Local brain activity persists during apparently generalized postictal EEG suppression

Epilepsy & Behavior 62 (2016) 218–224

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Local brain activity persists during apparently generalized postictal EEG suppression Dirk-Matthias Altenmüller a, Andreas Schulze-Bonhage a, Christian E. Elger b, Rainer Surges b,⁎ a b

Epilepsy Center, Department of Neurosurgery, Medical Center, University of Freiburg, Faculty of Medicine, Breisacher Strasse 64, 79106 Freiburg im Breisgau, Germany Department of Epileptology, University Hospital Bonn, Sigmund-Freud-Str. 25, 53127 Bonn, Germany

a r t i c l e

i n f o

Article history: Received 24 May 2016 Revised 1 July 2016 Accepted 2 July 2016 Available online xxxx Keywords: Postictal generalized EEG suppression Sudden unexpected death in epilepsy Intracranial EEG SUDEP

a b s t r a c t Objectives: Postictal generalized EEG suppression (PGES) frequently occurs after generalized convulsive seizures (GCS) and may be involved in the pathophysiology of sudden unexpected death in epilepsy (SUDEP). It is usually determined using conventional scalp EEG which is likely to miss cerebral activity in deeper brain structures. Here, we examined intracranial EEG activity after GCS to unravel the pattern and extent of local brain activity during apparent PGES on scalp EEG (s-PGES). Methods: We retrospectively reviewed electroencephalographic data of people with chronic epilepsy who had GCS during presurgical video-EEG monitoring using simultaneous intracranial and scalp EEG (10–20 system) electrodes. Results: Twenty-five GCS (20 with s-PGES) of 15 patients with an average number of 88 ± 42 intracranial electrode contacts were included. The majority of GCS with s-PGES (18 of 20) displayed persisting or reemerging intracranial EEG activity during apparent PGES on scalp EEG. Three patterns were identified: Pattern 1 (11 GCS, 6 patients) consisted of continuous local interictal activity; Pattern 2 (5 GCS, 5 patients) displayed suppressed EEG activity at all intracranial contacts in the early phase of s-PGES, but reemerging local brain activity before s-PGES dissolved; and Pattern 3 (2 GCS, 2 patients) showed persistent local ictal activity during s-PGES. Persisting intracranial EEG activity at PGES onset on scalp EEG was present in 10 ± 14% (range: 0 to 42%) of all intracranial contacts and mostly in the temporal lobe. Conclusions: Our results reveal that, during apparently generalized postictal EEG suppression, local brain activity persists or reemerges in most GCS. Possible implications of this localized neuronal activity in the context of SUDEP are discussed in the paper. © 2016 Elsevier Inc. All rights reserved.

1. Introduction Generalized convulsive seizures (GCS) are an established risk factor for sudden unexpected death in epilepsy (SUDEP) [1]. They commonly lead to prominent alterations of cardiorespiratory function which may lead, under specific circumstances, to the fatal SUDEP event [2]. The controversially discussed phenomenon of postictal generalized EEG suppression (PGES) may also play a role in the pathophysiology of SUDEP [2,3]. Postictal generalized EEG suppression frequently occurs after GCS in children and adults [2,4–12] and appears to be invariably found in GCS resulting in SUDEP [13]. Furthermore, the PGES duration

Abbreviations: PGES, postictal generalized EEG suppression; s-PGES, postictal generalized EEG suppression on scalp EEG; GCS, generalized convulsive seizures; SUDEP, sudden unexpected death in epilepsy. ⁎ Corresponding author: Tel.: +49 228 28 714 778; fax: +49 228 28 719 351. E-mail addresses: [email protected] (D.-M. Altenmüller), [email protected] (A. Schulze-Bonhage), [email protected] (C.E. Elger), [email protected] (R. Surges).

http://dx.doi.org/10.1016/j.yebeh.2016.07.008 1525-5050/© 2016 Elsevier Inc. All rights reserved.

was reported to be a potential predictor of SUDEP [6]. Subsequent studies, however, have not replicated these findings [14] but shown the inconsistent occurrence of PGES in people with multiple GCS, thereby reconciling the controversial findings [5]. Postictal generalized EEG suppression is usually determined with conventional scalp EEG electrodes, allowing the estimation of regional or global brain activity only. Thus, persistent or reoccurring local activity in deeper brain regions during putative PGES phases is likely to be missed. In this context, it is interesting to note that an incomplete involvement of the brain has been shown during clinically generalized seizures. For instance, secondarily generalized tonic–clonic seizures were shown not to be “truly” generalized in terms of electroencephalographic activity, as in about 25% of the seizures and patients, the authors identified a certain number of intracranial EEG contacts that never displayed seizure activity [15]. Here, we hypothesized that local brain activity persists or reoccurs during apparent PGES as assessed by scalp EEG. To investigate this, we have reviewed clinical and EEG data of patients with chronic epilepsy who underwent presurgical video-EEG monitoring using both intracranial and scalp EEG-electrodes who had GCS with PGES.

D.-M. Altenmüller et al. / Epilepsy & Behavior 62 (2016) 218–224

2. Material & methods 2.1. Patients We retrospectively reviewed video-EEG data of people with chronic focal epilepsy who underwent invasive presurgical video-EEG monitoring at the Epilepsy Center of the Medical Center — University of Freiburg (Germany) from January 2011 to December 2012. In addition to this population [n = 38], we also included patients with epilepsy due to hypothalamic hamartoma who underwent intracranial videoEEG monitoring from July 1999 to December 2012 [n = 5]. Inclusion criteria were the use of both intracranial and scalp EEG-electrodes and recording of GCS with postictal generalized EEG suppression on the scalp EEG (s-PGES) during video-EEG telemetry. Approval from the local ethics committee to conduct this study was received. 2.2. Presurgical evaluation Standard presurgical assessment comprised cerebral 3-Tesla MRI, neuropsychological testing, and noninvasive video-EEG telemetry using scalp EEG (10–20 or 10–10 system with additional temporal or sphenoidal electrodes, if indicated) prior to invasive video-EEG telemetry. Intracranial electrodes were implanted based on MRI findings and prior electroclinical findings during noninvasive video-EEG telemetry. During invasive recordings, simultaneous scalp EEG electrodes were placed according to the International 10–20 system. Antiepileptic drugs were gradually tapered off in all patients. 2.3. EEG recordings and data analysis The EEG data acquisition prior to 2010 was performed with a Neurofile NT digital video-EEG system (IT-med, Usingen, Germany) using up to 128 channels and a sampling rate of at least 256 Hz and a 16-bit analog to digital converter; data were band pass filtered between 0.08 and 120 Hz. Starting at the beginning of 2010, video-EEG monitoring was done with a Neuvo system (Compumedics, Abbotsford, Australia). Data were recorded using DC-coupled amplifiers and a sample rate of 500 Hz, resulting in a bandwidth of 0–200 Hz. Digitization was done using 24-bit analog–digital converters with an input range of ±0.6667 V.

219

The EEG recordings were analyzed by two experienced boardcertified (German Society for Clinical Neurophysiology and Functional Imaging) epileptologists (DMA, RS). The s-PGES was defined as a generalized attenuation of electroencephalographic activity on scalp EEG of any duration greater than 1 s below 10 μV immediately or within the first 30 s after an ictal scalp EEG pattern has ceased. The s-PGES was assessed in conventional longitudinal bipolar montage, allowing for artifacts due to pulse, movement, ECG, and muscle activity [14]. In order to ascertain EEG suppression and to properly identify artifacts, EEG traces were also routinely viewed in unipolar montages (“double check”) and, in case of doubt, at EEG sensitivities greater than 10 μV/mm. Data are given as mean ± SD, if not indicated otherwise. In most included patients, we analyzed only the first GCS with s-PGES in greater detail. In addition, however, in a subgroup of patients with multiple seizures (no. 1, 3), we investigated all two or the first five seizures with s-PGES to test for consistency of the postictal intracranial EEG patterns during consecutive GCS within a given patient. 3. Results Forty-three patients were considered for this study (Fig. 1). Eight patients had no scalp EEG electrodes, and a further 16 patients had no GCS during video-EEG recordings. Four additional patients had GCS without s-PGES. A total of 59 GCS in 19 patients were recorded. The s-PGES occurred in 43 (72.9%) of these 59 GCS in 15 (78.9%) of 19 patients. These fifteen patients (6 women, 9 men; age: 30.1 ± 15.4 years) were included in the final analysis. The clinical characteristics of the patients are detailed in Table 1. In nine patients, stereotactically implanted intracerebral depth electrodes were used for intracranial EEG recordings; in four patients, subdural electrodes were combined with intracerebral depth electrodes; and two patients had only subdural electrodes implanted. A total of 25 GCS were analyzed, of which 20 displayed PGES on scalp EEG (s-PGES). Six GCS with sPGES arose from wakefulness, whereas 14 seizures with s-PGES started during sleep. The duration of GCS with s-PGES amounted to 143 ± 89 s (range: 64–472 s) and the GCS without s-PGES to 122 ± 40 s (range: 93–183 s). The PGES lasted 29 ± 26 s (range: 2–82 s). Subsequent analysis of simultaneous intracranial EEG recordings revealed that, in the majority of seizures, intracranial EEG displayed

Fig. 1. Scheme of EEG electrode position and flowchart of patient selection. (A) Example of scheme of EEG electrode position (patient no. 1) and (B) flowchart of selection and inclusion of patients. *We considered patients who underwent invasive video-EEG telemetry from January 2011 to December 2012 [n = 38] and all patients with hypothalamic hamartoma and intracranial EEG recordings at any period [n = 5].

220

D.-M. Altenmüller et al. / Epilepsy & Behavior 62 (2016) 218–224

Table 1 Clinical characteristics of patients. Patient no.

Sex/agea/age at epilepsy onset

Etiology

Seizure focus

Intervention

Intracranial electrodes (total number of electrode contacts)

FUb/outcomec

1

F/25/7

Ganglioglioma II° TL L

Temp. lat. ant. L

Lesionectomy

12/1B

2

M/36/28

Unknown origin

Temp. lat. post. L

Multiple subpial transection

3

F/16/14

Multiple FCDs

Multifocal

None

4

M/17/3

FCD type IIad temporooccipital L

Temporooccipital L

Resection of the occipital pole L (partial resection of the FCD)

Subdural el.: temp. lat., temp. + occipital basal L (46 c) Intracerebral el.: amygd. + hipp. L (10 c) Subdural el.: temp. lat., temp. + occipital basal; pre-/postcentral gyrus L (100 c) Intracerebral el.: amygd. + hipp., sup. temp. gyrus L (20 c) Intracerebral el.: frontobasal; temp. basal post., amygd. R (96 c) Subdural el.: temp. lat. + basal; occipital lat. +

Unknown origin, suspected FCD L

Multifocal

HS and suspected FCD temp. lat. R

Temp. lat. + mesial R

2/3-Resection of the TL R +

5

6

F/18/5

F/31/1

None

7

M/10/7

FCD type Ibd frontal L

Inf. frontal gyrus L

AHE R Lesionectomy

8

F/60/49

Suspected FCDs TL L + R

Temporopolar L + R

Lesionectomy temporopolar L

9

M/46/6

Suspected FCD frontal R

Frontal R

None

10

M/52/28

Gliotic lesion of unknown origin parietoopercular L

Parietoopercular L

Topectomy parietoopercular L

11

M/29/4

FCD type Ibd temporooccipital L

Temporooccipital L

Lesionectomy temporooccipital L

12

M/25/19

Temporopolar L

Resection temp. pole L

13

F/20/1

Gliotic lesion TL L (suspected FCD) HH

HH

14

M/16/10

HH

HH + temporomesial R

Stereotactic implantation of 125 I-seeds into the hamartoma Stereotactic implantation of 125 I-seeds into the hamartoma

15

M/50/9

HH + HS L

HH + temporomesial L

2/3-Resection of the TL L + AHE L

basal + mesial; parietal post.; frontoparietal operculum L (82 c) Intracerebral el.: hipp. L (10 c) Intracerebral el.: frontal incl. SMA + cingulate gyrus; temp. incl. amygd. + hipp.; insula L (167 c) Intracerebral el.: temp. lat. + mesial (incl. amygd. + hipp.) R (126 c) Subdural el.: frontal incl. precentral gyrus L (72 c) Intracerebral el.: temp. lat. + mesial (incl. amygd. + hipp.) L + R (113 c) Subdural el.: frontal lat., basal + mesial R (102 c) Subdural el.: temp. lat. + basal; opercular; occipital basal L (50 c) Intracerebral el.: amygd. + hipp. L + R, temp. lat. R (29 c) Intracerebral el.: temp. lat. + mesial (incl. amygd. + hipp.); occipital; parietal inf.; post. insula L (167 c) Intracerebral el.: temp. lat. + mesial (incl. amygd. + hipp.); insula L (86 c) Intracerebral el.: HH L + R; cingulate gyrus R (6 c) Intracerebral el.: HH median; thalamus R; sup. frontal gyrus R; temp. lat. + mesial (incl. amygd. + hipp.) R (24 c) Intracerebral el.: HH median, inf. + medial frontal gyrus L; temp. lat. + mesial (incl. amygd. + hipp.) L (21 c)

3/2B

n.a. 6/1A

n.a.

6/1A 14/4B 46/1B n.a. 12/2A

12/1A

3/2B 60/4B 150/3B

44/1B

AHE, amygdalohippocampectomy; amygd., amygdala; ant.; anterior; c, electrode contacts; el., electrode(s); FCD, focal cortical dysplasia; HH, hypothalamic hamartoma; hipp., hippocampus; HS, hippocampal sclerosis; inf., inferior; L, left; lat., lateral; n.a., not applicable; post., posterior; R, right; SMA, supplementary motor area; sup., superior; temp., temporal; TL, temporal lobe. a At telemetry. b Follow-up in months. c According to the Engel classification. d According to [16].

ongoing cerebral local activity during s-PGES according to three patterns as follows. Pattern 1 (11 GCS, 6 patients) was characterized by persisting local interictal activity at a variable number of intracranial electrode contacts during s-PGES, as exemplified in Fig. 2 (for extended time scale, see Fig. e-1). In the remaining intracranial electrode contacts which showed suppression of EEG activity with seizure cessation, brain activity reemerged after a variable time before PGES resolved on scalp EEG (Fig. 3A, B). Pattern 2 (5 GCS, 5 patients) displayed suppression of all available intracranial electrode contacts during initial s-PGES, but interictal activity commonly reemerged before PGES resolved on the scalp EEG (Fig. 3A, B). Interictal activity during s-PGES predominantly consisted of nonrhythmic waves in the range of about 0.5–30 Hz or of epileptiform discharges. In some intracranial electrode contacts, the individually prevailing preictal activity was reestablished in the course of s-PGES. Pattern 3 (2 GCS, 2 patients) consisted of persistent ictal activity despite s-PGES (Fig. 3A, B). Only in a minority of seizures (2 of 20 seizures in 2 patients; Fig. 3A) was local brain activity suppressed in all available intracranial electrode contacts throughout the entire period of s-PGES. Given an average number of 88 ± 42 intracranial electrode contacts per patient, persisting EEG activity was detected in about

10 ± 14% (range: 0 to 42%) of the intracranial contacts in all 20 GCS with s-PGES (see also Fig. e-2) and commonly consisted of local background activity or epileptiform discharges (Fig. 2). Because of an inherent spatial sampling bias, the localization of the detected persistent interictal or ictal activity naturally depends on the implantation site of the intracerebral or subdural electrodes. Bearing this important limitation in mind, persistent local brain activity was mostly found in parts of the temporal lobe but also within the frontal, parietal, and occipital lobes as well as the insular cortex and hypothalamus (Fig. 3C). Four patients (no. 3, 4, 14, 15) had GCS with s-PGES (8 seizures with detailed analysis) and without s-PGES (5 seizures). As the presence of bilateral and symmetric tonic arm extension during a GCS was recently identified as a predictor of s-PGES [4], we have analyzed this feature in all seizures included in this study (Supplementary Table e-1). There were no obvious differences between GCS with or without s-PGES with respect to seizure semiology, seizure duration, and ictal surface or intracranial EEG pattern in these 4 patients. In all GCS without s-PGES, persistent postictal brain activity was found in several but not all intracranial electrode contacts. In view of the anatomical vicinity of the presumed

D.-M. Altenmüller et al. / Epilepsy & Behavior 62 (2016) 218–224

221

Fig. 2. Original recording of scalp and intracranial EEG following a generalized convulsive seizure in patient no. 1. Selected intracranial EEG channels; for electrode position, see Fig. 1 (A). Note persistent local interictal activity in the delta range in the left hippocampus (e.g., intracranial contacts HL5 to HL8) and subsequent onset of spikes and sharp slow wave complexes in the left lateral temporal neocortex (e.g., intracranial contacts D2 to D5), the left mesial temporal pole including the amygdala (e.g., intracranial contacts HL1 to HL4), and the left hippocampus (e.g., intracranial contacts HL5 to HL8) during apparent postictal generalized EEG suppression (PGES) as assessed by scalp EEG recording. Scalp EEG displays electrode artifacts (on the left side) and muscle artifacts (on the right side). Some intracranial EEG electrodes show pulse artifacts (e.g., A5, B8). Scalp EEG (upper panel): bipolar montage, lowpass filter 15 Hz, high-pass filter 1.6 Hz, 50-Hz Notch-Filter, scaling 100 μV/10 s. Intracranial EEG (lower panel; selected channels; for electrode positions, see Fig. 1): bipolar montage, low-pass filter 120 Hz, high-pass filter 1.6 Hz, 50-Hz Notch-Filter, scaling 200 μV/10 s.

Fig. 3. Local brain activity persists in more than 50% of the seizures and patients during putative postictal generalized EEG suppression (PGES) as assessed by scalp EEG. (A) Numbers of seizures (left panel) and patients (right panel) displaying local brain activity as assessed by intracranial EEG electrodes during putative PGES in scalp EEG. (B) Averaged time course of scalp EEG (Sc-EEG) and intracranial EEG (In-EEG) activity of the seizures displaying the indicated pattern. Scale bar indicates 50 s. (C) Localization of persistent local activity (seizures and patients are counted several times if applicable).

222

D.-M. Altenmüller et al. / Epilepsy & Behavior 62 (2016) 218–224

hypothalamic seizure generator and brain structures involved in autonomic regulation (i.e., thalamus and hypothalamus) and possibly the pathophysiology of SUDEP, we have also included electroclinical data of 3 patients with epilepsy due to hypothalamic hamartoma. During s-PGES, no specific intracranial EEG pattern was identified in this particular subgroup (Fig. 4; patient no. 13 to 15). Interestingly, in one patient (no. 3) with multiple seizures, the 5 consecutive GCS with s-PGES analyzed in detail showed some variability in the intracranial EEG pattern (Fig. 4; seizure no. 3/1 to 3/5). Overall, we did not find any clear relationship between localization or lateralization of the seizure onset zone as defined by intracranial EEG and the features of intracranial EEG patterns during s-PGES or the localization of persistent nonepileptic or epileptic (interictal or ictal) activity. Importantly, in none of the 15 patients did postictal cardiorespiratory function appear to be compromised (in terms of asystole or apnea as assessed by one-lead ECG and respiratory movements and noises on simultaneous video recordings).

4. Discussion Postictal generalized EEG suppression (PGES) is a common phenomenon following GCS and occurs in about 15–90% of all GCS in adults and children [4–12,14]. The origin of PGES is still unknown but may be caused by a sudden breakdown of cerebral activity due to sustained periictal hypoxemia and accumulating metabolites and transmitters (e.g., adenosine, GABA, and potassium) which, in turn, suppress neuronal function. Alternatively, a neuronal network at a strategical position, e.g., within the thalamus or the brain stem, could be switched on or off during the GCS which, in turn, switches off neuronal activity in a concerted, generalized manner. Our data, however, rather argue against such a centralized or generalized switch. Importantly, PGES appears to be an electroencephalographic hallmark of SUDEP, as it invariably occurred in monitored SUDEP events [13]. The clinical impact of PGES, however, remains unclear. One hypothesis is that PGES reflects a generalized breakdown of neuronal function which secondarily leads to

Fig. 4. Individual time courses of periictal cerebral activity as assessed by scalp and intracranial EEG electrodes. Time course of scalp EEG (Sc-EEG) and intracranial EEG (In-EEG) activity of individual seizures. Scale bar indicates 50 s.

D.-M. Altenmüller et al. / Epilepsy & Behavior 62 (2016) 218–224

suppression of cardiorespiratory function and the fatal SUDEP event. In monitored non-SUDEP events, however, onset of PGES was associated with the severity of periictal hypoxemia, and PGES did not lead to secondary cardiorespiratory dysfunction [17,18]. Our results allow an alternative explanation of how apparent PGES integrates into the pathophysiology of SUDEP. We have shown that local brain activity seems to persist in the majority of GCS after seizure cessation, although scalp EEG indicates postictal generalized EEG suppression. Overall, the extent of persisting local EEG activity appears to be small and amounts on average to about 10% of the available electrode contacts. Furthermore, local brain activity progressively reemerges towards the end of s-PGES in the remaining electrode contacts. Gradually reemerging local brain activity in the course of s-PGES was also seen in most of the remaining seizures. Because of the inherent spatial sampling bias as the major study limitation of our analysis, however, our data are unable to provide the true extent of ongoing brain activity during apparently generalized EEG suppression but can only demonstrate that different cerebral activity patterns do actually persist in various regions to a variable extent. In general, brain activity does not need to involve extended areas to have clinical effects. For instance, electrical stimulation of very limited brain areas can induce apnea, bradycardia, and asystole [19–22]. Thus, a particular type of local activity at a strategical position, e.g., in the hippocampus, cingulated gyrus, or brain stem, may have effects on cardiorespiratory function at variable degrees (e.g., [23]). In the scenario of PGES, the effect of such local brain activities may be particularly strong, because most other brain regions usually exerting regulatory or compensatory functions are suppressed. Nevertheless, in none of our patient was cardiorespiratory dysfunction apparent in the postictal period so that this hypothesis remains speculative and cannot be directly supported by electroclinical data. Alternatively, persisting local brain activity may also be protective rather than detrimental, for instance by stimulating cardiac activity. Finally, our results allow the interpretation that the degree of postictal cerebral suppression varies from seizure to seizure (see, e.g., patient no. 3) and that only true and sustained suppression of cerebral activity secondarily facilitates or induces cardiorespiratory dysfunction. Such full-blown suppression of intracranial EEG activity (developing in a generalized manner or asymmetrically first in one then in the other hemisphere) was previously reported in a fatal GCS but also in nonfatal GCS [3,24]. Taken together, our study reveals that, during apparently generalized postictal EEG suppression, local brain activity persists or reemerges in most GCS. Presently, it is unknown if this localized neuronal activity may induce protective or detrimental autonomic effects, while regulatory or compensatory functions of most other brain regions are suppressed, thereby possibly linking PGES with SUDEP. Our results indicate that s-PGES is not a homogeneous phenomenon but, in fact, can reflect diverse states of postictal brain activity, suggesting that the predictive value of s-PGES as a biomarker of increased risk of future SUDEP is limited. Larger scale studies are required to understand the conditions and the clinical importance of the different intracranial patterns related to apparent s-PGES. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2016.07.008. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Author contributions Dirk-Matthias Altenmüller contributed to study design and concept, acquisition and analysis of data, drafting the manuscript for content, and interpretation of data. Andreas Schulze-Bonhage contributed to drafting the manuscript for content and interpretation of data.

223

Christian E. Elger contributed to drafting the manuscript for content and interpretation of data. Rainer Surges contributed to study design and concept, acquisition and analysis of data, drafting the manuscript for content, and interpretation of data. The corresponding author completed the statistical analysis. Disclosures D.-M. Altenmüller has received speaker honoraria from UCB Pharma. A. Schulze-Bonhage has received personal fees from honorarium for presentations from Cyberonics, Inc., Desitin, Eisai, and UCB and honorarium for advice from Eisai and Precisis. C.E. Elger has served as a paid consultant for UCB Pharma, Desitin, and Pfizer. He has received speaker fees from Cyberonics, Desitin, Eisai, Medtronics, Novartis, and UCB Pharma. He is an employee of the Life and Brain Institute Bonn. R. Surges has received speaker fees from Cyberonics, EISAI, Novartis, and UCB Pharma. References [1] Hesdorffer DC, Tomson T, Benn E, Sander JW, Nilsson L, Langan Y, et al. Do antiepileptic drugs or generalized tonic–clonic seizure frequency increase SUDEP risk? A combined analysis. Epilepsia 2012;53:249–52. [2] Surges R, Sander JW. Sudden unexpected death in epilepsy: mechanisms, prevalence, and prevention. Curr Opin Neurol 2012;25:201–7. [3] Bozorgi A, Lhatoo SD. Seizures, cerebral shutdown, and SUDEP. Epilepsy Curr 2013; 13:236–40. [4] Alexandre V, Mercedes B, Valton L, Maillard L, Bartolomei F, Szurhaj W, et al. Risk factors of postictal generalized EEG suppression in generalized convulsive seizures. Neurology 2015;85:1598–603. [5] Lamberts RJ, Gaitatzis A, Sander JW, Elger CE, Surges R, Thijs RD. Postictal generalized EEG suppression: an inconsistent finding in people with multiple seizures. Neurology 2013;81:1252–6. [6] Lhatoo SD, Faulkner HJ, Dembny K, Trippick K, Johnson C, Bird JM. An electroclinical case–control study of sudden unexpected death in epilepsy. Ann Neurol 2010;68: 787–96. [7] Moseley BD, So E, Wirrell EC, Nelson C, Lee RW, Mandrekar J, et al. Characteristics of postictal generalized EEG suppression in children. Epilepsy Res 2013;106: 123–7. [8] Freitas J, Kaur G, Fernandez GB-V, Tatsuoka C, Kaffashi F, Loparo KA, et al. Age-specific periictal electroclinical features of generalized tonic–clonic seizures and potential risk of sudden unexpected death in epilepsy (SUDEP). Epilepsy Behav 2013; 29:289–94. [9] Poh M-Z, Loddenkemper T, Reinsberger C, Swenson NC, Goyal S, Madsen JR, et al. Autonomic changes with seizures correlate with postictal EEG suppression. Neurology 2012;78:1868–76. [10] Pavlova M, Singh K, Abdennadher M, Katz ES, Dworetzky BA, White DP, et al. Comparison of cardiorespiratory and EEG abnormalities with seizures in adults and children. Epilepsy Behav 2013;29:537–41. [11] Semmelroch M, Elwes RDC, Lozsadi DA, Nashef L. Retrospective audit of postictal generalized EEG suppression in telemetry. Epilepsia 2012;53:e21–4. [12] Tao JX, Yung I, Lee A, Rose S, Jacobsen J, Ebersole JS. Tonic phase of a generalized convulsive seizure is an independent predictor of postictal generalized EEG suppression. Epilepsia 2013;54:858–65. [13] Ryvlin P, Nashef L, Lhatoo SD, Bateman LM, Bird J, Bleasel A, et al. Incidence and mechanisms of cardiorespiratory arrests in epilepsy monitoring units (MORTEMUS): a retrospective study. Lancet Neurol 2013;12:966–77. [14] Surges R, Strzelczyk A, Scott CA, Walker MC, Sander JW. Postictal generalized electroencephalographic suppression is associated with generalized seizures. Epilepsy Behav 2011;21:271–4. [15] Schindler K, Leung H, Lehnertz K, Elger CE. How generalised are secondarily “generalised” tonic–clonic seizures? J Neurol Neurosurg Psychiatry 2007;78: 993–6. [16] 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. [17] Lamberts RJ, Laranjo S, Kalitzin SN, Velis DN, Rocha I, Sander JW, et al. Postictal generalized EEG suppression is not associated with periictal cardiac autonomic instability in people with convulsive seizures. Epilepsia 2013;54:523–9. [18] Seyal M, Hardin KA, Bateman LM. Postictal generalized EEG suppression is linked to seizure-associated respiratory dysfunction but not postictal apnea. Epilepsia 2012; 53:825–31. [19] Altenmüller D-M, Zehender M, Schulze-Bonhage A. High-grade atrioventricular block triggered by spontaneous and stimulation-induced epileptic activity in the left temporal lobe. Epilepsia 2004;45:1640–4. [20] Leung H, Schindler K, Kwan P, Elger C. Asystole induced by electrical stimulation of the left cingulate gyrus. Epileptic Disord 2007;9:77–81.

224

D.-M. Altenmüller et al. / Epilepsy & Behavior 62 (2016) 218–224

[21] Kaada BR, Jasper H. Respiratory responses to stimulation of temporal pole, insula, and hippocampal and limbic gyri in man. AMA Arch Neurol Psychiatry 1952;68: 609–19. [22] Dlouhy BJ, Gehlbach BK, Kreple CJ, Kawasaki H, Oya H, Buzza C, et al. Breathing inhibited when seizures spread to the amygdala and upon amygdala stimulation. J Neurosci 2015;35:10281–9.

[23] Surges R, Jordan A, Elger CE. Ictal modulation of cardiac repolarization, but not of heart rate, is lateralized in mesial temporal lobe epilepsy. PLoS One 2013;8, e64765. [24] Bird JM, Dembny KaT, Sandeman D, Butler S. Sudden unexplained death in epilepsy: an intracranially monitored case. Epilepsia 1997;38:S52–6.