torres CV, Vitali A, Tellez-Zenteno JF. Effectiveness of vagal nerve stimulation in medication-resistant epilepsy. Comparison between patients with and without medication changes. Acta Neurochir (Wien) 2017;159:131–6. [24] Lam S, Lin Y, Curry DJ, Reddy GD, Warnke PC. Revision surgeries following vagus nerve stimulator implantation. J Clin Neurosci 2016;30:83–7. [25] Couch JD, Gilman AM, Doyle WK. Long-term expectations of vagus nerve stimulation: a look at battery replacement and revision surgery. Neurosurgery 2016;78:42–6. [26] Vonck K, Dedeurwaerdere S, De Groote L, Thadani V, Claeys P, Gossiaux F, et al. Generator replacement in epilepsy patients treated with vagus nerve stimulation. Seizure 2005;14:89–99. [27] 4th Tatum WO, Ferreira JA, Benbadis SR, Heriaud LS, Gieron M, Rodgers-Neame NT, et al. Vagus nerve stimulation for pharmacoresistant epilepsy: clinical symptoms with end of service. Epilepsy Behav 2004;5:128–32. [28] Giordano F, Zicca A, Barba C, Guerrini R, Genitori L. Vagus nerve stimulation: surgical technique of implantation and revision and related morbidity. Epilepsia 2017;58(Suppl. 1):85–90. J. Chrastina et al. / Seizure 59 (2018) 41–47 47 205 Annex 13: Chrastina J, Kocvarova J, Novak Z, Dolezalova I, Svoboda M, Brazdil M. Older age and longer epilepsy duration do not predict worse seizure reduction outcome after vagus nerve stimulation. Journal of neurological surgery part A – Central European neurosurgery. 2018;79(2):152-158. 206 Older Age and Longer Epilepsy Duration Do Not Predict Worse Seizure Reduction Outcome after Vagus Nerve Stimulation Jan Chrastina1 Jitka Kocvarova2 Zdenek Novak1 Irena Dolezalova2 Michal Svoboda3 Milan Brazdil2,4 1Department of Neurosurgery, Masaryk University Medical Faculty, Faculty Hospital St. Anne’s, Brno, Czech Republic 2First Department of Neurology, Faculty Hospital St. Anne’s, Masaryk University Medical Faculty, St. Anne’s Hospital Brno, Brno, Czech Republic 3Institute of Biostatistics and Analyses, Medical Faculty Masaryk University, Brno, Czech Republic 4Central European Institute of Technology (CEITEC), Masaryk University, Brno, Brno, Czech Republic J Neurol Surg A Address for correspondence Jan Chrastina, MD, PhD, Department of Neurosurgery Masaryk University Medical Faculty, Faculty Hospital St. Anne’s, Pekarska 53, Brno 656 91, Czech Republic (e-mail: jan.chrastina@fnusa.cz). Keywords ► epilepsy ► vagus nerve stimulation ► seizures reduction ► responder Abstract Introduction We analyzed the results of vagus nerve stimulation (VNS) on older patients and patients with long-lasting epilepsy and included severely intellectually disabled patients. Patients and Methods A total of 103 adults with VNS implanted from 2005 to 2014 were studied. The responder rates, that is, the percentage of VNS patients who responded to VNS, classified as seizure reduction ! 50% (50R) and seizure reduction ! 90% (90R), were compared in defined age groups (< 40 and ! 40 years, and < 50 and ! 50 years) and epilepsy duration groups (< 20 and ! 20 years, < 30 and ! 30 years, and < 40 and ! 40 years) at the 1-year follow-up visit and the last follow-up visit (at least 2 years after surgery). The age distributions and responder rates were also studied in patients with an intellectual disability. Results The analysis did not confirm a significantly lower 50R or 90R rate in patients ! 40, ! 50, or ! 60 years when compared with their younger counterparts, but the 50R rate increase during follow-up care was the lowest in patients ! 50 and ! 60 years. The highest percentage of patients with an intellectual disability in the group < 40 years of age did not adversely affect the 50R rate. Longer epilepsy duration was not confirmed as a negativepredictorofVNSoutcome.Therewasasignificantlyhigher50Rrateinpatientswith epilepsy duration ! 20 years (at the last follow-up visit) and a higher 90R rate in patients with epilepsy duration ! 30 years (at the 1-year follow-up visit). The increase in the 50R rate during follow-up care was lower in patients with epilepsy durations ! 30 and ! 40 years. Conclusions The study did not find worse VNS outcomes, as defined by the 50R or 90R rate, in older adult patients or in patients with a longer epilepsy duration. The increasing stimulation effect over time is less marked in older patients and in patients with longer epilepsy duration. received January 9, 2017 accepted after revision August 2, 2017 © Georg Thieme Verlag KG Stuttgart · New York DOI https://doi.org/ 10.1055/s-0037-1607396. ISSN 2193-6315. Original Article Downloadedby:UniversiteLaval.Copyrightedmaterial. 207 Introduction Vagus nerve stimulation (VNS) is a safe and effective palliative surgical treatment for refractory epilepsy not amenable to resection. The effect of VNS on seizure frequency and severity was confirmed by three randomized controlled trials.1–3 In general, a > 50% seizure reduction can be expected in at least 50% of patients after 2 years of treatment with mild side effects.4,5 The effect of stimulation is not immediate, but it increases over time after device implanta- tion.6 The U.S. Food and Drug Administration (FDA) approved VNS in 1997 as an adjunctive therapy for intractable partial epilepsy in patients ! 12 years of age who are ineligible for resection; the indications for VNS are even wider including generalized epilepsies,7 Lennox-Gastaut syndrome,8 and very young children with epilepsy.9,10 Complete freedom from seizures is rarely achieved using VNS, and 25% of patients had no benefit.3 The standard evaluation scales (Engel and International League Against Epilepsy) are not suitable for VNS outcome studies. The percentage of seizure reduction and the frequency of responders (patients with a seizure reduction of at least 50% from baseline) are usually reported. Because of the progress in epilepsy surgery and the general awareness of surgical results and safety, older patients with chronic or lifetime intractable seizures as well as patients with an intellectual disability with the aim of seizure palliation and easier caregiving are more frequently referred for surgery. VNS implantation is not a difficult surgery and has few complications; it is therefore an attractive option for such patients. However, secondary epileptogenesis after long-lasting multiple seizures with a progressive course of the epileptic disorder, potentially leading to unsatisfactory treatment results, must be con- sidered.11 Brain injuries caused by falls during seizures and possible cerebrovascular accidents in middle-aged and older patients may also negatively influence the VNS outcome. Although the impact of mental disability on VNS outcome is not clear,12,13 its negative role cannot be excluded. This article presents the results of VNS in older patients and patients with longer epilepsy duration with the potential implication for VNS indication in older patients with longterm epilepsy. The secondary aim was to analyze the potential impact on VNS outcomes with the inclusion of patients with severe intellectual disabilities. Patients and Methods A total of 103 patients (47 men and 56 women, all > 18 years) with refractory epilepsy not suitable for resection had VNS system implantation from 2005 to 2014. The patients referred had multiple seizure types: complex partial seizure, complex partialseizure(CPS)withgeneralization,primarilygeneralized seizure, generalized tonic-clonic seizure (GTCS), simple partial seizure, and simple partial seizure with generalization. All of the patients underwent a complete presurgical evaluation at the Brno Epilepsy Center including video electroencephalogram (EEG) monitoring with interictal and ictal recordings, special epilepsy magnetic resonance imaging (MRI) protocol, and neuropsychology and functional imaging (single-photon emission computed tomography [SPECT], positron emission tomography, subtraction ictal SPECT coregistered with MRI) performed according to clinical needs. Before VNS, four patients underwent resective surgery without invasive exploration(oneextratemporalandthreetemporalresections; two for a low-gradetumor), and callosotomy wasperformed in two patients, all with unsatisfactory control of seizures. Invasive EEG was required in 12 patients and failed to identify a resectable epileptogenic focus in 6 patients. In the remaining six patients, the results of stereoelectroencephalography allowed resective surgery but without a worthwhile reduction of seizures. VNS implantation was performed by one of two neurosurgeons (Z.N. or J.C.). The only surgical complication was transient hoarseness in four patients. After implantation, the patients were followed at the First Department of Neurology at regular intervals (2, 4, 6, and 12 months; then yearly). Stimulation adjustments and antiepileptic treatment changes were made by the responsible physician. The study data were obtained retrospectively from medical records. Based on post-VNS seizure reduction, the patient was graded as a nonresponder (NR) (seizure frequency reduction < 50%), 50R (seizure frequency reduction ! 50%), or 90R (seizure frequency reduction ! 90%). The percentage of 50R and 90R rates was compared between patients < 40 years and ! 40 years and between patients aged < 50 years and ! 50 years of age at the 1-year follow-upvisit and thelast follow-upvisit(at least 2 years after surgery, maximum 10 years, median 5 years). There were 65 patients < 40 years, 27 patients ! 40 years but < 50 years, 11 patients ! 50 years, and 5 patients ! 60 years. The percentage of patients with severe intellectual disabilities (unable to take care of themselves, requiring institutional care) was 9.1% in patients ! 50 years (1 patient), 17.4% in patients < 50 years (16 patients), 7.9% in patients ! 40 years (3 patients), and 21.5% in patients < 40 years (14 patients). The causes of severe intellectual impairments were perinatal problems (most patients), meningoencephalitis, extensive developmental anomaly of the brain, tuberous sclerosis, and idiopathic. The effect of epilepsy duration on seizure outcome was analyzed by comparing 50R and 90R rates between patients with epilepsy duration < 20 years and ! 20 years, < 30 years and ! 30 years, and < 40 years and ! 40 years at the 1-year follow-up visit and the last follow-up visit. There were 38 patients with epilepsy duration < 20 years and 64 patients with epilepsy duration ! 20 years, 64 patients with epilepsy duration < 30 years and 38 patients with epilepsy duration ! 30 years, 84 patients with epilepsy duration < 40 years and 18 with epilepsy duration ! 40 years, and 6 patients with epilepsy duration ! 50 years. One patient with an uncertain epilepsy duration (the data from the patient and the patient’s medical records differed) was excluded from the analysis. Journal of Neurological Surgery—Part A Results of Vagus Nerve Stimulation Chrastina et al. Downloadedby:UniversiteLaval.Copyrightedmaterial. 208 Continuous parameters were described using medians (5% and 95%) and were tested using the Mann-Whitney test. Categorical parameters were described by absolute (relative) incidence, and the Fisher exact test was used. The analysis was supported by prediction of 50R and 90R at the 1-year follow-up visit and the last follow-up visit using logistic regression with age at implantation and seizure duration for the whole group. Results The outcomes were available for 103 patients at the 1-year follow-up visit and for 94 patients at the last follow-up visit. During the study period, the VNS system was explanted in seven NRs; two of them underwent palliative resection (both Engel III), and anterior thalamic nucleus stimulation was used in three patients. ►Table 1 summarizes the first part of the results, 50R rates at the 1-year follow-up visit and the last follow-up visit in the defined age and epilepsy duration categories. The statistical analysis did not prove a significant adverse effect of age on 50R rates for the age thresholds of 40 and 50 years. On the contrary, 50R rates were nonsignificantly higher in patients ! 40 years both at the 1-year follow-up and the last follow-up visit and in patients ! 50 years at the 1-year follow-up visit only. In the small group of five patients ! 60 years, the 50R rate was 60% at the 1-year follow-up visit and the last follow-up visit. However, instead of the expected increase in the 50R rate over time between the 1-year follow-up visit and the last follow-up visit, there was a decrease of the 50R rate in the patients ! 50 years of age (À3.7%). However, this decreased 50R rate in the group of 11 patients ! 50 years (7/11 [63.7%] to 6/10 [60%]) was caused by one patient (50R at the 1-year follow-up visit) who was lost to follow-up care for the final control. When removing this patient from the calculations, the 50R rate in patients ! 50 years at the 1-year and the last follow-up visits remained stable at 60% because all other 50Rs remained in this category. The adverse effect of longer epilepsy duration on 50R rates has not been proven for all the analyzed epilepsy duration thresholds. On the contrary, a significantly higher 50R rate was found in patients with epilepsy duration ! 20 years at the last follow-up visit. Moreover, the median duration of epilepsy in 50R is significantly higher than in NRs (28.5 years and 20.0 years, respectively; p ¼ 0.029). However, the increase in 50R rate during follow-up care is lower in patients with longer epilepsy duration: patients < 30 years 17.1%, > 30 years 2%, patients < 40 years 13.7%, and > 40 years 1.4%. In six patients with epilepsy duration ! 50 years, the 50R rate was 50% at both follow-up visits. ►Table 2 summarizes the 90R rates in the defined age and epilepsy duration categories at the 1-year follow-up and last follow-up visits. No significant difference was found in 90R rates between the 1-year and last follow-up visits in all age groups. Surprisingly, there was a significantly higher 90R rate in patients with epilepsy duration ! 30 years at the first follow-up visit (p ¼ 0.026). The difference between the median duration of epilepsy in 90Rs (32.5 years) and the rest of the group (24.0 years) does not reach statistical significance (p ¼ 0.058). An increase in the percentage of 90Rs during follow-up care is observed in all age and epilepsy duration groups, although this finding is definitively weakened by the small number of 90Rs. The results of VNS in terms of 50R and 90R in intellectually disabled patients are summarized in ►Table 3. The median Table 1 Effect of age and epilepsy duration on percentage of vagus nerve stimulation ! 50% seizure reduction at 1-year and final follow-up visits 50R 1-year follow-up p value 50R Final follow-up p value No Yes No Yes Age, y (range) 33.0 (20.0–54.0) 35.0 (19.0–61.0) 0.594 31.0 (20.0–60.0) 34.0 (19.0–58.0) 0.417 < 40 (%) 32 (49.2) 33 (50.8) 0.542 23 (38.3) 37 (61.7) 0.501 ! 40 (%) 16 (42.1) 22 (57.9) 10 (29.4) 24 (70.6) < 50 (%) 44 (47.8) 48 (52.2) 0.537 29 (34.5) 55 (65.5) 0.737 ! 50 (%) 4 (36.4) 7 (63.6) 4 (40.0) 6 (60.0) Epilepsy duration, y (range) 22.0 (5.0–54.0) 28.5 (9.0–50.0) 0.029 18.0 (4.0–54.0) 26.5 (8.0–48.5) 0.120 < 20 (%) 22 (57.9) 16 (42.1) 0.104 18 (48.6) 19 (51.4) 0.046 ! 20 (%) 26 (40.6) 38 (59.4) 15 (26.8) 41 (73.2) < 30 (%) 35 (54.7) 29 (45.3) 0.064 22 (37.3) 37 (62.7) 0.660 ! 30 (%) 13 (34.2) 25 (65.8) 11 (32.4) 23 (67.6) < 40 (%) 41 (48.8) 43 (51.2) 0.604 27 (35.1) 50 (64.9) 0.999 ! 40 (%) 7 (38.9) 11 (61.1) 6 (37.5) 10 (62.5) Abbreviation: 50R, ! 50% seizure reduction. Journal of Neurological Surgery—Part A Results of Vagus Nerve Stimulation Chrastina et al. Downloadedby:UniversiteLaval.Copyrightedmaterial. 209 age in the disabled patient group was significantly lower than in the group of patients without an intellectual disability (25.0 years and 34.0 years, respectively). Although the difference does not reach the level of statistical significance, the rates of 50R at both the 1-year follow-up and last followup visit were higher in the intellectually disabled patients. Therefore, the higher percentage of these patients in the younger age group does not adversely affect the percentage of 50R rates in younger patients. The results of the logistic regression for the entire group with age at implantation and epilepsy duration for prediction of 50R and 90R for 1-year and last follow-up visits are summarized in ►Tables 4–7. This analysis confirmed only longer epilepsy duration as a predictor of a higher percentage of 50R at the 1-year follow-up visit (odds ratio: 1.067; 95% confidence interval, 1.011–1.127; p ¼ 0.019), but this predictive value of longer epilepsy duration was lost at the last follow-up visit. Higher age was not confirmed as a negative predictive factor. The changes of seizure pattern and severity were inconsistent in all the studied age and seizure duration groups, for example, frequency reduction without change of seizure severity, lesser clustering or cumulation, shorter seizure duration, reduction or elimination of falls, marked reduction of CPSwith stable or minimally reduced frequency of GTCS or marked reduction of GTCS with stable frequency of CPS. Thus no statistical analysis was possible. Discussion Despite new antiepileptic drugs and advances in epilepsy surgery, a significant number of patients remain unresponsive to both pharmacotherapy and resective surgery. VNS, a safe and adjustable technique reducing the frequency and severity of intractable seizures with few adverse effects, is an attractive option. Although multiple factors potentially influencing VNS outcome have been studied (e.g., age, epilepsy duration, frequency, semiology, etiology, interictal EEG, and radiologic findings),3,14,15 the effect of VNS cannot be predicted reliably. Even the studies analyzing easily definable factors, such as age and epilepsy duration, produced contradictory results. Part of the difficulty in defining the best patient population may be the poor understanding of VNS action mechanisms. Moreover, single-center VNS series do not usually include homogeneous patient populations in terms of rare epileptic syndromes16 or even in terms of age and epilepsy duration (e.g., middle-aged and older patients, patients with long-lasting seizures). The problem with the inclusion of intellectually disabled patients is that the percentage of them is substantially higher in younger age groups as confirmed by their median age being significantly lower than that of the intellectually unaffected patients included in the study. One possible explanation is that there is still some reluctance to refer older institutionalized patients with an intellectual disability and chronic seizures for surgical treatment. But the higher percentage of intellectually disabled patients in the younger age group did not adversely affect the rate of 50R in younger patients. Most articles analyzing the impact of patient age at implantation on seizure reduction have dealt with children and young adolescents. Lagae et al showed that age at VNS implantation is theonlyfactor predictingseizureoutcomeinchildhoodepilepsy with particularly favorable results in children < 5 years.17 The Table 2 Effect of age and epilepsy duration on percentage of vagus nerve stimulation ! 90% seizure reduction at 1-year and final follow-up visits 90R 1-year follow-up p value 90R Final follow-up p value No Yes No Yes Age, y (range) 34.0 (20.0–58.0) 37.0 (20.0–61.0) 0.525 33.0 (20.0–60.0) 37.0 (19.0–50.0) 0.759 < 40 (%) 60 (92.3) 5 (7.7) 0.999 53 (88.3) 7 (11.7) 0.743 ! 40 (%) 36 (94.7) 2 (5.3) 31 (91.2) 3 (8.8) < 50 (%) 86 (93.5) 6 (6.5) 0.558 75 (89.3) 9 (10.7) 0.999 ! 50 (%) 10 (90.9) 1 (9.1) 9 (90.0) 1 (10.0) Epilepsy duration, y (range) 24.0 (5.0–50.0) 32.5 (21.0–61.0) 0.058 24.0 (5.0–54.0) 32.0 (18.0–44.0) 0.117 < 20 (%) 38 (100.0) 0 (0.0) 0.082 36 (97.3) 1 (2.7) 0.081 ! 20 (%) 58 (90.6) 6 (9.4) 48 (85.7) 8 (14.3) < 30 (%) 63 (98.4) 1 (1.6) 0.026 56 (94.9) 3 (5.1) 0.069 ! 30 (%) 33 (86.8) 5 (13.2) 28 (82.4) 6 (17.6) < 40 (%) 80 (95.2) 4 (4.8) 0,285 70 (90.9) 7 (9.1) 0.650 ! 40 (%) 16 (88.9) 2 (11.1) 14 (87.5) 2 (12.5) Abbreviation: 90R, ! 90% seizure reduction. Journal of Neurological Surgery—Part A Results of Vagus Nerve Stimulation Chrastina et al. Downloadedby:UniversiteLaval.Copyrightedmaterial. 210 Table 4 Prediction of ! 50% seizure reduction, logistic regression with age at implantation and seizure duration, at the 1-year follow-up visit OR 95% CI p value Age 0.952 0.894–1.014 0.128 Epilepsy duration 1.067 1.011–1.127 0.019 Abbreviations: CI, confidence interval; OR, odds ratio. Table 3 Intellectually disabled patients: characteristics and vagus nerve stimulation outcomes Intellectual disability p value No Yes Age, y (range) 34.0 (20.0–60.0) 25.0 (18.0–50.0) 0.024 < 40 (%) 53 (60.2) 14 (82.4) 0.102 ! 40 (%) 35 (39.8) 3 (17.6) < 50 (%) 78 (88.6) 16 (94.1) 0.688 ! 50 (%) 10 (11.4) 1 (5.9) Epilepsy duration (range) 24.0 (5.0–54.0) 24.0 (16.0–50.0) 0.712 < 20 35 (40.2) 5 (29.4) 0.587 ! 20 (%) 52 (59.8) 12 (70.6) < 30 (%) 54 (62.1) 12 (70.6) 0.590 ! 30 (%) 33 (37.9) 5 (29.4) < 40 (%) 71 (81.6) 15 (88.2) 0.730 ! 40 (%) 16 (18.4) 2 (11.8) 50 R, 1-year follow-up No (%) 42 (48.3) 6 (37.5) 0.587 Yes (%) 45 (51.7) 10 (62.5) 50R, final follow-up No (%) 30 (37.5) 3 (21.4) 0.365 Yes (%) 50 (62.5) 11 (78.6) 90R, 1-year follow-up No (%) 80 (92.0) 16 (100.0) 0,592 Yes (%) 7 (8.0) 0 (0.0) 90R, final follow-up No (%) 71 (88.8) 13 (92.9) 0.999 Yes (%) 9 (11.3) 1 (7.1) Abbreviation: 50R, ! 50% seizure reduction; 90R, ! 90% seizure reduction. Table 5 Prediction of ! 50% seizure reduction, logistic regression with ageat implantation and seizure duration, at finalfollow-upvisit OR 95% CI p value Age 0.986 0.927–1.049 0.659 Epilepsy duration 1.031 0.976–1.089 0.274 Abbreviations: CI, confidence interval; OR, odds ratio. Table 6 Prediction of ! 90% seizure reduction, logistic regression with age at implantation and seizure duration, at the 1-year follow-up visit OR 95% CI p value Age 0.916 0.757–1.107 0.362 Epilepsy duration 1.140 0.954–1.362 0.149 Abbreviations: CI, confidence interval; OR, odds ratio. Table 7 Prediction of ! 90% seizure reduction, logistic regression with ageat implantation and seizure duration, at finalfollow-upvisit OR 95% CI p value Age 0.937 0.821–1.070 0.338 Epilepsy duration 1.084 0.959–1.226 0.195 Abbreviations: CI, confidence interval; OR, odds ratio. Journal of Neurological Surgery—Part A Results of Vagus Nerve Stimulation Chrastina et al. Downloadedby:UniversiteLaval.Copyrightedmaterial. 211 potential explanation of better results in younger patients from pediatric groups may be the higher threshold for excitatory stimuli and seizure-induced changes in immaturebrains than in adult brains (based on an experimental study of amygdaloid kindling in a cat model).18 Alexopoulos et al found that age < 12 years at implantation (FDA-defined threshold) is actually associated with a better prognosis when considering median seizure frequency reduction.19 Similarly, Meng et al showed a higher percentage of 50R in patients < 12 years (71.4%) than in their older counterparts (60.6%).20 For different age thresholds, Wheless and Maggio found better VNS results in patients < 18 years.21 However, other reports do not confirm a better VNS outcome in younger patients. Thompson et al showed no significant differences in seizure frequency reduction, epilepsy duration, overall clinical improvement, or improvement in quality of life based on age < 12 or ! 12 years.22 In a Czech multicenter study, lower efficacy rates of VNS in patients < 16 years were observed.23 An Israeli retrospective multicenter open-label study comparing seizure reduction in patients < 22 and ! 22 years found higher 50R rates in older patients.24 Wheeler et al studied seizure outcome in age categories 0 to 20 years, 21 to 40 years, and > 40 years at VNS implantation. The percentage of patients with at least a worthwhile reduction of seizures (Engel I, II, and III) was 51% for 0 to 20 years, 69% for 21 to 40 years, and 80% for patients > 40 years. The outcomes in patients > 50 years were comparable. However, significantly higher percentages of Engel I and II were found in patients with better mental functioning; this finding does not agree with our results.25 A multicenter study of 45 adults > 50 years found that their response to VNS was similar to that of younger adults without increased morbidity.26 Our data do not confirm a statistically significant negative impact of age (for age thresholds 40 and 50 years using the Mann-Whitney test, and for the whole group using logistic regression) on VNS 50R rates. Statistically nonsignificant 50R rates were actually higher in patients ! 40 years at both the 1-year and the last follow-up visits and in patients ! 50 years at the 1-year follow-up visit. An increased VNS effect, as defined byan increasing 50R rate, was not observed inpatients ! 50 and ! 60 years. The rate of 90R in the selected age categories was also comparable. This finding matches the data presented by Englott et al, who observed no relationship between the age at implantation and seizure freedom rate.27 Our results do not confirm higher age to be a negative predictor of VNS effect. Another study aim was to analyze the effect of VNS in patients with long-lasting epilepsy. The expected worse outcomes in patients with long-lasting epilepsy were supported by some literature results. An analysis of the Cyberonics VNS patient registry data found a significantly higher percentages of patients with 100% seizure reduction and ! 90% seizure reduction in patients with VNS within 6 years of seizure onset.28 According to Englott et al, there is a trend among patients with shorter epilepsy duration (threshold 10 years) before VNS for a higher frequency of seizure freedom.27 Although the short 3-month follow-up care is a serious limitation in a study by Ranfroe and Wheless, the percentage of seizure-free patients after VNS with epilepsy duration < 5 years was 15%, and 4.4% in patients with epilepsy duration > 5 years.29 However, the association of longer epilepsy duration with worse outcomes was not confirmed by other studies. In a twocenterstudy (BelgiumandtheUnited States), thepercentage of Engel I to III outcomes was 61% in patients with epilepsy duration < 10 years and 70% in patients with epilepsy duration ! 10 years, but this difference was not statistically significant.25 Lagae et al did not confirm epilepsy duration as a prognostic factor in children and young adults.17 In a study of patients with extremely variable epilepsy durations (from 3 months to 57 years), Colicchio et al stated that longer epilepsy duration has a positive effect on seizure outcome.30 Our results do not confirm an adverse effect of epilepsy duration on the rate of 50R. On the contrary, the median duration of epilepsy in 50Rs is significantly higher than in NRs, and a significantly higher percentage of 50R was found in patients with epilepsy duration ! 30 years at the last followup visit. However, the increase in 50R rates during follow-up care is lower in patients with longer epilepsy duration. This correlates with the results of a logistic regression that confirmed longer epilepsy duration as a predictor of a higher percentage of 50R at the 1-year follow-up visit but not at the last follow-up visit. Because the mechanism of action of VNS is still largely unknown, it is difficult to provide an undisputable explanation for the less marked increase of 50R rates comparing the 1-year follow-up and last follow-up visits in older patients and patients with prolonged epilepsy duration. The research into the mechanisms of action of various strategies for electrical modulation of the brain suggests a crucial role of different neurotransmitter molecules and channels, such as glutamate, γ-aminobutyric acid (GABA), adenosine, brainderived neurotrophic factors, and calcium and sodium channels. Electrical modulation of the brain might also promote neurogenesis in subjects with pharmacoresistant epilepsy in whom this process is decreased.31 Walker et al demonstrated in an animal model that an increase in GABA transmission or a decrease in glutamate transmission reduces susceptibility to limbic motor seizures.32 Also, recent preclinical experiments indicate that activation of the noradrenergic system in the locus coeruleus (the primary site of vagal afferent termination) is critical for the antiepileptic effect of VNS.33 Therefore, the hypothetically reduced potential for the process of neurogenesis and decreased synthesis of neurotransmitters in older patients or patients exhausted by longlasting epilepsy provide a potential explanation for the lower increase of responder rates during follow-up care. Conclusions Older age and prolonged duration of epilepsy were not confirmed as negative predictors of VNS outcome as evaluated by the percentage of 50R and 90R at the 1-year and last follow-up visit, with even significantly higher 50R rates in Journal of Neurological Surgery—Part A Results of Vagus Nerve Stimulation Chrastina et al. Downloadedby:UniversiteLaval.Copyrightedmaterial. 212 patients with epilepsy duration ! 20 years at the last followup visit. The highest percentage of intellectually disabled patients in patients < 40 years did not adversely influence the 50R rate in this group. However, there are less prominent increases in VNS effect over time, as evaluated by 50R and 90R rates, in older patients and in patients with longer histories of epilepsy. References 1 The Vagus Nerve Stimulation Study Group. A randomized controlled trial of chronic vagus nerve stimulation for treatment of medically intractable seizures. Neurology 1995;45(02):224–230 2 Handforth A, DeGiorgio CM, Schachter SC, et al. Vagus nerve stimulation therapy for partial-onset seizures: a randomized active-control trial. Neurology 1998;51(01):48–55 3 Englot DJ, Chang EF, Auguste KI. Vagus nerve stimulation for epilepsy: a meta-analysis of efficacy and predictors of response. J Neurosurg 2011;115(06):1248–1255 4 Elliott RE, Morsi A, Kalhorn SP, et al. Vagus nerve stimulation in 436 consecutive patients with treatment-resistant epilepsy: long-term outcomes and predictors of response. Epilepsy Behav 2011;20(01):57–63 5 Cukiert A. Vagus nerve stimulation for epilepsy: an evidencebased approach. Prog Neurol Surg 2015;29:39–52 6 Amar AP, DeGiorgio CM, Tarver WB, Apuzzo ML. Long-term multicenter experience with vagus nerve stimulation for intractable partial seizures: results of the XE5 trial. Stereotact Funct Neurosurg 1999;73(1-4):104–108 7 Labar D, Murphy J, Tecoma E. Vagus nerve stimulation for medication-resistant generalized epilepsy. E04 VNS Study Group. Neurology 1999;52(07):1510–1512 8 Frost M, Gates J, Helmers SL, et al. Vagus nerve stimulation in children with refractory seizures associated with Lennox-Gastaut syndrome. Epilepsia 2001;42(09):1148–1152 9 Helmers SL, Wheless JW, Frost M, et al. Vagus nerve stimulation therapy in pediatric patients with refractory epilepsy: retrospective study. J Child Neurol 2001;16(11):843–848 10 Zamponi N,Rychlicki F, CorpaciL, CesaroniE,TrignaniR.Vagus nerve stimulation (VNS) is effective in treating catastrophic 1 epilepsy in very young children. Neurosurg Rev 2008;31(03):291–297 11 Morrell F. Varieties of human secondary epileptogenesis. J Clin Neurophysiol 1989;6(03):227–275 12 Danielsson S, Viggedal G, Gillberg C, Olsson I. Lack of effects of vagus nerve stimulation on drug-resistant epilepsy in eight pediatric patients with autism spectrum disorders: a prospective 2-year follow-up study. Epilepsy Behav 2008;12(02):298–304 13 Andriola MR, Vitale SA. Vagus nerve stimulation in the developmentally disabled. Epilepsy Behav 2001;2(02):129–134 14 Janszky J, Hoppe M, Behne F, Tuxhorn I, Pannek HW, Ebner A. Vagus nerve stimulation: predictors of seizure freedom. J Neurol Neurosurg Psychiatry 2005;76(03):384–389 15 Arcos A, Romero L, Gelabert M, et al. Can we predict the response in the treatment of epilepsy with vagus nerve stimulation? Neurosurg Rev 2014;37(04):661–668 16 Cukiert A. Commentary: Rates and predictors of seizure freedom with vagus nerve stimulation for intractable epilepsy. Neurosurgery 2016;79(03):354–355 17 Lagae L, Verstrepen A, Nada A, et al. Vagus nerve stimulation in children with drug-resistant epilepsy: age at implantation and shorter duration of epilepsy as predictors of better efficacy? Epileptic Disord 2015;17(03):308–314 18 Fernández-Guardiola A, Martínez A, Valdés-Cruz A, MagdalenoMadrigal VM, Martínez D, Fernández-Mas R. Vagus nerve prolonged stimulation in cats: effects on epileptogenesis (amygdala electrical kindling): behavioral and electrographic changes. Epilepsia 1999;40(07):822–829 19 Alexopoulos AV, Kotagal P, Loddenkemper T, Hammel J, Bingaman WE. Long-term results with vagus nerve stimulation in children with pharmacoresistant epilepsy. Seizure 2006;15(07):491–503 20 Meng FG, Jia FM, Ren XH, et al. Vagus nerve stimulation for pediatric and adult patients with pharmaco-resistant epilepsy. Chin Med J (Engl) 2015;128(19):2599–2604 21 Wheless JW, Maggio V. Vagus nerve stimulation therapy in patients younger than 18 years. Neurology 2002;59(06, Suppl 4): S21–S25 22 Thompson EM, Wozniak SE, Roberts CM, Kao A, Anderson VC, Selden NR. Vagus nerve stimulation for partial and generalized epilepsy from infancy to adolescence. J Neurosurg Pediatr 2012; 10(03):200–205 23 Kuba R, Brázdil M, Kalina M, et al. Vagus nerve stimulation: longitudinal follow-up of patients treated for 5 years. Seizure 2009;18(04):269–274 24 Menascu S, Kremer U, Schiller Y, et al. The Israeli retrospective multicenter open-label study evaluating vagus nerve stimulation efficacy in children and adults. Isr Med Assoc J 2013;15(11): 673–677 25 Wheeler M, De Herdt V, Vonck K, et al. Efficacy of vagus nerve stimulation for refractory epilepsy among patient subgroups: a re-analysis using the Engel classification. Seizure 2011;20(04): 331–335 26 Sirven JI, Sperling M, Naritoku D, et al. Vagus nerve stimulation therapy for epilepsy in older adults. Neurology 2000;54(05): 1179–1182 27 Englot DJ, Rolston JD, Wright CW, Hassnain KH, Chang EF. Rates and predictors of seizure freedom with vagus nerve stimulation for intractable epilepsy. Neurosurgery 2016;79(03):345–353 28 Helmers SL, Griesemer DA, Dean JC, et al. Observations on the use of vagus nerve stimulation earlier in the course of pharmacoresistant epilepsy: patients with seizures for six years or less. Neurologist 2003;9(03):160–164 29 Renfroe JB, Wheless JW. Earlier use of adjunctive vagus nerve stimulation therapy for refractory epilepsy. Neurology 2002;59 (06, Suppl 4):S26–S30 30 Colicchio G, Policicchio D, Barbati G, et al. Vagal nerve stimulation for drug-resistant epilepsies in different age, aetiology and duration. Childs Nerv Syst 2010;26(06):811–819 31 Rocha L. Interaction between electrical modulation of the brain and pharmacotherapy to control pharmacoresistant epilepsy. Pharmacol Ther 2013;138(02):211–228 32 Walker BR, Easton A, Gale K. Regulation of limbic motor seizures by GABA and glutamate transmission in nucleus tractus solitarius. Epilepsia 1999;40(08):1051–1057 33 Shen H, Fuchino Y, Miyamoto D, Nomura H, Matsuki N. Vagus nerve stimulation enhances perforant path-CA3 synaptic transmission via the activation of β-adrenergic receptors and the locus coeruleus. Int J Neuropsychopharmacol 2012;15(04):523–530 Journal of Neurological Surgery—Part A Results of Vagus Nerve Stimulation Chrastina et al. Downloadedby:UniversiteLaval.Copyrightedmaterial. 213 Annex 14: Chrastina J, Doležalová I, Novák Z, Pešlová E, Brázdil M. Pregnancy outcomes in refractory epilepsy with vagus nerve stimulation: long-term single center experience. (submitted) 214 Article Title: Pregnancy outcomes in refractory epilepsy patients with vagus nerve stimulation: long-term single-center experiences Running Title: Pregnancy outcomes and vagus stimulation Key Words: Vagus nerve stimulation, Epilepsy, Pregnancy, Obstetric interventions, Psychomotor development Chrastina J.1 , Doležalová I.2 , Novák Z. 1 , Pešlová E. 2 , Brázdil M. 2,3 1 Department of Neurosurgery, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic 2 Brno Epilepsy Center, Department of Neurology, St. Anne’s University Hospital, Faculty of Medicine, Masaryk University, Brno, Czech Republic 3 Behavioral and Social Neuroscience Research Group, CEITEC – Central European Institute of Technology, Masaryk University, Brno, Czech Republic Corresponding author: Jan Chrastina, Associate Professor, M.D., Ph.D. Department of Neurosurgery, Masaryk University Medical Faculty St. Anne’s Hospital Brno Pekařská 53 656 91 Brno Czech Republic phone: +420 543 182 697 fax: +420 543 182 687 e mail: jan.chrastina@fnusa.cz The work has not been previously published or presented None of the authors has any conflict of interest to disclose 215 Abstract Introduction: Vagus nerve stimulation (VNS) has been employed worldwide as an adjunctive therapy in drug-resistant epilepsy patients. However, the mechanisms of VNS action potentially increase the risk of obstetric complications. The study presents long-term single-center experiences with pregnancies and childbirth in women with VNS for refractory epilepsy based on prospectively collected epileptological data and a retrospective analysis of pregnancy, childbirth, and data about long-term child development. Material and Methods: From a group of patients with VNS implanted for refractory epilepsy between October 1999 and January 2018, all the women of childbearing age (younger than 40 years) were identified. After checking their hospital records for data about any pregnancies women with confirmed childbirth during active VNS stimulation were interviewed based on a prepared questionnaire regarding their gynecological history, the course of pregnancy and childbirth, gestational week, birth weight and length, any congenital anomalies of the child, and the child’s psychomotor development, school performance, and health problems. Results: From the group of 257 patients implanted with VNS for refractory epilepsy, four women (1.5%) became pregnant and gave birth (all on polypharmacotherapy). The mean interval from VNS implantation to birth was 44.3 months. Slight seizure worsening during the last trimester was reported in one woman. In one patient, acute Caesarean section was required due to placental separation. Planned birth induction and Caesarean section were used in another two women because of their seizure disorder. No malfunction of the stimulation system was detected during pregnancy or after birth. No congenital malformations were observed. The two children who were of school age at the time of this study require special schooling. Conclusions: The study results confirmed a high rate of obstetric interventions in VNS patients. Although no teratogenic effect of VNS has been proven, the higher incidence of VNS-exposed children needing special education requires attention. 216 Introduction: Vagus nerve stimulation (VNS) has been employed worldwide as adjunctive therapy in drugresistant epileptic patients. The data about VNS implantation for indications other than epilepsy (e.g. depression, some headache syndromes) are less extensive (Ansari et al., 2007). VNS is used in all ages, including the potential childbearing female population. The generally accepted mechanism of VNS action is the modulation of synaptic activity and therefore excitability in widespread cortical and subcortical regions of the brain, including some components of the central autonomic system such as the hypothalamus. Therefore, VNS therapy may influence neuroendocrine activity (e.g. the production of oxytocin and its storage and release from the pituitary gland) and thus the potential to affect pregnancy and fetal development cannot be ignored (Sabers et al., 2018). An experimental study by Sato et al. in non-pregnant rats demonstrated that efferent VNS can influence uterine blood flow and contractions (Sato et al., 1996). Although efferent VNS fibers form only 20% of the vagus nerve trunk, with substantially fewer fibers conveyed to the uterus than to the heart, the theoretical possibility of VNS-induced alterations of uterine functions during pregnancy and delivery cannot be excluded (Ortega-Villalobos et al., 1990). All these mechanisms may suggest that VNS can increase the risk of obstetric complications during pregnancy and at the time of the delivery. Possible teratogenic effects on the newborn must be considered. However, seizures during pregnancy can also have severe hypoxia-related consequences for the mother and the fetus, so the function of a medical device reducing seizure frequency or severity (including VNS) should be maintained during pregnancy (Christensen et al., 2018). In 2017, the most extensive clinical data regarding the impact of VNS on pregnancy was published by Sabers et al. The authors summarized the outcome of 26 pregnancies in 25 women, 14 of whom were registered in the European and International Registry of Antiepileptic Drugs and Pregnancy (EURAP); the authors also included patients who were either registered in the UK and Australian national databases or reported separately by EURAP coordinators. The study suggested an increased rate of obstetric intervention and no signs of VNS-related teratogenicity (Sabers et al., 2017). The aim of our study is to present a complex review of long-term single-center experiences, with the first VNS implantation in 1999, of pregnancies and childbirth in women treated with VNS for refractory epilepsy. The study is based on the prospectively collected follow-up data of VNS patients, supplemented with retrospectively obtained clinical data about the course of pregnancy and childbirth including obstetric problems and with data about long-term child development based on a questionnaire prepared for the purposes of the study. Material and methods: A retrospective analysis was performed of patients with VNS implanted for refractory epilepsy between October 1999 and January 2018. Women of childbearing age (younger than 40 years) were identified and their hospital records were checked for data about the possibility of pregnancy and childbirth during the period of active VNS. Following their informed consent, the women with confirmed childbirth during active VNS stimulation were interviewed by phone by two of the study authors (ID, EP) based on a pre-interview prepared questionnaire. The questions concerned age at conception, gynecological history, conception (e.g. planned/unplanned), any history of preceding abortions, any urological problems of the partner, the course of pregnancy including concomitant diseases, the course of childbirth including complications, gestational week at childbirth, sex, birth weight and birth length of the child, congenital anomalies of the child, breastfeeding duration, and the child’s psychomotor development, school performance, and health problems. Data regarding epilepsy type, previous surgeries including stereoelectroencephalography, pharmacological treatment including changes during pregnancy and breastfeeding, VNS response before and during 217 pregnancy and after childbirth, and stimulation parameter changes were obtained from the prospectively built database of VNS patients. Results: During the study period from October 1999 and January 2018, 257 patients underwent VNS implantation for refractory epilepsy and another three patients for refractory depression. Among them, 71 women who were younger than 40 years at VNS implantation were identified. In this group, four women with active VNS became pregnant and gave birth to a child. The ages at childbirth were 22, 26, 33, and 38 years. The types of epilepsy were rightsided multifocal, bitemporal, left pericentral, and idiopathic generalized, respectively. Before VNS implantation, one patient underwent repeated stereoencephalography and subsequently left temporal resection without significant effect on seizure frequency or severity (bitemporal epilepsy). The data about the patient history and the course of pregnancy are summarized in Table 1. Conception was unplanned in one case (a woman with a history of four spontaneous abortions before VNS implantation, probably due to uterine cervix insufficiency after conization). The gynecological history was negative in all other included women. None of the partners reported a history of urological problems. The mean interval between VNS implantation and birth was 44.3 months (10 to 81 months). The course of pregnancy was uneventful in three patients. One patient experienced placental separation requiring acute Caesarean section. One patient had a spontaneous natural birth. Planned obstetric interventions were used in otherwise uneventful pregnancies in two women (birth induction, Caesarean section), in both because of their seizure disorder. Table 2 summarizes the basic data regarding the childbirth parameters, the immediate postpartum course, and the subsequent psychomotor development, child health problems, and the course of school education if applicable (age). There were no congenital malformations observed in any of the children and no history of serious somatic disease during the entire follow-up period. Table 3 summarizes the basic data regarding the VNS response before pregnancy and seizure course during pregnancy and after birth. No malfunction of the stimulation system was detected during pregnancy or after birth. In only one patient there was slight increase of seizures frequency during the last pregnancy trimester. No changes of the stimulation parameters were required in any patients. Discussion: Before discussing the impact of pregnancy on the effect of VNS in patients with refractory epilepsy, a short note about the effect of pregnancy on seizure frequency in epilepsy patients should be made. The largest prospective study published so far, based on 3,806 pregnancies, concluded that seizure control remained unchanged throughout pregnancy in 70.5%. In 12.0% of patients, there was an increase in seizure frequency. In 15.8% of patients, a decrease in seizure frequency was reported (Battino et al., 2013). The deterioration in seizure control during pregnancy has been attributed to pharmacokinetic, metabolic, hormonal, physiological, and psychological factors, as well as possible lack of compliance in some cases (Morrell et al., 1992). VNS has been used as an adjunctive therapy for drug-resistant epilepsy for more than 20 years. Despite this, until the paper published by Sabers et al. (2017), only case reports and small patient cohorts published also in gynecological and obstetric journals and pointing primarily to birth-related complications such as spontaneous delivery complicated by mild preeclampsia (Houser et al., 2010) were available about the influence of VNS on pregnancy. According to the extensive study by Sabers et al. (2017) describing the outcome of 26 218 pregnancies in 25 women, a total of 35 pregnancies in women receiving VNS treatment appear to have been reported so far (Sabers et al., 2017). After Sabers’ publication, a singlecenter study by Suller Marti et al. was published with a study period nearly identical to ours (1998 to 2018). Among 114 patients with VNS implanted during the study period, the authors found four patients with a total of seven pregnancies, including the first report of a woman implanted with VNS having three pregnancies. In contrast to our group, three of their four patients had genetic generalized epilepsy and one was treated for focal epilepsy due to periventricular nodular heterotopia. It is important to note that none of the patients underwent any epileptosurgical intervention (stereoelectroencephalography, resection, or disconnection surgery) before VNS implantation (Suller Marti et al., 2019). The paper by Sabers et al. does not discuss epileptosurgical procedures prior to VNS (Sabers et al., 2017). The most extensive study about the impact of VNS on pregnancy and childbirth published before 2018 was the multicenter study by Rodríguez-Osorio et al. In the study, the outcome of four out of the five pregnancies reported in four included women were positive. The spontaneous abortion was probably more related to antiepileptic drugs than to VNS. None of the patients included underwent any epileptosurgical procedure other than VNS. The authors also discuss the course of 10 pregnancies in women with VNS (nine in epilepsy patients and one in a patient with treatment-refractory depression) published before their study. There were seven positive and three negative pregnancy outcomes (two elective abortions, one spontaneous abortion) (Rodríguez-Osorio et al., 2017). Of the 26 pregnancies reported by Sabers et al. one terminated in spontaneous abortion (Sabers et al., 2017). The gestational week in our group was more than 36 weeks in two babies, but in two babies the gestational ages at birth were 31 weeks (acute Caesarean section due to placenta separation) and 34 weeks (planned Caesarean section due to epilepsy; no seizure worsening during pregnancy was reported in this patient). In the study by Rodríguez-Osorio et al., in only one patient was the gestational week less than 36 (birth at week 35 due to fetal instability secondary to maternal- fetal Rh incompatibility requiring Caesarean section) (RodríguezOsorio et al., 2017). Sabers et al. reported a gestational age at birth of at least 36 weeks in all but one patient (35 weeks) (Sabers et al., 2017). However, it is difficult to separate the potential negative effects of VNS on pregnancy from other adverse factors including refractory seizure disorder and polypharmacotherapy. Some data could be potentially drawn from VNS implanted for different indications than epilepsy. However, data about the pregnancy course in VNS treatment for refractory depression are limited to a case report published by Husain et al. In this patient, VNS led to a substantial reduction in depressive symptoms and improved function throughout pregnancy and labor, with an uneventful delivery of a healthy baby (Husain et al., 2005). In the paper by Salerno et al., a woman with VNS implanted for depression also had epilepsy and obesity. Moreover, malfunctioning of VNS was detected during pregnancy. Although device functioning was not optimal during the critical period of organogenesis, no morphological abnormalities of the fetus were detected (Salerno et al., 2016). Regarding the incidence of obstetric interventions in our group, there was one spontaneous delivery, one planned induced birth, one elective Caesarean section (both reported as due to epilepsy), and one acute Caesarean section (because of placental separation). Our data matches the results published by Rodríguez-Osorio et al. – two vaginal births and two Caesarean sections due to obstetric problems – premature rupture of membranes and maternal- fetal Rh incompatibility (Rodríguez-Osorio et al., 2017). A high incidence of obstetric intervention in patients with VNS was reported by Sabers et al., with 38.5% Caesarean sections, 7.7% both vacuum extractors and induced labors, a combined 53.9% of labors with intervention. This rate is slightly higher than the reported rate in EURAP data – 219 48.2%. Almost half of the cases who underwent Caesarean section had an increase in seizure frequency during the third trimester; this fact has undoubtedly influenced the obstetrician’s decision to opt for a Caesarean section. The percentage of obstetric interventions was reported to be higher in AED polytherapy than in monotherapy (Sabers et al., 2017; Battino et al., 2013). Similarly, Suller Marti et al. pointed to the high incidence of obstetric complications in VNS patients: three patients with obstetric complications requiring Caesarean sections (Suller Marti et al., 2019). To support their data, the authors cited a paper by Judkins et al discussing the possible relationship between obstetric complications and the modified uterine input received from the vagus nerve (Judkins et al., 2018). In the immediate postpartum period, no problems were reported in babies born at term. In both preterm babies, the birth weight and length were low and incubation was required. Both mothers were VNS non responders (0%, 30% seizure reduction) and were exposed to polytherapy (see Table 3). No major malformation was observed in any of the children. Similarly, no major fetal malformation was reported by Rodríguez-Osorio et al. (Rodríguez-Osorio et al., 2017). In the group reported by Sabers et al., one major malformation was reported (unilateral congenital glaucoma). According to the authors and to the paper published by Tomson et al. this incidence is within the range expected in AED treated women (Sabers et al., 2017; Tomson et al., 2011). From other data discussing the potential negative effect of VNS on fetal development, an experimental study in rabbits failed to demonstrate any teratogenic effects or other abnormalities that could be attributed to VNS (Danielsson and Lister, 2009). Another experimental paper published by Judkins et al. proved that neither pup viability nor number of cells labeled for pro-inflammatory cytokines in nucleus tractus solitarii or hypoglossal motor nucleus was impaired by VNS (Judkins et al., 2018). The length of the follow-up period for VNS-exposed children varies among studies; generally not much attention is paid to problems other than health-related problems. Cases included in EURAP database had 12 months of follow-up care; no longer term follow-up data are available (Sabers et al., 2017). In a paper by Rodríguez-Osorio, three healthy children were followed until the age of 14 months to 4 years. For the oldest child, only a statement about vaginal delivery at 38 weeks of gestation with no complication is provided (Rodríguez-Osorio et al., 2017). The age of the healthy babies followed by Suller Marti et al. varied between 2 months and 12 years. A baby from that study with dysmorphic features, delayed developmental milestones, and possible heart aneurysm was lost to follow-up care (Suller Marti et al., 2019). Neither of these studies provides more extensive data about further psychomotor development, any intellectual problems, or special schooling requirements. In our data, the psychomotor development was normal in the two younger children; however, neither had yet reached school age. In both school-age children, there were some problems with the education process (postponed elementary school due to ADHD and need for special education due to mental delay with autistic disorder). Although epilepsy was reported to be the cause of the planned Caesarean section (in the 34th gestational week, with short-term incubation required) in the mother of the child with ADHD, there was no increase of seizures during the pregnancy. In the child requiring special education, the course of pregnancy and birth were normal, the mother was a VNS responder, and there was even a decrease of seizure frequency during the pregnancy period. There were no signs of VNS malfunction in our patients during pregnancy and no changes were made to the stimulation parameters. The same situation was observed by Suller Marti et al. (Suller Marti et al., 2019); however, Sabers et al reported changed stimulation parameters in 20% of the studied patients (Sabers et al., 2017). Although this indication is not approved by the FDA and no such patient was found in our study, the exceptional problem of VNS 220 system implantation during pregnancy should be noted. Jazebi et al. published a case of a 21year-old primigravid woman who underwent safe and successful VNS implantation and immediate activation of the device for medically intractable seizures in her 32nd week of pregnancy with dramatically improved seizure control and delivery of a healthy baby (Jazebi et al., 2017). Although the study is based on a prospectively built group of patients followed regularly in a super-specialized epilepsy surgery center, some data were obtained retrospectively and this can be considered as a limitation of the study. Conclusions: Based on the number of implanted patients, the presented study of VNS-exposed pregnancies is one of the two largest single-center studies reported to date. The study results confirmed a high rate of obstetric intervention in VNS patients, although the cause was not VNS in any of the patients and can be explained either by the severity of the epilepsy or by obstetric problems. Although there is no evidence for VNS-induced teratogenicity, the higher incidence of VNS-exposed children requiring special education requires attention. No case of VNS malfunction or failure caused by pregnancy or obstetric intervention was found in the studied group. 221 Literature: Ansari S, Chaudhri K, Al Moutaery KA. Vagus nerve stimulation: Indications and limitations. Acta Neurochir Suppl 2007; 97(Pt.2): 81-6. Battino D, Tomson T, Bonizzoni E, et al. Seizure control and treatment changes in pregnancy: observations from the EURAP epilepsy pregnancy registry. Epilepsia 2013; 54(9): 1621–7. Christensen J, Vestergaard C, Hammer Bech B. Maternal death in women with epilepsy. Smaller scope studies. Neurology 2018; 91(18): e1716–20. Danielsson I, Lister L. A pilot study of the teratogenicity of vagus nerve stimulation in a rabbit model. Brain Stim 2009; 2(1): 41-9 . Houser MV, Hennessy MD, Howard BC. Vagal nerve stimulator use during pregnancy for treatment of refractory seizure disorder. Obstet Gynecol. 2010; 115(2 Pt 2):417-9. Husain MM, Stegman D, Trevino K. Pregnancy and delivery while receiving vagus nerve stimulation for the treatment of major depression: a case report. Ann Gen Psychiatry 2005;4:16. Jazebi N, Moghimi N, Lall R, Osadebey E, Ortega-Barnett J, Masel T. Successful implantation and immediate activation of Vagus Nerve Stimulation (VNS) during pregnancy in a patient with intractable epilepsy: A case illustration and review of the literature. J Clin Neurosci. 2017; Aug 42:114-5. Judkins A, Johnson RL, Murray ST, Yellon SM, Wilson CG. Vagus nerve stimulation in pregnant rats and effects on inflammatory markers in the brainstem of neonates. Pediatr Res 2018; 83(2): 514-9. Morrell MJ. Hormones and epilepsy through the lifetime. Epilepsia 1992; 33 (Suppl.4): S49- 61. Ortega-Villalobos M, García-Bazán M, Soláno-Flores LP, Ninomiya-Alarcón JG, GuevaraGurmán R, Wayner MJ. Vagus nerve afferent and efferent innervation of the rat uterus: an electrophysiological and HRP study. Brain Res Bull 1990; 25(3): 365-71. Rodríguez-Osorio X, López-González FJ, Garamendi Í, et al. VNS and pregnancy: A multicentric experience of four cases. Acta Neurol Scand. 2017; 136(4): 372-4. Sabers A, Battino D, Bonizzoni E, et al. Maternal and fetal outcomes associated with vagus nerve stimulation during pregnancy. Epilepsy Res 2017; 137: 159-62. Sabers A, Buchgreitz L, Neuhuber W. Does vagus nerve stimulation influence pregnancy outcomes? Brain Stimul 2018; 11(3): 618-9. Salerno G, Passamonti C, Cecchi A, Zamponi N. Vagus nerve stimulation during pregnancy: an instructive case. Childs Nerv Syst. 2016;32(1):209-11. Sato Y, Hotta H, Nakayama H, Suzuji H. Sympathetic and parasympathetic regulation of the uterine blood flow and contraction in the rat. J Auton Nerv Syst 1996; 59(3): 151-8. Suller Marti A, Mirsattari SM, Steven DA, et al. Experience on the use of vagus nerve stimulation during pregnancy. Epilepsy Res 2019 Oct;156:106186. Epub 2019 Aug 9. Tomson T, Battino D, Bonizzoni E, et al. Dose-dependent risk of malformations with antiepileptic drugs. An analysis of the data from the EURAP epilepsy and pregnancy. Lancet Neurol 2011; 10(7): 609-17. 222 Supported by AZV NV 19-04-00343 PRESEnCE (Prediction of Stimulation Efficacy in Epilepsy) 223 Questions 1 – 3 1. What is the potential mechanism of VNS action influencing pregnancy outcomes and courses? Answer: Generally speaking, the mechanism of VNS action is the modulation of excitability in multiple cortical and subcortical regions of the brain, including components of the central autonomic system such as the hypothalamus. VNS therapy may therefore influence the neuroendocrine activity (e.g. the production of oxytocin and its storage and release from the pituitary gland). Although efferent VNS fibers form only 20% of the vagus nerve trunk and only a minority of these fibers are conveyed to the uterus, the theoretical possibility of VNS-induced alterations of uterine functions cannot be excluded. 2. Should VNS be switched off during pregnancy and childbirth? Answer: Definitely not. The sudden cessation of VNS may lead to increased seizure frequency and severity with potential negative consequences for mother and child. 3. Are there any data supporting the negative effect of VNS on the course of childbirth or on child outcome? Answer: The data from a multicenter study by Sabers et al. (2017) and from less extensive single center studies prove a higher incidence of obstetric intervention in patients with VNS implanted for epilepsy, but this rate is only slightly higher than the reported rate in EURAP (European and International Registry of Antiepileptic Drugs and Pregnancy) for the general population of pregnancies and antiepileptic drugs. To date no study has proved the negative impact of VNS on the child, although attention should be paid to longer term outcomes. 224 Table 1: The course of pregnancy and childbirth Patient Age Conception Interval VNS implantation - birth (months) Gynecological history incl. abortions Course of pregnancy Birth 1. 26 planned 81 negative uneventful planned induced 2. 33 unplanned 10 four spontaneous abortions due to cervix insufficiency uneventful Caesarean section planned (epilepsy) 3. 38 planned 63 negative, one healthy child before VNS implantation placental separation acute Caesarean section 4. 22 planned 23 negative uneventful spontaneous natural 225 Table 2: Child related data Patien t Gestationa l week Birth length(cm) / weight (grams)/ sex Postbirth adaptatio n Breast feeding Presen t age (years) Psychomot. developmen t Schooling 1. 39 49/ 3500/ male no problems yes, 6 months 2 normal no (age) 2. 34 47/ 2150/ male one week incubator no - insuff. lactation 10 ADHD Elementar y school – delayed for ADHD 3. 31 33/ 1315/ male 2 months incubator no - long term incubato r 1 normal (fetal hypotrophy) no (age) 4. 39 48/ 2700/ male no problems no - reason unknow n 10 mental delay, autistic disorder Special school 226 Table 3: Dynamics of VNS effect during pregnancy and childbirth Patient AED before conception AED changes during pregnancy VNS response Changes of stimulation parameters during pregnancy Change of seizure frequency during pregnancy Change of seizure frequency during birth 1. lamotrigine, levetiracetam, carbamazepine, perampanel no NR (40% response, effect on seizure severity) not done decrease none 2. levetiracetam, lamotrigine no NR (30% effect on seizure intensity) not done none seizure free one postbirth week 3. valproic acid, primidone, zonisamide, clonazepam no NR (0%) – VNS explantation later not done slight increase during last trimester none 4. carbamazepine, valproic acid, lamotrigine no R (60%) not done decrease none 227 Annex 15: Brázdil M, Doležalová I, Koriťáková E, Chládek J, Roman R, Pail M, Jurák P, Shaw DJ, Chrastina J. EEG reactivity predicts individual efficacy of vagal nerve stimulation in intractable epileptics. Frontiers in Neurology. 2019;10:392-392. 228 ORIGINAL RESEARCH published: 02 May 2019 doi: 10.3389/fneur.2019.00392 Frontiers in Neurology | www.frontiersin.org 1 May 2019 | Volume 10 | Article 392 Edited by: Fernando Cendes, Campinas State University, Brazil Reviewed by: Francesco Marrosu, University of Cagliari, Italy Roger Walz, Federal University of Santa Catarina, Brazil *Correspondence: Milan Brázdil mbrazd@med.muni.cz Specialty section: This article was submitted to Epilepsy, a section of the journal Frontiers in Neurology Received: 09 December 2018 Accepted: 01 April 2019 Published: 02 May 2019 Citation: Brázdil M, Doležalová I, Koritáková E, Chládek J, Roman R, Pail M, Jurák P, Shaw DJ and Chrastina J (2019) EEG Reactivity Predicts Individual Efficacy of Vagal Nerve Stimulation in Intractable Epileptics. Front. Neurol. 10:392. doi: 10.3389/fneur.2019.00392 EEG Reactivity Predicts Individual Efficacy of Vagal Nerve Stimulation in Intractable Epileptics Milan Brázdil1,2 *, Irena Doležalová1 , Eva Koritáková3 , Jan Chládek2,4 , Robert Roman2 , Martin Pail1 , Pavel Jurák4 , Daniel J. Shaw2 and Jan Chrastina1 1 Departments of Neurology and Neurosurgery, Medical Faculty of Masaryk University, Brno Epilepsy Center, St. Anne’s University Hospital, Brno, Czechia, 2 Behavioral and Social Neuroscience Research Group, CEITEC–Central European Institute of Technology, Masaryk University, Brno, Czechia, 3 Institute of Biostatistics and Analyses, Faculty of Medicine, Masaryk University, Brno, Czechia, 4 Institute of Scientific Instruments, The Czech Academy of Sciences, Brno, Czechia Background: Chronic vagal nerve stimulation (VNS) is a well-established non-pharmacological treatment option for drug-resistant epilepsy. This study sought to develop a statistical model for prediction of VNS efficacy. We hypothesized that reactivity of the electroencephalogram (EEG) to external stimuli measured during routine preoperative evaluation differs between VNS responders and non-responders. Materials and Methods: Power spectral analyses were computed retrospectively on pre-operative EEG recordings from 60 epileptic patients with VNS. Thirty five responders and 25 non-responders were compared on the relative power values in four standard frequency bands and eight conditions of clinical assessment—eyes opening/closing, photic stimulation, and hyperventilation. Using logistic regression, groups of electrodes within anatomical areas identified as maximally discriminative by n leave-one-out iterations were used to classify patients. The reliability of the predictive model was verified with an independent data-set from 22 additional patients. Results: Power spectral analyses revealed significant differences in EEG reactivity between responders and non-responders; specifically, the dynamics of alpha and gamma activity strongly reflected VNS efficacy. Using individual EEG reactivity to develop and validate a predictive model, we discriminated between responders and non-responders with 86% accuracy, 83% sensitivity, and 90% specificity. Conclusion: We present a new statistical model with which EEG reactivity to external stimuli during routine presurgical evaluation can be seen as a promising avenue for the identification of patients with favorable VNS outcome. This novel method for the prediction of VNS efficacy might represent a breakthrough in the management of drug-resistant epilepsy, with wide-reaching medical and economic implications. Keywords: vagal nerve stimulation, neurostimulation, epilepsy, efficacy prediction, EEG reactivity, epilepsy treatment 229 Brázdil et al. EEG Predicts VNS Efficacy INTRODUCTION Resective surgery is currently the best therapeutic option for treatment of patients with drug-resistant epilepsy, but a substantial number of intractable patients remains who are ineligible for such treatment or for whom resective surgery fails to abolish seizures. Chronic vagal nerve stimulation (VNS) has become a well-established alternative, offering a palliative method of treatment for drug-resistant epilepsy; it rarely results in complete seizure freedom (∼5% of treated patients), but provides substantial (≥50%) seizure reduction in 50–60% of individuals. Unfortunately, however, seizure frequency remains unchanged after VNS therapy in ∼25% of patients (1, 2). Identifying individuals who will benefit from VNS therapy prior to the implantation of the VNS device would improve patient selection, minimize unnecessary surgical procedures, and reduce associated financial expenses dramatically; and yet there exists no method with which to predict individual efficacy pre-intervention (2). Achieving a pre-operative classification of individual patients as VNS responders or non-responders (i.e., patients with ≥50% or <50% seizure reduction, respectively) would represent a major breakthrough in the treatment of drugresistant epilepsy. It is presumed that VNS increases seizure threshold by activating neuronal networks in the thalamus and other limbic structures (3, 4), but the precise mechanism of VNS action is not yet understood fully. Both synchronization and desynchronization of the electroencephalogram (EEG) has been proposed as a possible mechanism behind the antiepileptic effect of VNS (5, 6), and recent neurophysiological studies focusing on EEG parameters lend support to this: Fraschini et al. report a significant correlation between VNS-induced global desynchronization in gamma bands and positive clinical outcome in temporal lobe epilepsy patients (7). Similarly, Bodin et al. revealed a lower level of global EEG synchronization in delta and alpha frequency bands during the ON phase of VNS in responders (8). Theoretically, differential alterations in brain rhythms from VNS therapy between responders and non-responders might reflect inter-individual variability in the (non-specific) susceptibility of EEG to be synchronized or desynchronized by external stimulation. It follows that differences in this susceptibility might underlie individual VNS efficacy. We tested the hypothesis that EEG reactivity to standard external stimuli used during routine pre-operative EEG assessment differs between VNS responders and non-responders, with the aim of developing a reliable statistical model for prediction of VNS efficacy. MATERIALS AND METHODS Study Design We performed retrospective analyses of EEG data collected from all adult patients implanted with a VNS device for drugresistant epilepsy in the Brno Epilepsy Center between 2005 and 2015. Data from patients implanted between 2005 and 2012 were used for investigation of EEG reactivity to external stimuli and subsequently for development of the statistical model (Cohort 1). Additional data from patients implanted with VNS between 2013 and 2015 were used as independent data-set for validation of the statistical model (Cohort 2). All the data were acquired during routine outpatient pre-operative assessment, 20 min recording at morning with two standard eyes opening and closing activation procedures (i.e., 10 s period with eyes open), photic stimulation (PS), and hyperventilation (HV). Each EEG recording was filtered into individual frequency bands and segmented into specific intervals representing the eyes opening and closing, PS and HV periods. The relative powers of EEG spectrum in distinct time intervals for a particular frequency band and brain area were then calculated. Based on their individual responses to VNS, patients were categorized as Responders or Non-responders. In Cohort 1, Responders and Non-responders were first compared on the relative power values, and then a statistical model for prediction of VNS efficacy was developed. Subsequently, the validity of the statistical model was tested in Cohort 2. The study was conducted in St. Anne’s University Hospital and approved by the local ethics committee. All patients gave their informed written consent for the use of their pre-operative data. Patients’ Description All patients were implanted with a VNS system (Cyberonics, Houston) according to a standard implantation procedure (9). Before implantation, all patients underwent a comprehensive assessment protocol for epilepsy surgery candidates, including a detailed history and neurological examination, magnetic resonance imaging (MRI), interictal PET, neuropsychological testing, and scalp video-EEG monitoring. In some patients, ictal and interictal SPECT (SISCOM) and invasive video-EEG monitoring have been completed if necessary. Based on the results of all the investigations, patients indicated for VNS and included in this study were ruled out as suitable candidates for resective epilepsy surgery. Our analyses were applied to data acquired from patients who fulfilled the following criteria: The duration of VNS treatment was at least 2 years; the efficacy of VNS treatment was determined in regular visits every 3 or 6 months; and artifact-free pre-operative interictal EEG was available for eye opening and closing, PS and HV periods. Demographic information and data regarding the type and number of antiepileptic drugs (AEDs) at the time of implantation were obtained by review of patients’ charts. The efficacy of VNS was categorized using a classification system reported by McHugh (10). The cut-off value for seizure-reduction between responders and non-responders was 50%. Patients were defined as Responders or Non-responder only if they were categorized as such for the entire follow-up period. EEG Analysis First, we compared relative EEG powers between Responders and Non-responders in Cohort 1. Interictal scalp EEG was recorded on a 64-channel Alien Deymed system with international 10–20 electrode placement and a sampling frequency of 128 Hz. Standard antialiasing filters were applied before digitalization. Occasional artifacts were rejected manually and further Frontiers in Neurology | www.frontiersin.org 2 May 2019 | Volume 10 | Article 392 230 Brázdil et al. EEG Predicts VNS Efficacy processing was performed with artifact-free EEG periods. The resulting EEG signals were filtered into four frequency bands: theta (4–7.5 Hz), alpha (8–12 Hz), beta (14–30 Hz), and gamma (31–45 Hz). A Hilbert transform was then used to estimate the envelopes of pre-defined pass-band frequency oscillations as a function of time (Figure 1). The EEG records were segmented into the following conditions (time intervals): 1 –Rest#1 (2 min) 2 –Eyes opening/closing (10 s) 3 –Rest#2 (immediately after eye closure; 10 s) 4 –Photic stimulation (PS; 2.5 min) 5 –Hyperventilation (HV; 4 min) 6 –Eyes opening/closing (10 s) 7 –Rest#3 (immediately after eye closure; 10 s) 8 –Rest#4 (2 min) Further analysis was focused on oscillatory power changes in these conditions. Absolute mean power of the EEG spectrum was computed as a mean value of the passband power envelope inside each interval separately, for each scalp electrode. Subsequently, relative mean power (RPW) was calculated as a percentage decrease or increase of mean power relative to baseline, i.e., event-related desynchronization or synchronization, respectively. As a baseline we selected Rest#1 (11). We then evaluated differences between Responders and Non-responders by comparing relative power in seven conditions of clinical assessment: Open/Close#1, Rest#2, PS, HV, Open/Close#2, Rest#3, and Rest#4. Statistical Analysis Demographic data were compared between Responders and Non-responders using Fisher’s exact test or the Mann-Whitney test. Statistical comparisons of RPW between Responders and Non-responders were also performed with Mann-Whitney tests. Using false discovery rate (FDR) (12), p-values for all electrodes were corrected for multiple comparisons in each time interval separately. Differences were considered significant when p ≤ 0.05. Prediction of Response to VNS Developing the Statistical Model The statistical model for prediction of VNS efficacy was developed from data obtained in Cohort 1 in the following steps: Firstly, electrodes were grouped into seven anatomical regions as follows: (1) left frontal—Fp1, F3, Fz; (2) right frontal—Fp2, F4, Fz; (3) left anterotemporal—F7, T3; (4) right anterotemporal—F8, T4; (5) central—C3, Cz, C4; (6) left posterior quadrant—P3, Pz, T5, O1; (7) right posterior quadrant—P4, Pz, T6, O2. Specifically, mean values were calculated for the respective groups of electrodes. This resulted FIGURE 1 | Pre-processing of EEG signal. The EEG was segmented into eight time intervals, and then filtered into four frequency bands: theta, alpha, beta, and gamma. Subsequent analyses focused on the oscillatory power changes within these intervals, which were analyzed separately for each frequency band. Frontiers in Neurology | www.frontiersin.org 3 May 2019 | Volume 10 | Article 392 231 Brázdil et al. EEG Predicts VNS Efficacy in 196 electrode group variables (7 conditions × 4 frequency bands × 7 anatomical regions). In the second step, the maximally discriminative electrode group variables were then selected using stepwise logistic regression performed in leave-one-out (LOO) manner; specifically, with n patients, logistic regression was performed over n iterations, each with one subject omitted. This approach allowed us to avoid overestimating the classification results, which occurs when classification is performed on electrode groups selected using all subjects simultaneously. Thirdly, electrode group variables identified most frequently as maximally discriminative after the n LOO iterations were used for classification using three classifiers; namely, logistic regression (LR), linear support vector machines (SVM), and linear discriminant analysis (LDA). In this step, we used a LOO cross-validation to split the data into training and testing sets: one patient was chosen randomly as a testing subject and the remaining patients were employed for training the classifier. The testing subject was then classified as belonging to the Responder or to the Non-responder class, and the resulting class label was compared with the true classification label. This procedure was repeated using each of the subjects as the testing subject sequentially, and the overall classification performance measures of accuracy, sensitivity, and specificity were calculated (see the black schematics in Figure 2). Fourth, using one-sample binomial test we performed a comparison of the achieved classification accuracies with a classification by chance. We also attempted to predict VNS efficacy based on 532 possible single electrode variables (7 conditions × 4 frequency bands × 19 electrodes; see Supplementary Material 1). Validation of the Statistical Model The validity of the aforementioned statistical model was verified on the independent data-set obtained from Cohort 2. This validation cohort consisted of patients suffering from drugresistant epilepsy who were implanted with VNS in our Center between 2013 and 2015. We obtained EEG data recorded using an identical EEG system and pre-surgical assessment protocol, and processed mathematically in the exact same way described above. Patients were classified as Responders and Nonresponders with our statistical model using two approaches (see the gray schematics in Figure 2). Both began with data reduction, whereby we selected only data corresponding to the maximally discriminative electrode group variables identified based on Cohort 1. In the first approach, this data reduction was followed subsequently by a classification of m subjects from Cohort 2 as Responders or Non-responders based on a single classifier, which was trained on all n subjects from the Cohort 1. In contrast, in the second approach we classified the reduced data of Cohort 2 using majority voting of n classifiers trained on n-1 subsets of Cohort 1. The resulting indices for accuracy, sensitivity, and specificity were then compared to those from the classification of Cohort 1. RESULTS Participants Cohort 1 The VNS device was implanted in 110 patients in our center between 2005 and 2012. We excluded 50 patients (45%) from further analyses−17 (15%) because of poor-quality pre-surgical FIGURE 2 | The statistical model for prediction of VNS efficacy based on EEG analysis. Schematics representing the development of the statistical model on data from Cohort 1 outlining the selection of the most discriminative electrodes groups as well as classifier training (CT) and testing (T) performed in LOO manner (the black schematics); and two approaches taken for the validation procedure, whereby our statistical model was applied to an independent data-set from Cohort 2 (the gray schematics). Equivalent steps taken in each of the two validation approaches are visualized using solid lines, including the reduction of Cohort 2 by selecting the maximally discriminative electrodes identified in Cohort 1. The two approaches differ in the steps represented by dashed lines—i.e., the classification (C) of the reduced dataset using: (A) a single classifier trained on all n subjects from Cohort 1, and (B) voting of n classifiers trained on n-1 subsets of Cohort 1. Frontiers in Neurology | www.frontiersin.org 4 May 2019 | Volume 10 | Article 392 232 Brázdil et al. EEG Predicts VNS Efficacy EEG, 13 (12%) due to attrition, and 20 (18%) who switched between VNS outcomes during the follow-up period (see Supplementary Material 2 for details). Demographics for the 60 patients included in the analyses are summarized in Table 1. According to the response to VNS, patients were subdivided into 35 (58%) Responders and 25 (42%) Non-responders. When examining demographic data, we observed significant differences between Responders and Non-responders in patients’ age at epilepsy onset, duration of epilepsy before VNS implantation, and treatment by valproic acid (Table 1). Cohort 2 In total, the VNS device was implanted in 25 patients between 2013 and 2015. Three patients were excluded from our analyses due to poor-quality pre-surgical EEG recordings. Data from the remaining 22 patients were used for independent validation: 12 (54.5%) Responders and 10 (45.5%) Non-responders. The demographic data are summarized in Table 1. The Cohort 2 did not differ from the Cohort 1 in the demographic characteristics, apart from the anticipated difference in duration of VNS (p < 0.001) and subtle differences in antiepileptic treatment (specifically, eslicarbazepine, lacosamide, and zonisamide were more frequent in Cohort 2; p = 0.029, p = 0.003, and p = 0.012, respectively). RPW Differences Between Responders and Non-responders In Cohort 1, computation of RPWs and their dynamics in the recording sites revealed significant changes for both Responders TABLE 1 | Demographic and treatment data for Cohort 1 and Cohort 2. Cohort 1 Cohort 2 Combined (n = 60) Non-responders (n = 25) Responders (n = 35) p Combined (n = 22) Non-responders (n = 10) Responders (n = 12) p Type of epilepsy, n (%) TLE 14 (23) 6 (24) 8 (23) 1.000 6 (27) 3 (30) 3 (25) 0.801 Extra-TLE 43 (72) 18 (72) 25 (72) 15 (68) 6 (60) 9 (75) IGE 3 (5) 1 (4) 2 (6) 1 (5) 1 (10) 0 (0) Gender, n (%) Females 34 (57) 17 (68) 17 (49) 0.188 11 (50) 4 (40) 7 (58) 0.670 Males 26 (43) 8 (32) 18 (51) 11 (50) 6 (60) 5 (42) Age (years) at VNS implantation (median, min-max) 33 (15–65) 30 (15–65) 36 (18–63) 0.134 31 (22–71) 26 (22–42) 40 (22–71) 0.069 Age (years) at epilepsy onset (median, min-max) 9 (1–51) 13 (1–27) 6 (1–51) 0.014 10 (0–59) 8 (0–18) 12 (4–59) 0.123 Duration (years) of epilepsy before vagal nerve stimulator implantation (median, min-max) 22 (4–60) 15 (4–55) 26 (7–60) 0.019 20 (2–49) 20 (14–34) 19 (2–49) 0.872 Duration (years) of VNS (median, min-max) 6 (3–11) 6 (3–10) 6 (3–11) 0.581 3 (2–3) 3 (2–3) 3 (2–3) 0.628 Treatment at the time of VNS implantation, n (%) BRV 2 (3) 2 (8) 0 (0) 0.169 CBZ 32 (53) 15 (60) 17 (49) 0.439 9 (41) 3 (30) 6 (50) 0.415 CLB 1 (2) 1 (4) 0 (0) 0.417 CLZ 13 (22) 6 (24) 7 (20) 0.758 4 (18) 3 (30) 1 (8) 0.293 ESL 3 (5) 2 (8) 1 (3) 0.565 5 (23) 4 (40) 1 (8) 0.135 GBP 1 (2) 0 (0) 1 (3) 1.000 LCM 6 (10) 1 (4) 5 (14) 0.386 9 (41) 6 (60) 3 (25) 0.192 LEV 36 (60) 16 (64) 20 (57) 0.790 10 (45) 4 (40) 6 (50) 0.691 LTG 27 (45) 11 (44) 16 (46) 1.000 9 (41) 4 (40) 5 (42) 1.000 PGB 5 (8) 2 (8) 3 (9) 1.000 1 (5) 0 (0) 1 (8) 1.000 PHE 1 (2) 0 (0) 1 (3) 1.000 PHT 4 (7) 1 (4) 3 (9) 0.634 1 (5) 1 (10) 0 (0) 0.455 PRM 3 (5) 1 (4) 2 (6) 1.000 2 (9) 1 (10) 1 (8) 1.000 TPM 13 (22) 5 (20) 8 (23) 1.000 2 (9) 1 (10) 1 (8) 1.000 VPA 14 (23) 2 (8) 12 (34) 0.028 5 (23) 3 (30) 2 (17) 0.624 ZNS 8 (13) 5 (20) 3 (9) 0.259 9 (41) 4 (40) 5 (42) 1.000 Number of AEDs used at the time of VNS implantation, n (%) 1 4 (7) 1 (4) 3 (9) 0.974 1 (5) 0 (0) 1 (8) 0.495 2 17 (28) 8 (32) 9 (26) 7 (32) 3 (30) 4 (33) 3 26 (43) 11 (44) 15 (43) 7 (32) 2 (20) 5 (42) 4 12 (20) 5 (20) 7 (20) 5 (23) 3 (30) 2 (17) 5 1 (2) 0 (0) 1 (3) 2 (9) 2 (20) 0 (0) AEDs, antiepileptic drugs; BRV, brivaracetam; CBZ, carbamazepine; CLB, clobazam; CLZ, clonazepam; ESL, eslicarbazepine; Extra-TLE, extratemporal lobe epilepsy; GBP, gabapentin; LCM, lacosamide; LEV- levetiracetam; LTG, lamotrigine; NA, not applicable; PGB, pregabalin; PHE, phenobarbital; PHT, phenytoin; PRM, primidone; TLE, temporal lobe epilepsy; TPM, topiramate; VNS, vagal nerve stimulation; VPA, valproic acid; ZNS, zonisamide. Frontiers in Neurology | www.frontiersin.org 5 May 2019 | Volume 10 | Article 392 233 Brázdil et al. EEG Predicts VNS Efficacy and Non-responders in all investigated frequency bands, all defined conditions, and the majority of electrodes (Figure 3). These changes were largely equivalent for both Responders and Non-responders within specific frequency bands and conditions. In the alpha and gamma frequency bands, however, there were striking dissimilarities between the two patient groups in conditions, with the most prominent differences during photic stimulation (PS) and hyperventilation (HV). Comparing Responders and Non-responders with respect to the RPW in all pre-defined frequency bands and conditions, we revealed significant differences in alpha and gamma frequency bands (Figure 4). When analyzing the alpha band, significant differences between Responders and Non-responders were found in the following conditions: 4-PS, 5-HV, and 8–Rest#4. The differences were present over different brain regions in each condition: In 4-PS, there were differences over central and anterior areas (absence of desynchronization in Responders); in the 5-HV interval we observed differences over central and posterior regions (significantly higher synchronization in Responders); and in 8-R, differences were localized to right anterior and left posterior areas (higher power in Responders but persisting desynchronization in Non-responders). When focusing on the gamma frequency band, we observed significant differences (gamma synchronization in Responders and desynchronization in Non-responders) in RPW within 4PS, 5-HV and 7-Rest#3: In the first two conditions, differences were distributed across almost the whole scalp. The differences in RPWs within 7-Rest#3 were observed within right anterior and left posterior areas. Prediction of VNS Response—Statistical Model Based on the EEG data of Cohort 1, the statistical model was developed. Eight groups of electrodes were selected as the most discriminative in this statistical model (visualized in Figure 5). The best classification results based on these eight most discriminative groups of electrodes were obtained using the LR classifier, achieving 86.7% accuracy, with 88.6% sensitivity and 84% specificity (Table 2). This classification accuracy was significantly higher than those achieved by chance (p < 0.001). The SVM classifier achieved lower accuracy (75%) but it was still significantly higher than that achieved by chance (p = 0.004). The lowest classification performance measures were obtained in classification using LDA (accuracy 65%). Detailed visualizations of classification accuracy are provided in Figure 6. FIGURE 3 | Relative mean powers (RPWs) for all frequency bands. Each head represents RPW in a given condition and frequency band. The RPW percentage scale is displayed on the right-hand side. Where a statistically significant difference exists in a given condition relative to baseline (Rest#1), the electrode is marked by a white dot. If no significant difference exists, the electrode is marked by a black dot. Frontiers in Neurology | www.frontiersin.org 6 May 2019 | Volume 10 | Article 392 234 Brázdil et al. EEG Predicts VNS Efficacy FIGURE 4 | Relative mean power (RPW) differences between Responders and Non-responders. Statistically significant or non-significant differences in a given electrode are marked by a white or black dot, respectively. The significant differences between Responders and Non-responders were identified in alpha and gamma frequency ranges only (heads in red boxes). The final statistical model is described in more detail in Supplementary Material 3. Validation of the Statistical Model The data of patients in Cohort 2 were used for independent validation. When comparing the results of our classification model applied to Cohort 2 against real-life outcome, the best results were again obtained for the model using the LR classifier achieving an accuracy of 86.4%, sensitivity of 83.3%, and specificity of 90%. The complete results achieved using classifier voting are summarized in Table 2. Classification performance based on a single classifier was lower (data not shown). DISCUSSION Over 80,000 epilepsy patients have been treated worldwide using vagal nerve stimulation (VNS), a standard of modern nonpharmacological treatment. Importantly, the number of epileptic patients who are eligible for VNS therapy is approximately three million subjects worldwide. In this light, a method for the reliable prediction of individual efficacy of VNS therapy is desperately needed. Recently, several authors have attempted to identify predictors of VNS outcome (1, 2, 4, 13). Despite these efforts, however, clinical predictors of individual responsiveness to VNS therapy remain elusive. Similar estimates of efficacy are reported for diverse neurostimulation techniques in the treatment of drugresistant epilepsy (i.e., VNS, Deep Brain Stimulation of Anterior Thalamic Nuclei, Brain Responsive RNS System, transcutaneous VNS, or external Trigeminal Nerve Stimulation), which might suggest a more consequential impact of external stimuli per se on epileptic activity. For this reason, we retrospectively evaluated routine EEG data acquired before implantation in a large cohort of VNS patients. Using standard computations of power spectral analyses of interictal EEG, we reveal significant differences between responders and non-responders in two pre-defined frequency bands (alpha and gamma) and four conditions of standard clinical assessment. Based on RPWs and their dynamics, we have developed and validated a statistical model for prediction of VNS efficacy that discriminated between responders and nonresponders with almost 90% accuracy. Our primary finding is that VNS responders and nonresponders differ significantly in EEG power dynamics within alpha and gamma frequency bands prior to therapy. Whilst both patient groups demonstrated equivalent alpha desynchronization during eyes opening, they differed in alpha reactivity to photic stimulation and hyperventilation; specifically, responders showed no decrease in alpha power during the former but Frontiers in Neurology | www.frontiersin.org 7 May 2019 | Volume 10 | Article 392 235 Brázdil et al. EEG Predicts VNS Efficacy FIGURE 5 | Electrode groups and individual electrodes selected for statistical modeling. Electrode groups (A) and individual electrodes (B) selected for the development of the statistical model, in each condition and frequency band. TABLE 2 | Classification performance indices for electrode group variables. Cohort 1−60 patients Cohort 2−22 patients Accuracy Sensitivity Specificity p Accuracy Sensitivity Specificity p LR 86.7 88.6 84.0 <0.001 86.4 83.3 90.0 0.006 SVM 75.0 62.9 92.0 0.004 77.3 58.3 100.0 0.043 LDA 65.0 48.6 88.0 0.147 45.5 16.7 80.0 0.879 LDA, linear discriminant analysis; LR, logistic regression; p, p-value calculated using one-sample binominal test; SVM, linear support vector machines. an enormous increase during the latter. This reactivity pattern stands in contrast to that observed in healthy individuals, in whom photic stimulation typically leads to alpha attenuation and standardized hyperventilation has been shown to decrease alpha power (14). Interestingly, we also observed significant increases in gamma power during both photic stimulation and hyperventilation in responders relative to non-responders. Hyperventilation-induced physiological changes are thought to be a consequence of increased neuronal excitability resulting from the hypocapnia-induced alkalosis (15, 16). Significantly enhanced alpha and gamma activities during hyperventilation in responders might reflect distinct properties of responders’ brains vis-a-vis neuronal excitability and synaptic transmission. Nevertheless, hyperventilation is a complicated practice which, in epileptic subjects, results in unpredictable responses. The reasons for this remain unclear (e.g., metabolic hypersynchronization, altered neurotransmitters, etc.). It also remains unclear why the alpha (and less expressed gamma) desynchronization persist in our non-responders at the end of EEG recordings (Rest#4). It seems that the protocol itself, particularly photic stimulation and hyperventilation, induces some change from the restingstate baseline for non-responders. Differential dynamics of power changes after the stimulation—faster in responders and slower in non-responders—might represent another characteristic in which these groups of patients differ substantially. Interestingly, valproic acid (VPA) was used more frequently in pharmacological treatment prior to implantation in our VNS responders compared to the non-responders. We might therefore ask whether differences in EEG reactivity observed in our study is related to some pharmacological imprint. Although this speculative explanation cannot be fully excluded, it seems unlikely that pharmacological impact on EEG is the main factor driving our findings. Still more unlikely is substantial VPA impact on the prediction of VNS efficacy, bearing in mind that only one third of responders were treated with VPA and the accuracy of individual VNS efficacy prediction was almost 90%. Frontiers in Neurology | www.frontiersin.org 8 May 2019 | Volume 10 | Article 392 236 Brázdil et al. EEG Predicts VNS Efficacy FIGURE 6 | Classification accuracy based on a number of selected groups of electrodes. The bottom row of numbers shows a minimum number of LOO iterations (# iterations) in which a given number of electrode groups (# groups) were selected as most discriminative (e.g., eight electrode groups were selected as maximally discriminative in 35 or more LOO iterations). The best LR classification accuracy and corresponding results for SMV and LDA (shown in Table 2) are depicted by circles and a dashed line. Since Hans Berger’s initial observation in 1933, the best known example of EEG reactivity is alpha attenuation (alpha blocking and desynchronization). This is observed typically when subjects open their eyes (the Berger effect), but alpha also disappears when subjects become drowsy and it can be blocked by numerous kinds of external stimuli (visual or auditory) or mental operations (e.g., imagery, visualization, mental arithmetics). This implies that alpha reactivity as a modality-independent, general phenomenon, reflecting the functional modes of thalamo-cortical and cortico-cortical loops that facilitate/inhibit the transmision of information in the brain (17, 18). Particularly noteworthy is the high inter-individual variability in alpha reactivity; it has been shown to differ between extraverts and introverts, for example, and there is less pronounced alpha desynchronization in people with high intelligent quotient during several cognitive tasks (19, 20). Alpha reactivity has also been reported to decrease with aging (21–24) and in patients with mild cognitive impairment and Alzheimer disease (25), and a lack of alpha reactivity has been used to predict the long-term deterioration of higher functions in subjects with cognitive decline (26). From this viewpoint, our discovery of differential alpha reactivity in VNS responders vs. non-responders further emphasizes the multifold functions of the diffuse alpha system (27). In our study, significant differences in EEG power dynamics within the gamma frequency band between VNS responders and non-responders very likely reflects distinct reactivity of true brain gamma. When interpreting thoroughly our results in this frequency band, however we shall keep in mind the study of Whitham et al. (28); these authors showed recently that even the normal resting EEG might reveal significant contamination with electromyography (EMG) activity in this frequency band. Further, the level of EMG contamination increased dramatically when subjects perform various experimental tasks and are sitting during EEG recordings (29). More recently, however, Boytsova et al. showed that EMG contamination does not necessarily hide high-frequency EEG and does not preclude qualitative detections of electroencephalographic correlates of mental activities in beta and low gamma frequency ranges (30). Unfortunately, despite the availability of several techniques for the reduction of muscular artifacts from EEG traces, at a present none are able to guarantee that analyzed data are completely free of high-frequency artifacts (29). In our study, EEG was recorded under standard conditions: patients laid comfortably and were instructed to relax (including their facial muscles), no task was performed, and recordings from patients containing artifacts identified with careful visual inspection (e.g., muscular activity) were excluded (17 out of 110 patients). As such, we strongly believe an increased gamma power during both photic stimulation and hyperventilation in responders relative to non-responders is not resulting from distinct muscle artifacts contaminations. Both brain rhythms—alpha and gamma—are considered to represent a kind of universal code consistent with their putative role in brain signaling (27). Both are generated in a widely distributed system, with a major role of thalamocortical circuits in their origin (31–33). The differential impact of external stimuli on alpha and gamma between VNS responders and non-responders might be mediated by differences in neuronal interconnectivity and different levels of neurotransmitters within underlying cerebral matrices (24). Such differences might influence the effect of external stimuli delivered to thalamocortical circuits and other brain networks via the vagal nerve (34, 35). Consistent with this notion, many consider the mechanism of VNS action to be modulation of synaptic activity in the thalamus and thalamocortical projections, increased plasticity in GABA receptors, and modulation of GABAergic activity that is related directly to Frontiers in Neurology | www.frontiersin.org 9 May 2019 | Volume 10 | Article 392 237 Brázdil et al. EEG Predicts VNS Efficacy gamma oscillations (6, 36). Indeed, one of the most plausible factors involved in gamma variation between our populations of responders vs. non-responders is represented by the role of GABA neurotransmission and divergent dynamics of inhibitory interneuron networks within the central nervous system (37, 38). The close relationship between different types of external stimulation and brain reactivity we have observed might be indicative of a common mechanism underlying the various forms of neurostimulation in epilepsy treatment. Finally, as seen in the developed and validated statistical model, for accurate prediction of VNS efficacy distinct recording electrode groups, different conditions (especially hyperventilation, but also eyes opening/closing and resting periods), and all frequency bands were selected as the most discriminative ones. Using our approach we achieved high accuracy, sensitivity and specificity in both well-defined and predominant VNS patients. We also evaluated a statistical model based on single electrodes, which had slightly superior classification performance in Cohort 1 and predominant VNS patients but failed in the independent validation set of Cohort 2. This indicates that groups of electrodes are better suited for VNS efficacy prediction—by integrating data from larger regions, such groupings appear to produce more robust results. To conclude, we have revealed that EEG reactivity to external stimuli used during routine pre-operative EEG investigation differs between VNS responders and non-responders. Moreover, we have developed and validated a statistical tool that can predict with extremely high accuracy whether or not individual drug-resistant epileptic patients will benefit from VNS treatment (Patent Number EP3437692-A1). This electrophysiological marker could prove invaluable when providing patients with expected postoperative prognosis. Further research is required before this can be achieved, however; our findings come from a retrospective and monocenter study, with a limited number of patients comprising the independent validation. Our results must therefore be replicated in prospective, multicentre, and well-designed clinical study in order to obtain a clear-cut statistical power. ETHICS STATEMENT Ethic committee in St. Anne’s University hospital approved this study. All patients gave their informed consent. AUTHOR CONTRIBUTIONS MB, RR, PJ, JChr, and DS: identification of research topic, preparation of study design, interpretation of results. ID and MP: preparation of manuscript, design of study, interpretation of results. JChl and PJ: mathematical analysis. EK: statistical analysis. FUNDING The results of this research have been acquired within CEITEC 2020 (LQ1601) project with financial contribution made by the Ministry of Education, Youth and Sports of the Czech Republic (MEYS CR) within special support paid from the National Programme for Sustainability II funds, MEYS CR project LO1212, and the Ministry of Health of the Czech Republic, grant nr. NV19-04-00343. ACKNOWLEDGMENTS We would like to thank Matthew C. Walker for his valuable help and advice with manuscript. SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fneur. 2019.00392/full#supplementary-material REFERENCES 1. Englot DJ, Chang EF, Auguste KI. Vagus nerve stimulation for epilepsy: a meta-analysis of efficacy and predictors of response. J Neurosurg. (2011) 115:1248–55. doi: 10.3171/2011.7.JNS11977 2. Englot DJ, Rolston JD, Wright CW, Hassnain KH, Chang EF. Rates and predictors of seizure freedom with vagus nerve stimulation for intractable epilepsy. Neurosurgery. (2016) 79:345–53. doi: 10.1227/NEU.0000000000001165 3. Theodore WH, Fisher RS. Brain stimulation for epilepsy. Lancet Neurol. (2004) 3:111–8. doi: 10.1016/S1474-4422(03)00664-1 4. Alexander GM, McNamara JO. Vagus nerve stimulation elevates seizure threshold in the kindling model. Epilepsia. (2012) 53:2043–52. doi: 10.1111/j.1528-1167.2012.03646.x 5. Jaseja H. EEG-desynchronization as the major mechanism of anti-epileptic action of vagal nerve stimulation in patients with intractable seizures: clinical neurophysiological evidence. Med Hypotheses. (2010) 74:855–6. doi: 10.1016/j.mehy.2009.11.031 6. Marrosu F, Santoni F, Puligheddu M, Barberini L, Maleci A, Ennas F, et al. Increase in 20-50 Hz (gamma frequencies) power spectrum and synchronization after chronic vagal nerve stimulation. Clin Neurophysiol. (2005) 116:2026–36. doi: 10.1016/j.clinph.2005.06.015 7. Fraschini M, Puligheddu M, Demuru M, Polizzi L, Maleci A, Tamburini G, et al. VNS induced desynchronization in gamma bands correlates with positive clinical outcome in temporal lobe pharmacoresistant epilepsy. Neurosci Lett. (2013) 536:14–8. doi: 10.1016/j.neulet.201 2.12.044 8. Bodin C, Aubert S, Daquin G, Carron R, Scavarda D, McGonigal A, et al. Responders to vagus nerve stimulation (VNS) in refractory epilepsy have reduced interictal cortical synchronicity on scalp EEG. Epilepsy Res. (2015) 113:98–103. doi: 10.1016/j.eplepsyres.2015.03.018 9. Landre E. Vagus nerve stimulation and refractory partial epilepsies. Rev Neurol. (2004) 160:S280–7. 10. McHugh JC, Singh HW, Phillips J, Murphy K, Doherty CP, Delanty N. Outcome measurement after vagal nerve stimulation therapy: proposal of a new classification. Epilepsia. (2007) 48:375–8. doi: 10.1111/j.1528-1167.2006.00931.x 11. Pfurtscheller G, Aranibar A. Event-related cortical desynchronization detected by power measurements of scalp EEG. Electroencephalogr Clin Neurophysiol. (1977) 42:817–26. Frontiers in Neurology | www.frontiersin.org 10 May 2019 | Volume 10 | Article 392 238 Brázdil et al. EEG Predicts VNS Efficacy 12. Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J R Stat Soc B. (1995) 57:289–300. 13. Liu H, Yang Z, Huang L, Qu W, Hao H, Li L. Heart-rate variability indices as predictors of the response to vagus nerve stimulation in patients with drug-resistant epilepsy. Epilepsia. (2017) 58:1015–22. doi: 10.1111/epi.13738 14. Van der Worp HB, Kraaier V, Wieneke GH, Van Huffelen AC. Quantitative EEG during progressive hypocarbia and hypoxia. Hyperventilationinduced EEG changes reconsidered. Electroencephalogr Clin Neurophysiol. (1991) 79:335–41. 15. Wyke BD. Brain function and blood sugar: observations based on a case of islet cell adenoma of the pancreas. Electroencephalogr Clin Neurophysiol. (1952) 4:339–50. 16. Neurbauer AC, Sange G, Pfurtscheller G. Psychometric intelligence and event-related desynchronization during performance of a letter matching task. In: Pfurtscheller G, Lopes da Silva FH, editors. Handbook of Electroencephalography nad Clinical Neurophysiology. Amsterdam: Willy (1999). p. 219–31. 17. Steriade M, Llinas RR. The functional states of the thalamus and the associated neuronal interplay. Physiol Rev. (1988) 68:649–742. 18. Pfurtscheller G, Lopes da Silva FH. Event-related EEG/MEG synchronization and desynchronization: basic principles. Clin Neurophysiol. (1999) 110:1842–57. 19. Venturini R, De Pascalis V, Imperiali MG, Martini PS. EEG alpha reactivity and extraversion-introversion. Pers Indiv Differ. (1981) 2:215–20. 20. Neubauer A, Freudenthaler HH, Pfurtscheller G. Intelligence and spatiotemporal patterns of event-related desynchronization (ERD). Intelligence. (1995) 20:249–66. 21. Duffy FH, Albert MS, McAnulty G. Brain electrical activity in patients with presenile and senile dementia of the Alzheimer type. Ann Neurol. (1984) 16:439–48. 22. Kononen M, Partanen JV. Blocking of EEG alpha activity during visual performance in healthy adults. A quantitative study. Electroencephalogr Clin Neurophysiol. (1993) 87:164–6. 23. Marciani MG, Gotman J. Effect of drug-withdrawal on location of seizure onset. Epilepsia. (1986) 27:423–31. 24. Gaal ZA, Boha R, Stam CJ, Molnar M. Age-dependent features of EEGreactivity–spectral, complexity, and network characteristics. Neurosci Lett. (2010) 479:79–84. doi: 10.1016/j.neulet.2010.05.037 25. Babiloni C, Lizio R, Vecchio F, Frisoni GB, Pievani M, Geroldi C, et al. Reactivity of cortical alpha rhythms to eye opening in mild cognitive impairment and Alzheimer’s disease: an EEG study. J Alzheimers Dis. (2010) 22:1047–64. doi: 10.3233/JAD-2010-100798 26. van der Hiele K, Bollen EL, Vein AA, Reijntjes RH, Westendorp RG, van Buchem MA, et al. EEG markers of future cognitive performance in the elderly. J Clin Neurophysiol. (2008) 25:83–9. doi: 10.1097/WNP.0b013e31816a5b25 27. Basar E, Schurmann M, Basar-Eroglu C, Karakas S. Alpha oscillations in brain functioning: an integrative theory. Int J Psychophysiol. (1997) 26:5–29. 28. Whithain EM, Pope KJ, Fitzgibbon SP, Lewis T, Clark CR, Loveless S, et al. Scalp electrical recording during paralysis: Quantitative evidence that EEG frequencies above 20 Hz are contaminated by EMG. Clin Neurophysiol. (2007) 118:1877–88. doi: 10.1016/j.clinph.2007.04.027 29. Muthukumaraswamy SD. High-frequency brain activity and muscle artifacts in MEG/EEG: a review and recommendations. Front Hum Neurosci. (2013) 7:138. doi: 10.3389/fnhum.2013.00138 30. Boytsova JA, Danko SG, Medvedev SV. When EMG contamination does not necessarily hide high-frequency EEG: scalp electrical recordings before and after Dysport injections. Exp Brain Res. (2016) 234:3091–106. doi: 10.1007/s00221-016-4708-3 31. Andersen P, Andersson SA, Lomo T. Thalamo-cortical relations during spontaneous barbiturate spindles. Electroencephalogr Clin Neurophysiol. (1968) 24:90. 32. Lopes da Silva FH, van Lierop THMT, Schrijer CF, Storm van Leeuwen W. Organization of thalamic and cortical alpha rhythms: Spectra and coherences. Electroencephalogr Clin Neurophysiol. (1973) 35:627–39. 33. Timofeev I, Steriade M. Fast (mainly 30-100 Hz) oscillations in the cat cerebellothalamic pathway and their synchronization with cortical potentials. J Physiol. (1997) 504:153–68. 34. Bartolomei F, Bonini F, Vidal E, Trebuchon A, Lagarde S, Lambert I, et al. How does vagal nerve stimulation (VNS) change EEG brain functional connectivity? Epilepsy Res. (2016) 126:141–6. doi: 10.1016/j.eplepsyres.2016.06.008 35. Wostyn S, Staljanssens W, De Taeye L, Strobbe G, Gadeyne S, Van Roost D, et al. EEG derived brain activity reflects treatment response from vagus nerve stimulation in patients with epilepsy. Int J Neural Syst. (2017) 27:1650048. doi: 10.1142/S0129065716500489 36. De Herdt V, De Waele J, Raedt R, Wyckhuys T, El Tahry R, Vonck K, et al. Modulation of seizure threshold by vagus nerve stimulation in an animal model for motor seizures. Acta Neurol Scand. (2010) 121:271–6. doi: 10.1111/j.1600-0404.2009.01223.x 37. Whittington MA, Traub RD, Kopell N, Ermentrout B, Buhl EH. Inhibition-based rhythms: experimental and mathematical observations on network dynamics. Int J Psychophysiol. (2000) 38:315–36. doi: 10.1016/S0167-8760(00)00173-2 38. Traub RD, Whittington MA, Colling SB, Buzsáki G, Jefferys JG. Analysis of gamma rhythms in the rat hippocampus in vitro and in vivo. J Physiol. (1996) 493(Pt 2):471–84. Conflict of Interest Statement: The patent application was applied (patent applicant Masaryk University; inventors are the authors of the manuscript: MB, ID, EK, JChl, RR, JChr, MP). The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Copyright © 2019 Brázdil, Doležalová, Koritáková, Chládek, Roman, Pail, Jurák, Shaw and Chrastina. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Neurology | www.frontiersin.org 11 May 2019 | Volume 10 | Article 392 239 Annex 16: Plešinger F, Halámek J, Chládek J, Jurák P, Doležalová I, Chrastina J, Brázdil M. Response to vagal stimulation by heart-rate feature in drug-resistant epileptic patients. (submitted) 240 Abstract— Vagal Nerve Stimulation (VNS) is an option in the treatment of drug-resistant epilepsy. However, approximately a quarter of VNS subjects does not respond to the therapy. In this retrospective study, we introduce heart-rate features to distinguish VNS responders and non-responders. Standard pre-implantation measurements of 66 patients were segmented in relation to specific stimuli (open/close eyes, photic stimulation, hyperventilation, and rests between). Median interbeat intervals were found for each segment and normalized (NMRR). Five NMRRs were significant; the strongest feature achieved significance with p=0.013 and AUC=0.66. Low mutual correlation and independence on EEG signals mean that presented features could be considered as an addition for models predicting VNS response using EEG. Clinical Relevance— Responders and non-responders to VNS can be distinguished (with low AUC) also by heart activity. I. INTRODUCTION Drug-resistant epilepsy can be treated by resective surgery (if localized) or by vagal nerve stimulation (VNS). Although the VNS is less interventing to a patient, it is less effective and usually does not lead to complete seizure freedom. Approximately half of the patients show a decreased amount of seizures by 50%; a quarter of the patients does not respond to the VNS [1], [2] at all. In general, epilepsy research usually uses electroencephalograph (EEG) signals to measure brain activity. This also applies to works focused on explaining the mechanism of the VNS effect [3]–[5] investigating synchronization/desynchronization of EEG in specific frequency bands. Yet up to this date, features and models predicting response to the VNS are missing [2] with exception to [6]. However, presented studies were focused on information from the brain while information describing heart activity, electrocardiogram (ECG), remains unused except for [7] using HRV analysis from 24-hours ECG recordings. * This research has been financially supported by grant AZV NV 19-04- 00343 F.Plesinger, J.Chladek, P.Jurak, J. Halamek, is with the Institute of Scientific Instruments of the CAS, Brno, 612 64 Czechia (e-mail: fplesinger@ isibrno.cz). I. Dolezalova, J. Chrastina, M. Brazdil, J.Chladek is with Behavioral and Social Neuroscience Research Group, CEITEC – Central European Institute of Technology, Masaryk University, Brno, Czechia and with (excepting J. Chladek) Brno Epilepsy Center, Department of Neurology, St. Anne’s University Hospital and the Medical Faculty at Masaryk University, Brno, Czechia In this retrospective study, we investigated features derived from ECG signals; we explored their ability to distinguish between responders and non-responders to VNS therapy in drug-resistant epileptic patients. II. DATA Subjects to VNS were measured in the Brno Epilepsy Center (Brno, Czech Republic) between 2005 and 2015 as a standard pre-operative assessment. These measurements were part of a study [6] focused on relations between scalp EEG signals and VNS response. After the VNS implantation, patients were followed-up (≥ 2 years), and individual outcomes were evaluated based on seizure reduction [8]: patients with < 50% reduction were considered as non-responders while others were considered VNS responders. A. Patients In this analysis, we used a subgroup of patient cohort presented in the source study [6]. Patients were removed from the presented analysis when measurement protocol was not respected, or unexpected events occurred (coughing during hyperventilation, sleep). Finally, a total of 66 patients (age 34 ± 12 years) consisting of 30 men and 36 women with 36 responders and 30 non-responders formed the cohort for this analysis (Tab. 1). TABLE I. PATIENT COHORT Gender Patient cohort All Responders Non-responders All 66 36 30 Females 36 18 18 Males 30 18 12 B. Protocol description A standard protocol (Fig. 1) was used for measurements: • rest at the beginning with closed eyes (Rest 1) • open eyes for 9 seconds • the second rest (Rest 2) • photic stimulation with increasing and decreasing frequency from 5 to 50 Hz • the third rest (Rest 3) • hyperventilation using nose • hyperventilation using mouth • the fourth rest (Rest 4) • second block with open eyes Response to Vagal Stimulation by Heart-rate Features in Drug-resistant Epileptic Patients* F. Plesinger, J. Halamek, J. Chladek, P. Jurak, I. Dolezalova, J. Chrastina, M. Brazdil 241 Figure 1. Protocol design with mean segment names and durations C. Recordings ECG signals were recorded together with EEG signals using Alien Deymed device with standard 10-20 electrode placement. Recordings were sampled at 128 Hz. If more then one ECG channel was available, the first one was used. Recording device provided antialiasing filtering for all input signals. III. METHOD A. Measurement segmentation Each measurement was split into segments based on individual marks. Because this analysis works with heart activity, a sufficient length of time segments is needed. Therefore, the duration of rest segments is higher in comparison to the rest segments used in the source study [6]. In total, each recording was split into nine segments. B. The segment-specific median RR interval (MRR) Heartbeats (QRS complexes) were found for each measurement segment. For this purpose, we used a technique based on amplitude envelopes developed for the ECG Holter study [9]. This technique transforms input ECG signal into amplitude envelope in band 5-25 Hz, searches for local maxima, and applies several rules to deny or accept specific peak as a QRS. Then an array of inter-beat (RR) intervals for each segment is computed as a difference of QRS positions. Finally, the MRR is computed as a median of the array. C. Baseline normalization As RR intervals are specific to patients, they need to be normalized to a baseline. It is a crucial part of this study since the significance of heart-rate features strongly relates to used baseline. As a baseline, we used the rest between the end of deep breathing and the last opening of eyes (Rest 4). Therefore, normalized MRR (NMRR) of each segment is computed as 𝑁𝑀𝑅𝑅 𝑠𝑒𝑔 = 𝑀𝑅𝑅 𝑠𝑒𝑔 𝑀𝑅𝑅 𝑅𝑒𝑠𝑡 4 where seg refers to a specific measurement segment. D. Statistical analysis Statistical analysis of NMRR values was processed using Python 3.7 and a SciPy [10] package. Non-parametric Wilcoxon–Mann–Whitney test [11] was used to evaluate the ability of each NMRR to separate responders and nonresponders. P-values lower than 0.05 were considered significant. Also, the Area under the Receiver-Operator Curve (AUC) was computed for each measurement segment. A linear relationship between significant features was examined using a correlation matrix. IV. RESULTS A. Median RR values in segments MRR values do not show significant differences between VNS responders and non-responders in any of the segments (Tab. 2). For completeness, the last row of tab.2 shows MRR over the whole measurement, which also remains non- significant. TABLE II. MEDIAN RR VALUES FOR SEGMENTS Segment Mean of Median RR values [s] p-value Responders [N=36] Non-responders [N=30] Rest 1 0,84 ± 0,15 0,83 ± 0,17 0,38 Open eyes 1 0,86 ± 0,15 0,85 ± 0,17 0,43 Rest 2 0,84 ± 0,15 0,83 ± 0,16 0,36 Photic stimulation 0,85 ± 0,15 0,84 ± 0,16 0,32 Rest 3 0,85 ± 0,15 0,84 ± 0,16 0,26 Hyperventilation nose 0,82 ± 0,16 0,81 ± 0,16 0,32 Hypervenilation mouth 0,81 ± 0,16 0,80 ± 0,15 0,39 Rest 4 0,84 ± 0,16 0,85 ± 0,16 0,45 Open eyes 2 0,85 ± 0,15 0,84 ± 0,15 0,34 Overall 0,84 ± 0,15 0,83 ± 0,16 0,38 242 B. Normalized Median RR values in segments On the other hand, different results are obtained when MRR values are normalized to a baseline (Fig. 2 and Tab. 3). We used the Rest 4 as the baseline segment; this normalization revealed five significant features – NMRRs from Rest 1, Rest 3, Hyperventilation by the nose, Hyperventilation by mouth, and the second segment with Open eyes. TABLE III. NORMALIZED MEDIAN RR VALUES FOR SEGMENTS Segment Mean Normalized Median RR p-value Responders [N=36] Non-responders [N=30] Rest 1 1,01 ± 0,07 0,99 ± 0,05 0.038* Open eyes 1 1,03 ± 0,07 1,00 ± 0,07 0.141 Rest 2 1,01 ± 0,06 0,99 ± 0,04 0.075 Photic stimulation 1,01 ± 0,06 0,99 ± 0,04 0.110 Rest 3 1,01 ± 0,05 0,99 ± 0,03 0.013* Hyperventilation nose 0,98 ± 0,05 0,96 ± 0,05 0.033* Hypervenilation mouth 0,97 ± 0,04 0,95 ± 0,05 0.028* Open eyes 2 1,02 ± 0,06 0,99 ± 0,06 0.023* Figure 2 shows differences in NMRR values between responders and non-responders, accompanied by p and AUC values. In those five significant features, we examined mutual correlations (Fig. 3); the strongest correlations were found between two hyperventilation segments (by mouth and nose). Figure 3. Correlation matrix of significant normalized median RR intervals (NMRR). Figure 2. Ability of median RR (MRR) intervals to separate responders ( blue) and non-responders (orange). MRR values were normalized to MRR from segment Rest-4. Blue and orange stripes are defined by 1st and 3rd quartile of specific boxplots. Grey areas mark significant features. 243 V. DISCUSSION A. Selecting a proper normalization segment Results showed that features based only on heart activity are weak, but significant for recognizing VNS response. However, this significance depends on the segment used for the normalization. We started with normalization using the median RR interval over the whole measurement as we expected that this could refer to the regular heart rate of the specific patient in the corresponding daytime. However, as shown in the last row of Tab. 4, this normalization led to the only one significant feature – Rest 4. Then we tried normalizing to different measurement segments. Since normalizing to most segments brings none or just one significant feature, we found that normalizing to the Rest 4 brings a surprising amount of significant features. TABLE IV. SIGNIFICANT FEATURES BY A BASELINE SEGMENT Baseline segment Significant NMRRs Count The most significant NMRR Name p- value Rest 1 1 Rest 4 0.038 Open eyes 1 0 - Rest 2 0 - Photic stimulation 0 - Rest 3 1 Rest 4 0.013 Hyperventilation nose 1 Rest 4 0.033 Hyperventilation mouth 1 Rest 4 0.028 Rest 4 5 Rest 3 0.013 Open eyes 2 1 Rest 4 0.023 Overall 1 Rest 4 0.013 B. Possible explanation Presented observation tells that the long-term response to VNS might be linked with the patient’s heart reaction to hyperventilation. More specifically, it might be hypothesized that the epileptic activity enhanced by hyperventilation impacts differentially on brain structures involved in controlling heart rate (e.g., insula or other autonomous brain structures) in responders versus non-responders. An alternative explanation could be that the reactivity of autonomous brain structures to hyperventilation per se is different in responders versus non-responders. C. Level of significance All significant features show rather weak AUC values. On the other hand, we are looking at features based purely on the ECG signal, and we are using it for an EEG-specific domain. D. Features usability We showed that correlations between significant features in Fig.3 are rather low, which means that several of them could be used together in models predicting VNS response. E. Method benefits and limitations The presented method shows significant features based purely on the ECG signal with low mutual correlation. Therefore, presented features could be valuable companions to more common EEG features in models predicting VNS. Presented ECG features are expectedly weaker then EEG features used in [6]. The size of the study cohort consisting of 66 patients is another limitation. Therefore, the presented method should be tested on an independent cohort. VI. CONCLUSION In this retrospective study, we have shown that features based only on the ECG signal can significantly distinguish responders and non-responders to vagal nerve stimulation in drug-resistant epileptic patients. We have also shown that the selection of the proper normalization baseline is crucial for feature significance. REFERENCES [1] D. J. Englot, J. D. Rolston, C. W. Wright, K. H. Hassnain, and E. F. Chang, “Rates and Predictors of Seizure Freedom With Vagus Nerve Stimulation for Intractable Epilepsy,” Neurosurgery, vol. 79, no. 3, pp. 345–353, Sep. 2016. [2] D. J. Englot, E. F. Chang, and K. I. Auguste, “Vagus nerve stimulation for epilepsy: a meta-analysis of efficacy and predictors of response,” J. Neurosurg., vol. 115, no. 6, pp. 1248– 1255, Dec. 2011. [3] H. Jaseja, “EEG-desynchronization as the major mechanism of anti-epileptic action of vagal nerve stimulation in patients with intractable seizures: Clinical neurophysiological evidence,” Med. Hypotheses, 2010. [4] F. Marrosu et al., “Increase in 20-50 Hz (gamma frequencies) power spectrum and synchronization after chronic vagal nerve stimulation,” Clin. Neurophysiol., 2005. [5] C. Bodin et al., “Responders to vagus nerve stimulation (VNS) in refractory epilepsy have reduced interictal cortical synchronicity on scalp EEG,” Epilepsy Res., 2015. [6] M. Brázdil et al., “EEG Reactivity Predicts Individual Efficacy of Vagal Nerve Stimulation in Intractable Epileptics,” Front. Neurol., vol. 10, May 2019. [7] H. Liu, Z. Yang, L. Huang, W. Qu, H. Hao, and L. Li, “Heart-rate variability indices as predictors of the response to vagus nerve stimulation in patients with drug-resistant epilepsy,” Epilepsia, 2017. [8] J. C. McHugh, H. W. Singh, J. Phillips, K. Murphy, C. P. Doherty, and N. Delanty, “Outcome Measurement after Vagal Nerve Stimulation Therapy: Proposal of a New Classification,” Epilepsia, vol. 48, no. 2, pp. 375–378, Feb. 2007. [9] F. Plesinger, P. Nejedly, I. Viscor, J. Halamek, and P. Jurak, “Automatic detection of atrial fibrillation and other arrhythmias in holter ECG recordings using rhythm features and neural networks,” in Comput Cardiol (Rennes IEEE), 2017, vol. 44, pp. 1–4. [10] E. Jones, T. Oliphant, P. Peterson, and Others, “SciPy.org,” SciPy: Open source scientific tools for Python2, 2001. . [11] H. B. Mann and D. R. Whitney, “On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other,” Ann. Math. Stat., vol. 18, no. 1, pp. 50–60, Mar. 1947. 244 Annex 17: Doležalová I, Kunst J, Kojan M, Chrastina J, Baláž M, Brázdil M. Anterior thalamic deep brain stimulation in epilepsy and persistent psychiatric side effects following discontinuation. Epilepsy Behavior Reports 2019;12:100344. 245 Case Report Anterior thalamic deep brain stimulation in epilepsy and persistent psychiatric side effects following discontinuation Irena Doležalová a, ⁎, Jonáš Kunst b , Martin Kojan a,c , Jan Chrastina d , Marek Baláž a , Milan Brázdil a,d a First Department of Neurology, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic b Faculty of Medicine, Masaryk University Brno, Czech Republic c Central European Institute of Technology (CEITEC), Brno, Czech Republic d Department of Neurosurgery, St. Anne's University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic a b s t r a c ta r t i c l e i n f o Article history: Received 4 August 2019 Received in revised form 27 October 2019 Accepted 28 October 2019 Available online 5 November 2019 We report a case of a patient with drug-resistant epilepsy treated with deep brain stimulation of the anterior nucleus of the thalamus (ANT-DBS). The patient developed psychiatric side effects (PSEs), namely irritability, hostility, aggressiveness, and paranoia, after implantation and stimulation initiation. The stimulation was discontinued and the PSEs were mitigated, but the patient did not return to her pre-implantation state, as documented by repeated psychiatric reports and hospitalizations. To our knowledge, this is the first report of a patient who developed long-term PSEs that did not disappear after stimulation discontinuation. We suppose that ANTDBS caused a persistent perturbation of the thalamic neuronal networks that are responsible for long-term PSEs. © 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Keywords: Deep brain stimulation of the anterior nucleus of the thalamus Long-term psychiatric side effects Case report 1. Introduction Deep brain stimulation of the anterior nucleus of the thalamus (ANT-DBS) is a novel and promising treatment method for patients with drug-resistant epilepsy. More than 70% of patients implanted with ANT-DBS benefit significantly from this method, i.e., they report seizure-reduction rates higher than 50% [1]. We have only limited knowledge about short- and long-term ANT-DBS side effects because of relatively low numbers of implanted patients. When focusing on the adverse events reported in a study of stimulation of the anterior nuclei of thalamus (SANTE study), the patients reported paresthesia (18% patients), pain in the implant side (10.9% patients), and infection at the implant site (9.1% patients) [2]. During a five-year follow-up course, device-related adverse events were reported [1]. When focusing on adverse events, depression was present in 37.3% patients (3 events in 3 subjects were considered to be devicerelated), memory impairment was present in 27.3% patients (approximately a third of memory impairment was confirmed by neuropsychological examination), 11.8% patients reported suicidal ideation (one subject committed suicide; the suicide was not thought to be device related), and 7 patients died during the study — 1 probable sudden unexpected death in epilepsy (SUDEP), 2 definite SUDEP, and 1 possible SUDEP [1]. The anterior nucleus of the thalamus (ANT) has connections with limbic structures, anterior cingulate cortex, and orbitomedial prefrontal cortex; thus, it plays a vital role in memory processes and in emotional and executive functions [3]. A disruption or alteration in baseline circuits can be associated with ANT-DBS psychiatric side effects (PSEs). PSEs can appear immediately after stimulation initialization or can develop over a more extended period. However, in all previously published reports, the PSEs were time-related to stimulation. Järvenpää et al. [9] suggested decreasing the voltage or changing the stimulation contacts to suppress PSEs. At the moment, there is no official recommendation for the management of PSEs in patients with ANT-DBS. We report a patient in whom ANT-DBS caused PSEs that persisted despite stimulation discontinuation. To our knowledge, this is the first report of a patient in whom ANT-DBS caused PSEs not directly linked to the stimulation itself. 2. Case report The patient is a female born in 1970 with a family history of mesial temporal lobe epilepsy. The patient was treated only for lower back pain and had no history of psychiatric illness. Family history regarding psychiatric disease was also absent. The patient developed epilepsy at the age of 17; the epilepsy was characterized as focal with independent left and right temporal interictal epileptiform discharges and bitemporal seizure onsets during scalp EEG monitoring. Magnetic Epilepsy & Behavior Reports 12 (2019) 100344 Abbreviations: ANT, anterior nucleus of the thalamus; ANT-DBS, deep brain stimulation of the anterior nucleus of the thalamus; MRI, magnetic resonance imaging; PET, positron emission tomography; PSEs, psychiatric side effects. ⁎ Corresponding author at: First Department of Neurology, St. Anne's University Hospital, Pekařská 53, 656 91 Brno, Czech Republic. E-mail address: irena.dolezalova@fnusa.cz (I. Doležalová). https://doi.org/10.1016/j.ebr.2019.100344 2589-9864/© 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Contents lists available at ScienceDirect Epilepsy & Behavior Reports journal homepage: www.elsevier.com/locate/ebcr 246 resonance imaging (MRI) revealed right-sided hippocampal sclerosis. Positron emission tomography (PET) showed bitemporal hypometabolism. The Wada test demonstrated a crucial role of rightsided mesial temporal structures for verbal memory. The patient was judged to be a poor surgical candidate and neuromodulation was favored. Therefore, the implantation of a vagus nerve stimulator (VNS) was performed at the age of 29. The vagus nerve stimulator was ineffective, which led to its explantation after seven years. The patient was offered ANT-DBS implantation at the age of 41 years. The patient experienced 8 focal seizures with impairment of awareness per month. Before implantation, the psychological examination revealed no severe mental pathology, such as anxiety, depression, or psychosis. However, there were some psychological problems, specifically echolalia, perseveration, and difficulties with management of stressful situations, as well as global deterioration of cognitive function (IQ 72, the most severe alteration in execution and memory). The patient was on stable doses of antiseizure drugs for two years before implantation. She was treated with pregabalin 600 mg/day, zonisamide 400 mg/day, and lacosamide 400 mg/day. The monopolar stimulation of proximal contacts (contact 3 on the left and contact 11 on the right) was initiated one month after ANT-DBS implantation (Fig. 1a and b); the stimulation parameters were as follows: amplitude 2.5 V, stimulation frequency 140 Hz, pulse width 90 ms, 5-minute off-time, 1minute on-time. The patient did not note a decrease in seizure frequency, but she did report reduction in seizure severity and duration. Behavioral changes started to appear gradually three months after implantation. The family complained about patient irritability, hostility, and aggressiveness, which were accompanied by newly evolved paranoia. The patient reported that DBS influenced her manners and forced her to walk backward, which worsened when someone was speaking about DBS. A psychological and psychiatric examination performed five months after stimulation initialization revealed incoherent thoughts and behavioral (unrest, agitation) and emotional alteration. Personality and behaviorial disorders were diagnosed. Any possibility of a new structural abnormality was excluded by post-operative magnetic resonance imaging (MRI). We discontinued stimulation and introduced quetiapine (50 mg/day) and subsequently risperidone (1.5 mg/day). This resulted in partial alleviation of the PSEs, but the patient did not reach her preimplantation state over the next seven years documented by repeated psychological and psychiatric examinations. The patient's state required recurrent psychiatric hospitalizations conditioned by intermittent worsening with psychosis. The patient subjectively reported persistent reduction in seizure severity. 3. Discussion The anterior nucleus of the thalamus plays a crucial role in seizure generation, as shown in animal and human studies. Stimulation of the anterior nucleus of the thalamus leads to substantial seizure reduction in the majority of patients. The results of two studies suggest that the positioning of the DBS electrode has a significant impact on seizure reduction [4,5]. Patients with electrodes located more anterior and superior [5] or in the antero-ventral part of the thalamus tend to have better responses [4]. The thalamus is a key structure in two core neurocognitive networks, the salience network and the default-mode network, that were proven to play important roles in cognition, emotion, and execution, and that are altered in many psychiatric conditions, including depression, bipolar disorder, schizophrenia, substance use disorder, obsessive–compulsive disorder, and anxiety disorder [6,7]. It is therefore not surprising that interference with the functioning of this structure could be associated with PSEs. ANT-DBS was found to be independently associated with de novo psychopathology in a group of surgically treated patients [8]. Järvenpää et al. [9] reported a group of 22 patients treated with ANT-DBS; four patients developed reversible PSEs. The stimulation induced depression in two patients. The other two patients had symptoms of paranoia and anxiety. In all their patients, the PSEs completely disappeared when they lowered the output current or changed the stimulated contacts [9]. Based on their experience, the authors supposed the stimulation of other thalamic nuclei or tracts close to the ANT to be responsible for PSEs, as the stimulated contacts were not in the ANT itself but beneath. These findings contrast with our experience. As shown in Fig. 1, the appropriate contact on the left was chosen, the more distal contact was more suitable on the right. Moreover, the PSEs did not disappear after stimulation discontinuation. They persisted for seven years despite intensive treatment with repeated hospitalization. The explanation for this situation is complicated; we can merely speculate about possible reasons. Our hypothesis is that the insertion of electrodes and subsequent stimulation can cause alteration in thalamic circuitry with potential long-term consequences in predisposed individuals such as our patient. We suspect some lesional effect because the PSEs persisted despite stimulation discontinuation similar to thalamic lesions caused by ischemia. Most authors localized the ANT with surrounding nuclei to the territory of the tuberothalamic artery, the closure of which is characterized by severe wide-ranging neuropsychological deficits [10]. Strokes in this arterial territory are associated with the fluctuation of consciousness in the early stage of ischemia. Subsequently, Fig. 1. The position of deep brain stimulation (DBS) electrodes to anterior nucleus thalami (ANT). Panel a illustrates the relation between DBS electrodes and ANT in the coronal section. There are 4 contacts on each electrode (contacts are labeled from the most distal to the most proximal as 0, 1, 2, and 3 on the left, and as 8, 9, 10, and 11 on the right). We stimulated the most proximal contacts on both sides, i.e., contact 3 on the left and contact 11 on the right. Panel b shows the estimated distribution of the electrical field (yellow) and its relation to ANT (blue). On the right, the correct contact for stimulation was chosen. On the left, the more distal contact was more suitable. The figures were obtained using SureTune software (Medtronic, Minneapolis, MN, USA). L — left, R — right. 2 I. Doležalová et al. / Epilepsy & Behavior Reports 12 (2019) 100344 247 disorientation, euphoria, lack of insight, apathy, lack of spontaneity, and emotional unconcern can develop [11]. We excluded an acute surgical complication and doubt an adverse effect due to antiseizure drugs was suddenly responsible for the psychiatric complaints in our patient. ANT-DBS is a promising novel technique for drug-resistant epilepsy treatment, but may be occasionally associated with significant PSEs. It is necessary to inform patients about the possibility of PSEs when offering ANTI-DBS. Because only limited information is available about chronic ANT-DBS PSEs more experience in larger groups of patients is necessary to identify those who are at risk for long-term consequences. Funding The results of this research have been acquired with the support of the Ministry of Health of the Czech Republic, grant no. NV19-04-00343. Ethical statement We confirm that we have read the Journal's position on issues involved in ethical publications and affirm that this report is consistent with those guidelines. Declaration of competing interest Neither of the authors has any conflict of interest to disclose. References [1] Salanova V, Witt T, Worth R, Henry TR, Gross RE, Nazzaro JM, et al. Long-term efficacy and safety of thalamic stimulation for drug-resistant partial epilepsy. Neurology 2015;84:1017–25. [2] Fisher R, Salanova V, Witt T, Worth R, Henry T, Gross R, et al. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia 2010; 51:899–908. [3] Child ND, Benarroch EE. Anterior nucleus of the thalamus: functional organization and clinical implications. Neurology 2013;81:1869–76. [4] Krishna V, King NK, Sammartino F, Strauss I, Andrade DM, Wennberg RA, et al. Anterior nucleus deep brain stimulation for refractory epilepsy: insights into patterns of seizure control and efficacious target. Neurosurgery 2016;78:802–11. [5] Lehtimaki K, Mottonen T, Jarventausta K, Katisko J, Tahtinen T, Haapasalo J, et al. Outcome based definition of the anterior thalamic deep brain stimulation target in refractory epilepsy. Brain Stimul 2016;9:268–75. [6] Menon V. Large-scale brain networks and psychopathology: a unifying triple network model. Trends Cogn Sci 2011;15:483–506. [7] Peters SK, Dunlop K, Downar J. Cortico-striatal-thalamic loop circuits of the salience network: a central pathway in psychiatric disease and treatment. Front Syst Neurosci 2016;10. [8] Novais F, Pestana LC, Loureiro S, Andrea M, Figueira ML, Pimentel J. Predicting de novo psychopathology after epilepsy surgery: a 3-year cohort study. Epilepsy Behav 2019;90:204–8. [9] Järvenpää S, Peltola J, Rainesalo S, Leinonen E, Lehtimaki K, Jarventausta K. Reversible psychiatric adverse effects related to deep brain stimulation of the anterior thalamus in patients with refractory epilepsy. Epilepsy Behav 2018;88: 373–9. [10] Bogousslavsky J, Regli F, Assal G. The syndrome of unilateral tuberothalamic artery territory infarction. Stroke 1986;17:434–41. [11] Schmahmann JD. Vascular syndromes of the thalamus. Stroke 2003;34:2264–78. 3I. Doležalová et al. / Epilepsy & Behavior Reports 12 (2019) 100344 248 Annex 18: Rektor I, Dolezalova I, Chrastina J, Jurak P, Halamek J, Balaz M, Brazdil M. High-frequency oscillations in the human anterior nucleus of the thalamus. Brain Stimulation. 2016;9(4):629-631. 249 Letter to the Editor High-Frequency Oscillations in the Human Anterior Nucleus of the Thalamus Dear Editor: Deep brain stimulation of the anterior nucleus of the thalamus (ANTDBS) has recently been introduced in therapy for refractory epilepsy and is approved for clinical use in Europe [1]. ANT is a part of the Papez circuitry and is a key structure in the intrathalamic pathways; it also projects to the cingulate gyrus and further to the limbic structures and wide regions of the neocortex, and via the mammillary circuit to the brain stem [2]. ANT stimulation produces EEG changes in the frontal and temporal areas and inhibits seizures [3]. The role played by the ANT in human epileptic seizures and the mechanisms leading to the antiseizure effects of ANT-DBS have not yet been fully elucidated. Knowledge of processes occurring in the ANT in human epilepsy might improve the understanding of its role. Here we report the first description of interictal and ictal EEG recording in the human ANT. DBS electrodes were implanted in the ANT of six pharmacoresistent patients with temporal and extratemporal epilepsies who had not responded to vagus nerve stimulation. (Patient characteristics and outcome see Supplementary material I; Table S1). The recordings were approved by the ethics committee; the patients gave their informed consent. A two-step procedure that is usual in Parkinson’s disease was used for DBS in epilepsy, i.e. the DBS electrodes were left externalized for several days before the internalization and the implantation of the battery. Video-EEG monitoring was performed during the 3-day period between the two steps of the operation. Depth EEG via the DBS electrodes was performed using the clinical 128channel EEG system TruScan (Deymed Diagnostic, Alien Technic, Czech Republic) with a sampling rate of 128 and 256 Hz and standard antialiasing filters. To record high frequency oscillations (HFO), in four of six patients a 192-channel research machine (M&I; Brainscope, Czech Republic) was used for ≥10 minutes for awake resting EEG recordings with a sampling rate of 5000 Hz. HFO were detected and scored manually in SignalPlant and ScopeWin visualization softwares. Recordings were filtered and checked in the bandpass of 80–2000 Hz. (More about Methods see in the Supplementary material II). Inter-ictal recordings: No inter-ictal epileptiform discharges were recorded from the DBS electrodes. The rarity of the epileptic discharges is not specific to the ANT; the absence of epileptiform activity had also been observed in the basal ganglia [4]. The recording with a high-sampling EEG revealed HFO of up to 240 Hz, in one case up to 500 Hz, in all four patients (Fig. 1 and Supporting Material I, Table S2). The occurrence of HFOs in the ANT was bilateral. However, a detailed analysis of individual oscillations in monopolar and bipolar montages displayed different phases of HFOs in the left and right ANT (Supplementary material III, Figs. S2 and S3). Ictal recordings: We recorded eight clinical seizures in four patients with DBS electrodes implanted in the ANT. An early ictal broadband increase of power occurred in all seizures. A specific ictal epileptiform activity was recorded in one seizure preceding the onset of the clinical seizure by 4 seconds (Supplementary material III, Fig. S4). The first broad-band power increase preceded the onset of the clinical seizure symptoms in three patients by between 32 seconds and four seconds (Fig. 1). This is, to our knowledge, the first report of HFO occurrence in a human subcortical structure. Inter-ictal HFO, ripples, in one case fast ripples, were displayed in all patients in whom a high sample rate EEG could be recorded. It has been shown in animal models that the ripples are generated in the cortex and propagate to the thalamus where they excite the inhibitory reticular neurons inhibiting thalamocortical projections [5]. The electrophysiological state of ANT is altered in patients with epilepsy; atypical firing properties of ANT cells were observed in intraoperative microelectrode recordings [6]. A close connection between epilepsy and HFO in the 80– 600 Hz range has been repeatedly described [7,8]. As epilepsy is the only reason to implant an electrode in the ANT and no control recordings are possible we cannot be sure about the nature of HFOs. The characteristics of the HFOs recorded in the ANT indicate that this phenomenon may be pathological. The frequency of spontaneous ANT HFO (up to 240 and 500 Hz) was higher than the frequency of physiological ripples in the human hippocampus (80– 160 Hz); it overlapped with the frequency of clearly pathological HFO. Fast ripples (200–600 Hz) could be observed in normal neocortex, but not in normal hippocampus and parahippocampal structures [8,9]. The ANT has tight anatomical and functional connection with the hippocampus. One limit of our work is inherent to the DBS procedure. Notably, patients have no intracranial exploration of cortical epileptogenic areas, and consequently some links, such as the synchronization between the cortex and the ANT, could not be studied. The surface recordings were limited by the necessity of avoiding the area surrounding the implantation sites and by the location of the electrodes under a bandage; thus, the use of scalp EEG data was very limited. The seizures might have originated at unrecorded cortical sites and propagated to the ANT, but the recorded interictal and ictal phenomena, including the HFO and the epileptiform EEG activity, were generated in this nucleus. The depth electrodes are submerged in the brain tissue and record from their immediate vicinity. The local fields were displayed as EEG, and its modifications occurred both in reference and in bipolar montage, excluding with high probability the volume conduction from other structures, namely from the cortex, or transsynaptic propagation along the cortical–subcortical pathways [10]. Based on our recordings of HFO, of early ictal ANT electrophysiological modifications, and of ictal epileptiform pattern, 1935-861X/© 2016 Published by Elsevier Inc. Brain Stimulation 9 (2016) 629–631 Contents lists available at ScienceDirect Brain Stimulation journal homepage: www.brainstimjrnl.com 250 we suggest that the ANT may participate directly in the network elaborating clinical seizures in human epilepsies. Acknowledgments The authors wish to express their thanks to Dr. Robert Fisher, Stanford, USA, and Dr. Premysl Jiruska, Prague, Czech Republic, for their valuable advices. Supported by the Ministry of Education, Youth and Sports of the Czech Republic under the project CEITEC 2020 (LQ1601). Appendix. Supplementary material Supplementary data to this article can be found online at doi:10.1016/j.brs.2016.04.010. Ivan Rektor*, Irena Doležalová, Jan Chrastina Brno Epilepsy Centre and Movement Disorders Centre, First Department of Neurology and Department of Neurosurgery, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic Central European Institute of Technology, Masaryk University, Centre of Neuroscience, Brno, Czech Republic Pavel Jurák, Josef Halámek Czech Academy of Sciences, Institute of Scientific Instruments, Brno, Czech Republic Marek Baláž, Milan Brázdil Brno Epilepsy Centre and Movement Disorders Centre, First Department of Neurology and Department of Neurosurgery, St. Anne’s University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic Central European Institute of Technology, Masaryk University, Centre of Neuroscience, Brno, Czech Republic * Corresponding author. Masaryk University, Epilepsy Centre Brno, First Department of Neurology, Faculty of Medicine, St. Anne’s University Hospital, Pekarˇská 53, 656 91 Brno, Czech Republic Tel.: +420 543 182 623; fax: +420 543 182 624 E-mail address: irektor@med.muni.cz (I. Rektor) Received 4 April 2016 Available online 14 April 2016 http://dx.doi.org/10.1016/j.brs.2016.04.010 Figure 1. Patient 1. Preictal and ictal recording with high-sampling EEG, bipolar montage, contacts P3-4, right ANT. Time window 90 seconds before and 50 second after the onset of clinical seizure – marked by vertical yellow line at 0s. A: Enlargement of the signal by zooming at 30–20 s before clinical seizure onset. B: Raw EEG signal, time window 90 seconds before to 50 second after the onset of clinical seizure marked by vertical yellow line at 0s. C: Normalized Time Frequency Map (TFM). All values in each TFM frequency row (horizontal line) were normalized to the mean value. Notice enhanced broad-band oscillations with HFO, 30–20 s before the onset of clinical seizure. 630 Letter to the Editor / Brain Stimulation 9 (2016) 629–631 251 References [1] Fisher R, Salanova V, Witt T, Worth R, Henry T, Gross R, et al. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia 2010;51:899–908. [2] Zhong XL, Yu JT, Zhang Q, Wang ND, Tan L. Deep brain stimulation for epilepsy in clinical practice and in animal models. Brain Res Bull 2011;85:81– 8. [3] Kerrigan JF, Litt B, Fisher RS, Cranstoun S, French JA, Blum DE, et al. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia 2004;45:346–54. [4] Rektor I, Kuba R, Brazdil M, Halámek J, Jurák P. Ictal and peri-ictal oscillations in the human basal ganglia in temporal lobe epilepsy. Epilepsy Behav 2011;20:512–17. [5] Grenier FO, Timofeev I, Steriade M. Focal synchronization of ripples (80–200 Hz) in neocortex and their neuronal correlates. J Neurophysiol 2001;86:1884–98. [6] Hodaie M, Cordella R, Lozano AM, Wennberg R, Dostrovsky JO. Bursting activity of neurons in the human anterior thalamic nucleus. Brain Res 2006;1115:1–8. [7] Jacobs J, Zijlmans M, Zelmann R, Chatillon CE, Hall J, Olivier A, et al. Highfrequency electroencephalographic oscillations correlate with outcome of epilepsy surgery. Ann Neurol 2010;67:209–20. [8] Jiruska P, Bragin A. High-frequency activity in experimental and clinical epileptic foci. Epilepsy Res 2011;97:300–7. [9] Engel J Jr, Bragin A, Staba R, Mody I. High-frequency oscillations: what is normal and what is not? Epilepsia 2009;50:598–604. [10] Wennberg RA, Lozano AM. Intracranial volume conduction of cortical spikes and sleep potentials recorded with deep brain stimulating electrodes. Clin Neurophysiol 2003;114:1403–18. 631Letter to the Editor / Brain Stimulation 9 (2016) 629–631 252 Annex 19: Doležalova I., Pešlová E, Michnová M, Nečasová T, Kočvarová J, Musilová K, Rektor I. Brázdil M. Epileptochirurgická léčba zlepšuje kvalitu života – výsledky dotazníkové studie. Ceská a Slovenská Neurologie a Neurochirurgie. 2016;79/112(4):430-439. 253 430 Cesk Slov Neurol N 2016; 79/112(4): 430–439 KRÁTKÉ SDĚLENÍ SHORT COMMUNICATION Epileptochirurgická léčba zlepšuje kvalitu života – výsledky dotazníkové studie Epilepsy Surgery Improves Quality of Life – Results of a Questionnaire Study Souhrn Cíl: Tradičním cílem epileptochirurgické léčby je dosažení bezzáchvatovosti. V současnosti se naše pozornost stále více soustřeďuje na dopad epileptochirurgické léčby na ostatní aspekty života pacienta, nejvíce je řešen její vztah k zaměstnanosti a ke kvalitě života. Metody: Pro účely této studie byl na našem pracovišti vytvořen dotazník, který obsahoval 13 uzavřených otázek, jejichž obsah je možné rozdělit do čtyř částí. První okruh otázek zjišťoval obecné informace o nemocném. V druhé části jsme se dotazovali na informace týkající se vlastní operace. Třetí část se zaměřovala na oblast sociální problematiky, konkrétně otázku zaměstnanosti před operací a po ní a držení řidičského oprávnění po operaci. Čtvrtý okruh otázek se týkal subjektivního hodnocení přínosu operace pro pacienta. Odpovědi pacientů byly následně statisticky vyhodnoceny. Výsledky: Do studie bylo zařazeno 91 respondentů – 56 mužů (61,5 %) a 35 žen (38,5 %), kteří kompletně vyplnili zaslaný dotazník. U 59 pacientů (64,8 %) po operaci epileptické záchvaty úplně vymizely. Před operací bylo zaměstnáno celkem 46 pacientů (50,5 %), po operaci došlo k mírnému vzestupu zaměstnanosti u čtyř pacientů (4,4 %), tento rozdíl nedosáhl statistické významnosti (p = 0,596). Polovina, konkrétně 49 pacientů (53,8 %), udává pooperačně vzestup kvality života. Poslední otázka zjišťovala, zda by pacient znovu podstoupil operační zákrok, přičemž na tuto otázku většina – 78 pacientů (85,7 %) – odpověděla kladně. Závěr: Na základě našeho dotazníkového šetření můžeme říci, že ač velká část pacientů podstoupila úspěšný epileptochirurgický výkon, který je zcela zbavil záchvatů, zaměstnání se podařilo najít jen malé části z nich. I přes tento fakt je epileptochirurgická léčba ze strany našich operantů hodnocena kladně, přináší jim často zvýšení kvality života a většina z nich by znovu podstoupila operační zákrok Abstract Aim: Seizure-freedom is the traditional goal of epilepsy surgery. At presents, our attention is more and more concerned with the effects of epilepsy surgery on other aspects, mainly the impact of epilepsy surgery on employment status and quality of life. Methods: A questionnaire was designed for the purposes of our study; this questionnaire consists of 13 questions focusing on topics related to epilepsy surgery. These questions could be divided into four subfields: 1. demographic data and patients´ characteristics, 2. information related to surgery, 3. employment and social support, 4. subjective evaluation of surgery impact on the patient’s life. Results: Ninety-one respondents were included in the study – 56 men (61.5%) and 35 women (38.5%), who correctly completed the questionnaires. Fifty-nine patients (64.8%) were completely seizure-free after surgery. Before the surgery, 46 patients (50.5%) were employed, there was a mild increase in employment after surgery that did not reach statistical significance (p = 0.596). Half of the patients (49, 54%) reported an increase in quality of life after surgery. One question asked whether the patient would undergo the surgery again if he/she could change the past; 78 patients (85.7%) agreed they would. Conclusion: The majority of patients undergoing a surgery for drug-resistant epilepsy were seizure-free but only a minority managed subsequently to find an employment. Despite this, epilepsy surgery is assessed positively in the vast majority of patients, it provides them with an increase in quality of life and the majority would undergo the surgery again. Autoři deklarují, že v souvislosti s předmětem studie nemají žádné komerční zájmy. The authors declare they have no potential conflictsofinterestconcerningdrugs,products,or services used in the study. Redakční rada potvrzuje, že rukopis práce splnil ICMJE kritéria pro publikace zasílané do biomedicínských časopisů. The Editorial Board declares that the manuscript met the ICMJE “uniform requirements” for biomedical papers. I. Doležalová1 , E. Pešlová2 , M. Michnová1 , T. Nečasová3 , J. Kočvarová1 , K. Musilová1 , I. Rektor1,4 , M. Brázdil1,4 1 Centrum pro epilepsie Brno, I. neurologická klinika LF MU a FN u sv. Anny v Brně 2 LF MU, Brno 3 Institut biostatistiky a analýz, LF MU, Brno 4 CEITEC – Středoevropský technologický institut, MU, Brno  MUDr. Irena Doležalová, Ph.D. I. neurologická klinika LF MU a FN u sv. Anny v Brně Pekařská 53 656 91 Brno e-mail: irena.dolezalova@fnusa.cz Přijato k recenzi: 29. 10. 2015 Přijato do tisku: 15. 2. 2016 Klíčová slova epilepsie – farmakorezistence – chirurgická léčba – bezzáchvatovost – kvalita života – zaměstnanost pacientů po operaci – invalidní důchod Key words epilepsy – drug-resistant – surgery – seizure-freedom – quality of life – employment after surgery – disability pension proLékaře.cz | 29.4.2020 254 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 431 Úvod Epilepsie je onemocnění, které postihuje kolem 1 % populace v rozvinutých zemích, je tedy významným problémem jak medicínským, tak i společenským. Pacienti, kteří mají záchvaty i přes správnou léčbu, jsou definováni jako pacienti farmakorezistentní. Farmakorezistentní epilepsie s sebou přináší významná omezení v běžném životě, konkrétně se jedná o zvýšené riziko fyzického poranění či předčasného úmrtí v průběhu záchvatu, sociální problémy a snížení kvality života [1,2]. Přesné vymezení farmakorezistence bylo provedeno v roce 2010 Mezinárodní ligou proti epilepsii (International League against Epilepsy; ILAE). Farmakorezistence je v současnosti definována jako selhání dvou antiepileptik, ať už v monoterapii či v kombinaci [3,4]. U pacientů, kteří plní výše uvedené kritérium, by měly být posouzeny možnosti operační léčby [5]. „Klasickým“ měřítkem úspěšnosti epileptochirurgické léčby je vymizení epileptických záchvatů, stále více epileptochirurgických center se však v posledních letech soustřeďuje na dopad operace i na ostatní aspekty oblasti života nemocného [6]. Otázkou zůstává, zda společně s efektem operace dochází u pacienta i ke zlepšení kvality jeho života či možností jeho pracovního uplatnění a k vyšší spokojenosti v osobním a společenském životě [7]. Prezentovaná studie se věnuje výsledkům chirurgické léčby u pacientů s farmakorezistentní epilepsií, kteří podstoupili operační zákrok v Centru pro epilepsie Brno. Vyjma standardního hodnocení vymizení či přetrvávání záchvatů se soustřeďujeme na problematiku sociální (zaměstnanost, invalidita) a problematiku kvality života. Metodika Pro účely této studie byl na našem pracovišti vytvořen dotazník (příloha 1), který obsahoval 13 uzavřených otázek. Dotazník byl vytvořen tak, aby otázky byly jednoduché a srozumitelné a pacient je byl schopen vyplnit samostatně bez pomoci lékaře či psychologa. Obsah dotazníku je možné rozdělit do čtyř okruhů. První okruh zjišťoval obecné informace o nemocném (pohlaví, věk v době operace, nejvyšší dosažené vzdělání a věk, ve kterém se u pacienta epilepsie objevila). V druhém okruhu jsme se dotazovali na informace týkající se vlastní operace (kolik let uplynulo od operace, zda je pacient bez záchvatů). Třetí část se zaměřovala na oblast sociální problematiky (otázka zaměstnanosti před operací a po ní, pobírání invalidního důchodu před operací a po ní, držení řidičského oprávnění). Čtvrtý okruh otázek se týkal subjektivního hodnocení přínosu operace pro pacienta. V rámci tohoto okruhu byly položeny dvě otázky. První se týkala kvality života po operaci. Druhá otázka byla hypotetická a zněla, zda by se pacient nechal znovu operovat, kdyby mohl vrátit čas. Dotazník byl rozeslán celkem 137 pacientům, kteří byli operováni v rámci Centra pro epilepsie Brno, podmínkou bylo, že od operace uběhla doba nejméně jednoho roku. Dodržení tohoto časového intervalu jsme považovali za důležité pro validní hodnocení výsledků šetření. Sběr dat probíhal v období prosinec 2012–březen 2013. Statistika Všechny proměnné byly nejdříve popisně sumarizovány a srovnány Fisherovým exaktním testem nebo McNemarovým testem. Pro modelování byl použit logistický regresní model. Nejprve byly vyhodnoceny jednorozměrné modely. Dále byly vytvořeny maximální modely, a to zpětnou, dopřednou Příloha 1. Dotazník – Kvalita života po epileptochirurgické léčbě. Otázka č. 1. Jaké je Vaše pohlaví? • muž • žena Otázka č. 2. Kolik Vám bylo v době operace let? • 18–25 let • 26–35 let • 36–45 let • 46–60 let • více než 61 let Otázka č. 3. Jaké je Vaše nejvyšší dosažené vzdělání? • základní • učební obor bez maturity • středoškolské s maturitou • vyšší odborné • vysokoškolské Otázka č. 4. V kterém životním období (věku) jste měl/měla první epileptický záchvat? • v útlém dětství (do 1 roku věku) • v dětství (1–10 let) • v dospívání (11–18 let) • v dospělosti (18–60 let) • později než v 60 letech Otázka č. 5. Jaká doba uplynula od Vaší operace? • více než 1 rok • více než 2 roky • více než 5 let • více než 10 let Otázka č. 6. Vymizely u Vás záchvaty po operaci? • ano • je jich méně • ne • ano, ale mám po operaci jiné obtíže, u kterých si myslím, že souvisí s epilepsií/s vlastní operací (specifikujte jaké) ......................................................................................... Otázka č. 7. Byl/byla jste před operací zaměstnán/zaměstnána? • ano – zaměstnanec nebo osoba samostatně výdělečně činná (OSVČ) • ne Otázka č. 8. Pobírala jste před operací invalidní důchod? • ano • ne Otázka č. 9. Jste nyní po operaci zaměstnán/zaměstnána? • ano – zaměstnanec nebo OSVČ • ne Otázka č. 10. Pobíráte nyní po operaci invalidní důchod? • ano • ne Otázka č. 11. Získal/získala jste po operaci řidičské oprávnění (řidičský průkaz)? • ano • ne Otázka č. 12. Jak byste hodnotil svoji kvalitu života po operaci ve srovnání s kvalitou života před operací? Moje kvalita života se: • zvýšila • snížila • je stále stejná • nevím Otázka č. 13. Představte si, že byste mohl/mohla vrátit čas. Nechal/nechala byste se znovu operovat? • ano • ne • nevím proLékaře.cz | 29.4.2020 255 432 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 i kombinovanou selekcí příznaků. Výstupem modelů jsou tabulky, ve kterých jsou uvedeny poměry šancí (OR) s intervalem spolehlivosti (IS) a p hodnotou. V tabulkách je u kategoriálních proměnných uvedena nejprve riziková a poté referenční kategorie. U výsledného vícenásobného modelu je uvedena hodnota pseudo R2, tzv. Nagelkerke R2, který je počítán jako podíl věrohodností nulového a plného modelu. Jeho největší možnou hodnotou je 1, pokud model perfektně predikuje výsledek s věrohodností 1. Analýza byla provedena v softwaru IBM SPSS Statistics 22 a software R3.1.3. Všechna srovnání byla provedena na hladině významnosti α = 0,05. Výsledky Demografická data Do studie bylo zařazeno 91 respondentů, 56 mužů (61,5 %) a 35 žen (38,5 %), kteří kompletně vyplnili zaslaný dotazník. Většina, konkrétně 62 pacientů (68,1 %), byla z věkové kategorie 36–60 let, 28 pacientů (30,8 %) bylo z věkové kategorie 18–35 let, jen jeden pacient byl starší 60 let (1,1 %). Pouze u devíti pacientů (9,9 %) se objevily epileptické záchvaty do jednoho roku věku, u 32 v dětství (35,2 %), u 25 v dospívání (27,5 %) a u 25 v dospělosti (27,5 %). Čtyřicet pět pacientů (49,5 %) absolvovalo učební obor bez maturity, 23 pacientů (25,3 %) mělo středoškolské vzdělání s maturitou, 11 pacientů (12,1 %) Tab. 1. Demografická data pacientů a výsledky chirurgické léčby. Pohlaví muž n (%) 56 (62) žena n (%) 35 (39) Věk v době operace 18–25 let n (%) 6 (7 %) 26–35 let n (%) 22 (24 %) 36–45 let n (%) 32 (35 %) 46–60 let n (%) 30 (33 %) více než 61 let n (%) 1 (1 %) Nejvyšší dosažené vzdělání základní n (%) 11 (12 %) učební obor bez maturity n (%) 49 (50 %) středoškolské s maturitou n (%) 23 (25 %) vyšší odborné nebo vysokoškolské n (%) 12 (10 %) Vznik epilepsie do 1 roku věku n (%) 9 (10 %) 1–10 let n (%) 32 (35 %) 11–18 let n (%) 25 (27 %) 19–64 let n (%) 25 (27 %) nad 65 let n (%) 0 (0 %) Doba od operace více než 1 rok 8 (9 %) více než 2 roky 25 (27 %) více než 5 let 31 (34 %) více než 10 let 27 (30 %) Výsledky chirurgické léčby – vymizení epileptických záchvatů ano 58 (64 %) je jich méně 25 (27 %) ne 6 (7 %) Zaměstnananost před a po operaci 0 % 20 % 40 % 60 % 80 % 100 % 0 % 20 % 40 % 60 % 80 % 100 % zaměstnání před operací zaměstnání po operaci ne n (%) 45 (50 %) 41 (45 %) 46 (50 %) 50 (55 %) ano n (%) Zaměstnanost před a po operaci – rozdíl mezi muži a ženami zaměstnání před operací zaměstnání po operaci zaměstnání před operací zaměstnání po operaci Muži Ženy 25 (46 %) 20 (57 %) 22 (63 %) 19 (34 %) 31 (54 %) 37 (66 %) 15 (43 %) 13 (37 %) Graf 1. Změny v zaměstnanosti před a po operaci. Graf 1a) Hodnoceny změny v zaměstnanosti před a po operaci bez ohledu na pohlaví respondentů. Graf 1b) Hodnoceny změny v zaměstnanosti před a po operaci z hlediska pohlaví respondentů. proLékaře.cz | 29.4.2020 256 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 433 mělo základní vzdělání, 12 pacientů (13,2 %) mělo vzdělání vyšší odborné nebo vysokoškolské. Jak bylo uvedeno výše, u všech pacientů byl dodržen časový limit nejméně jednoho roku od operace, 33 pacientů (36,3 %) bylo méně než pět let po operaci, 31 (34,1 %) méně než 10 let od operace, zbylých 27 pacientů (29,7 %) bylo více než 10 let po operaci (tab. 1). Výsledky chirurgické léčby – hodnoceno vymizení epileptických záchvatů U 59 pacientů (64,8 %) vymizely úplně epileptické záchvaty po operaci, u dalších 26 (28,5 %) se frekvence záchvatů výrazným způsobem snížila, u šesti pacientů (6,6 %) nedošlo k významné změně frekvence epileptických záchvatů (tab. 1). Je zajímavé, že ze skupiny pacientů, u kterých došlo k úplnému vymizení záchvatů, si 20 (33,9 %) stěžovalo na jiné obtíže. Část pacientů udávala obtíže, které jsou v přímé souvislosti s operačním výkonem, konkrétně poruchy zorného pole (šest pacientů (6,6 %)), vznik pooperační hemiparézy (dva pacienti (2,2 %)), či poruchy paměti (tři pacienti (3,3 %)). Druhá skupina pacientů udávala obtíže, o jejichž kauzálním vztahu k operaci můžeme spíše debatovat. Nejčastěji se jednalo o bolesti hlavy (šest pacientů (6,6 %)), ojediněle (méně než u tří pacientů v celém souboru) byly přítomny psychické problémy, poruchy sluchu a spánku, vyšší únavnost. Dopady chirurgické léčby epilepsie na jednotlivé aspekty života po operaci Zaměstnanost, invalidita, držení řidičského oprávnění Před operací bylo zaměstnáno (v zaměstnaneckém poměru či jako osoba samostatně výdělečně činná (OSVČ)) celkem 46 (50,5 %) pacientů, po operaci došlo k mírnému vzestupu zaměstnanosti o čtyři pacienty (4,4 %), tedy po operaci bylo zaměstnáno 50 pacientů (54,9 %), tento rozdíl v zaměstnanosti nedosáhl statistické významnosti (p = 0,596) (graf 1a). Nárůst zaměstnanosti se týkal pouze mužů. Počet zaměstnaných mužů po operaci vzrostl z 31 (55,4 %) na 37 (66,1 %), tj. o 10,7 %, ani zde však nebyl rozdíl statisticky signifikantní (p = 0,264). Upozorňujeme však na důležitou skutečnost, že po operaci našlo nově zaměstnání 13 mužů, avšak sedm mužů z nejrůznějších důvodů pracovat přestalo, došlo tedy k celkovému vzestupu zaměstnanosti o šest mužů. U žen naopak došlo k mírnému poklesu zaměstnanosti po operaci, před operací bylo zaměstnáno 15 žen (42,9 %), po operaci pak pouze 13 (37,1 %) (p = 0,773) (graf 1b). I zde zdůrazňujeme fakt, že po operaci nově našlo zaměstnání pět žen, ale sedm žen po operaci pracovat přestalo, došlo tedy k celkovému poklesu zaměstnanosti o dvě ženy. Před operací mělo invalidní důchod 43 pacientů (47,3 %), po operaci vzrostl jejich počet na 54 (59,3 %), tento vzestup však nebyl statisticky významný (p = 0,054) (graf 2a). Nově invalidní důchod získalo 19 pacientů, osm pacientů ale po operaci invalidní důchod ztratilo. Rovněž v „pobírání“ invalidních důchodů byl přítomen rozdíl mezi jednotlivými pohlavími (graf 2b). U žen došlo k vzestupu v „pobírání“ invalidního důchodu z 18 (51,4 %) na 25 (71,4 %), tento rozdíl dosáhl statistické významnosti (p = 0,070). U mužů nebyl tento nárůst tak významný, z 25 mužů (44,6 %) „pobírajících“ invalidní důchod před operací vzrostl tento počet na 29 (51,8 %) (p = 0,453). Vzhledem k faktu, že v našem souboru byli pouze pacienti s aktivní epilepsií, neměl nikdo před operací oprávnění k řízení motorových vozidel. Po operaci 19 pacientů (20,9 %) toto oprávnění získalo, což je statisticky významný nárůst (p < 0,001). V další analýze jsme se pokusili určit jednotlivé proměnné, které souvisí se zaměstnaností pacienta po operaci (tab. 2). Byl zjištěn statisticky významný vztah mezi zaměstnáním po operaci a následujícími charakteristikami pacientů: pohlaví (p = 0,009), vzdělání (p = 0,020), zaměstnání před operací (p = 0,006), pobírání důchodu před ope- 0 % 20 % 40 % 60 % 80 % 100 % 0 % 20 % 40 % 60 % 80 % 100 % ne n (%)ano n (%) Invalidita před a po operaci invalidní důchod před operací invalidní důchod po operaci Invalidita před a po operaci – rozdíl mezi muži a ženami invalidní důchod před operací invalidní důchod po operaci invalidní důchod před operací invalidní důchod po operaci Muži Ženy 48 (53 %) 43 (47 %) 37 (41 %) 54 (59 %) 31 (55 %) 25 (45 %) 27 (48 %) 29 (5 %) 17 (49 %) 18 (51 %) 10 (29 %) 25 (71 %) Graf 2. Změny v invaliditě před a po operaci. Graf 2a) Hodnoceny změny v invaliditě před a po operaci bez ohledu na pohlaví respondentů. Graf 2b) Hodnoceny změny v invaliditě před a po operaci z hlediska pohlaví respondentů. proLékaře.cz | 29.4.2020 257 434 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 rací (p = 0,021), pobírání důchodů po operaci (p = 0,001), získání řidičského oprávnění po operaci (p = 0,004). Následně byla provedenajednorozměrná logistickáregrese,která hodnotila vliv jednotlivých charakteristik pacienta na zaměstnání po operaci (v logistické regresi byly použity pouze proměnné, které popisují stav pacienta před operací) (tab. 3). Pomocí výsledků získaných jednorozměrnou logistickou regresí byl vytvořen finální model, jehož cílem se predikce zaměstnání po operaci. Na základě tohoto modelu můžeme říci, že po operaci byli častěji zaměstnáni pacienti s vyšším než základním vzděláním (OR = 6,5; IS = 1,2–36,0; p = 0,033), muži než ženy (OR = 3,3; IS = 1,3–8,6; p = 0,013), pacienti, kteří měli zaměstnání již před operací (OR = 3,0; IS = 1,2–7,5; p = 0,021). Kvalita života po operaci Poslední část našeho dotazníku se věnovala subjektivnímu hodnocení kvality života po operaci. Polovina, 49 pacientů (53,8 %), udává pooperačně vzestup kvality života, což je statisticky významný rozdíl proti kvalitě života před operací. V další analýze jsme hledali faktory, které souvisí se změnou kvality života po operaci (tab. 4). Jediná proměnná, konkrétně zaměstnání po operaci, vykazuje statisticky významný vztah s kvalitou života po operaci. Z celkového počtu 50 pacientů, kteří jsou po operaci zaměstnáni, jich 32 (64,0 %) udává, že mají po operaci zlepšenou kvalitu života. Naopak z celkového počtu 41 pacientů, kteří po operaci nepracují, udává zlepšenou kvalitu života pouze 17 (41,5 %) pacientů (p = 0,037). Na základě údajů, které byly o pacientech známy před provedením operace, jsme se pokusili sestavit model logistické regrese, který by predikoval, zda bude mít pacient po operaci zlepšenou kvalitu života (tab. 5). Vzhledem k faktu, že se nám nepodařilo při použití jednorozměrné logistické regrese identifikovat charakteristiky pacientů, které by vysvětlovaly zlepšení života po operaci, nebylo možné sestavit smysluplný vícerozměrný logistický model. Poslední otázka našeho dotazníku zněla: „Představte si, že byste mohl/mohla vrátit čas. Nechal/nechala byste se znovu operovat?“ Většina, 78 (85,7 %) pacientů, odpověděla, že ano, 13 (14 %) odpovědělo ne. Dvě proměnné, konkrétně stupeň dosaženého vzdělání a kvalita života po operaci, vykazovaly statisticky významný vztah k odpovědi na tuto otázku (tab. 6). Většina, konkrétně 71 (88,8 %) z celkového počtu 80 pacientů s vyšším vzděláním než základním, by se nechala operovat znovu, naopak pouze sedm (63,6 %) z celkového počtu 11 pacientů se základním vzděláním by znovu podstoupilo operační výkon (p = 0,048). Čtyřicet devět pacientů (100,0 %), u kterých došlo ke zvýšení kvality života po operaci, by se nechalo operovat znovu. „Pouze“ 29 ze 42 pacientů (69,1 %), u kterých po operaci nedošlo ke zvýšení kvality života, by opětovně podstoupilo operaci (p < 0,001). Rovněž i u této otázky jsme se pokoušeli najít pomocí jedTab. 2. Zaměstnanost po operaci – vztah k ostatním proměnným. 53 % zaměstnání po operaci ano n (%) ne n (%) p hodnota Věk v době operace 18–25 let 3 (50 %) 3 (50 %) 0,885 26–35 let 12 (54,5 %) 10 (45,5 %) 36–45 let 19 (59 %) 13 (41) 46–60 let 16 (53 %) 14 (47) více než 60 let 0 (0 %) 1 (100 %) Pohlaví muž 37 (66 %) 19 (34 %) 0,009 žena 13 (37 %) 22 (63 %) Vzdělání základní 2 (18 %) 9 (82 %) 0,020 středoškolské a vyšší 48 (60 %) 32 (40 %) Vznik epilepsie v dětství (< 10 let věku) 26 (63 %) 15 (37 %) 0,278v dospívání (< 18 let věku) 11 (44 %) 14 (56 %) v dospělosti (> 18 let) 13 (52 %) 12 (48 %) Zaměstnání před operací ano 32 (70 %) 14 (30 %) 0,006 ne 18 (40 %) 27 (60 %) Důchod před operací ano 18 (42 %) 25 (58 %) 0,021 ne 32 (67 %) 16 (33 %) Důchod po operaci ano 18 (42 %) 25 (58 %) 0,001 ne 18 (40 %) 27 (60 %) Záchvaty po operaci přetrvávají 18 (56 %) 14 (44 %) 1,000 vymizely 32 (54) 27 (46 %) Řidičské oprávnění ano 16 (84 %) 3 (16 %) 0,004 ne 34 (47 %) 38 (53 %) Tab. 3. Vliv jednotlivých proměnných na zaměstnanost pacientů po operaci – jednorozměrná logistická regrese. Prediktor OR IS (95%) p hodnota Věk 1,006 (0,962; 1,051) 0,791 Pohlaví (muž/žena) 3,296 (1,366; 7,953) 0,008 Vzdělání (střední a vyšší/základní) 8,625 (1,602; 46,447) 0,012 Vznik epilepsie (dětství/dospělost) 1,600 (0,583; 4,392) 0,362 (dospívání/dospělost) 0,725 (0,238; 2,208) 0,572 Zaměstnání před operací (ano/ne) 3,429 (1,442; 8,152) 0,005 Důchod před operací (ne/ano) 2,778 (1,184; 6,517) 0,019 proLékaře.cz | 29.4.2020 258 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 435 Diskuze Pacientům s farmakorezistentní epilepsií by dle doporučení ILAE měla být nabídnuta možnost operačního řešení tohoto onemocnění [8]. „Klasickým“ a všeobecně akceptovaným měřítkem úspěšnosti operační léčby je vymizení epileptických záchvatů, avšak v současnosti se naše pozornost stále více soustřeďuje i na jiné cíle, které v běžném životě člověka hrají neméně důležitou roli. V naší práci se detailněji věnujeme problematice zaměstnanosti a kvalitě života, ale podobný význam hraje i možnost většího zapojení se do společenského dění, lepší přístup ke vzdělání či získání větší samostatnosti a nezávislosti. Jak bylo uvedeno výše, „klasickým“ měřítkem úspěšnosti chirurgické léčby epilepsie je dosažení bezzáchvatovosti, čehož se nám podařilo dosáhnout u 64,8 % pacientů z našeho souboru (jednalo se o neselektovanou skupinu pacientů jak s temporální, tak extratemporální epilepsií, u níž je obecně horší prognóza). Toto číslo dobře koresponduje s výsledky publikovanými Wiebem et al, kteří prokázali efektivitu chirurgické léčby u pacientů s temporální epilepsií v rámci randomizované kontrolované klinické studie [9]. Ve své práci náhodně rozdělili 80 pacientů s farmakorezistentní temporální epilepsií na pacienty, kteří podstoupili chirurgickou léčbu, a na pacienty, kteří byli ponecháni „pouze“ na léčbě farmakologické. Po operaci bylo bez záchvatů 58 % pacientů, ze skupiny konzervativně léčených pacientů to bylo jen 8 %. První oblastí, kterou jsme hodnotili v našem dotazníku, byla otázka zaměstnanosti pacientů před operací a po ní. V rámci naší studie došlo pouze k minimálnímu vzestupu zaměstnanosti po operaci, před operací bylo zaměstnáno 50 % pacientů, po operaci 55 %, tj. zaměstnanost se zvýšila o 5 %. V zahraniční literatuře jsou přítomny diskrepance při hodnocení zaměstnanosti pacientů po epileptochirurgickém výkonu. Existují práce, které dokumentují jak její nárůst, tak i její pokles [10–12]. Ve studii publikované Zarrolim et al, která analyzovala změnu zaměstnanosti u cca 350 pacientů s temporální epilepsií, došlo k nárůstu zaměstnanosti po operaci o 16 %, před operací nemělo zaměstnání 41 % pacientů, po operaci se počet nezaměstnaných snížil na 25% [10]. Ve studii publikované Lendtem et al došlo k poklesu v nezaměstnanosti o 16 % [11]. Naopak v práci Reevse et al došlo k poklesu zaměstnanosti po operaci o 3 % [12]. Dle nafaktorem bylo pouze vzdělání (OR = 9,4; IS = 1,4–61,9; p = 0,020), proto vícerozměrný model nebyl sestavován. norozměrné logistické regrese faktory, které by mohly předoperačně predikovat odpověď na tuto otázku (tab. 7). Významným Tab. 4. Změna kvality života po operaci. kvalita života po operaci zvýšená jiná odpověď* p hodnota Věk v době operace 18–25 let n (%) 4 (67 %) 2 (33 %) 0,785 26–35 let n (%) 13 (59 %) 9 (41 %) 36–45 let n (%) 15 (47 %) 17 (53 %) 46–60 let n (%) 16 (53 %) 14 (47 %) > 60 let n (%) 1 (100 %) 0 (0 %) Pohlaví muž n (%) 29 (52 %) 27 (48 %) 0,670 žena n (%) 20 (57 %) 15 (43 %) Vzdělání základní n (%) 4 (36 %) 7 (64 %) 0,334 středoškolské a vyšší n (%) 45 (56 %) 35 (44 %) Vznik epilepsie v dětství (< 10 let) 21 (51 %) 20 (49 %) 0,894v dospívání (< 18 let) 14 (56 %) 11 (44 %) v dospělosti (> 18 let) 14 (56 %) 11 (44 %) Zaměstnání před operací ano n (%) 26 (57 %) 20 (44 %) 0,676 ne n (%) 23 (51 %) 22 (49 %) Důchod před operací ano n (%) 22 (51 %) 21 (49 %) 0,677 ne n (%) 27 (56 %) 21 (44 %) Zaměstnání po operaci ano n (%) 32 (64 %) 18 (36 %) 0,037 ne n (%) 17 (41,5 %) 24 (58,5 %) Důchod po operaci ano n (%) 25 (46 %) 29 (54 %) 0,091 ne n (%) 24 (65 %) 13 (35 %) Záchvaty po operaci přetrvávají n (%) 14 (44 %) 18 (56 %) 0,189 vymizely n (%) 35 (59 %) 24 (41 %) Řidičské oprávnění po operaci ano n (%) 12 (62 %) 7 (37 %) 0,442 ne n (%) 37 (51 %) 35 (49 %) * snížená, stejná, nevím. Tab. 5. Vliv jednotlivých proměnných na kvalitu života po operaci – jednorozměrná logistická regrese. Prediktor OR IS (95%) p hodnota Věk 0,992 (0,949; 1,037) 0,713 Pohlaví (muž/žena) 0,806 (0,344; 1,885) 0,618 Vzdělání (středoškolsé a vyšší/základní) 3,354 (0,817; 13,778) 0,093 Vznik epilepsie (dětství/dospělost) 0,825 (0,304; 2,241) 0,706 (dospívání/dospělost) 1,000 (0,327; 3,055) 1,000 Zaměstnání před operací (ano/ne) 1,243 (0,545; 2,839) 0,605 Důchod před operací (ano/ne) 1,227 (0,573; 2,804) 0,627 proLékaře.cz | 29.4.2020 259 436 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 šeho názoru mohou být tyto rozdíly podmíněny mnoha faktory: rozdílnými společenskými a sociálními systémy v jednotlivých státech, rozdílným výběrem pacientů a rovněž rozdílným designem jednotlivých studií. Zde bychom rádi vyzvedli jeden, ač na první pohled nepatrný, avšak v rámci hodnocení zaměstnanosti pacientů po operaci důležitý faktor – dobu od operace. My jsme zařadili do naší studie již pacienty jeden rok po operaci, ale například Lendt et al hodnotili pacienty až čtyři roky po operaci [11]. Je pravděpodobné, že rok od operace je příliš krátká doba, aby se pacient přizpůsobil nový nárokům, které jsou na něj kladeny v souvislosti s hledáním zaměstnání. Pacient, který podstoupil operaci před čtyřmi roky, měl delší čas k adaptaci na změnu života a na změnu svého společenského postavení, což není změna, kterou je člověk schopen učinit ze dne na den. Jedná se bezesporu o dlouhodobější proces, při kterém je nutné si osvojit kompletně nový soubor návyků a přístupů. Pokud bychom tedy měli hodnotit dopad chirurgické léčby na zaměstnanost pacientů, bylo by vhodnější zvolit delší minimální časový interval pro validitu hodnocení. Naše studie zaznamenala nárůst počtu pacientů pobírajících invalidní důchod po operaci ve srovnání s předoperačním obdobím o 12 %, tento nárůst však nedosáhl statistické významnosti. Obdobný trend byl zdokumentován rovněž ve Švédsku, kde počet pacientů s invalidním důchodem vzrostl o 7 % v prvních dvou letech po operaci, ve třetím roce dokonce až o 10 % [13]. V rámci naší analýzy jsme identifikovali faktory, které souvisí se zaměstnaností po operaci. Jedná se o následující demografická data a charakteristiky pacientů: zaměstnanost před operací (pacienti, kteří mají zaměstnání před operací, jsou častěji zaměstnaní i po operaci), pohlaví (muži jsou zaměstnáni častěji než ženy), vzdělání (pacienti s vyšším vzděláním jsou zaměstnáni častěji), přiznaná invalidita před operací a po ní (pacienti s přiznaným důchodem jsou zaměstnáni méně často). Se zaměstnáním po operaci souvisí i držení řidičského oprávnění. Obdobné faktory ovlivňující zaměstnanost po operaci, přestože se jedná o odlišné socio-ekonomické prostředí (Spojené státy americké vs. Česká a Slovenská republika), identifikovali Zarroli et al [10]. Dle jejich výsledků bylo zaměstnání před operací významným prediktorem zaměstnání po operaci a stejně jako v naší studii část žen, která Tab. 6. Odpověď, na otázku: „Představte si, že byste mohl/mohla vrátit čas. Nechal/nechala byste se znovu operovat?“. „Nechal/a byste se znovu operovat?“ Ano, nechal/a. Ne, nenechal/a. p hodnota Věk v době operace 18–25 let n (%) 5 (83 %) 1 (17 %) 1,000 26–35 let n (%) 19 (86 %) 3 (14 %) 36–45 let n (%) 27 (84 %) 5 (16 %) 46–60 let n (%) 26 (87 %) 4 (13 %) > 60 let n (%) 1 (100 %) 0 (0 %) Pohlaví muž n (%) 48 (86 %) 8(14 %) 1,000 žena n (%) 30 (85,7 %) 5 (14 %) Vzdělání základní n(%) 7 (64 %) 4 (36 %) 0,048 středoškolské a vyšší n (%) 71 (89 %) 9 (11 %) Vznik epilepsie v dětství (< 10 let) 34 (83 %) 7 (17 %) 0,192v dospívání (< 18 let) 20 (80 %) 5 (20 %) v dospělosti (> 18 let) 24 (96 %) 1 (4 %) Zaměstnání před operací ano n (%) 42 (91 %) 4 (9 %) 0,145 ne n (%) 36 (80 %) 9 (20 %) Důchod před operací ano n (%) 36 (83 %) 9 (16 %) 0,766 ne n (%) 42 (88 %) 6 (13 %) Zaměstnání po operaci ano n (%) 45 (90 %) 5 (10 %) 0,237 ne n (%) 33 (80,5 %) 8 (19,5 %) Důchod po operaci ano n (%) 45 (83 %) 9 (17 %) 0,549 ne n (%) 33 (89 %) 4 (11 %) Záchvaty po operaci přetrvávají n (%) 26 (81 %) 6 (19 %) 0,369 vymizely n (%) 33 (89 %) 4 (11 %) Řidičské oprávnění po operaci ano n (%) 19 (100 %) 0 (0 %) 0,063 ne n (%) 59 (82 %) 13 (18 %) Kvalita života po operaci zvýšená 49 (100 %) 0 (0 %) < 0,001 jiná odpověď 29 (69 %) 13 (31 %) Tab. 7. Vliv jednotlivých proměnných na odpověď na otázku „Nechal/a byste se operovat znovu?“ – jednorozměrná logistická regrese. Prediktor OR IS (95%) p hodnota Věk 1,008 (0,947; 1,073) 0,807 Pohlaví (muž/žena) 1,000 (0,299; 3,343) 1,000 Vzdělání (středoškolské a vyšší/základní) 9,429 (1,434; 61,986) 0,020 Vznik epilepsie (dětství/dospělost) 0,202 (0,023; 1,754) 0,147 (dospívání/dospělost) 0,167 (0,018; 1,546) 0,115 Zaměstnání před operací (ano/ne) 2,625 (0,745; 9,246) 0,133 Důchod před operací (ano/ne) 1,361 (0,419; 4,420) 0,608 proLékaře.cz | 29.4.2020 260 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 437 významné srovnání s českou prací Preisse a Vojtěcha, kteří se zabývali změnou kvality života po operaci u 50 pacientů s farmakorezistentní, většinou temporální epilepsií, operovaných v nemocnici Na Homolce. Tito autoři našli rovněž významné zvýšení kvality života, zvláště v následujících sférách: percepce zdraví, sociální funkce a strach ze záchvatu [30]. V naší studii byla přítomna pouze jediná proměnná – zaměstnání pacienta po operaci – která statisticky korelovala se zvýšením kvality života po operaci. Z celkového počtu 50 pacientů, kteří měli po operaci zaměstnání, udávalo zlepšení kvality života 64,0 %. Z počtu 41 pacientů, kteří po operaci zaměstnání neměli, uvádělo zlepšení kvality života pouze 41,5 %. Obdobná korelace mezi kvalitou života po operaci a zaměstnáním po operaci byla nalezena i v několika dalších studiích [31–34]. Byla popsána celá řada proměnných, které souvisí s kvalitou života po operaci. Z údajů, které jsou známy již před vlastní operací, kvalitu života po operaci ovlivňují následující: psychologické a psychiatrické komorbidity (přítomnost psychiatrických onemocnění, špatná výkonnost v psychologických testech), nerealistická očekávání od operace, očekávání, která nebyla po operaci naplněna [31,35,36]. Z pooperačních faktorů ovlivňuje kvalitu života především dosažená dobrá kompenzace záchvatů [26]. Zdá se však, že úplné vymizení epileptických záchvatů je pro kvalitu života pacienta mnohem důležitější než pouhá jejich redukce. Dalším faktorem zvyšujícím kvalitu života pacienta je držení řidičského oprávnění [34]. Naopak kvalitu života po operaci snižovala přítomnost nežádoucích účinků antiepileptické medikace, psychické obtíže, obtíže s verbální pamětí, těžký průběh záchvatů, přítomnost jiného somatického onemocnění [32–34,37–39]. Je pro nás překvapením, že se vymizení epileptických záchvatů v naší práci neuplatňovalo jako faktor, který by ovlivňoval kvalitu života po operaci. V naší studii udávalo zvýšenou kvalitu života 59,3 % pacientů, kteří po ope- racizáchvatyneměli,a43,8%pacientů,ukterých epileptické záchvaty přetrvávaly i po operaci. Můžeme se domnívat, že je tento jev částečně podmíněn jinými zdravotními obtížemi, jež pacienti popisovali v našem dotazníku. Druhým možným vysvětlením je „syndrom břemene normality“ (burden of normality syndrome) popsaný Wilsonovou v roce 2001 [40]. „Syndrom břemene normality“ se vyskytuje u pacientů po vyléčení z japo operaci a vymizení epileptických záchvatů [10,17–22]. V rámci naší práce bylo zaměstnáno 56,3 % pacientů, u kterých záchvaty vymizely, a 54,2 % pacientů, u kterých záchvaty po operaci přetrvávaly. Dle Zarroli et al je právě úplné vymizení či velmi podstatná redukce záchvatů nejdůležitějším faktorem, který určuje, zda bude po operaci pacient zaměstnán či nikoliv [10]. Právě pacienti, u kterých dojde po operaci k úplnému vymizení či podstatné redukci záchvatů, nachází častěji práci než pacienti, u kterých dochází k méně významné změně frekvence záchvatů. Obdobný vztah je přítomen i v mnoha dalších publikacích. Další faktor ovlivňující zaměstnanost pacientů, jež se v naší studii nepodařilo potvrdit, byl věk pacienta v době operace [18,23,24]. Mladší pacienti mají dle výsledků jiných publikací lepší šanci najít zaměstnání. Ve studii Sperlinga et al z poloviny 90. let minulého století byl průměrný věk pacienta, kterému se po operaci podařilo najít zaměstnání 31,0 ± 8 let. Naopak průměrný věk pacienta, který po operaci zaměstnání nenašel, byl 42,4 ± 10,9 let [18]. Obdobné výsledky v roce 2009 publikovali i George et al [23]. Výše popsaný jev je pravděpodobně podmíněn částečně společensky (mladší lidé obecně hledají snáze zaměstnaní), ale částečně to souvisí jistě i se schopností mladších pacientů se snáze učit novým věcem a přizpůsobovat se více nárokům. Naopak starší pacienti se již mohli vžít a osvojit si „roli nemocného“ [25]. Otázkou je, proč jsme v naší skupině nenašli ani věk v době operace, ani vymizení záchvatů jako důležitý prediktor pro vyšší zaměstnanost pooperačně. Opětovně se domníváme, že by bylo potřeba delšího časového odstupu od operace. Navíc nejsme v našich modelech schopni provést korekci ani na somatické a psychiatrické komorbidity, roli zde samozřejmě může hrát i sociální prostředí a velikost jednotlivých souborů. Zaznamenali jsme významný nárůst kvality života po operaci: 53,8 % pacientů udávalo, že se jejich kvalita života po operaci zvýšila. Tento výsledek odpovídá výsledkům většiny studií, které se zabývaly touto problematikou. Metaanalýza publikovaná Seiamem et al v roce 2011 analyzovala výsledky 39 publikací, které se věnovaly změně kvality života u pacientů s epilepsií [26]. Většina prací, konkrétně 36 (92,3 %), našla po operaci zvýšenou kvalitu života. Zbylé tři práce (7,7 %) nezaznamenaly po operaci zvýšení kvality života [27–29]. Z našeho pohledu je byla před operací zaměstnána, zaměstnání po operaci ztratila (v práci Zarroli et al je však přítomen celkový nárůst zaměstnanosti i u žen). Výše uvedená ztráta zaměstnání byla odůvodněna faktem, že se část žen po operaci rozhodla zůstat trvale v domácnosti a věnovat se péči o rodinu a děti. Předpokládáme, že obdobná tendence je přítomna i v rámci České a Slovenské republiky. Pobírání invalidního důchodu či státní podpory (dle zvyklostí jednotlivých zemí) je v literatuře asociováno s nižší zaměstnaností po operaci [14,15]. Částečně je toto zcela jistě podmíněno faktem, že pacienti pobírající invalidní důchod mohou mít výraznější kognitivní a psychické obtíže v souvislosti s epilepsií, mohou rovněž trpět významnějšími somatickými onemocněními bez jednoznačného vztahu k epilepsii. Může se však projevovat i snížená motivace této skupiny pacientů najít zaměstnaní. Dalším faktorem, který má vztah k zaměstnání pacientů po operaci, je získání řidičského oprávnění (jedná se o řidičské oprávnění pouze pro osobní účely, nikoliv o řidičské oprávnění, které umožňuje řízení motorových vozidel v pracovněprávním vztahu). Po operaci získalo řidičské oprávnění jen 20,9 % pacientů, z nichž však bylo po operaci zaměstnáno 84,2 %. Z pacientů, kteří řidičské oprávnění nezískali, bylo po operaci zaměstnáno jen 47,2 %. Výše uvedený vztah byl potvrzen celou řadou zahraničních studií. Řidičské oprávnění se ukazuje jako nezávislý prediktor zaměstnanosti pacientů i v modelech, v nichž byl statisticky odstraněn vliv případné mentální retardace, psychických či somatických komorbidit [12,16]. V literatuře se objevují názory, že řidičské oprávnění, ač jen pro osobní účely, ulehčuje pacientům hledání zaměstnání, pacienti s řidičským oprávněním jsou snáze schopni dorazit na pracovní pohovor, případně i do práce, nemusí se spoléhat na veřejné dopravní prostředky, na pomoc rodiny či blízkých. Objevují se rovněž spekulace, že pacienti, kteří po operaci získají řidičské oprávnění, disponují povahovými rysy, které jim usnadňují hledání zaměstnání, zvláště je vyzdvihována jejich ochota riskovat a čelit případnému zklamání, ochota učit se novým věcem. Tyto spekulace nelze potvrdit ani vyvrátit výsledky této či jiné studie. Bylo pro nás překvapivé, že jsme nepotvrdili vztah zaměstnanosti po operaci a některých dalších faktorů, které jsou často přítomny v ostatních studiích. Prvním takovýmto vztahem je vztah zaměstnanosti proLékaře.cz | 29.4.2020 261 438 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 11. Lendt M, Helmstaedter C, Elger CE. Pre- and postoperative socioeconomic development of 151 patients with focal epilepsies. Epilepsia 1997;38(12):1330–7. 12. Reeves AL, So EL, Evans RW, et al. Factors associated with work outcome after anterior temporal lobectomy for intractable epilepsy. Epilepsia 1997;38(6): 689–95. 13. Asztely F, Ekstedt G, Rydenhag B, et al. Long-term follow-up of the first 70 operated adults in the Goteborg Epilepsy Surgery Series with respect to seizures, psychosocial outcome and use of antiepileptic drugs. J Neurol Neurosurg Psychiatry 2007;78(6):605–9. 14. Sung C, Muller V, Jones JE, et al. Vocational rehabilitation service patterns and employment outcomes of people with epilepsy. Epilepsy Res 2014;108(8):1469–79. doi: 10.1016/j.eplepsyres.2014.06.016. 15. Capella ME. Inequities in the VR system: do they still exist? Rehabil Couns Bull 2002;45(3):143–53. 16. Elsharkawy AE, May T, Thorbecke R, et al. Long-term outcome and determinants of quality of life after temporal lobe epilepsy surgery in adults. Epilepsy Res 2009;86(2–3):191–9. doi: 10.1016/j.eplepsy- res.2009.06.008. 17. Augustine EA, Novelly RA, Mattson RH, et al. Occupational adjustment following neurosurgical treatment of epilepsy. Ann of Neurol 1984;15(1):68–72. 18. Sperling MR, Saykin AJ, Roberts FD, et al. Occupational outcome after temporal lobectomy for refractory epilepsy. Neurology 1995;45(5):970–7. 19. Chin PS, Berg AT, Spencer SS, et al. Employment outcomes following resective epilepsy surgery. Epilepsia 2007;48(12):2253–7. 20. Chin PS, Berg AT, Spencer SS, et al. Patient-perceived impact of resective epilepsy surgery. Neurology 2006;66(12):1882–7. 21. Elsharkawy AE, Alabbasi AH, Pannek H, et al. Long-term outcome after temporal lobe epilepsy surgery in 434 consecutive adult patients. J Neurosurg 2009;110(6):1135–46. doi: 10.3171/2008.6.JNS17 613. 22. Kral T, von Lehe M, Podlogar M, et al. Focal cortical dysplasia: long-term seizure outcome after surgical treatment. J Neurol Neurosurg Psychiatry 2007;78(8):853–6. 23. George L, Lyer RS, James R, et al. Employment outcome and satisfaction after anterior temporal lobectomy for refractory epilepsy: a developing country‘s perspective. Epilepsy Behav 2009;16(3):495–500. doi: 10.1016/j.yebeh.2009.08.020. 24. Thorbecke R, May TW, Koch-Stoecker S, et al. Effects of an inpatient rehabilitation program after temporal lobe epilepsy surgery and other factors on employment 2 years after epilepsy surgery. Epilepsia 2014;55(5):725–33. doi: 10.1111/epi.12573. 25. Wilson SJ, Bladin PF, Saling MM. Paradoxical results in the cure of chronic illness: the „burden of normality“ as exemplified following seizure surgery. Epilepsy Behav 2004;5(1):13–21. 26. Seiam AH, Dhaliwal H, Wiebe S. Determinants of quality of life after epilepsy surgery: systematic review and evidence summary. Epilepsy Behav 2011;21(4):441–5. doi: 10.1016/j.yebeh.2011.05.005. 27. Meldolesi GN, Di Gennaro G, Quarato PP, et al. Changes in depression, anxiety, anger, and personality after resective surgery for drug-resistant temporal lobe epilepsy: a 2-year follow-up study. Epilepsy Res 2007;77(1):22–30. 28. Cankurtaran ES, Ulug B, Saygi S, et al. Psychiatric morbidity, quality of life, and disability in mesial temporal lobe epilepsy patients before and after anterior temporal lobectomy. Epilepsy Behav 2005;7(1):116–22. 29. Bjornaes H, Stabell KE, Heminghyt E, et al. Resective surgery for intractable focal epilepsy in patients with low IQ: predictors for seizure control and outcome with respect to seizures and neuropsycholo- Závěr Pacienti s farmakorezistentní epilepsií vytváří různorodou skupinu, u které by měla být posouzena možnost operační léčby. Operace nabízí pacientovi vyšší pravděpodobnost zbavení se epileptických záchvatů ve srovnání s léčbou farmakologickou. Po operaci stojí pacient před celou řadou dalších problémů a úskalí, dochází k významné změně jeho životní role z „nemocného“ člověka na člověka „zdravého“, na něhož jsou kladeny vyšší nároky. Na základě našeho dotazníkového šetření můžeme říci, že ač velká část pacientů podstoupila úspěšný epileptochirurgický výkon, který je zcela zbavil záchvatů, zaměstnání se podařilo najít jen malé části z nich. Je zde vidět prostor pro aktivitu sociálních pracovníků, pro rekvalifikační programy, pro programy zvyšování kvalifikace či programy motivace zaměstnavatelů, které by umožnily lepší zapojení našich bývalých pacientů do pracovního procesu. I přes tyto nedostatky „následné rehabilitace“ je epileptochirurgická léčba ze strany našich operantů hodnocena kladně, přináší jim v naprosté většině případů zvýšení kvality života a většina z nich by znovu podstoupila operační zákrok. Literatura 1. Tellez-Zenteno JF, Ronquillo LH, Wiebe S. Sudden unexpected death in epilepsy: evidence-based analysis of incidence and risk factors. Epilepsy Res 2005;65(1–2): 101–15. 2. Tomson T, Beghi E, Sundqvist A, et al. Medical risks in epilepsy: a review with focus on physical injuries, mortality, traffic accidents and their prevention. Epilepsy Res 2004;60(1):1–16. 3. Kwan P, Arzimanoglou A, Berg AT, et al. Definition of Drug Resistant Epilepsy: Consensus Proposal by the Ad Hoc Task Force of the ILAE Commission on Therapeutic Strategies. Aktuel Neurol 2010;37(8):372–81. doi: 10.1111/j.1528-1167.2009.02397.x. 4. Brázdil M. Proměny epileptochirurgie ve 21. století. Neurol Praxi 2015;16(2):77–9. 5. Spencer S, Huh L. Outcomes of epilepsy surgery in adults and children. Lancet Neurology 2008;7(6):525–37. doi: 10.1016/S1474-4422(08)70109-1. 6. Kuba R, Brázdil M, Novák Z, et al. Dlouhodobá účinnost resekčních epileptochirurgických zákroků 5 let od operace. Neurol Praxi 2008;9(3):166–70. 7. Kuba R, Brázdil M, Novák Z, et al. Kvalita života po resekční operaci pro farmakorezistentní epilepsii. Čes Slov Psychiat 2007;103:175–83. 8. Katz A, Marks DA, McCarthy G, et al. Does interictal spiking change prior to seizures? Clin Neurophysiol 1992;79(2):153–6. 9. Wiebe S, Blume WT, Girvin JP, et al. Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group. A randomized, controlled trial of sur- geryfortemporal-lobeepilepsy.NEnglJMed2001;345(5): 311–8. 10. Zarroli K, Tracy JI, Nei M, et al. Employment after anterior temporal lobectomy. Epilepsia 2011;52(5):925–31. doi: 10.1111/j.1528-1167.2011.03098.x. kéhokoliv závažného chronického onemocnění (týká se pacientů po transplantacích, onkologických pacientů, pacientů s kardiovaskulárním onemocněním). Pacienti a jejich rodiny, ač vyléčeni z chronického onemocnění, nejsou s výsledky léčby spokojeni. Pacienti si stěžují, že jim léčba nic pozitivního nepřinesla, a v některých případech dokonce litují, že léčbu podstoupili, přestože vedla k jejich uzdravení. Známky „syndromu břemene normality“ nalézáme v různém stupni až u 66 % pacientů v prvních dvou letech po epileptochirurgickém výkonu. Klinicky se může manifestovat psychickými obtížemi (nadměrné očekávání od sebe i od ostatních, lítost nad předchozím životem s nemocí), změnami behaviorálními (nadměrná denní aktivita ve snaze „dohnat vše zameškané“ nebo naopak stažení se do sebe, snížení aktivity) a sociálními (narušení vztahu s rodinou, partnerem, přáteli, obtíže s hledáním zaměstnání) [25,30,40]. Častěji se setkáváme se „syndromem tíhy normality“ u pacientů, kteří mají s léčbou spojeny nerealistická očekávání (vyřešení sociálních a rodinných problémů, zvýšení inteligence) než u pacientů, kteří chtějí operací dosáhnout realistických, dobře definovaných cílů (vymizení epileptických záchvatů, získání řidičského oprávnění). Poslední otázka v našem dotazníku byla hypotetická, dotazovala se pacienta, zda by se nechal znovu operovat, kdyby mohl vrátit čas. Naprostá většina pacientů, konkrétně 85,7 %, odpovědělo, že by operaci podstoupilo znovu. Tuto hodnotu považujeme za srovnatelnou s výsledky publikovanými z Mayo kliniky v Rochesteru, kde by 85 % pacientů podstoupilo epileptochirurgickou léčbu [41]. Operaci by znovu podstoupili všichni pacienti, kteří po operaci udávají zlepšení kvality života. Uvědomujeme si četné limitace naší studie. Jedná se o relativně malý počet respondentů, pacienti byli dotazováni v různém časovém intervalu od operačního výkonu. Jak bylo zmíněno již v úvodu, nebyl použit standardizovaný dotazník, což sice umožnilo získání odpovědí zcela anonymní formou (při dotazování nebyl přítomen ani lékař, ani psycholog, jehož postoj a pohled na chirurgii epilepsie by mohl ovlivňovat odpovědi pacienta), na druhé straně neumožňuje zcela přesné srovnání s výsledky ostatních prací. Přes všechny výše zmíněné nedostatky si myslíme, že je tato práce schopna reflektovat vztah našich pacientů k epileptochirurgické léčbě a benefity, které jim může tato léčba nabídnout. proLékaře.cz | 29.4.2020 262 EPILEPTOCHIRURGICKÁ LÉČBA ZLEPŠUJE KVALITU ŽIVOTA  VÝSLEDKY DOTAZNÍKOVÉ STUDIE Cesk Slov Neurol N 2016; 79/112(4): 430–439 439 38. Aydemir N, Ozkara C, Canbeyli R, et al. Changes in quality of life and self-perspective related to surgery in patients with temporal lobe epilepsy. Epi Behav 2004;5(5):735–42. 39. Chovaz CJ, McLachlan RS, Derry PA, et al. Psychosocial function following temporal lobectomy – influence of seizure control and learned helplessness. Seizure 1994;3(3):171–6. 40. Wilson S, Bladin P, Saling M. The „burden of normality“: concepts of adjustment after surgery for seizures. J Neurol Neurosurg Psychiatry 2001;70(5):649–56. 41. Wass CT, Rajala MM, Hughes JM, et al. Long-term follow-up of patients treated surgically for medically intractable epilepsy: results in 291 patients treated at Mayo Clinic Rochester between July 1972 and March 1985. Mayo Clin Proc 1996;71(11):1105–13. 34. Gilliam F, Kuzniecky R, Meador K, et al. Patient-oriented outcome assessment after temporal lobectomy for refractory epilepsy. Neurology 1999;53(4): 687–94. 35. Derry PA, McLachlan RS. Causal attributions for seizures – relation to preoperative psychological adjustment and postoperative psychosocial function in temporal-lobe epilepsy. J Epilepsy 1995;8(1):74–82. 36. Wheelock I, Peterson C, Buchtel HA. Presurgery expectations, postsurgery satisfaction, and psychosocial adjustment after epilepsy surgery. Epilepsia 1998;39(5):487–94. 37. Tanriverdi T, Olivier NP, Olivier A. Quality of life after extratemporal epilepsy surgery: a prospective clinical study. Clin Neurol Neurosur 2008;110(1):30–7. gical and psychosocial functioning. Epilepsia 2004;45(2): 131–9. 30. Preiss J, Vojtěch Z. Kvalita života po resekční operaci pro farmakorezistentní epilepsii. Čes Slov Psychiat 2007;103(41):175–83. 31. Maganti R, Rutecki P, Bell B, et al. Epilepsy surgery outcome among US veterans. Epilepsy Behav 2003;4(6):723–8. 32. Rausch R, Kraemer S, Pietras CJ, et al. Early and late cognitive changes following temporal lobe surgery for epilepsy. Neurology 2003;60(6):951–9. 33. Poochikian-Sarkissian S, Sidani S, Wennberg R, et al. Seizure freedom reduces illness intrusiveness and improves quality of life in epilepsy. Can J Neurol Sci 2008;35(3):280–6. proLékaře.cz | 29.4.2020 263 References: Adcock, J.E., Wise, R.G., Oxbury, J.M., Oxbury, S.M., Matthews, P.M., 2003. Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy. Neuroimage. 18(2), 423–438. https://doi.org/10.1016/S1053-8119(02)00013- 7 Alessio, A., Pereira, F.R., Sercheli, M.S., Rondina, J.M., Ozelo, H.B., Bilevicius, E., Pedro, T., Covolan, R.J., Damasceno, B.P., Cendes, F., 2013. Brain plasticity for verbal and visual memories in patients with mesial temporal lobe epilepsy and hippocampal sclerosis: an fMRI study. Hum Brain Mapp. 34(1), 186–199. https://doi.org/10.1002/hbm.21432 Alkawadri, R., So, N.K., Van Ness, P.C., Alexopoulos, A.V., 2013. Cingulate epilepsy: report of 3 electroclinical subtypes with surgical outcomes. JAMA Neurol. 70(8), 995–1002. https://doi.org/10.1001/jamaneurol.2013.2940 Allen, P.J., Fish, D.R., Smith, S.J.M., 1992. Very high-frequency rhythmic activity during SEEG suppression in frontal lobe epilepsy. Electroencephalogr Clin Neurophysiol. 82(2), 155–159. https://doi.org/10.1016/0013-4694(92)90160-J Aminoff, M.J., Scheinman, M.M., Griffin, J.C., Herre, J.M., 1988. Electrocerebral accompaniments of syncope associated with malignant ventricular arrhythmias. Ann Intern Med. 108(6), 791–796. https://doi.org/10.7326/0003-4819-108-6-791 Andermann, F., 2003. Pseudotemporal vs neocortical temporal epilepsy: things aren’t always where they seem to be. Neurology. 61(6), 732–733. https://doi.org/10.1212/WNL.61.6.732 Andrade, D.M., Zumsteg, D., Hamani, C., Hodaie, M., Sarkissian, S., Lozano, A.M., Wennberg, R.A., 2006. Long-term follow-up of patients with thalamic deep brain stimulation for epilepsy. Neurology. 66(10), 1571–1573. https://doi.org/10.1212/01.wnl.0000206364.19772.39 Andriola, M.R., Vitale, S.A., 2001. Vagus nerve stimulation in the developmentally disabled. Epilepsy Behav. 2(2), 129–134. https://doi.org/10.1006/ebeh.2001.0160 Arnold, S., Schlaug, G., Niemann, H., Ebner, A., Lüders, H., Witte, O.W., Seitz, R.J., 1996. Topography of interictal glucose hypometabolism in unilateral mesiotemporal epilepsy. Neurology. 46(5), 1422–1430. https://doi.org/10.1212/WNL.46.5.1422 Aubert, S., Wendling, F., Regis, J., McGonigal, A., Figarella-Branger, D., Peragut, J.-C., Girard, N., Chauvel, P., Bartolomei, F., 2009. Local and remote epileptogenicity in focal cortical dysplasias and neurodevelopmental tumours. Brain. 132(Pt 11), 3072–3086. https://doi.org/10.1093/brain/awp242 Aull-Watschinger, S., Pataraia, E., Czech, T., Baumgartner, C., 2008. Outcome predictors for surgical treatment of temporal lobe epilepsy with hippocampal sclerosis. Epilepsia. 49(8), 1308– 1316. https://doi.org/10.1111/j.1528-1167.2008.01732.x Avbersek, A., Sisodiya, S., 2010. Does the primary literature provide support for clinical signs used to distinguish psychogenic nonepileptic seizures from epileptic seizures? J Neurol Neurosurg Psychiatry. 81(7), 719–725. http://dx.doi.org/10.1136/jnnp.2009.197996 264 Axmacher, N., Elger, C.E., Fell, J., 2008. Ripples in the medial temporal lobe are relevant for human memory consolidation. Brain. 131(Pt 7), 1806–1817. https://doi.org/10.1093/brain/awn103 Babajani-Feremi, A., Noorizadeh, N., Mudigoudar, B., Wheless, J.W., 2018. Predicting seizure outcome of vagus nerve stimulation using MEG-based network topology. Neuroimage Clin. 19, 990–999. https://doi.org/10.1016/j.nicl.2018.06.017 Bartolomei, F., Chauvel, P., Wendling, F., 2008. Epileptogenicity of brain structures in human temporal lobe epilepsy: a quantified study from intracerebral EEG. Brain. 131(Pt 7), 1818–1830. https://doi.org/10.1093/brain/awn111 Baxendale, S.A., van Paesschen, W., Thompson, P.J., Connelly, A., Duncan, J.S., Harkness, W.F., Shorvon, S.D., 1998. The relationship between quantitative MRI and neuropsychological functioning in temporal lobe epilepsy. Epilepsia. 39(2), 158–166. https://doi.org/10.1111/j.1528- 1157.1998.tb01353.x Bechara, A., Damasio, H., Damasio, A.R., 2000. Emotion, decision making and the orbitofrontal cortex. Cereb Cortex. 10(3), 295–307. https://doi.org/10.1093/cercor/10.3.295 Bell, M.L., Rao, S., So, E.L., Trenerry, M., Kazemi, N., Stead, S.M., Cascino, G., Marsh, R., Meyer, F.B., Watson, R.E., Giannini, C., Worrell, G.A., 2009. Epilepsy surgery outcomes in temporal lobe epilepsy with a normal MRI. Epilepsia. 50(9), 2053–2060. https://doi.org/10.1111/j.1528-1167.2009.02079.x Ben-Menachem, E., 2002. Vagus-nerve stimulation for the treatment of epilepsy. Lancet Neurol. 1(8), 477–482. https://doi.org/10.1016/S1474-4422(02)00220-X Ben-Menachem, E., Mañon-Espaillat, R., Ristanovic, R., Wilder, B.J., Stefan, H., Mirza, W., Tarver, W.B., Wernicke, J.F., First International Vagus Nerve Stimulation Study Group, 1994. Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. Epilepsia. 35, 616–626. https://doi.org/10.1111/j.1528-1157.1994.tb02482.x Benninger, C., Matthis, P., Scheffner, D., 1984. EEG development of healthy boys and girls. Results of a longitudinal study. Electroencephalogr Clin Neurophysiol. 57(1), 1–12. https://doi.org/10.1016/0013-4694(84)90002-6 Bien, C.G., Szinay, M., Wagner, J., Clusmann, H., Becker, A.J., Urbach, H., 2009. Characteristics and surgical outcomes of patients with refractory magnetic resonance imaging-negative epilepsies. Arch Neurol. 66(12), 1491–1499. https://doi.org/10.1001/archneurol.2009.283 Bernasconi, A., Cendes, F., Theodore, W.H., Gill, R.S., Koepp, M.J., Hogan, R.E., Jackson, G.D., Federico, P., Labate, A., Vaudano, A.E., Blümcke, I., Ryvlin, P., Bernasconi, N., 2019. Recommendations for the use of structural magnetic resonance imaging in the care of patients with epilepsy: a consensus report from the International League Against Epilepsy Neuroimaging Task Force. Epilepsia. 60(6), 1054–1068. https://doi.org/10.1111/epi.15612 Bonini, F., McGonigal, A., Trébuchon, A., Gavaret, M., Bartolomei, F., Giusiano, B., Chauvel, P., 2014. Frontal lobe seizures: from clinical semiology to localization. Epilepsia. 55(2), 264–277. https://doi.org/10.1111/epi.12490 Boscolo Galazzo, I., Mattoli, M.V., Pizzini, F.B., De Vita, E., Barnes, A., Duncan, J.S., Jäger, H.R., Golay, X., Bomanji, J.B., Koepp, M., Groves, A.M., Fraioli, F., 2016. Cerebral metabolism and perfusion in MR-negative individuals with refractory focal epilepsy assessed by simultaneous 265 acquisition of (18)F-FDG PET and arterial spin labeling. Neuroimage Clin. 11, 648–657. https://doi.org/10.1016/j.nicl.2016.04.005 Bragin, A., Azizyan, A., Almajano, J., Engel Jr., J., 2009. The cause of the imbalance in the neuronal network leading to seizure activity can be predicted by the electrographic pattern of the seizure onset. J Neurosci. 29(11), 3660–3671. https://doi.org/10.1523/JNEUROSCI.5309-08.2009 Bragin, A., Engel Jr., J., Wilson, C.L., Fried, I., Mathern, G.W., 1999. Hippocampal and entorhinal cortex high-frequency oscillations (100-500 Hz) in human epileptic brain and in kainic acid-treated rats with chronic seizures. Epilepsia. 40(2), 127–137. https://doi.org/10.1111/j.1528- 1157.1999.tb02065.x Bragin, A., Wilson, C.L., Engel Jr, J., 2007. Voltage depth profiles of high-frequency oscillations after kainic acid-induced status epilepticus. Epilepsia. 48(Suppl 5), 35–40. https://doi.org/10.1111/j.1528-1167.2007.01287.x Brázdil, M., Pail, M., Halámek, J., Plešinger, F., Cimbálník, J., Roman, R., Klimeš, P., Daniel, P., Chrastina, J., Brichtová, E., Rektor, I., Worrell, G.A., Jurák, P., 2017. Very high-frequency oscillations: novel biomarkers of the epileptogenic zone. Ann Neurol. 82(2), 299–310. https://doi.org/10.1002/ana.25006 Brázdil, M., Doležalová, I., Koritáková, E., Chládek, J., Roman, R., Pail, M., Jurák, P., Shaw, D.J., Chrastina, J., 2019. EEG reactivity predicts individual efficacy of vagal nerve stimulation in intractable epileptics. Front Neurol. 10, 392. https://doi.org/10.3389/fneur.2019.00392 Brigo, F., Nardone, R., Ausserer, H., Storti, M., Tezzon, F., Manganotti, P., Bongiovanni, L.G., 2013. The diagnostic value of urinary incontinence in the differential diagnosis of seizures. Seizure. 22(2), 85–90. https://doi.org/10.1016/j.seizure.2012.10.011 Buzsáki, G., Horváth, Z., Urioste, R., Hetke, J., Wise, K., 1992. High-frequency network oscillation in the hippocampus. Science. 256(5059), 1025–1027. http://doi.org/10.1126/science.1589772 Carne, R.P., Cook, M.J., MacGregor, L.R., Kilpatrick, C.J., Hicks, R.J., O’Brien, T.J., 2007a. “Magnetic resonance imaging negative positron emission tomography positive” temporal lobe epilepsy: FDG-PET pattern differs from mesial temporal lobe epilepsy. Mol Imaging Biol. 9(1), 32–42. https://doi.org/10.1007/s11307-006-0073-0 Carne, R.P., O’Brien, T.J., Kilpatrick, C.J., MacGregor, L.R., Hicks, R.J., Murphy, M.A., Bowden, S.C., Kaye, A.H., Cook, M.J., 2004. MRI-negative PET-positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain. 127(Pt 10), 2276–2285. https://doi.org/10.1093/brain/awh257 Carne, R.P., O’Brien, T.J., Kilpatrick, C.J., MacGregor, L.R., Litewka, L., Hicks, R.J., Cook, M.J., 2007b. ‘MRI-negative PET-positive’ temporal lobe epilepsy (TLE) and mesial TLE differ with quantitative MRI and PET: a case control study. BMC Neurol. 7, 16. https://doi.org/10.1186/1471- 2377-7-16 Charlson, F.J., Baxter, A.J., Dua, T., Degenhardt, L., Whiteford, H.A., Vos, T., 2016. Excess mortality from mental, neurological, and substance use disorders in the global burden of disease study 2010, in: Patel, V., Chisholm, D., Dua, T., Laxminarayan, R., Medina-Mora, M.E. (Eds.), Mental, Neurological, and Substance Use Disorders: Disease Control Priorities, Third Edition (Volume 4). The International Bank for Reconstruction and Development / The World Bank, Washington (DC). 266 Child, N.D., Benarroch, E.E., 2013. Anterior nucleus of the thalamus: functional organization and clinical implications. Neurology. 81(21), 1869–1876. https://doi.org/10.1212/01.wnl.0000436078.95856.56 Chrastina, J., Kočvarová, J., Novák, Z., Doležalová, I., Svoboda, M., Brázdil, M., 2018a. Older age and longer epilepsy duration do not predict worse seizure reduction outcome after vagus nerve stimulation. J Neurol Surg A Cent Eur Neurosurg. 79(2), 152–158. https://doi.org/10.1055/s-0037- 1607396 Chrastina, J., Novák, Z., Zeman, T., Kočvarová, J., Pail, M., Doležalová, I., Jarkovský, J., Brázdil, M., 2018b. Single-center long-term results of vagus nerve stimulation for epilepsy: a 10-17 year follow-up study. Seizure. 59, 41–47. https://doi.org/10.1016/j.seizure.2018.04.022 Chung, M.Y., Walczak, T.S., Lewis, D.V., Dawson, D.V., Radtke, R., 1991. Temporal lobectomy and independent bitemporal interictal activity - what degree of lateralization is sufficient? Epilepsia. 32(2), 195–201. https://doi.org/10.1111/j.1528-1157.1991.tb05244.x Cooper, I.S., Amin, I., Gilman, S., 1973. The effect of chronic cerebellar stimulation upon epilepsy in man. Trans Am Neurol Assoc. 98, 192–196. Cooper, I.S., Upton, A.R., Amin, I., 1980. Reversibility of chronic neurologic deficits. Some effects of electrical stimulation of the thalamus and internal capsule in man. Appl Neurophysiol. 43(3-5), 244–258. https://doi.org/10.1159/000102263 Crompton, D.E., Berkovic, S.F., 2009. The borderland of epilepsy: clinical and molecular features of phenomena that mimic epileptic seizures. Lancet Neurol. 8(4), 370–381. https://doi.org/10.1016/S1474-4422(09)70059-6 Curio, G., Mackert, B.-M., Burghoff, M., Neumann, J., Nolte, G., Scherg, M., Marx, P., 1997. Somatotopic source arrangement of 600 Hz oscillatory magnetic fields at the human primary somatosensory hand cortex. Neurosci Lett. 234(2-3), 131–134. https://doi.org/10.1016/S0304- 3940(97)00690-3 Danielsson, S., Viggedal, G., Gillberg, C., Olsson, I., 2008. Lack of effects of vagus nerve stimulation on drug-resistant epilepsy in eight pediatric patients with autism spectrum disorders: a prospective 2-year follow-up study. Epilepsy Behav. 12(2), 298–304. https://doi.org/10.1016/j.yebeh.2007.10.007 Davis, R., Emmonds, S.E., 1992. Cerebellar stimulation for seizure control: 17-year study. Stereotact Funct Neurosurg. 58(1-4), 200–208. https://doi.org/10.1159/000098996 De Herdt, V., Boon, P., Ceulemans, B., Hauman, H., Lagae, L., Legros, B., Sadzot, B., Van Bogaert, P., van Rijckevorsel, K., Verhelst, H., Vonck, K., 2007. Vagus nerve stimulation for refractory epilepsy: a Belgian multicenter study. Eur J Paediatr Neurol. 11(5), 261–269. https://doi.org/10.1016/j.ejpn.2007.01.008 Delbeke, D., Lawrence, S.K., Abou-Khalil, B.W., Blumenkopf, B., Kessler, R.M., 1996. Postsurgical outcome of patients with uncontrolled complex partial seizures and temporal lobe hypometabolism on 18FDG-positron emission tomography. Invest Radiol. 31(5), 261–266. https://doi.org/10.1097/00004424-199605000-00003 Doležalová, I., Brázdil, M., Chrastina, J., Hemza, J., Hermanová, M., Janoušová, E., Pažourková, M., Kuba, R., 2016a. Differences between mesial and neocortical magnetic-resonance-imaging- 267 negative temporal lobe epilepsy. Epilepsy Behav. 61, 21–26. https://doi.org/10.1016/j.yebeh.2016.04.027 Doležalová, I., Brázdil, M., Hermanová, M., Horáková, I., Rektor, I., Kuba, R., 2013a. Intracranial EEG seizure onset patterns in unilateral temporal lobe epilepsy and their relationship to other variables. Clin Neurophysiol. 124(6), 1079–1088. https://doi.org/10.1016/j.clinph.2012.12.046 Doležalová, I., Brázdil, M., Hermanová, M., Janoušová, E., Kuba, R., 2014. Effect of partial drug withdrawal on the lateralization of interictal epileptiform discharges and its relationship to surgical outcome in patients with hippocampal sclerosis. Epilepsy Res. 108(8), 1406–1416. https://doi.org/10.1016/j.eplepsyres.2014.06.009 Doležalová, I., Pešlová, E., Michnová, M., Nečasová, T., Kočvarová, J., Musilová, K., Rektor, I., Brázdil, M., 2016b. Epilepsy surgery improves quality of life - results of a questionnaire study. Česká a Slovenská neurologie a neurochirurgie. 79(4), 430–439. Doležalová, I., Brázdil, M., Rektor, I., Tyrlíková, I., Kuba, R., 2013b. Syncope with atypical trunk convulsions in a patient with malignant arrhythmia. Epileptic Disord. 15(2), 171–174. https://doi.org/10.1684/epd.2013.0564 Doležalová, I., Kunst, J., Kojan, M., Chrastina, J., Baláž, M., Brázdil, M., 2019. Anterior thalamic deep brain stimulation in epilepsy and persistent psychiatric side effects following discontinuation. Epilepsy Behavior Rep. 12, 100344. https://doi.org/10.1016/j.ebr.2019.100344 Doležalová, I., Schachter, S., Chrastina, J., Hemza, J., Hermanová, M., Rektor, I., Pažourková, M., Brázdil, M., 2017. Atypical handedness in mesial temporal lobe epilepsy. Epilepsy Behav. 72, 78– 81. https://doi.org/10.1016/j.yebeh.2017.01.034 Drzezga, A., Arnold, S., Minoshima, S., Noachtar, S., Szecsi, J., Winkler, P., Römer, W., Tatsch, K., Weber, W., Bartenstein, P., 1999. F-18-FDG PET studies in patients with extratemporal and temporal epilepsy: evaluation of an observer-independent analysis. J Nucl Med. 40(5), 737–746. Durwen, H.F., Elger, C.E., Helmstaedter, C., Penin, H., 1989. Circumscribed improvement of cognitive performance in temporal-lobe epilepsy patients with intractable seizures following reduction of anticonvulsant medication. J Epilepsy. 2(3), 147–153. https://doi.org/10.1016/0896- 6974(89)90029-7 Duun-Henriksen, J., Madsen, R.E., Remvig, L.S., Thomsen, C.E., Sorensen, H.B., Kjaer, T.W., 2012. Automatic detection of childhood absence epilepsy seizures: toward a monitoring device. Pediatr Neurol. 46(5), 287–292. https://doi.org/10.1016/j.pediatrneurol.2012.02.018 Elliott, R.E., Morsi, A., Kalhorn, S.P., Marcus, J., Sellin, J., Kang, M., Silverberg, A., Rivera, E., Geller, E., Carlson, C., Devinsky, O., Doyle, W.K., 2011. Vagus nerve stimulation in 436 consecutive patients with treatment-resistant epilepsy: long-term outcomes and predictors of response. Epilepsy Behav. 20(1), 57–63. https://doi.org/10.1016/j.yebeh.2010.10.017 Enatsu, R., Bulacio, J., Nair, D.R., Bingaman, W., Najm, I., Gonzalez-Martinez, J., 2014. Posterior cingulate epilepsy: clinical and neurophysiological analysis. J Neurol Neurosurg Psychiatry. 85(1), 44–50. http://dx.doi.org/10.1136/jnnp-2013-305604 Engel Jr., J., 2008. Progress in epilepsy: reducing the treatment gap and the promise of biomarkers. Curr Opin Neurol. 21(2), 150–154. https://doi.org/10.1097/WCO.0b013e3282f4edc3 268 Englot, D.J., Birk, H., Chang, E.F., 2017. Seizure outcomes in nonresective epilepsy surgery: an update. Neurosurg Rev. 40(2), 181–194. https://doi.org/10.1007/s10143-016-0725-8 Englot, D.J., Chang, E.F., Auguste, K.I., 2011. Vagus nerve stimulation for epilepsy: a metaanalysis of efficacy and predictors of response. J Neurosurg. 115(6), 1248–1255. https://doi.org/10.3171/2011.7.JNS11977 Englot, D.J., Rolston, J.D., Wright, C.W., Hassnain, K.H., Chang, E.F., 2016. Rates and predictors of seizure freedom with vagus nerve stimulation for intractable epilepsy. Neurosurgery. 79(3), 345–353. https://doi.org/10.1227/NEU.0000000000001165 Faught, E., Kuzniecky, R.I., Hurst, D.C., 1992. Ictal EEG wave forms from epidural electrodes predictive of seizure control after temporal lobectomy. Electroencephalogr Clin Neurophysiol. 83(4), 229–235. https://doi.org/10.1016/0013-4694(92)90116-Y Fisher, R., Salanova, V., Witt, T., Worth, R., Henry, T., Gross, R., Oommen, K., Osorio, I., Nazzaro, J., Labar, D., Kaplitt, M., Sperling, M., Sandok, E., Neal, J., Handforth, A., Stern, J., DeSalles, A., Chung, S., Shetter, A., Bergen, D., Bakay, R., Henderson, J., French, J., Baltuch, G., Rosenfeld, W., Youkilis, A., Marks, W., Garcia, P., Barbaro, N., Fountain, N., Bazil, C., Goodman, R., McKhann, G., Babu Krishnamurthy, K., Papavassiliou, S., Epstein, C., Pollard, J., Tonder, L., Grebin, J., Coffey, R., Graves, N., Sante Study Group, 2010. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia. 51(5), 899–908. https://doi.org/10.1111/j.1528-1167.2010.02536.x Fisher, R.S., Webber, W.R., Lesser, R.P., Arroyo, S., Uematsu, S., 1992. High-frequency EEG activity at the start of seizures. J Clin Neurophysiol. 9(3), 441–448. https://doi.org/10.1097/00004691- 199207010-00012 Foerster, O., Altenburger, H., 1935. Elektrobiologische Vorgänge an der menschlichen Hirnrinde. Deutsche Zeitschrift für Nervenheilkunde. 135, 277–288. https://doi.org/10.1007/BF01732786 Frauscher, B., Bartolomei, F., Kobayashi, K., Cimbalnik, J., van ‘t Klooster, M.A., Rampp, S., Otsubo, H., Höller, Y., Wu, J.Y., Asano, E., Engel Jr., J., Kahane, P., Jacobs, J., Gotman, J., 2017. High-frequency oscillations: the state of clinical research. Epilepsia. 58(8), 1316–1329. https://doi.org/10.1111/epi.13829 Frauscher, B., von Ellenrieder, N., Zelmann, R., Doležalová, I., Minotti, L., Olivier, A., Hall, J., Hoffmann, D., Nguyen, D.K., Kahane, P., Dubeau, F., Gotman, J., 2018. Atlas of the normal intracranial electroencephalogram: neurophysiological awake activity in different cortical areas. Brain. 141(4), 1130–1144. https://doi.org/10.1093/brain/awy035 Fujiwara, H., Greiner, H.M., Lee, K.H., Holland-Bouley, K.D., Seo, J.H., Arthur, T., Mangano, F.T., Leach, J.L., Rose, D.F., 2012. Resection of ictal high-frequency oscillations leads to favorable surgical outcome in pediatric epilepsy. Epilepsia. 53(9), 1607–1617. https://doi.org/10.1111/j.1528- 1167.2012.03629.x Gastaut, H., 1949. Enregistrement sous-cortical de l'activité électrique spontanée et provoquée du lobe occipital humain. Electroencephalogr Clin Neurophysiol. 1(1-4), 205–221. https://doi.org/10.1016/0013-4694(49)90177-7 Geller, E.B., 2018. Responsive neurostimulation: review of clinical trials and insights into focal epilepsy. Epilepsy Behav. 88S, 11–20. https://doi.org/10.1016/j.yebeh.2018.06.042 269 Gotman, J., Koffler, D.J., 1989. Interictal spiking increases after seizures but does not after decrease in medication. Electroencephalogr Clin Neurophysiol. 72(1), 7–15. https://doi.org/10.1016/0013-4694(89)90026-6 Gotman, J., Marciani, M.G., 1985. Electroencephalographic spiking activity, drug levels, and seizure occurrence in epileptic patients. Ann Neurol. 17(6), 597–603. https://doi.org/10.1002/ana.410170612 Graf, M., Niedermeyer, E., Schiemann, J., Uematsu, S., Long, D.M., 1984. Electrocorticography: information derived from intraoperative recordings during seizure surgery. Clin Electroencephalogr. 15(2), 83–91. https://doi.org/10.1177/155005948401500204 Haegelen, C., Perucca, P., Châtillon, C.-E., Andrade-Valença, L., Zelmann, R., Jacobs, J., Collins, D.L., Dubeau, F., Olivier, A., Gotman, J., 2013. High-frequency oscillations, extent of surgical resection, and surgical outcome in drug-resistant focal epilepsy. Epilepsia. 54(5), 848–857. https://doi.org/10.1111/epi.12075 Haller, S., Zaharchuk, G., Thomas, D.L., Lovblad, K.O., Barkhof, F., Golay, X., 2016. Arterial spin labeling perfusion of the brain: emerging clinical applications. Radiology. 281(2), 337–356. https://doi.org/10.1148/radiol.2016150789 Handforth, A., DeGiorgio, C.M., Schachter, S.C., Uthman, B.M., Naritoku, D.K., Tecoma, E.S., Henry, T.R., Collins, S.D., Vaughn, B.V., Gilmartin, R.C., Labar, D.R., Morris 3rd ., G.L., Salinsky, M.C., Osorio, I., Ristanovic, R.K., Labiner, D.M., Jones, J.C., Murphy, J.V., Ney, G.C., Wheless, J.W., 1998. Vagus nerve stimulation therapy for partial-onset seizures: a randomized active-control trial. Neurology. 51(1), 48–55. https://doi.org/10.1212/WNL.51.1.48 Hardy, S.G., Miller, J.W., Holmes, M.D., Born, D.E., Ojemann, G.A., Dodrill, C.B., Hallam, D.K., 2003. Factors predicting outcome of surgery for intractable epilepsy with pathologically verified mesial temporal sclerosis. Epilepsia. 44(4), 565–568. https://doi.org/10.1046/j.1528- 1157.2003.39202.x Harroud, A., Bouthillier, A., Weil, A.G., Nguyen, D.K., 2012. Temporal lobe epilepsy surgery failures: a review. Epilepsy Res Treat. 2012, 201651. https://doi.org/10.1155/2012/201651 Hayne, R., Meyers, R., 1950. An improved model of a human stereotaxic instrument. J Neurosurg. 7(5), 463–466. https://doi.org/10.3171/jns.1950.7.5.0463 Helmstaedter, C., Kurthen, M., Linke, D.B., Elger, C.E., 1994. Right-hemisphere restitution of language and memory functions in right-hemisphere language-dominant patients with left temporal-lobe epilepsy. Brain. 117(Pt 4), 729–737. https://doi.org/10.1093/brain/117.4.729 Helmstaedter, C., Pohl, C., Elger, C.E., 1995. Relations between verbal and nonverbal memory performance - evidence of confounding effects particularly in patients with right temporal-lobe epilepsy. Cortex. 31(2), 345–355. https://doi.org/10.1016/S0010-9452(13)80367-X Hirsch, L.J., Spencer, S.S., Williamson, P.D., Spencer, D.D., Mattson, R.H., 1991. Comparison of bitemporal and unitemporal epilepsy defined by depth electroencephalography. Ann Neurol. 30(3), 340–346. https://doi.org/10.1002/ana.410300305 Hodaie, M., Wennberg, R.A., Dostrovsky, J.O., Lozano, A.M., 2002. Chronic anterior thalamus stimulation for intractable epilepsy. Epilepsia. 43(6), 603–608. https://doi.org/10.1046/j.1528- 1157.2002.26001.x 270 Holmes, M.D., Miles, A.N., Dodrill, C.B., Ojemann, G.A., Wilensky, A.J., 2003. Identifying potential surgical candidates in patients with evidence of bitemporal epilepsy. Epilepsia. 44(8), 1075–1079. https://doi.org/10.1046/j.1528-1157.2003.58302.x Hopfengärtner, R., Kasper, B.S., Graf, W., Gollwitzer, S., Kreiselmeyer, G., Stefan, H., Hamer, H., 2014. Automatic seizure detection in long-term scalp EEG using an adaptive thresholding technique: a validation study for clinical routine. Clin Neurophysiol. 125(7), 1346–1352. https://doi.org/10.1016/j.clinph.2013.12.104 Ibrahim, G.M., Sharma, P., Hyslop, A., Guillen, M.R., Morgan, B.R., Wong, S., Abel, T.J., Elkaim, L., Cajigas, I., Shah, A.H., Fallah, A., Weil, A.G., Altman, N., Bernal, B., Medina, S., Widjaja, E., Jayakar, P., Ragheb, J., Bhatia, S., 2017. Presurgical thalamocortical connectivity is associated with response to vagus nerve stimulation in children with intractable epilepsy. Neuroimage Clin. 16, 634–642. ttps://doi.org/10.1016/j.nicl.2017.09.015 Isnard, J., Guénot, M., Ostrowsky, K., Sindou, M., Mauguière, F., 2000. The role of the insular cortex in temporal lobe epilepsy. Ann Neurol. 48(4), 614–623. http://dx.doi.org/10.1002/1531- 8249(200010)48:4<614::AID-ANA8>3.0.CO;2-S Jacobs, J., LeVan, P., Châtillon, C.-É., Olivier, A., Dubeau, F., Gotman, J., 2009. High frequency oscillations in intracranial EEGs mark epileptogenicity rather than lesion type. Brain. 132(Pt 4), 1022–1037. https://doi.org/10.1093/brain/awn351 Janszky, J., Janszky, I., Schulz, R., Hoppe, M., Behne, F., Pannek, H.W., Ebner, A., 2005. Temporal lobe epilepsy with hippocampal sclerosis: predictors for long-term surgical outcome. Brain. 128(Pt 2), 395–404. https://doi.org/10.1093/brain/awh358 Järvenpää, S., Peltola, J., Rainesalo, S., Leinonen, E., Lehtimäki, K., Järventausta, K., 2018. Reversible psychiatric adverse effects related to deep brain stimulation of the anterior thalamus in patients with refractory epilepsy. Epilepsy Behav. 88, 373–379. https://doi.org/10.1016/j.yebeh.2018.09.006 Jaseja, H., 2010. EEG-desynchronization as the major mechanism of anti-epileptic action of vagal nerve stimulation in patients with intractable seizures: clinical neurophysiological evidence. Med Hypotheses. 74(5), 855–856. https://doi.org/10.1016/j.mehy.2009.11.031 Jasper, H., Penfield, W., 1949. Electrocorticograms in man: effect of voluntary movement upon the electrical activity of the precentral gyrus. Archiv für Psychiatrie und Nervenkrankheiten. 183, 163–174. https://doi.org/10.1007/BF01062488 Jehi, L., Friedman, D., Carlson, C., Cascino, G., Dewar, S., Elger, C., Engel Jr., J., Knowlton, R., Kuzniecky, R., McIntosh, A., O’Brien, T.J., Spencer, D., Sperling, M.R., Worrell, G., Bingaman, B., Gonzalez-Martinez, J., Doyle, W., French, J., 2015. The evolution of epilepsy surgery between 1991 and 2011 in nine major epilepsy centers across the United States, Germany, and Australia. Epilepsia. 56(10), 1526–1533. https://doi.org/10.1111/epi.13116 Jokeit, H., Luerding, R., Ebner, A., 2000. Cognitive impairment in temporal-lobe epilepsy. Lancet. 355(9208), 1018–1019. https://doi.org/10.1016/S0140-6736(05)74765-6 Juhász, C., Chugani, D.C., Muzik, O., Shah, A., Shah, J., Watson, C., Canady, A., Chugani, H.T., 2001. Relationship of flumazenil and glucose PET abnormalities to neocortical epilepsy surgery outcome. Neurology. 56(12), 1650–1658. https://doi.org/10.1212/WNL.56.12.1650 271 Kasowski, H.J., Stoffman, M.R., Spencer, S.S., Spencer, D.D., 2003. Surgical management of parietal lobe epilepsy. Adv Neurol. 93, 347–356. Kerrigan, J.F., Litt, B., Fisher, R.S., Cranstoun, S., French, J.A., Blum, D.E., Dichter, M., Shetter, A., Baltuch, G., Jaggi, J., Krone, S., Brodie, M., Rise, M., Graves, N., 2004. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia. 45(4), 346–354. https://doi.org/10.1111/j.0013-9580.2004.01304.x Kim, B.S., Lee, S.T., Yun, T.J., Lee, S.K., Paeng, J.C., Jun, J., Kang, K.M., Choi, S.H., Kim, J.H., Sohn, C.H., 2016. Capability of arterial spin labeling MR imaging in localizing seizure focus in clinical seizure activity. Eur J Radiol. 85(7), 1295–1303. https://doi.org/10.1016/j.ejrad.2016.04.015 Ko, D., Heck, C., Grafton, S., Apuzzo, M.L., Couldwell, W.T., Chen, T., Day, J.D., Zelman, V., Smith, T., DeGiorgio, C.M., 1996. Vagus nerve stimulation activates central nervous system structures in epileptic patients during PET H2(15)O blood flow imaging. Neurosurgery. 39(2), 426–430; discussion 430–431. https://doi.org/10.1097/00006123-199608000-00061 Kojan, M., Doležalová, I., Koriťáková, E., Mareček, R., Řehák, Z., Hermanová, M., Brázdil, M., Rektor, I., 2018. Predictive value of preoperative statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy with hippocampal sclerosis. Epilepsy Behav. 79, 46– 52. https://doi.org/10.1016/j.yebeh.2017.11.014 Krendl, R., Lurger, S., Baumgartner, C., 2008. Absolute spike frequency predicts surgical outcome in TLE with unilateral hippocampal atrophy. Neurology. 71(6), 413–418. https://doi.org/10.1212/01.wnl.0000310775.87331.90 Krucoff, M.O., Chan, A.Y., Harward, S.C., Rahimpour, S., Rolston, J.D., Muh, C., Englot, D.J., 2017. Rates and predictors of success and failure in repeat epilepsy surgery: a meta-analysis and systematic review. Epilepsia. 58(12), 2133–2142. https://doi.org/10.1111/epi.13920 Kuba, R., Brázdil, M., Kalina, M., Procházka, T., Hovorka, J., Nežádal, T., Hadač, J., Brožová, K., Sebroňová, V., Komárek, V., Marusič, P., Ošlejšková, H., Zárubová, J., Rektor, I., 2009. Vagus nerve stimulation: longitudinal follow-up of patients treated for 5 years. Seizure. 18(4), 269–274. https://doi.org/10.1016/j.seizure.2008.10.012 Kuba, R., Tyrlíková, I., Chrastina, J., Slaná, B., Pažourková, M., Hemza, J., Brázdil, M., Novák, Z., Hermanová, M., Rektor, I., 2011. “MRI-negative PET-positive” temporal lobe epilepsy: invasive EEG findings, histopathology, and postoperative outcomes. Epilepsy Behav. 22(3), 537– 541. https://doi.org/10.1016/j.yebeh.2011.08.019 Kugler, J., 1991. Electroencephalography 60 years later. Recenti Prog Med. 82(3), 163–165. Kulju, T., Haapasalo, J., Lehtimäki, K., Rainesalo, S., Peltola, J., 2018. Similarities between the responses to ANT‐DBS and prior VNS in refractory epilepsy. Brain Behav. 8(6), e00983. https://doi.org/10.1002/brb3.983 Labar, D., 2004. Vagus nerve stimulation for 1 year in 269 patients on unchanged antiepileptic drugs. Seizure. 13(6), 392–398. https://doi.org/10.1016/j.seizure.2003.09.009 Lee, C.-Y., Tabesh, A., Nesland, T., Jensen, J.H., Helpern, J.A., Spampinato, M.V., Bonilha, L., 2014. Human brain asymmetry in microstructural connectivity demonstrated by diffusional kurtosis imaging. Brain Res. 1588, 73–80. https://doi.org/10.1016/j.brainres.2014.09.002 272 Lee, K.J., Jang, K.S., Shon, Y.M., 2006a. Chronic deep brain stimulation of subthalamic and anterior thalamic nuclei for controlling refractory partial epilepsy. Acta Neurochir Suppl. 99, 87– 91. https://doi.org/10.1007/978-3-211-35205-2_17 Lee, S.-A., Spencer, D.D., Spencer, S.S., 2000. Intracranial EEG seizure-onset patterns in neocortical epilepsy. Epilepsia. 41(3), 297–307. https://doi.org/10.1111/j.1528-1157.2000.tb00159.x Lee, S.K., Lee, S.-Y., Yun, C.-H., Lee, H.-Y., Lee, J.-S., Lee, D.-S., 2006b. Ictal SPECT in neocortical epilepsies: clinical usefulness and factors affecting the pattern of hyperperfusion. Neuroradiology 48(9), 678–684. https://doi.org/10.1007/s00234-006-0106-z Leijten, F.S.S., Dutch TeleEpilepsy Consortium. 2018. Multimodal seizure detection: a review. Epilepsia. 59(Suppl 1), 42–47. https://doi.org/10.1111/epi.14047 Li, M.C.H., Cook, M.J., 2018. Deep brain stimulation for drug-resistant epilepsy. Epilepsia. 59(2), 273–290. https://doi.org/10.1111/epi.13964 Liava, A., Mai, R., Tassi, L., Cossu, M., Sartori, I., Nobili, L., Lo Russo, G., Francione, S., 2014. Paediatric epilepsy surgery in the posterior cortex: a study of 62 cases. Epileptic Disord. 16, 141– 164. https://doi.org/10.1684/epd.2014.0648 Lim, S.-N., Lee, S.-T., Tsai, Y.-T., Chen, I.-A., Tu, P.-H., Chen, J.-L., Chang, H.-W., Su, Y.-C., Wu, T., 2007. Electrical stimulation of the anterior nucleus of the thalamus for intractable epilepsy: a long-term follow-up study. Epilepsia. 48(2), 342–347. https://doi.org/10.1111/j.1528- 1167.2006.00898.x Lin, J.-T., Ziegler, D.K., Lai, C.-W., Bayer, W., 1982. Convulsive syncope in blood donors. Ann Neurol. 11(5), 525–528. https://doi.org/10.1002/ana.410110513 Liu, H., Yang, Z., Huang, L., Qu, W., Hao, H., Li, L., 2017. Heart-rate variability indices as predictors of the response to vagus nerve stimulation in patients with drug-resistant epilepsy. Epilepsia. 58(6), 1015–1022. https://doi.org/10.1111/epi.13738 Looney, D., Kidmose, P., Park, C., Ungstrup, M., Rank, M., Rosenkranz, K., Mandic, D., 2012. The in-the-ear recording concept: user-centered and wearable brain monitoring. IEEE Pulse. 3(6), 32–42. ttps://doi.org/10.1109/MPUL.2012.2216717 Loring, D.W., Strauss, E., Hermann, B.P., Perrine, K., Trenerry, M.R., Barr, W.B., Westerveld, M., Chelune, G.J., Lee, G.P., Meador, K.J., 1999. Effects of anomalous language representation on neuropsychological performance in temporal lobe epilepsy. Neurology. 53(2), 260–264. https://doi.org/10.1212/WNL.53.2.260 Ma, B.B., Rao, V.R., 2018. Responsive neurostimulation: candidates and considerations. Epilepsy Behav. 88, 388–395. https://doi.org/10.1016/j.yebeh.2018.09.032 Maillard, L., Vignal, J.-P., Gavaret, M., Guye, M., Biraben, A., McGonigal, A., Chauvel, P., Bartolomei, F., 2004. Semiologic and electrophysiologic correlations in temporal lobe seizure subtypes. Epilepsia. 45(12), 1590–1599. https://doi.org/10.1111/j.0013-9580.2004.09704.x Malow, B.A., Blaxton, T.A., Sato, S., Bookheimer, S.Y., Kufta, C.V., Figlozzi, C.M., Theodore, W.H., 1996. Cortical stimulation elicits regional distinctions in auditory and visual naming. Epilepsia. 37(3), 245–252. https://doi.org/10.1111/j.1528-1157.1996.tb00020.x Marciani, M.G., Gotman, J., 1986. Effects of drug-withdrawal on location of seizure onset. Epilepsia. 27(4), 423–431. https://doi.org/10.1111/j.1528-1157.1986.tb03562.x 273 Marrosu, F., Santoni, F., Puligheddu, M., Barberini, L., Maleci, A., Ennas, F., Mascia, M., Zanetti, G., Tuveri, A., Biggio, G., 2005. Increase in 20-50 Hz (gamma frequencies) power spectrum and synchronization after chronic vagal nerve stimulation. Clin Neurophysiol. 116(9), 2026–2036. https://doi.org/10.1016/j.clinph.2005.06.015 Martin, P., Winston, G.P., Bartlett, P., de Tisi, J., Duncan, J.S., Focke, N.K., 2017. Voxel-based magnetic resonance image postprocessing in epilepsy. Epilepsia. 58(9), 1653–1664. https://doi.org/10.1111/epi.13851 McHugh, J.C., Singh, H.W., Phillips, J., Murphy, K., Doherty, C.P., Delanty, N., 2007. Outcome measurement after vagal nerve stimulation therapy: proposal of a new classification. Epilepsia. 48(2), 375–378. https://doi.org/10.1111/j.1528-1167.2006.00931.x McIntosh, A.M., Kalnins, R.M., Mitchell, L.A., Fabinyi, G.C.A., Briellmann, R.S., Berkovic, S.F., 2004. Temporal lobectomy: long-term seizure outcome, late recurrence and risks for seizure recurrence. Brain. 127(Pt 9), 2018–2030. https://doi.org/10.1093/brain/awh221 Melani, F., Zelmann, R., Mari, F., Gotman, J., 2013. Continuous High Frequency Activity: a peculiar SEEG pattern related to specific brain regions. Clin Neurophysiol. 124(8), 1507–1516. https://doi.org/10.1016/j.clinph.2012.11.016 Mikkelsen, K.B., Kappel, S.L., Mandic, D.P., Kidmose, P., 2015. EEG recorded from the ear: characterizing the Ear-EEG method. Front Neurosci. 9, 438. https://doi.org/10.3389/fnins.2015.00438 Mirski, M.A., Ferrendelli, J.A., 1984. Interruption of the mammillothalamic tract prevents seizures in guinea pigs. Science. 226(4670), 72–74. https://doi.org/10.1126/science.6433485 Mirski, M.A., Ferrendelli, J.A., 1986. Anterior thalamic mediation of generalized pentylenetetrazol seizures. Brain Res. 399(2), 212–223. https://doi.org/10.1016/0006-8993(86)91511-8 Mirski, M.A., McKeon, A.C., Ferrendelli, J.A., 1986. Anterior thalamus and substantia nigra: two distinct structures mediating experimental generalized seizures. Brain Res. 397(2), 377–380. https://doi.org/10.1016/0006-8993(86)90642-6 Mirski, M.A., Rossell, L.A., Terry, J.B., Fisher, R.S., 1997. Anticonvulsant effect of anterior thalamic high frequency electrical stimulation in the rat. Epilepsy Res. 28(2), 89–100. https://doi.org/10.1016/S0920-1211(97)00034-X Mithani, K., Mikhail, M., Morgan, B.R., Wong, S., Weil, A.G., Deschenes, S., Wang, S., Bernal, B., Guillen, M.R., Ochi, A., Otsubo, H., Yau, I., Lo, W., Pang, E., Holowka, S., Snead, O.C., Donner, E., Rutka, J.T., Go, C., Widjaja, E., Ibrahim, G.M., 2019. Connectomic profiling identifies responders to vagus nerve stimulation. Ann Neurol. 86(5), 743–753. https://doi.org/10.1002/ana.25574 Morris 3rd , G.L., Mueller, W.M., The Vagus Nerve Stimulation Study Group E01-E05, 1999. Longterm treatment with vagus nerve stimulation in patients with refractory epilepsy. Neurology. 53(8), 1731–1735. https://doi.org/10.1212/WNL.53.8.1731 Nagasawa, T., Juhász, C., Rothermel, R., Hoechstetter, K., Sood, S., Asano, E., 2012. Spontaneous and visually driven high-frequency oscillations in the occipital cortex: intracranial recording in epileptic patients. Hum Brain Mapp. 33(3), 569–583. https://doi.org/10.1002/hbm.21233 Neal, A., Ostrowsky-Coste, K., Jung, J., Lagarde, S., Maillard, L., Kahane, P., Touraine, R., Catenoix, H., Montavont, A., Isnard, J., Arzimanoglou, A., Bartolomei, F., Guenot, M., Rheims, 274 S., 2020. Epileptogenicity in tuberous sclerosis complex: a stereoelectroencephalographic study. Epilepsia. 61(1), 81–95. https://doi.org/10.1111/epi.16410 Neligan, A., Shorvon, S.D., 2011. Prognostic factors, morbidity and mortality in tonic-clonic status epilepticus: a review. Epilepsy Res. 93(1), 1–10. https://doi.org/10.1016/j.eplepsyres.2010.09.003 Nemtsas, P., Birot, G., Pittau, F., Michel, C.M., Schaller, K., Vulliemoz, S., Kimiskidis, V.K., Seeck, M., 2017. Source localization of ictal epileptic activity based on high-density scalp EEG data. Epilepsia. 58(6), 1027–1036. https://doi.org/10.1111/epi.13749 Neurological Disorders Collaborator Group, 2017. Global, regional, and national burden of neurological disorders during 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Neurol. 16, 877–897. https://doi.org/10.1016/S1474-4422(17)30299-5 Nguyen, R., Téllez-Zenteno, J.F., 2009. Injuries in epilepsy: a review of its prevalence, risk factors, type of injuries and prevention. Neurol Int. 1(1), e20. https://doi.org/10.4081/ni.2009.e20 Novais, F., Pestana, L.C., Loureiro, S., Andrea, M., Figueira, M.L., Pimentel, J., 2019. Predicting de novo psychopathology after epilepsy surgery: a 3-year cohort study. Epilepsy Behav. 90, 204– 208. https://doi.org/10.1016/j.yebeh.2018.11.037 Oner, A.Y., Eryurt, B., Ucar, M., Capraz, I., Kurt, G., Bilir, E., Tali, T., 2015. pASL versus DSC perfusion MRI in lateralizing temporal lobe epilepsy. Acta Radiol. 56(4), 477–481. https://doi.org/10.1177/0284185114531128 Onorati, F., Regalia, G., Caborni, C., Migliorini, M., Bender, D., Poh, M.Z., Frazier, C., Kovitch Thropp, E., Mynatt, E.D., Bidwell, J., Mai, R., LaFrance Jr., W.C., Blum, A.S., Friedman, D., Loddenkemper, T., Mohammadpour-Touserkani, F., Reinsberger, C., Tognetti, S., Picard, R.W., 2017. Multicenter clinical assessment of improved wearable multimodal convulsive seizure detectors. Epilepsia. 58(11), 1870–1879. https://doi.org/10.1111/epi.13899 Orosz, I., McCormick, D., Zamponi, N., Varadkar, S., Feucht, M., Parain, D., Griens, R., Vallée, L., Boon, P., Rittey, C., Jayewardene, A.K., Bunker, M., Arzimanoglou, A., Lagae, L., 2014. Vagus nerve stimulation for drug-resistant epilepsy: a European long-term study up to 24 months in 347 children. Epilepsia. 55(10), 1576–1584. https://doi.org/10.1111/epi.12762 Osorio, I., Overman, J., Giftakis, J., Wilkinson, S.B., 2007. High frequency thalamic stimulation for inoperable mesial temporal epilepsy. Epilepsia. 48(8), 1561–1571. https://doi.org/10.1111/j.1528- 1167.2007.01044.x Osorio, I., Schachter, S., 2011. Extracerebral detection of seizures: a new era in epileptology? Epilepsy Behav. 22(Suppl 1), S82–87. https://doi.org/10.1016/j.yebeh.2011.09.012 Pail, M., Řehulka, P., Cimbálník, J., Doležalová, I., Chrastina, J., Brázdil, M., 2017. Frequencyindependent characteristics of high-frequency oscillations in epileptic and non-epileptic regions. Clin Neurophysiol. 128(1), 106–114. https://doi.org/10.1016/j.clinph.2016.10.011 Palmini, A., Andermann, F., Dubeau, F., Gloor, P., Olivier, A., Quesney, L.F., Salanova, V., 1993. Occipitotemporal epilepsies: evaluation of selected patients requiring depth electrodes studies and rationale for surgical approaches. Epilepsia. 34(1), 84–96. https://doi.org/10.1111/j.1528- 1157.1993.tb02380.x 275 Patil, S., Biassoni, L., Borgwardt, L., 2007. Nuclear medicine in pediatric neurology and neurosurgery: epilepsy and brain tumors. Semin Nucl Med. 37(5), 357–381. https://doi.org/10.1053/j.semnuclmed.2007.04.002 Pauri, F., Pierelli, F., Chatrian, G.-E., Erdly, W.W., 1992. Long-term EEG-video-audio monitoring: computer detection of focal EEG seizure patterns. Electroencephalogr Clin Neurophysiol. 82(2), 1–9. https://doi.org/10.1016/0013-4694(92)90175-H Penry, J.K., Dean, J.C., 1990. Prevention of intractable partial seizures by intermittent vagal stimulation in humans: preliminary results. Epilepsia. 31(Suppl 2), S40–43. https://doi.org/10.1111/j.1528-1157.1990.tb05848.x Petersen, M.C., Bickford, G., Sem-Jacobsen, C.W., Dodge Jr., H.W., 1953. The depth electrogram in schizophrenic patients. Proc Staff Meet Mayo Clin. 28, 170–175. Picot, M.-C., Jaussent, A., Neveu, D., Kahane, P., Crespel, A., Gelisse, P., Hirsch, E., Derambure, P., Dupont, S., Landré, E., Chassoux, F., Valton, L., Vignal, J.-P., Marchal, C., Lamy, C., Semah, F., Biraben, A., Arzimanoglou, A., Petit, J., Thomas, P., Macioce, V., Dujols, P., Ryvlin, P., 2016. Cost-effectiveness analysis of epilepsy surgery in a controlled cohort of adult patients with intractable partial epilepsy: A 5-year follow-up study. Epilepsia. 57(10), 1669–1679. https://doi.org/10.1111/epi.13492 Pizzo, F., Roehri, N., Catenoix, H., Medina, S., McGonigal, A., Giusiano, B., Carron, R., Scavarda, D., Ostrowsky, K., Lepine, A., Boulogne, S., Scholly, J., Hirsch, E., Rheims, S., Bénar, C.-G., Bartolomei, F., 2017. Epileptogenic networks in nodular heterotopia: a stereoelectroencephalography study. Epilepsia. 58(12), 2112–2123. https://doi.org/10.1111/epi.13919 Powell, H.W., Richardson, M.P., Symms, M.R., Boulby, P.A., Thompson, P.J., Duncan, J.S., Koepp, M.J., 2007. Reorganization of verbal and nonverbal memory in temporal lobe epilepsy due to unilateral hippocampal sclerosis. Epilepsia. 48(8), 1512–1525. https://doi.org/10.1111/j.1528- 1167.2007.01053.x Radhakrishnan, K., So, E.L., Silbert, P.L., Jack Jr., C.R., Cascino, G.D., Sharbrough, F.W., O’Brien, P.C., 1998. Predictors of outcome of anterior temporal lobectomy for intractable epilepsy - a multivariate study. Neurology. 51(2), 465–471. https://doi.org/10.1212/WNL.51.2.465 Radtke, R.A., Hanson, M.W., Hoffman, J.M., Crain, B.J., Walczak, T.S., Lewis, D.V., Beam, C., Coleman, R.E., Friedman, A.H., 1993. Temporal-lobe hypometabolism on pet - predictor of seizure control after temporal lobectomy. Neurology. 43(6), 1088–1092. https://doi.org/10.1212/WNL.43.6.1088 Ramgopal, S., Thome-Souza, S., Jackson, M., Kadish, N.E., Sánchez Fernández, I., Klehm, J., Bosl, W., Reinsberger, C., Schachter, S., Loddenkemper, T., 2014. Seizure detection, seizure prediction, and closed-loop warning systems in epilepsy. Epilepsy Behav. 37, 291–307. https://doi.org/10.1016/j.yebeh.2014.06.023 Reif, P.S., Strzelczyk, A., Rosenow, F., 2016. The history of invasive EEG evaluation in epilepsy patients. Seizure. 41, 191–195. https://doi.org/10.1016/j.seizure.2016.04.006 Reimer, G.R., Grimm, R.J., Dow, R.S., 1967. Effects of cerebellar stimulation on cobalt-induced epilepsy in the cat. Electroencephalogr Clin Neurophysiol. 23(5), 456–462. https://doi.org/10.1016/0013-4694(67)90188-5 276 Rektor, I., Doležalová, I., Chrastina, J., Jurák, P., Halámek, J., Balíž, M., Brázdil, M., 2016. Highfrequency oscillations in the human anterior nucleus of the thalamus. Brain Stimul. 9(4), 629–631. https://doi.org/10.1016/j.brs.2016.04.010 Rémi, J., Vollmar, C., de Marinis, A., Heinlin, J., Peraud, A., Noachtar, S., 2011. Congruence and discrepancy of interictal and ictal EEG with MRI lesions in focal epilepsies. Neurology. 77(14), 1383–1390. https://doi.org/10.1212/WNL.0b013e31823152c3 Renfroe, J.B., Wheless, J.W., 2002. Earlier use of adjunctive vagus nerve stimulation therapy for refractory epilepsy. Neurology. 59(6 Suppl 4), S26–S30. https://doi.org/10.1212/WNL.59.6_suppl_4.S26 Révész, D., Fröjd, V., Rydenhag, B., Ben-Menachem, E., 2018. Estimating long-term vagus nerve stimulation effectiveness: accounting for antiepileptic drug treatment changes. Neuromodulation. 21(8), 797–804. https://doi.org/10.1111/ner.12775 Richardson, M.P., Strange, B.A., Duncan, J.S., Dolan, R.J., 2003. Preserved verbal memory function in left medial temporal pathology involves reorganisation of function to right medial temporal lobe. Neuroimage. 20(Suppl 1), S112–119. https://doi.org/10.1016/j.neuroimage.2003.09.008 Rolls, E.T., 2000. The orbitofrontal cortex and reward. Cereb Cortex. 10(3), 284–294. https://doi.org/10.1093/cercor/10.3.284 Rosenow, F., Lüders, H., 2001. Presurgical evaluation of epilepsy. Brain. 124(Pt 9), 1683–1700. https://doi.org/10.1093/brain/124.9.1683 Rutecki, P., 1990. Anatomical, physiological, and theoretical basis for the antiepileptic effect of vagus nerve stimulation. Epilepsia. 31(Suppl 2), S1–S6. https://doi.org/10.1111/j.1528- 1157.1990.tb05843.x Ryvlin, P., Bouvard, S., Le Bars, D., De Lamérie, G., Grégoire, M.C., Kahane, P., Froment, J.C., Mauguière, F., 1998. Clinical utility of flumazenil-PET versus F-18 fluorodeoxyglucose-PET and MRI in refractory partial epilepsy. A prospective study in 100 patients. Brain. 121(Pt 11), 2067– 2081. https://doi.org/10.1093/brain/121.11.2067 Ryvlin, P., Gilliam, F.G., Nguyen, D.K., Colicchio, G., Iudice, A., Tinuper, P., Zamponi, N., Aguglia, U., Wagner, L., Minotti, L., Stefan, H., Boon, P., Sadler, M., Benna, P., Raman, P., Perucca, E., 2014. The long-term effect of vagus nerve stimulation on quality of life in patients with pharmacoresistant focal epilepsy: the PuLsE (Open Prospective Randomized Long-term Effectiveness) trial. Epilepsia. 55(6), 893–900. https://doi.org/10.1111/epi.12611 Ryvlin, P., Kahane, P., 2005. The hidden causes of surgery-resistant temporal lobe epilepsy: extratemporal or temporal plus? Curr Opin Neurol. 18(2), 125–127. https://doi.org/10.1097/01.wco.0000162852.22026.6f Ryvlin, P., Nashef, L., Tomson, T., 2013. Prevention of sudden unexpected death in epilepsy: a realistic goal? Epilepsia. 54(Suppl 2), 23–28. https://doi.org/10.1111/epi.12180 Ryvlin, P., Rheims, S., 2016. Predicting epilepsy surgery outcome. Curr Opin Neurol. 29(2), 182– 188. https://doi.org/10.1097/WCO.0000000000000306 Řehulka, P., Doležalová, I., Janoušová, E., Tomášek, M., Marusič, P., Brázdil, M., Kuba, R., 2014. Ictal and postictal semiology in patients with bilateral temporal lobe epilepsy. Epilepsy Behav. 41, 40–46. https://doi.org/10.1016/j.yebeh.2014.09.033 277 Sabers, A., Buchgreitz, L., Neuhuber, W., 2018. Does vagus nerve stimulation influence pregnancy outcomes? Brain Stimul. 11(3), 618–619. https://doi.org/10.1016/j.brs.2017.12.006 Salanova, V., Markand, O., Worth, R., 2005. Temporal lobe epilepsy: analysis of failures and the role of reoperation. Acta Neurol Scand. 111(2), 126–133. https://doi.org/10.1111/j.1600- 0404.2005.00371.x Salanova, V., Witt, T., Worth, R., Henry, T.R., Gross, R.E., Nazzaro, J.M., Labar, D., Sperling, M.R., Sharan, A., Sandok, E., Handforth, A., Stern, J.M., Chung, S., Henderson, J.M., French, J., Baltuch, G., Rosenfeld, W.E., Garcia, P., Barbaro, N.M., Fountain, N.B., Elias, W.J., Goodman, R.R., Pollard, J.R., Tröster, A.I., Irwin, C.P., Lambrecht, K., Graves, N., Fisher, R., SANTE Study Group, 2015. Long-term efficacy and safety of thalamic stimulation for drug-resistant partial epilepsy. Neurology. 84(10), 1017–1025. https://doi.org/10.1212/WNL.0000000000001334 Sarazin, M., Pillon, B., Giannakopoulos, P., Rancurel, G., Samson, Y., Dubois, B., 1998. Clinicometabolic dissociation of cognitive functions and social behavior in frontal lobe lesions. Neurology. 51(1), 142–148. https://doi.org/10.1212/WNL.51.1.142 Sato, Y., Hotta, H., Nakayama, H., Suzuki, H., 1996. Sympathetic and parasympathetic regulation of the uterine blood flow and contraction in the rat. J Auton Nerv Syst. 59(3), 151–158. https://doi.org/10.1016/0165-1838(96)00019-7 Schiller, Y., Cascino, G.D., Busacker, N.E., Sharbrough, F.W., 1998. Characterization and comparison of local onset and remote propagated electrographic seizures recorded with intracranial electrodes. Epilepsia 39, 380-388. Schulz, R., Lüders, H.O., Hoppe, M., Tuxhorn, I., May, T., Ebner, A., 2000. Interictal EEG and ictal scalp EEG propagation are highly predictive of surgical outcome in mesial temporal lobe epilepsy. Epilepsia. 41(5), 564–570. https://doi.org/10.1111/j.1528-1157.2000.tb00210.x Seiam, A.-H.R., Dhaliwal, H., Wiebe, S., 2011. Determinants of quality of life after epilepsy surgery: systematic review and evidence summary. Epilepsy Behav. 21(4), 441–445. https://doi.org/10.1016/j.yebeh.2011.05.005 Serdaroglu, A., Arhan, E., Kurt, G., Erdem, A., Hirfanoglu, T., Aydin, K., Bilir, E., 2016. Long term effect of vagus nerve stimulation in pediatric intractable epilepsy: an extended follow-up. Childs Nerv Syst. 32(4), 641–646. https://doi.org/10.1007/s00381-015-3004-z Shihabuddin, B., Abou-Khalil, B., Delbeke, D., Fakhoury, T., 2001. Orbito-frontal epilepsy masquerading as temporal lobe epilepsy - a case report. Seizure. 10(2), 134–138. https://doi.org/10.1053/seiz.2000.0477 Shorvon, S., Tomson, T., 2011. Sudden unexpected death in epilepsy. Lancet. 378(9808), 2028– 2038. https://doi.org/10.1016/S0140-6736(11)60176-1 Signorini, M., Paulesu, E., Friston, K., Perani, D., Colleluori, A., Lucignani, G., Grassi, F., Bettinardi, V., Frackowiak, R.S.J., Fazio, F., 1999. Rapid assessment of regional cerebral metabolic abnormalities in single subjects with quantitative and nonquantitative F-18 FDG PET: a clinical validation of statistical parametric mapping. Neuroimage. 9(1), 63–80. https://doi.org/10.1006/nimg.1998.0381 So, N., Gotman, J., 1990. Changes in seizure activity following anticonvulsant drug-withdrawal. Neurology. 40(3 Pt 1), 407–413. https://doi.org/10.1212/WNL.40.3_Part_1.407 278 Soleman, J., Knorr, C., Datta, A.N., Strozzi, S., Ramelli, G.P., Mariani, L., Guzman, R., 2018. Early vagal nerve stimulator implantation in children: personal experience and review of the literature. Childs Nerv Syst. 34(5), 893–900. https://doi.org/10.1007/s00381-017-3694-5 Spanedda, F., Cendes, F., Gotman, J., 1997. Relations between EEG seizure morphology, interhemispheric spread, and mesial temporal atrophy in bitemporal epilepsy. Epilepsia. 38(12), 1300–1314. https://doi.org/10.1111/j.1528-1157.1997.tb00068.x Spencer, S.S., Guimaraes, P., Katz, A., Kim, J., Spencer, D., 1992. Morphological patterns of seizures recorded intracranially. Epilepsia. 33(3), 537–545. https://doi.org/10.1111/j.1528- 1157.1992.tb01706.x Spencer, S.S., Spencer, D.D., Williamson, P.D., Mattson, R.H., 1981. Ictal effects of anticonvulsant medication withdrawal in epileptic patients. Epilepsia. 22(3), 297–307. https://doi.org/10.1111/j.1528-1157.1981.tb04113.x Syed, T.U., LaFrance Jr., W.C., Kahriman, E.S., Hasan, S.N., Rajasekaran, V., Gulati, D., Borad, S., Shahid, A., Fernandez-Baca, G., Garcia, N., Pawlowski, M., Loddenkemper, T., Amina, S., Koubeissi, M.Z., 2011. Can semiology predict psychogenic nonepileptic seizures? A prospective study. Ann Neurol. 69(6), 997–1004. https://doi.org/10.1002/ana.22345 Takahashi, M., Soma, T., Kawai, K., Koyama, K., Ohtomo, K., Momose, T., 2012. Voxel-based comparison of preoperative FDG-PET between mesial temporal lobe epilepsy patients with and without postoperative seizure-free outcomes. Ann Nucl Med. 26(9), 698–706. https://doi.org/10.1007/s12149-012-0629-9 Tassinari, C.A., Cantalupo, G., Högl, B., Cortelli, P., Tassi, L., Francione, S., Nobili, L., Meletti, S., Rubboli, G., Gardella, E., 2009. Neuroethological approach to frontolimbic epileptic seizures and parasomnias: the same central pattern generators for the same behaviours. Rev Neurol (Paris). 165(10), 762–768. https://doi.org/10.1016/j.neurol.2009.08.002 Téllez-Zenteno, J.F., Dhar, R., Wiebe, S., 2005. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain. 128(Pt 5), 1188–1198. https://doi.org/10.1093/brain/awh449 Téllez-Zenteno, J.F., Hernandéz Ronquillo, L., Moien-Afshari, F., Wiebe, S., 2010. Surgical outcomes in lesional and non-lesional epilepsy: a systematic review and meta-analysis. Epilepsy Res. 89(2-3), 310–318. https://doi.org/10.1016/j.eplepsyres.2010.02.007 Thakur, K.T., Albanese, E., Giannakopoulos, P., Jette, N., Linde, M., Prince, M.J., Steiner, T.J., Dua, T., 2016. Neurological disorders, in: Patel, V., Chisholm, D., Dua, T., Laxminarayan, R., Medina-Mora, M.E. (Eds.), Mental, Neurological, and Substance Use Disorders: Disease Control Priorities, Third Edition (Volume 4). The International Bank for Reconstruction and Development / The World Bank, Washington (DC). Theodore, W.H., Fisher, R.S., 2004. Brain stimulation for epilepsy. Lancet Neurol. 3(2), 111–118. https://doi.org/10.1016/S1474-4422(03)00664-1 Thurman, D.J., Hesdorffer, D.C., French, J.A., 2014. Sudden unexpected death in epilepsy: assessing the public health burden. Epilepsia. 55(10), 1479–1485. https://doi.org/10.1111/epi.12666 Usui, N., Terada, K., Baba, K., Matsuda, K., Nakamura, F., Usui, K., Tottori, T., Umeoka, S., Fujitani, S., Mihara, T., Inoue, Y., 2010. Very high frequency oscillations (over 1000 Hz) in human epilepsy. Clin Neurophysiol. 121(11), 1825–1831. https://doi.org/10.1016/j.clinph.2010.04.018 279 Usui, N., Terada, K., Baba, K., Matsuda, K., Nakamura, F., Usui, K., Yamaguchi, M., Tottori, T., Umeoka, S., Fujitani, S., Kondo, A., Mihara, T., Inoue, Y., 2011. Clinical significance of ictal high frequency oscillations in medial temporal lobe epilepsy. Clin Neurophysiol. 122(9), 1693–1700. https://doi.org/10.1016/j.clinph.2011.02.006 Usui, N., Terada, K., Baba, K., Matsuda, K., Usui, K., Tottori, T., Mihara, T., Inoue, Y., 2015. Significance of very-high-frequency oscillations (over 1,000hz) in epilepsy. Ann Neurol. 78(2), 295–302. https://doi.org/10.1002/ana.24440 Uthman, B.M., Wilder, B.J., Hammond, E.J., Reid, S.A., 1990. Efficacy and safety of vagus nerve stimulation in patients with complex partial seizures. Epilepsia. 31(Suppl 2), S44–S50. https://doi.org/10.1111/j.1528-1157.1990.tb05849.x Uthman, B.M., Wilder, B.J., Penry, J.K., Dean, C., Ramsay, R.E., Reid, S.A., Hammond, E.J., Tarver, W.B., Wernicke, J.F., 1993. Treatment of epilepsy by stimulation of the vagus nerve. Neurology. 43(7), 1338–1345. https://doi.org/10.1212/WNL.43.7.1338 van 't Klooster, M.A., van Klink, N.E.C., Leijten, F.S.S., Zelmann, R., Gebbink, T.A., Gosselaar, P.H., Braun, K.P.J., Huiskamp, G.J.M., Zijlmans, M., 2015. Residual fast ripples in the intraoperative corticogram predict epilepsy surgery outcome. Neurology. 85(2), 120–128. https://doi.org/10.1212/WNL.0000000000001727 van 't Klooster, M.A., van Klink, N.E.C., Zweiphenning, W.J.E.M., Leijten, F.S.S., Zelmann, R., Ferrier, C.H., van Rijen, P.C., Otte, W.M., Braun, K.P.J., Huiskamp, G.J.M., Zijlmans, M., 2017. Tailoring epilepsy surgery with fast ripples in the intraoperative electrocorticogram. Ann Neurol. 81(5), 664–676. https://doi.org/10.1002/ana.24928 van 't Klooster, M.A., Zijlmans, M., Leijten, F.S.S., Ferrier, C.H., van Putten, M.J.A.M., Huiskamp, G.J.M., 2011. Time-frequency analysis of single pulse electrical stimulation to assist delineation of epileptogenic cortex. Brain. 134(Pt 10), 2855–2866. https://doi.org/10.1093/brain/awr211 van Andel, J., Thijs, R.D., de Weerd, A., Arends, J., Leijten, F., 2016. Non-EEG based ambulatory seizure detection designed for home use: what is available and how will it influence epilepsy care? Epilepsy Behav. 57(Pt A), 82–89. https://doi.org/10.1016/j.yebeh.2016.01.003 van Andel, J., Zijlmans, M., Fischer, K., Leijten, F.S., 2009. Quality of life of caregivers of patients with intractable epilepsy. Epilepsia. 50(5), 1294–1296. https://doi.org/10.1111/j.1528- 1167.2009.02032.x Van Bogaert, P., Massager, N., Tugendhaft, P., Wikler, D., Damhaut, P., Levivier, M., Brotchi, J., Goldman, S., 2000. Statistical parametric mapping of regional glucose metabolism in mesial temporal lobe epilepsy. Neuroimage. 12(2), 129–138. https://doi.org/10.1006/nimg.2000.0606 Van Buren, J.M., Wood, J.H., Oakley, J., Hambrecht, F., 1978. Preliminary evaluation of cerebellar stimulation by double-blind stimulation and biological criteria in the treatment of epilepsy. J Neurosurg. 48(3), 407–416. https://doi.org/10.3171/jns.1978.48.3.0407 Villanueva, V., Peral, E., Albisua, J., de Felipe, J., Serratosa, J.M., 2004. [Prognostic factors in temporal lobe epilepsy surgery]. Neurologia. 19(3), 92–98. Vonck, K., Thadani, V., Gilbert, K., Dedeurwaerdere, S., De Groote, L., De Herdt, V., Goossens, L., Gossiaux, F., Achten, E., Thiery, E., Vingerhoets, G., Van Roost, D., Caemaert, J., De Reuck, J., Roberts, D., Williamson, P., Boon, P., 2004. Vagus nerve stimulation for refractory epilepsy: a 280 transatlantic experience. J Clin Neurophysiol. 21(4), 283–289. https://doi.org/10.1097/01.WNP.0000139654.32974.4E Wang, Y.-H., An, Y., Fan, X.-T., Lu, J., Ren, L.-K., Wei, P.-H., Cui, B.-X., Du, J.-L., Lu, C., Wang, D., Zhang, H.-Q., Shan, Y.-Z., Zhao, G.-G., 2018. Comparison between simultaneously acquired arterial spin labeling and (18)F-FDG PET in mesial temporal lobe epilepsy assisted by a PET/MR system and SEEG. Neuroimage Clin. 19, 824–830. https://doi.org/10.1016/j.nicl.2018.06.008 Weinand, M.E., Wyler, A.R., Richey, E.T., Phillips, B.B., Somes, G.W., 1992. Long-term ictal monitoring with subdural strip electrodes - prognostic factors for selecting temporal lobectomy candidates. J Neurosurg. 77(1), 20–28. https://doi.org/10.3171/jns.1992.77.1.0020 Williamson, A., Spencer, S.S., Spencer, D.D., 1995. Depth electrode studies and intracellular dentate granule cell recordings in temporal lobe epilepsy. Ann Neurol. 38(5), 778–787. https://doi.org/10.1002/ana.410380513 Williamson, P.D., Boon, P.A., Thadani, V.M., Darcey, T.M., Spencer, D.D., Spencer, S.S., Novelly, R.A., Mattson, R.H., 1992. Parietal lobe epilepsy: diagnostic considerations and results of surgery. Ann Neurol. 31(2), 193–201. https://doi.org/10.1002/ana.410310210 Won, H.J., Chang, K.-H., Cheon, J.-E., Kim, H.D., Lee, D.S., Han, M.H., Kim, I.-O., Lee, S.K., Chung, C.-K., 1999. Comparison of MR imaging with PET and ictal SPECT in 118 patients with intractable epilepsy. AJNR Am J Neuroradiol. 20(4), 593–599. Wright, G.D., McLellan, D.L., Brice, J.G., 1984. A double-blind trial of chronic cerebellar stimulation in twelve patients with severe epilepsy. J Neurol Neurosurg Psychiatry. 47(8), 769– 774. http://dx.doi.org/10.1136/jnnp.47.8.769 Zang, Y., Jiang, T., Lu, Y., He, Y., Tian, L., 2004. Regional homogeneity approach to fMRI data analysis. Neuroimage. 22(1), 394–400. https://doi.org/10.1016/j.neuroimage.2003.12.030 Zhang, Z., Lu, G., Zhong, Y., Tan, Q., Chen, H., Liao, W., Tian, L., Li, Z., Shi, J., Liu, Y., 2010. fMRI study of mesial temporal lobe epilepsy using amplitude of low-frequency fluctuation analysis. Hum Brain Mapp. 31(12), 1851–1861. https://doi.org/10.1002/hbm.20982 Zibrandtsen, I.C., Kidmose, P., Christensen, C.B., Kjaer, T.W., 2017. Ear-EEG detects ictal and interictal abnormalities in focal and generalized epilepsy - a comparison with scalp EEG monitoring. Clin Neurophysiol. 128(12), 2454–2461. https://doi.org/10.1016/j.clinph.2017.09.115 281