Clinical Neurophysiology 126 (2015) 1670–1676
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Dynamic changes of interictal post-spike slow waves toward seizure onset in focal cortical dysplasia type II Yosuke Sato a, Sam M. Doesburg b,c,d,e, Simeon M. Wong b,c, Tohru Okanishi a, Ryan Anderson b, Dragos A. Nita a, Ayako Ochi a, Hiroshi Otsubo a,⇑ a
Division of Neurology, Hospital for Sick Children, Toronto, Ont., Canada Department of Diagnostic Imaging, Hospital for Sick Children, Toronto, Ont., Canada Neuroscience & Mental Health Program, Hospital for Sick Children Research Institute, Toronto, Ont., Canada d Department of Medical Imaging, University of Toronto, Toronto, Ont., Canada e Department of Psychology, University of Toronto, Toronto, Ont., Canada b c
a r t i c l e
i n f o
Article history: Accepted 15 November 2014 Available online 25 November 2014 Keywords: Post-spike slow wave Interictal spike Focal cortical dysplasia Seizure onset zone Preictal period
h i g h l i g h t s Power of post-spike slow wave increased in all areas from interictal to preictal period. Power of post-spike slow wave within the seizure onset zone (SOZ) was significantly higher than that
outside SOZ during preictal period. Post-spike slow wave with high power became confined to SOZ toward seizure onset.
a b s t r a c t Objective: A post-spike slow wave (PSS) as part of a spike and slow wave is presumably related to inhibition of epileptic activity. In this study, we evaluated dynamic changes of PSS power toward seizure onset in patients with focal cortical dysplasia (FCD) type II. Methods: We collected data from 10 pediatric patients with FCD type II, who underwent invasive monitoring with subdural grids. The PSS were averaged based on spike-triggering in 30 s epochs during both interictal and preictal periods. We quantitatively measured and compared PSS power and distribution between interictal and preictal periods, both within and outside the seizure onset zone (SOZ). Results: PSS power was significantly higher in all areas during preictal period compared with interictal period. During preictal period, PSS power within SOZ was significantly higher than outside SOZ. From interictal to preictal period, the number of electrodes with high power PSS significantly increased within SOZ and decreased outside SOZ. Conclusions: Toward seizure onset, PSS power increased in all areas, predominantly within SOZ, and became confined into SOZ in a subset of FCD type II patients. Significance: Preictal PSS power increase and confinement into SOZ accompany transition to seizures. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The spike and slow wave pattern on EEG is well known to be an interictal epileptiform discharge in focal epilepsy (Alarcón et al., 2012). The spike and slow wave complex consists of an initial high-frequency neuronal excitation during the ‘‘spike’’ phase and
⇑ Corresponding author at: Division of Neurology, The Hospital for Sick Children, Toronto M5G 1X8, Ont., Canada. Tel.: +1 416 813 6295; fax: +1 416 813 6334. E-mail addresses: [email protected], [email protected] (H. Otsubo).
a following inhibition or disfacilitation of neuronal firing during the ‘‘slow wave’’ phase (Matsumoto and Marsan, 1964; Blumenfeld, 2005; Alarcón et al., 2012). As previously shown in animal models of focal epilepsy, interictal post-spike slow waves (PSS) are associated with inhibitory potentials that play a protective role to prevent the propagation of epileptic discharges (Lebovitz, 1979, Domann et al., 1991; de Curtis et al., 1998). De Curtis et al. demonstrated that the presence of interictal post-spike depression, which corresponds to PSS, correlated with an increased threshold for the generation of interictal spikes evoked in response to afferent stimulation in human focal epilepsy (de Curtis et al., 2005). The
http://dx.doi.org/10.1016/j.clinph.2014.11.012 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1671
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
dynamics of PSS power may take part in inhibitory control of interictal epileptic discharges. Focal cortical dysplasia (FCD) type II has been known to present with frequent interictal spike and slow waves in the lesion (Taylor et al., 1971; Ferrier et al., 2001; Francione et al., 2003). Palmini et al. indicated that the interictal discharges in FCD type II were intrinsically epileptogenic (Palmini et al., 1995). Focal seizures could be triggered by an evolution of preictal epileptic activity, mediated by increased neuronal excitation accompanied by a loss of inhibition (Lopes da Silva et al., 2003; Mormann et al., 2007). Lange et al. reported focal decreases in spiking rate which coincided with increased rates of bilateral spikes before seizures (Lange et al., 1983). Conversely, Katz et al. demonstrated that there were no remarkable changes in spike rates before seizures (Katz et al., 1991). We hypothesized that PSS power would show dynamic changes during transition toward seizure onset in focal epilepsy and that such changes may reflect the progressive inhibitory mechanisms relevant for impeding the evolution of interictal epileptiform discharges into focal ictal events. We compared PSS power within and outside the seizure onset zone (SOZ) during both interictal and preictal periods. This report is the first to characterize the PSS dynamics associated with seizure onset in focal epilepsies including characterization of both power and topographic distribution. 2. Methods 2.1. Subjects We retrospectively analyzed data collected between June 2009 and September 2013 from 10 pediatric patients (6 males and 4 females) with medically refractory focal epilepsy secondary to focal cortical dysplasia type II who underwent intracranial video EEG (IVEEG) monitoring prior to surgical resection. Age at data recording and surgery ranged from 3 to 14 years. The cortical dysplastic lesions were verified with magnetic resonance imaging and subsequently confirmed by pathology. The sites of electrode placement were individualized according to clinical history, seizure seminology, neuroimaging and scalp EEG findings as previously described (Otsubo et al., 1999). In order to investigate the hypothesis that the power of PSS is differentially distributed within and outside SOZ, we retained for analysis patients fulfilling the following criteria: SOZ was well localized, all seizures started from the same SOZ, interictal spike and slow waves were seen simultaneously both within and outside SOZ. In all patients, postsurgical neuropathological diagnosis revealed FCD type II. The patient clinical characteristics are summarized in Table 1. This study was approved by the Research Ethic Board at the Hospital for Sick Children.
2.2. IVEEG recordings The implantation of intracranial electrodes and extraoperative functional mapping were performed as described previously (Benifla et al., 2009). Center to center spacing of the contacts of the subdural electrodes was 10.5 mm (Ad-Tech, Racine, WI, USA). IVEEG data were acquired using a Harmonie system (Stellate, Montreal, PQ, Canada). The EEG signals were sampled at 1 kHz in two patients (Patient 1 and 2) and were sampled at 2 kHz in eight patients (Patients 3–10). The recordings were performed referentially with the referential electrode placed in a cortical region without epileptic discharges. 2.3. Determination of SOZ High-frequency oscillations (HFOs) are thought to be biomarkers of epileptogenic brain regions (Bragin et al., 1999; Staba et al., 2002) and have been shown to be topographically concentrated in SOZ during ictal and interictal periods (Fisher et al., 1992; Alarcon et al., 1995; Akiyama et al., 2005; Ochi et al., 2007). Given such findings, board certified clinical neurophysiologists determined SOZ for each patient by visual inspection and spectral analysis of intracranial EEG, in accordance with the current clinical practice at our institution. The details of these methods have been described elsewhere (Ochi et al., 2007; Jacobs et al., 2009). For each patient, three to five electrodes (4.2 ± 0.6; total 42) were classified as being within SOZ. The locations and number of the analyzed electrodes are shown in Table 1. 2.4. EEG data selection Two certified clinical neurophysiologists (AO, HO) determined electrographic seizure onset. The entire IVEEG recording was reviewed for the presence of seizures. All electrographic seizures were marked for times of unequivocal electrographic seizure onset defined as the earliest electrical changes recorded as the paroxysmal fast activity, repetitive spiking, or rhythmic slow wave. For each patient, at least one typical partial seizure was confirmed during IVEEG recordings. We selected the most typical partial seizure per patient. In all patients, electrographic seizure onset began from one to three electrodes, which were consistent with the electrodes defined as those in SOZ. A preictal epoch was extracted which consisted of 30 s of activity immediately preceding electrographic seizure onset. An interictal epoch was also extracted for each patient, and was comprised of 30 s of activity during Stage II non-REM sleep remote from seizures by at least 1 h. Since it was necessary to make recording conditions constant for comparing of PSS powers, EEG data from depth electrodes were excluded in the present analysis. Moreover, each selected epoch was visually inspected to
Table 1 Clinical characteristics of 10 patients with FCD type II. # of Pt
1 2 3 4 5 6 7 8 9 10
Age (years)
Gender
At onset
At surgery
1 3 5 3 2 3 10 2 8 10
4 12 7 7 3 3 14 4 9 12
Male Female Male Male Male Female Female Female Male Male
Location of FCD on MRI
Rt. TP Lt. P Lt. F Rt. F Lt. F Rt. F Rt. PO Lt. P Lt. F Rt. T
Extraoperative video EEG Location covered by electrodes
# of analyzed electrodes
Rt. FTP Lt. FP Lt. FTP Rt. FTP Lt. FTP Rt. FTP Rt. PO Lt. FTP Lt. FP Rt. OT
76 96 109 109 63 87 63 63 63 48
F = frontal; FCD = focal cortical dysplasia; Lt. = left; MRI = magnetic resonance imaging; O = occipital; P = parietal; Rt. = right; SOZ = seizure onset zone; T = temporal.
1672
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
ensure it was not contaminated by significant artifacts, such as environmental artifacts and muscle artifacts (Otsubo et al., 2008). Table 1 lists the number of the implanted electrodes, the analyzed electrodes, and the number of electrodes in SOZ. 2.5. Analysis of PSS power and distribution To suppress contamination by background activity for PSS analysis, we carried out spike-triggered averaging based on the selected spike and slow waves using EEG Harmonie Stellate software and a custom-made program written in MATLAB (The MathWorks, Version 7.1). The following procedures were performed in all patients during both interictal and preictal epochs. We selected and averaged ten 1-s epochs including the spike and slow waves that met the following criteria: (i) the spike and slow waves were verified as repetitive discharges typically seen in focal cortical dysplasia by visual inspection by a clinical neurophysiologist (YS) using a bipolar montage, 1–70 Hz bandpass filter on 10 s/page, (ii) we selected one electrode that showed the highest amplitude spike within SOZ, (iii) consecutive spike and slow waves did not overlap within 300 ms, (iv) we selected 10 epochs of one second from 300 ms before to 700 ms after the spike peak, (v) using a referential montage without filter, we averaged 10 epochs (Fig. 1A).
We estimated the spike duration up to 100 ms and focused on the period of PSS ranging from 50 to 250 ms after the spike peak. During this period, we quantitatively measured the amplitudes (lV) of the PSS (A(t)) using 0.5–5 Hz filter (Fig. 1B). We verified this bandpass filter according to the evidence that the frequencies of PSS varied between 1 and 4 Hz (Blumenfeld, 2005). The mean power (dB) of the PSS (P) were then obtained by the following formula:
P ¼ 10log10
t2 1X A2 ðtÞ s t¼t1
! ð1Þ
where t1 denotes the time 50 ms after the spike peak of the selected spike and slow wave, t2 indicates the time 250 ms after the spike peak of the selected spike and slow wave, and s is number of the data samples obtained during t1 and t2. To characterize the distribution of PSS power, we calculated the mean and the standard deviation (SD) of measured PSS power in each patient during each of interictal and preictal epochs. Next, for Z score normalization, interictal and preictal Z scores were calculated for each electrode of each patient: Z = ([individual value of PSS power] [mean value of PSS power])/(SD of PSS power). Finally, we created interictal and preictal Z score maps by superimposing topographically Z scores onto an intraoperative photograph
Fig. 1. Analysis of the post-spike slow wave (PSS) in patient 7 (A) Interictal spike and slow waves before averaging with 1–70 Hz bandpass filtering (BPF) using a bipolar montage. We selected one electrode (electrode 47 in this case) that showed the highest amplitude spike and slow wave (black arrows) in the seizure onset zone (SOZ). (B) The averaged 1-s epoch. The time point of the spike peak is marked by the vertical line. Ten spike and slow waves (300 ms before the spike peak and 700 ms after the spike peak) are selected for averaging without filtering on the referential montage. An arrow, also in panel C, marks the spike and wave in SOZ (electrode 47), which was used as a basis for averaging. (C) The averaged spike and waves are filtered (0.5–5 Hz). We analyze PSS powers during the period from 50 to 250 ms after the spike peak (200 ms, shaded in yellow).
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
of each patient’s brain surface. These procedures were described in detail elsewhere, and have been demonstrated as an effective means for mapping the topography of power in intracranial EEG recordings (Akiyama et al., 2006; Sato et al., 2013).
2.6. Statistical analysis For analysis of the changes of PSS power, we selected one electrode showing the maximal power of PSS each of within and outside SOZ for all patients. We compared the maximal power of PSS between interictal and preictal periods, both within and outside SOZ. We also analyzed a difference in the highest power of PSS between within and outside SOZ, during both interictal and preictal periods. Z score maps were employed to characterize the distribution of PSS power visually for each patient. The threshold of a Z-score was defined as 1.65, which is equivalent to the significance level of 0.1 in a two-tailed test based on the standard normal distribution. In this study, the threshold of 1.65 was used (Z > 1.65; p < 0.1). For further evaluation of potential differential distribution of PSS power between interictal and preictal periods, we counted the number of the electrodes with Z scores >1.65 of PSS powers both within and outside SOZ for all patients, and constructed the boxplots of Z scores of those electrodes across all patients. We compared the number of the electrodes with high Z score between interictal and preictal periods, each of within and outside SOZ. All comparisons of two samples were performed using a Wilcoxon signed-rank test. These statistical analyses were carried out by Microsoft Excel 2010 (Microsoft Corp., Seattle, WA) and R 3.0.2 (available at http://www.r-project.org).
1673
3.2. Dynamics of PSS power 3.2.1. Z score maps In all patients, Z score maps revealed that the electrodes expressing high PSS power tended to be localized in regions adjacent SOZ during interictal periods, whereas during preictal periods they tended to be localized within SOZ. Within SOZ, electrodes with the highest Z scores were identical between interictal and preictal periods in all patients (Fig. 3). Outside SOZ, the electrodes with the highest Z scores differed between interictal and preictal periods in all patients except patient 7. During preictal periods, the electrodes outside SOZ with the highest Z scores became close to the margin of SOZ in all patients. 3.2.2. Distribution of PSS power Table 3 details the number of electrodes with Z score >1.65 of PSS power within SOZ, which increased from interictal to preictal periods. In all patients, the number of electrodes with Z score >1.65 within SOZ significantly increased from interictal to preictal periods (p < 0.0001). The number of electrodes with Z score >1.65 outside SOZ significantly decreased in all patients except patient 9 whose number was unchanged (p < 0.001). In 3 patients (#1, 6, and 10), there was no electrode with Z score >1.65 outside SOZ during preictal periods. Fig. 4 described the range of Z score >1.65 between within and outside SOZ during interictal and preictal periods. Z scores within SOZ were higher than those outside SOZ during both periods. Since the mean values and SD of PSS power were higher during preictal periods than during interictal periods, the mean values of the Z scores during preictal periods were relatively lower than those of the Z scores during interictal periods. 4. Discussion
3. Results We analyzed a total of 10 seizures and 777 electrodes (mean ± SD, 77.7 ± 21.4 electrodes) in 10 patients. A total of 400 averaged 1-s epochs were analyzed for PSS power in this study.
3.1. Highest power of PSS In all 10 patients, the highest power of PSS both within and outside SOZ significantly increased from interictal to preictal periods (mean ± SE: within SOZ, interictal 40.7 ± 40.4 to preictal; 46.7 ± 45.0 dB, p < 0.001; outside SOZ, 39.9 ± 39.5–43.9 ± 43.3 dB, p < 0.001) (Fig. 2). During interictal period, five patients (#1, 5, 6, 7, and 9) showed higher power of PSS within SOZ than outside SOZ. During preictal periods, all 10 patients showed higher power of PSS within SOZ than outside SOZ (Table 2).
We investigated the characteristics of PSS during interictal and preictal periods, within and outside SOZ in patients with FCD type II. The main findings reported by this study are: (i) a significant increase in the PSS power both within and outside SOZ prior to seizure onset; (ii) a significant difference in the PSS power within and outside SOZ during preictal periods, with the PSS power within SOZ being much higher; and (iii) a peculiar topographic pattern of confinement of the high power PSS in SOZ prior to seizure onset (Fig. 5). 4.1. Increment of PSS powers toward seizure onset Spike-wave discharges in SOZ are assumed to be generated by the synchronous firing of neuronal assemblies and consist of an initial paroxysmal depolarizing shift underlying the generation of the EEG ‘‘spike’’ followed by a period of silence associated with
Fig. 2. The highest power of post-spike slow wave (PSS) between interictal and preictal periods, and between within and outside the seizure onset zone (SOZ) in all 10 patients. The PSS power both within and outside SOZ significantly increased from interictal to preictal periods in all patients. During interictal period, five patients (#1, 5, 6, 7, and 9) showed higher power of PSS within SOZ than outside SOZ. During preictal periods, all patients showed higher power of PSS within SOZ than outside SOZ. inter = interictal; pre = preictal; Pt = patient.
1674
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
Table 2 Highest power (dB) of post-spike slow wave both within SOZ and outside SOZ. # of Pt
Within SOZ
Outside SOZ
# of electrodes
Interictal
Preictal
# of electrodes
Interictal
Preictal
1 2 3 4 5 6 7 8 9 10
4 4 4 5 4 5 4 5 3 4
33.43 28.20 36.53 30.56 44.81 39.74 41.88 39.51 45.57 38.31
39.62 33.10 44.54 34.33 49.59 42.55 48.32 47.06 49.89 44.73
72 92 105 104 59 82 69 68 60 44
30.41 29.90 38.86 31.39 44.26 36.63 39.48 39.90 44.55 39.33
32.82 31.10 41.71 31.54 47.87 38.62 46.32 46.44 47.41 41.54
dB = log10(power(lV2)); SOZ = seizure onset zone.
Fig. 3. Interictal and preictal Z score maps of post-spike slow wave (PSS) power in three representative patients (Patient 5, 7, and 8). The columns (from left to right) show location of seizure onset zone (SOZ; green circles), interictal Z score maps, and preictal Z score maps. Z scores are topographically superimposed onto an intraoperative photograph of individual patient’s brain surface and grid. PSS with high power was confined to SOZ from interictal to preictal periods in all three patients. Similar findings were obtained for all other patients.
Table 3 # of electrodes with Z score >1.65 of PSS powers. # of Pt
1 2 3 4 5 6 7 8 9 10
Within SOZ**
Outside SOZ*
Interictal
Preictal
Interictal
Preictal
3 1 1 2 1 4 2 3 1 2
4 3 2 5 4 5 3 4 3 4
1 4 5 7 5 1 5 4 3 1
– 2 3 5 1 – 1 1 3 –
PSS = post-spike slow wave; SOZ = seizure onset zone. ** p < 0.0001. * p < 0.01.
neuronal hyperpolarization during the ‘‘wave’’ component. At the cellular level the ‘‘spike’’ is initiated through a process of gradual
enhancement of the synaptic excitation that leads to a Ca2+ dependent depolarization (Chamberlin et al., 1990; Hoffman and Haberly, 1991) and is mediated mainly by AMPA and NMDA glutamate receptors. The PSS is associated with a depression in excitability and follows the ‘‘spike’’ component. Since GABA-ergic inhibitory neurons are presumably preserved in SOZ, the activation of recurrent inhibitory networks may be responsible for dampening the neuronal excitability and may play a role in controlling the frequency of interictal spiking (Lebovitz, 1979, Domann et al., 1991; de Curtis et al., 1998). GABA-dependent synaptic inhibition contributes to PSS through mechanisms relying both on fast GABA-A receptor mediated and slow GABA-B receptor mediated processes (Prince, 1971; Hablitz, 1984). However, the GABA-A receptor mediated effects are short lasting even when reinforced by ‘‘slow’’ GABA-B receptor mediated mechanisms. Recent studies have proposed that the EEG ‘‘wave’’ related hyperpolarization, corresponding to PSS in our study, reflects the combined effect of potassium (K+) currents and synaptic disfacilitation rather than pure GABA-mediated synaptic inhibition (Steriade and Amzica, 1999; Neckelmann et al., 2000; Timofeev
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
1675
Fig. 4. Distribution of PSS powers with Z score >1.65 across all patients during interictal and preictal periods. Medians are denoted by solid black lines inside the box while the top and bottom box edges denote the first and third quartile. Whiskers denote the largest and smallest scores within 1.5 times the interquartile range while the outliers (open circles) are more than 1.5 times the interquartile range. Note that ‘‘n’’ means the number of samples for each boxplot, namely the total number of the electrodes showing post-spike slow wave (PSS) powers with Z scores >1.65 of all patients each within and outside the seizure onset zone (SOZ).
Fig. 5. Schematic representation of power and topographic changes of the postspike slow wave (PSS) toward seizure onset. PSS dynamics are characterized by: (i) PSS with increased power both within and outside seizure onset zone (SOZ) toward seizure onset, (ii) higher power PSS within SOZ, relative to outside SOZ, during preictal period, and (iii) confinement of higher power PSS into SOZ toward seizure onset.
and Steriade, 2004). In particular, calcium-dependent K+ currents in pyramidal cells are assumed to play a significant role in the mediation of the EEG ‘‘wave’’ related hyperpolarization and also contribute to depolarizing components of spike and slow waves (Timofeev et al., 2004). Others demonstrated that extrasynaptic mechanisms, such as depolarization block (Prince and Gutnick, 1971), pH changes (de Curtis et al., 1998), and DC shifts (Gumnit et al., 1970; Ikeda et al., 1999), may further contribute to the generation of PSS. At the cellular level the spontaneous development of spikewave seizures from the normal cortical slow-waves is a continuous process in which the difference between the depolarized active state and the hyperpolarized silent state of the membrane potential is constantly increasing. In parallel, the depolarizing shift from one membrane state to the other is becoming steeper, and during the depolarized state neocortical neurons generates fast bursts of action potentials corresponding to the high frequency EEG oscillations (Steriade et al., 1998). Preictal enhancement of PSS power may be corresponding to this gradual evolution of interictal discharges into seizures and can be used as a clinical parameter for resective epilepsy surgery.
4.2. Confinement of high power PSS into SOZ toward seizure onset Z score maps of PSS power revealed that interictal PSS with high power presented a broad topographic distribution. Most of the interictal spike and slow waves with high power PSS distant from SOZ would have been unrecognizable unless they were averaged.
Therefore those interictal PSS might not be diagnosed as interictal epileptic discharges in the conventional EEG. In contrast, preictal high power PSS was confined into SOZ in all patients. Preictal high power PSS even outside SOZ tended to be localized closer to SOZ. In the comparison of PSS power changes we used only one electrode showing the maximum power of PSS. This analysis approach was chosen because the electrode with maximum power would have most reflected the activity of PSS and the adjacent electrodes would have been directly and strongly influenced by the electrode with maximum power. As shown in Fig. 3, the electrode with maximum power of PSS existed each within and outside SOZ respectively. Also, the electrode with maximum power outside SOZ was closer to SOZ during preictal period compared to interictal period. These findings indicated that the physiological characteristics of PSS outside SOZ may differ between interictal and preictal periods. Further studies on this issue will be needed to resolve this issue more completely. Enatsu et al. demonstrated that the amplitudes of the slow waves induced by electrical stimulation in SOZ were significantly higher than those in the control region (Enatsu st al., 2012). Also, according to findings from recent study by Nayak et al. (2014), the slow waves induced by electrical stimulation may have similar implication as the slow waves of spontaneous spike-wave discharges, which correspond to PSS in this study. Such evidence is congruent with our findings that high power PSS was confined to SOZ during the preictal period. One possible mechanism to account for the specific preictal PSS power increase within SOZ may rely on the buffering and redistribution of extracellular K+ by the local glial cells within SOZ. A highly intense firing during very focal spike-wave seizures would increase the extracellular K+ in the epileptogenic region (Amzica et al., 2002). The buffering of extracellular K+ appears to be of critical importance for ensuring normal neuronal excitability (Janigro et al., 1997). In experimental cortical dysplasia, Bordey et al. demonstrated that the buffering capacity for extracellular K+ was lost in the dysplastic lesion, while it was proficient in the hyperexcitable zone (Bordey et al., 2001). In light of such evidence, the preictal confinement of high power PSS into SOZ may indicate a failure of extracellular K+ buffering due to the presence of abnormal glial cells in a very limited topographic area around SOZ.
5. Conclusions Our results support the hypothesis that PSS power exhibits dynamic changes during transition toward seizure onset in focal
1676
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
epilepsies. Preictal PSS power increases together with confinement into SOZ may parallel the cellular transition from interictal to ictal discharges in SOZ, and may be used in EEG analysis for definition of SOZ. Acknowledgements This study was supported fully or partly by research grants from Japan Epilepsy Research Foundation, Uehara Memorial Foundation, Narishige Neuroscience Research Foundation, and Young Investigator Awards from IFCN: the International Federation of Clinical Neurophysiology. This study was also conducted with the support of EpLink – the Epilepsy Research Program funded by the Ontario Brain Institute, an independent non-profit corporation, funded partially by the Ontario government. The opinions, results and conclusions are those of the authors and no endorsement by the Ontario Brain Institute is intended or should be inferred. Conflicts of interest: None of the authors have potential conflicts of interest to be disclosed. References Akiyama T, Otsubo H, Ochi A, Ishiguro T, Kadokura G, Ramachandrannair R, et al. Focal cortical high-frequency oscillations trigger epileptic spasms: confirmation by digital video subdural EEG. Clin Neurophysiol 2005;116:2819–25. Akiyama T, Otsubo H, Ochi A, Galicia EZ, Weiss SK, Donner EJ, et al. Topographic movie of ictal high-frequency oscillations on the brain surface using subdural EEG in neocortical epilepsy. Epilepsia 2006;47:1953–7. 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. Alarcón G, Martinez J, Kerai SV, Lacruz ME, Quiroga RQ, Selway RP, et al. In vivo neuronal firing patterns during human epileptiform discharges replicated by electrical stimulation. Clin Neurophysiol 2012;123:1736–44. Amzica F, Massimini M, Manfridi A. Spatial buffering during slow and paroxysmal sleep oscillations in cortical networks of glial cells in vivo. J Neurosci 2002;22:1042–53. Benifla M, Sala Jr F, Jane J, Otsubo H, Ochi A, Drake J, et al. Neurosurgical management of intractable rolandic epilepsy in children: role of resection in eloquent cortex. Clinical article. J Neurosurg Pediatr 2009;4:199–216. Blumenfeld H. Cellular and network mechanisms of spike-wave seizures. Epilepsia 2005;46:21–33. Bordey A, Lyons SA, Hablitz JJ, Sontheimer H. Electrophysiological characteristics of reactive astrocytes in experimental cortical dysplasia. J Neurophysiol 2001;85:1719–31. 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. Chamberlin NL, Traub RD, Dingledine R. Role of EPSPs in initiation of spontaneous synchronized burst firing in rat hippocampal neurons bathed in high potassium. J Neurophysiol 1990;64:1000–8. de Curtis M, Manfridi A, Biella G. Activity-dependent pH shifts and periodic recurrence of spontaneous interictal spikes in a model of focal epileptogenesis. J Neurosci 1998;18:7543–51. De Curtis M, Tassi L, Lo Russo G, Mai R, Cossu M, Francione S. Increased discharge threshold after an interictal spike in human focal epilepsy. Eur J Neurosci 2005;22:2971–6. Domann R, Uhlig S, Dorn T, Witte OW. Participation of interneurons in penicillininduced epileptic discharges. Exp Brain Res 1991;83:683–6. Enatsu R, Piao Z, O’Connor T, Horning K, Mosher J, Burgess R, et al. Cortical excitability varies upon ictal onset patterns in neocortical epilepsy: a corticocortical evoked potential study. Clin Neurophysiol 2012;123:252–60. Ferrier CH, Alarcon G, Engelsman J, Binnie CD, Koutroumanidis M, Polkey CE, et al. Relevance of residual histologic and electrocorticographic abnormalities for surgical outcome in frontal lobe epilepsy. Epilepsia 2001;42:363–71. Fisher RS, Webber WR, Lesser RP, Arroyo S, Uematsu S. High-frequency EEG activity at the start of seizures. J Clin Neurophysiol 1992;9:441–8.
Francione S, Nobili L, Cardinale F, Citterio A, Galli C, Tassi L. Intra-lesional stereoEEG activity in Taylor’s focal cortical dysplasia. Epileptic Disord 2003;52: 105–14. Gumnit RJ, Matsumoto H, Vasconetto C. DC activity in the depth of an experimental epileptic focus. Electroencephalogr Clin Neurophysiol 1970;28:333–9. Hablitz JJ. Picrotoxin-induced epileptiform activity in hippocampus: role of endogenous versus synaptic factors. J Neurophysiol 1984;51:1011–27. Hoffman WH, Haberly LB. Bursting-induced epileptiform EPSPs in slices of piriform cortex are generated by deep cells. J Neurosci 1991;11:2021–31. Ikeda A, Taki W, Kunieda T, Terada K, Mikuni N, Nagamine T, et al. Focal ictal direct current shifts in human epilepsy as studied by subdural and scalp recording. Brain 1999;122:827–38. Janigro D, Gasparini S, D’Ambrosio R, McKhann 2nd G, DiFrancesco D. Reduction of K+ uptake in glia prevents long-term depression maintenance and causes epileptiform activity. J Neurosci 1997;17:2813–24. Jacobs J, Levan P, Châtillon CE, Olivier A, Dubeau F, Gotman J. High frequency oscillations in intracranial EEGs mark epileptogenicity rather than lesion type. Brain 2009;132:1022–37. Katz A, Marks DA, McCarthy G, Spencer SS. Does interictal spiking change prior to seizures? Electroencephalogr Clin Neurophysiol 1991;79:153–6. Lange HH, Lieb JP, Engel Jr J, Crandall PH. Temporo-spatial patterns of pre-ictal spike activity in human temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol 1983;56:543–55. Lebovitz RM. Autorhythmicity of spontaneous interictal spike discharge at hippocampal penicillin foci. Brain Res 1979;172:35–55. Lopes Silva F, Blanes W, Kalitzin SN, Parra J, Suffczynski P, Velis DN. Epilepsies as dynamical diseases of brain systems: basic models of the transition between normal and epileptic activity. Epilepsia 2003;44:72–83. Matsumoto H, Marsan CA. Cortical cellular phenomena in experimental epilepsy: interictal manifestations. Exp Neurol 1964;9:286–304. Mormann F, Andrzejak RG, Elger CE, Lehnertz K. Seizure prediction: the long and winding road. Brain 2007;130:314–33. Nayak D, Valentín A, Selway RP, Alarcón G. Can single pulse electrical stimulation provoke responses similar to spontaneous interictal epileptiform discharges? Clin Neurophysiol 2014;125:1306–11. Neckelmann D, Amzica F, Steriade M. Changes in neuronal conductance during different components of cortically generated spike-wave seizures. Neuroscience 2000;96:475–85. 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. Otsubo H, Ochi A, Imai K, Akiyama T, Fujimoto A, Go C, et al. High-frequency oscillations of ictal muscle activity and epileptogenic discharges on intracranial EEG in a temporal lobe epilepsy patient. Clin Neurophysiol 2008;119:862–8. Otsubo H, Sharma R, Elliott I, Holowka S, Rutka JT, Snead 3rd OC. Confirmation of two magnetoencephalographic epileptic foci by invasive monitoring from subdural electrodes in an adolescent with right frontocentral epilepsy. Epilepsia 1999;40:608–13. Palmini A, Gambardella A, Andermann F, Dubeau F, da Costa JC, Olivier A, et al. Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol 1995;37:476–87. Prince DA, Gutnick MJ. Neuronal activities in epileptogenic foci of immature cortex. Trans Am Neurol Assoc 1971;96:88–91. Prince DA. Cortical cellular activities during cyclically occurring inter-ictal epileptiform discharges. Electroencephalogr Clin Neurophysiol 1971;31: 469–84. Sato Y, Fukuda M, Oishi M, Shirasawa A, Fujii Y. Ictal near-infrared spectroscopy and electrocorticography study of supplementary motor area seizures. J Biomed Opt 2013;18:76022. 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. Steriade M, Amzica F, Neckelmann D, Timofeev I. Spike-wave complexes and fast components of cortically generated seizures. II. Extra- and intracellular patterns. J Neurophysiol 1998;80:1456–79. Steriade M, Amzica F. Intracellular study of excitability in the seizure-prone neocortex in vivo. J Neurophysiol 1999;82:3108–22. Taylor DC, Falconer MA, Bruton CJ, Corsellis JA. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry 1971;34:369–87. Timofeev I, Grenier F, Steriade M. Contribution of intrinsic neuronal factors in the generation of cortically driven electrographic seizures. J Neurophysiol 2004;92:1133–43. Timofeev I, Steriade M. Review neocortical seizures: initiation, development and cessation. Neuroscience 2004;123:299–336.
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Dynamic changes of interictal post-spike slow waves toward seizure onset in focal cortical dysplasia type II Yosuke Sato a, Sam M. Doesburg b,c,d,e, Simeon M. Wong b,c, Tohru Okanishi a, Ryan Anderson b, Dragos A. Nita a, Ayako Ochi a, Hiroshi Otsubo a,⇑ a
Division of Neurology, Hospital for Sick Children, Toronto, Ont., Canada Department of Diagnostic Imaging, Hospital for Sick Children, Toronto, Ont., Canada Neuroscience & Mental Health Program, Hospital for Sick Children Research Institute, Toronto, Ont., Canada d Department of Medical Imaging, University of Toronto, Toronto, Ont., Canada e Department of Psychology, University of Toronto, Toronto, Ont., Canada b c
a r t i c l e
i n f o
Article history: Accepted 15 November 2014 Available online 25 November 2014 Keywords: Post-spike slow wave Interictal spike Focal cortical dysplasia Seizure onset zone Preictal period
h i g h l i g h t s Power of post-spike slow wave increased in all areas from interictal to preictal period. Power of post-spike slow wave within the seizure onset zone (SOZ) was significantly higher than that
outside SOZ during preictal period. Post-spike slow wave with high power became confined to SOZ toward seizure onset.
a b s t r a c t Objective: A post-spike slow wave (PSS) as part of a spike and slow wave is presumably related to inhibition of epileptic activity. In this study, we evaluated dynamic changes of PSS power toward seizure onset in patients with focal cortical dysplasia (FCD) type II. Methods: We collected data from 10 pediatric patients with FCD type II, who underwent invasive monitoring with subdural grids. The PSS were averaged based on spike-triggering in 30 s epochs during both interictal and preictal periods. We quantitatively measured and compared PSS power and distribution between interictal and preictal periods, both within and outside the seizure onset zone (SOZ). Results: PSS power was significantly higher in all areas during preictal period compared with interictal period. During preictal period, PSS power within SOZ was significantly higher than outside SOZ. From interictal to preictal period, the number of electrodes with high power PSS significantly increased within SOZ and decreased outside SOZ. Conclusions: Toward seizure onset, PSS power increased in all areas, predominantly within SOZ, and became confined into SOZ in a subset of FCD type II patients. Significance: Preictal PSS power increase and confinement into SOZ accompany transition to seizures. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The spike and slow wave pattern on EEG is well known to be an interictal epileptiform discharge in focal epilepsy (Alarcón et al., 2012). The spike and slow wave complex consists of an initial high-frequency neuronal excitation during the ‘‘spike’’ phase and
⇑ Corresponding author at: Division of Neurology, The Hospital for Sick Children, Toronto M5G 1X8, Ont., Canada. Tel.: +1 416 813 6295; fax: +1 416 813 6334. E-mail addresses: [email protected], [email protected] (H. Otsubo).
a following inhibition or disfacilitation of neuronal firing during the ‘‘slow wave’’ phase (Matsumoto and Marsan, 1964; Blumenfeld, 2005; Alarcón et al., 2012). As previously shown in animal models of focal epilepsy, interictal post-spike slow waves (PSS) are associated with inhibitory potentials that play a protective role to prevent the propagation of epileptic discharges (Lebovitz, 1979, Domann et al., 1991; de Curtis et al., 1998). De Curtis et al. demonstrated that the presence of interictal post-spike depression, which corresponds to PSS, correlated with an increased threshold for the generation of interictal spikes evoked in response to afferent stimulation in human focal epilepsy (de Curtis et al., 2005). The
http://dx.doi.org/10.1016/j.clinph.2014.11.012 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1671
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
dynamics of PSS power may take part in inhibitory control of interictal epileptic discharges. Focal cortical dysplasia (FCD) type II has been known to present with frequent interictal spike and slow waves in the lesion (Taylor et al., 1971; Ferrier et al., 2001; Francione et al., 2003). Palmini et al. indicated that the interictal discharges in FCD type II were intrinsically epileptogenic (Palmini et al., 1995). Focal seizures could be triggered by an evolution of preictal epileptic activity, mediated by increased neuronal excitation accompanied by a loss of inhibition (Lopes da Silva et al., 2003; Mormann et al., 2007). Lange et al. reported focal decreases in spiking rate which coincided with increased rates of bilateral spikes before seizures (Lange et al., 1983). Conversely, Katz et al. demonstrated that there were no remarkable changes in spike rates before seizures (Katz et al., 1991). We hypothesized that PSS power would show dynamic changes during transition toward seizure onset in focal epilepsy and that such changes may reflect the progressive inhibitory mechanisms relevant for impeding the evolution of interictal epileptiform discharges into focal ictal events. We compared PSS power within and outside the seizure onset zone (SOZ) during both interictal and preictal periods. This report is the first to characterize the PSS dynamics associated with seizure onset in focal epilepsies including characterization of both power and topographic distribution. 2. Methods 2.1. Subjects We retrospectively analyzed data collected between June 2009 and September 2013 from 10 pediatric patients (6 males and 4 females) with medically refractory focal epilepsy secondary to focal cortical dysplasia type II who underwent intracranial video EEG (IVEEG) monitoring prior to surgical resection. Age at data recording and surgery ranged from 3 to 14 years. The cortical dysplastic lesions were verified with magnetic resonance imaging and subsequently confirmed by pathology. The sites of electrode placement were individualized according to clinical history, seizure seminology, neuroimaging and scalp EEG findings as previously described (Otsubo et al., 1999). In order to investigate the hypothesis that the power of PSS is differentially distributed within and outside SOZ, we retained for analysis patients fulfilling the following criteria: SOZ was well localized, all seizures started from the same SOZ, interictal spike and slow waves were seen simultaneously both within and outside SOZ. In all patients, postsurgical neuropathological diagnosis revealed FCD type II. The patient clinical characteristics are summarized in Table 1. This study was approved by the Research Ethic Board at the Hospital for Sick Children.
2.2. IVEEG recordings The implantation of intracranial electrodes and extraoperative functional mapping were performed as described previously (Benifla et al., 2009). Center to center spacing of the contacts of the subdural electrodes was 10.5 mm (Ad-Tech, Racine, WI, USA). IVEEG data were acquired using a Harmonie system (Stellate, Montreal, PQ, Canada). The EEG signals were sampled at 1 kHz in two patients (Patient 1 and 2) and were sampled at 2 kHz in eight patients (Patients 3–10). The recordings were performed referentially with the referential electrode placed in a cortical region without epileptic discharges. 2.3. Determination of SOZ High-frequency oscillations (HFOs) are thought to be biomarkers of epileptogenic brain regions (Bragin et al., 1999; Staba et al., 2002) and have been shown to be topographically concentrated in SOZ during ictal and interictal periods (Fisher et al., 1992; Alarcon et al., 1995; Akiyama et al., 2005; Ochi et al., 2007). Given such findings, board certified clinical neurophysiologists determined SOZ for each patient by visual inspection and spectral analysis of intracranial EEG, in accordance with the current clinical practice at our institution. The details of these methods have been described elsewhere (Ochi et al., 2007; Jacobs et al., 2009). For each patient, three to five electrodes (4.2 ± 0.6; total 42) were classified as being within SOZ. The locations and number of the analyzed electrodes are shown in Table 1. 2.4. EEG data selection Two certified clinical neurophysiologists (AO, HO) determined electrographic seizure onset. The entire IVEEG recording was reviewed for the presence of seizures. All electrographic seizures were marked for times of unequivocal electrographic seizure onset defined as the earliest electrical changes recorded as the paroxysmal fast activity, repetitive spiking, or rhythmic slow wave. For each patient, at least one typical partial seizure was confirmed during IVEEG recordings. We selected the most typical partial seizure per patient. In all patients, electrographic seizure onset began from one to three electrodes, which were consistent with the electrodes defined as those in SOZ. A preictal epoch was extracted which consisted of 30 s of activity immediately preceding electrographic seizure onset. An interictal epoch was also extracted for each patient, and was comprised of 30 s of activity during Stage II non-REM sleep remote from seizures by at least 1 h. Since it was necessary to make recording conditions constant for comparing of PSS powers, EEG data from depth electrodes were excluded in the present analysis. Moreover, each selected epoch was visually inspected to
Table 1 Clinical characteristics of 10 patients with FCD type II. # of Pt
1 2 3 4 5 6 7 8 9 10
Age (years)
Gender
At onset
At surgery
1 3 5 3 2 3 10 2 8 10
4 12 7 7 3 3 14 4 9 12
Male Female Male Male Male Female Female Female Male Male
Location of FCD on MRI
Rt. TP Lt. P Lt. F Rt. F Lt. F Rt. F Rt. PO Lt. P Lt. F Rt. T
Extraoperative video EEG Location covered by electrodes
# of analyzed electrodes
Rt. FTP Lt. FP Lt. FTP Rt. FTP Lt. FTP Rt. FTP Rt. PO Lt. FTP Lt. FP Rt. OT
76 96 109 109 63 87 63 63 63 48
F = frontal; FCD = focal cortical dysplasia; Lt. = left; MRI = magnetic resonance imaging; O = occipital; P = parietal; Rt. = right; SOZ = seizure onset zone; T = temporal.
1672
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
ensure it was not contaminated by significant artifacts, such as environmental artifacts and muscle artifacts (Otsubo et al., 2008). Table 1 lists the number of the implanted electrodes, the analyzed electrodes, and the number of electrodes in SOZ. 2.5. Analysis of PSS power and distribution To suppress contamination by background activity for PSS analysis, we carried out spike-triggered averaging based on the selected spike and slow waves using EEG Harmonie Stellate software and a custom-made program written in MATLAB (The MathWorks, Version 7.1). The following procedures were performed in all patients during both interictal and preictal epochs. We selected and averaged ten 1-s epochs including the spike and slow waves that met the following criteria: (i) the spike and slow waves were verified as repetitive discharges typically seen in focal cortical dysplasia by visual inspection by a clinical neurophysiologist (YS) using a bipolar montage, 1–70 Hz bandpass filter on 10 s/page, (ii) we selected one electrode that showed the highest amplitude spike within SOZ, (iii) consecutive spike and slow waves did not overlap within 300 ms, (iv) we selected 10 epochs of one second from 300 ms before to 700 ms after the spike peak, (v) using a referential montage without filter, we averaged 10 epochs (Fig. 1A).
We estimated the spike duration up to 100 ms and focused on the period of PSS ranging from 50 to 250 ms after the spike peak. During this period, we quantitatively measured the amplitudes (lV) of the PSS (A(t)) using 0.5–5 Hz filter (Fig. 1B). We verified this bandpass filter according to the evidence that the frequencies of PSS varied between 1 and 4 Hz (Blumenfeld, 2005). The mean power (dB) of the PSS (P) were then obtained by the following formula:
P ¼ 10log10
t2 1X A2 ðtÞ s t¼t1
! ð1Þ
where t1 denotes the time 50 ms after the spike peak of the selected spike and slow wave, t2 indicates the time 250 ms after the spike peak of the selected spike and slow wave, and s is number of the data samples obtained during t1 and t2. To characterize the distribution of PSS power, we calculated the mean and the standard deviation (SD) of measured PSS power in each patient during each of interictal and preictal epochs. Next, for Z score normalization, interictal and preictal Z scores were calculated for each electrode of each patient: Z = ([individual value of PSS power] [mean value of PSS power])/(SD of PSS power). Finally, we created interictal and preictal Z score maps by superimposing topographically Z scores onto an intraoperative photograph
Fig. 1. Analysis of the post-spike slow wave (PSS) in patient 7 (A) Interictal spike and slow waves before averaging with 1–70 Hz bandpass filtering (BPF) using a bipolar montage. We selected one electrode (electrode 47 in this case) that showed the highest amplitude spike and slow wave (black arrows) in the seizure onset zone (SOZ). (B) The averaged 1-s epoch. The time point of the spike peak is marked by the vertical line. Ten spike and slow waves (300 ms before the spike peak and 700 ms after the spike peak) are selected for averaging without filtering on the referential montage. An arrow, also in panel C, marks the spike and wave in SOZ (electrode 47), which was used as a basis for averaging. (C) The averaged spike and waves are filtered (0.5–5 Hz). We analyze PSS powers during the period from 50 to 250 ms after the spike peak (200 ms, shaded in yellow).
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
of each patient’s brain surface. These procedures were described in detail elsewhere, and have been demonstrated as an effective means for mapping the topography of power in intracranial EEG recordings (Akiyama et al., 2006; Sato et al., 2013).
2.6. Statistical analysis For analysis of the changes of PSS power, we selected one electrode showing the maximal power of PSS each of within and outside SOZ for all patients. We compared the maximal power of PSS between interictal and preictal periods, both within and outside SOZ. We also analyzed a difference in the highest power of PSS between within and outside SOZ, during both interictal and preictal periods. Z score maps were employed to characterize the distribution of PSS power visually for each patient. The threshold of a Z-score was defined as 1.65, which is equivalent to the significance level of 0.1 in a two-tailed test based on the standard normal distribution. In this study, the threshold of 1.65 was used (Z > 1.65; p < 0.1). For further evaluation of potential differential distribution of PSS power between interictal and preictal periods, we counted the number of the electrodes with Z scores >1.65 of PSS powers both within and outside SOZ for all patients, and constructed the boxplots of Z scores of those electrodes across all patients. We compared the number of the electrodes with high Z score between interictal and preictal periods, each of within and outside SOZ. All comparisons of two samples were performed using a Wilcoxon signed-rank test. These statistical analyses were carried out by Microsoft Excel 2010 (Microsoft Corp., Seattle, WA) and R 3.0.2 (available at http://www.r-project.org).
1673
3.2. Dynamics of PSS power 3.2.1. Z score maps In all patients, Z score maps revealed that the electrodes expressing high PSS power tended to be localized in regions adjacent SOZ during interictal periods, whereas during preictal periods they tended to be localized within SOZ. Within SOZ, electrodes with the highest Z scores were identical between interictal and preictal periods in all patients (Fig. 3). Outside SOZ, the electrodes with the highest Z scores differed between interictal and preictal periods in all patients except patient 7. During preictal periods, the electrodes outside SOZ with the highest Z scores became close to the margin of SOZ in all patients. 3.2.2. Distribution of PSS power Table 3 details the number of electrodes with Z score >1.65 of PSS power within SOZ, which increased from interictal to preictal periods. In all patients, the number of electrodes with Z score >1.65 within SOZ significantly increased from interictal to preictal periods (p < 0.0001). The number of electrodes with Z score >1.65 outside SOZ significantly decreased in all patients except patient 9 whose number was unchanged (p < 0.001). In 3 patients (#1, 6, and 10), there was no electrode with Z score >1.65 outside SOZ during preictal periods. Fig. 4 described the range of Z score >1.65 between within and outside SOZ during interictal and preictal periods. Z scores within SOZ were higher than those outside SOZ during both periods. Since the mean values and SD of PSS power were higher during preictal periods than during interictal periods, the mean values of the Z scores during preictal periods were relatively lower than those of the Z scores during interictal periods. 4. Discussion
3. Results We analyzed a total of 10 seizures and 777 electrodes (mean ± SD, 77.7 ± 21.4 electrodes) in 10 patients. A total of 400 averaged 1-s epochs were analyzed for PSS power in this study.
3.1. Highest power of PSS In all 10 patients, the highest power of PSS both within and outside SOZ significantly increased from interictal to preictal periods (mean ± SE: within SOZ, interictal 40.7 ± 40.4 to preictal; 46.7 ± 45.0 dB, p < 0.001; outside SOZ, 39.9 ± 39.5–43.9 ± 43.3 dB, p < 0.001) (Fig. 2). During interictal period, five patients (#1, 5, 6, 7, and 9) showed higher power of PSS within SOZ than outside SOZ. During preictal periods, all 10 patients showed higher power of PSS within SOZ than outside SOZ (Table 2).
We investigated the characteristics of PSS during interictal and preictal periods, within and outside SOZ in patients with FCD type II. The main findings reported by this study are: (i) a significant increase in the PSS power both within and outside SOZ prior to seizure onset; (ii) a significant difference in the PSS power within and outside SOZ during preictal periods, with the PSS power within SOZ being much higher; and (iii) a peculiar topographic pattern of confinement of the high power PSS in SOZ prior to seizure onset (Fig. 5). 4.1. Increment of PSS powers toward seizure onset Spike-wave discharges in SOZ are assumed to be generated by the synchronous firing of neuronal assemblies and consist of an initial paroxysmal depolarizing shift underlying the generation of the EEG ‘‘spike’’ followed by a period of silence associated with
Fig. 2. The highest power of post-spike slow wave (PSS) between interictal and preictal periods, and between within and outside the seizure onset zone (SOZ) in all 10 patients. The PSS power both within and outside SOZ significantly increased from interictal to preictal periods in all patients. During interictal period, five patients (#1, 5, 6, 7, and 9) showed higher power of PSS within SOZ than outside SOZ. During preictal periods, all patients showed higher power of PSS within SOZ than outside SOZ. inter = interictal; pre = preictal; Pt = patient.
1674
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
Table 2 Highest power (dB) of post-spike slow wave both within SOZ and outside SOZ. # of Pt
Within SOZ
Outside SOZ
# of electrodes
Interictal
Preictal
# of electrodes
Interictal
Preictal
1 2 3 4 5 6 7 8 9 10
4 4 4 5 4 5 4 5 3 4
33.43 28.20 36.53 30.56 44.81 39.74 41.88 39.51 45.57 38.31
39.62 33.10 44.54 34.33 49.59 42.55 48.32 47.06 49.89 44.73
72 92 105 104 59 82 69 68 60 44
30.41 29.90 38.86 31.39 44.26 36.63 39.48 39.90 44.55 39.33
32.82 31.10 41.71 31.54 47.87 38.62 46.32 46.44 47.41 41.54
dB = log10(power(lV2)); SOZ = seizure onset zone.
Fig. 3. Interictal and preictal Z score maps of post-spike slow wave (PSS) power in three representative patients (Patient 5, 7, and 8). The columns (from left to right) show location of seizure onset zone (SOZ; green circles), interictal Z score maps, and preictal Z score maps. Z scores are topographically superimposed onto an intraoperative photograph of individual patient’s brain surface and grid. PSS with high power was confined to SOZ from interictal to preictal periods in all three patients. Similar findings were obtained for all other patients.
Table 3 # of electrodes with Z score >1.65 of PSS powers. # of Pt
1 2 3 4 5 6 7 8 9 10
Within SOZ**
Outside SOZ*
Interictal
Preictal
Interictal
Preictal
3 1 1 2 1 4 2 3 1 2
4 3 2 5 4 5 3 4 3 4
1 4 5 7 5 1 5 4 3 1
– 2 3 5 1 – 1 1 3 –
PSS = post-spike slow wave; SOZ = seizure onset zone. ** p < 0.0001. * p < 0.01.
neuronal hyperpolarization during the ‘‘wave’’ component. At the cellular level the ‘‘spike’’ is initiated through a process of gradual
enhancement of the synaptic excitation that leads to a Ca2+ dependent depolarization (Chamberlin et al., 1990; Hoffman and Haberly, 1991) and is mediated mainly by AMPA and NMDA glutamate receptors. The PSS is associated with a depression in excitability and follows the ‘‘spike’’ component. Since GABA-ergic inhibitory neurons are presumably preserved in SOZ, the activation of recurrent inhibitory networks may be responsible for dampening the neuronal excitability and may play a role in controlling the frequency of interictal spiking (Lebovitz, 1979, Domann et al., 1991; de Curtis et al., 1998). GABA-dependent synaptic inhibition contributes to PSS through mechanisms relying both on fast GABA-A receptor mediated and slow GABA-B receptor mediated processes (Prince, 1971; Hablitz, 1984). However, the GABA-A receptor mediated effects are short lasting even when reinforced by ‘‘slow’’ GABA-B receptor mediated mechanisms. Recent studies have proposed that the EEG ‘‘wave’’ related hyperpolarization, corresponding to PSS in our study, reflects the combined effect of potassium (K+) currents and synaptic disfacilitation rather than pure GABA-mediated synaptic inhibition (Steriade and Amzica, 1999; Neckelmann et al., 2000; Timofeev
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
1675
Fig. 4. Distribution of PSS powers with Z score >1.65 across all patients during interictal and preictal periods. Medians are denoted by solid black lines inside the box while the top and bottom box edges denote the first and third quartile. Whiskers denote the largest and smallest scores within 1.5 times the interquartile range while the outliers (open circles) are more than 1.5 times the interquartile range. Note that ‘‘n’’ means the number of samples for each boxplot, namely the total number of the electrodes showing post-spike slow wave (PSS) powers with Z scores >1.65 of all patients each within and outside the seizure onset zone (SOZ).
Fig. 5. Schematic representation of power and topographic changes of the postspike slow wave (PSS) toward seizure onset. PSS dynamics are characterized by: (i) PSS with increased power both within and outside seizure onset zone (SOZ) toward seizure onset, (ii) higher power PSS within SOZ, relative to outside SOZ, during preictal period, and (iii) confinement of higher power PSS into SOZ toward seizure onset.
and Steriade, 2004). In particular, calcium-dependent K+ currents in pyramidal cells are assumed to play a significant role in the mediation of the EEG ‘‘wave’’ related hyperpolarization and also contribute to depolarizing components of spike and slow waves (Timofeev et al., 2004). Others demonstrated that extrasynaptic mechanisms, such as depolarization block (Prince and Gutnick, 1971), pH changes (de Curtis et al., 1998), and DC shifts (Gumnit et al., 1970; Ikeda et al., 1999), may further contribute to the generation of PSS. At the cellular level the spontaneous development of spikewave seizures from the normal cortical slow-waves is a continuous process in which the difference between the depolarized active state and the hyperpolarized silent state of the membrane potential is constantly increasing. In parallel, the depolarizing shift from one membrane state to the other is becoming steeper, and during the depolarized state neocortical neurons generates fast bursts of action potentials corresponding to the high frequency EEG oscillations (Steriade et al., 1998). Preictal enhancement of PSS power may be corresponding to this gradual evolution of interictal discharges into seizures and can be used as a clinical parameter for resective epilepsy surgery.
4.2. Confinement of high power PSS into SOZ toward seizure onset Z score maps of PSS power revealed that interictal PSS with high power presented a broad topographic distribution. Most of the interictal spike and slow waves with high power PSS distant from SOZ would have been unrecognizable unless they were averaged.
Therefore those interictal PSS might not be diagnosed as interictal epileptic discharges in the conventional EEG. In contrast, preictal high power PSS was confined into SOZ in all patients. Preictal high power PSS even outside SOZ tended to be localized closer to SOZ. In the comparison of PSS power changes we used only one electrode showing the maximum power of PSS. This analysis approach was chosen because the electrode with maximum power would have most reflected the activity of PSS and the adjacent electrodes would have been directly and strongly influenced by the electrode with maximum power. As shown in Fig. 3, the electrode with maximum power of PSS existed each within and outside SOZ respectively. Also, the electrode with maximum power outside SOZ was closer to SOZ during preictal period compared to interictal period. These findings indicated that the physiological characteristics of PSS outside SOZ may differ between interictal and preictal periods. Further studies on this issue will be needed to resolve this issue more completely. Enatsu et al. demonstrated that the amplitudes of the slow waves induced by electrical stimulation in SOZ were significantly higher than those in the control region (Enatsu st al., 2012). Also, according to findings from recent study by Nayak et al. (2014), the slow waves induced by electrical stimulation may have similar implication as the slow waves of spontaneous spike-wave discharges, which correspond to PSS in this study. Such evidence is congruent with our findings that high power PSS was confined to SOZ during the preictal period. One possible mechanism to account for the specific preictal PSS power increase within SOZ may rely on the buffering and redistribution of extracellular K+ by the local glial cells within SOZ. A highly intense firing during very focal spike-wave seizures would increase the extracellular K+ in the epileptogenic region (Amzica et al., 2002). The buffering of extracellular K+ appears to be of critical importance for ensuring normal neuronal excitability (Janigro et al., 1997). In experimental cortical dysplasia, Bordey et al. demonstrated that the buffering capacity for extracellular K+ was lost in the dysplastic lesion, while it was proficient in the hyperexcitable zone (Bordey et al., 2001). In light of such evidence, the preictal confinement of high power PSS into SOZ may indicate a failure of extracellular K+ buffering due to the presence of abnormal glial cells in a very limited topographic area around SOZ.
5. Conclusions Our results support the hypothesis that PSS power exhibits dynamic changes during transition toward seizure onset in focal
1676
Y. Sato et al. / Clinical Neurophysiology 126 (2015) 1670–1676
epilepsies. Preictal PSS power increases together with confinement into SOZ may parallel the cellular transition from interictal to ictal discharges in SOZ, and may be used in EEG analysis for definition of SOZ. Acknowledgements This study was supported fully or partly by research grants from Japan Epilepsy Research Foundation, Uehara Memorial Foundation, Narishige Neuroscience Research Foundation, and Young Investigator Awards from IFCN: the International Federation of Clinical Neurophysiology. This study was also conducted with the support of EpLink – the Epilepsy Research Program funded by the Ontario Brain Institute, an independent non-profit corporation, funded partially by the Ontario government. The opinions, results and conclusions are those of the authors and no endorsement by the Ontario Brain Institute is intended or should be inferred. Conflicts of interest: None of the authors have potential conflicts of interest to be disclosed. References Akiyama T, Otsubo H, Ochi A, Ishiguro T, Kadokura G, Ramachandrannair R, et al. Focal cortical high-frequency oscillations trigger epileptic spasms: confirmation by digital video subdural EEG. Clin Neurophysiol 2005;116:2819–25. Akiyama T, Otsubo H, Ochi A, Galicia EZ, Weiss SK, Donner EJ, et al. Topographic movie of ictal high-frequency oscillations on the brain surface using subdural EEG in neocortical epilepsy. Epilepsia 2006;47:1953–7. 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. Alarcón G, Martinez J, Kerai SV, Lacruz ME, Quiroga RQ, Selway RP, et al. In vivo neuronal firing patterns during human epileptiform discharges replicated by electrical stimulation. Clin Neurophysiol 2012;123:1736–44. Amzica F, Massimini M, Manfridi A. Spatial buffering during slow and paroxysmal sleep oscillations in cortical networks of glial cells in vivo. J Neurosci 2002;22:1042–53. Benifla M, Sala Jr F, Jane J, Otsubo H, Ochi A, Drake J, et al. Neurosurgical management of intractable rolandic epilepsy in children: role of resection in eloquent cortex. Clinical article. J Neurosurg Pediatr 2009;4:199–216. Blumenfeld H. Cellular and network mechanisms of spike-wave seizures. Epilepsia 2005;46:21–33. Bordey A, Lyons SA, Hablitz JJ, Sontheimer H. Electrophysiological characteristics of reactive astrocytes in experimental cortical dysplasia. J Neurophysiol 2001;85:1719–31. 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. Chamberlin NL, Traub RD, Dingledine R. Role of EPSPs in initiation of spontaneous synchronized burst firing in rat hippocampal neurons bathed in high potassium. J Neurophysiol 1990;64:1000–8. de Curtis M, Manfridi A, Biella G. Activity-dependent pH shifts and periodic recurrence of spontaneous interictal spikes in a model of focal epileptogenesis. J Neurosci 1998;18:7543–51. De Curtis M, Tassi L, Lo Russo G, Mai R, Cossu M, Francione S. Increased discharge threshold after an interictal spike in human focal epilepsy. Eur J Neurosci 2005;22:2971–6. Domann R, Uhlig S, Dorn T, Witte OW. Participation of interneurons in penicillininduced epileptic discharges. Exp Brain Res 1991;83:683–6. Enatsu R, Piao Z, O’Connor T, Horning K, Mosher J, Burgess R, et al. Cortical excitability varies upon ictal onset patterns in neocortical epilepsy: a corticocortical evoked potential study. Clin Neurophysiol 2012;123:252–60. Ferrier CH, Alarcon G, Engelsman J, Binnie CD, Koutroumanidis M, Polkey CE, et al. Relevance of residual histologic and electrocorticographic abnormalities for surgical outcome in frontal lobe epilepsy. Epilepsia 2001;42:363–71. Fisher RS, Webber WR, Lesser RP, Arroyo S, Uematsu S. High-frequency EEG activity at the start of seizures. J Clin Neurophysiol 1992;9:441–8.
Francione S, Nobili L, Cardinale F, Citterio A, Galli C, Tassi L. Intra-lesional stereoEEG activity in Taylor’s focal cortical dysplasia. Epileptic Disord 2003;52: 105–14. Gumnit RJ, Matsumoto H, Vasconetto C. DC activity in the depth of an experimental epileptic focus. Electroencephalogr Clin Neurophysiol 1970;28:333–9. Hablitz JJ. Picrotoxin-induced epileptiform activity in hippocampus: role of endogenous versus synaptic factors. J Neurophysiol 1984;51:1011–27. Hoffman WH, Haberly LB. Bursting-induced epileptiform EPSPs in slices of piriform cortex are generated by deep cells. J Neurosci 1991;11:2021–31. Ikeda A, Taki W, Kunieda T, Terada K, Mikuni N, Nagamine T, et al. Focal ictal direct current shifts in human epilepsy as studied by subdural and scalp recording. Brain 1999;122:827–38. Janigro D, Gasparini S, D’Ambrosio R, McKhann 2nd G, DiFrancesco D. Reduction of K+ uptake in glia prevents long-term depression maintenance and causes epileptiform activity. J Neurosci 1997;17:2813–24. Jacobs J, Levan P, Châtillon CE, Olivier A, Dubeau F, Gotman J. High frequency oscillations in intracranial EEGs mark epileptogenicity rather than lesion type. Brain 2009;132:1022–37. Katz A, Marks DA, McCarthy G, Spencer SS. Does interictal spiking change prior to seizures? Electroencephalogr Clin Neurophysiol 1991;79:153–6. Lange HH, Lieb JP, Engel Jr J, Crandall PH. Temporo-spatial patterns of pre-ictal spike activity in human temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol 1983;56:543–55. Lebovitz RM. Autorhythmicity of spontaneous interictal spike discharge at hippocampal penicillin foci. Brain Res 1979;172:35–55. Lopes Silva F, Blanes W, Kalitzin SN, Parra J, Suffczynski P, Velis DN. Epilepsies as dynamical diseases of brain systems: basic models of the transition between normal and epileptic activity. Epilepsia 2003;44:72–83. Matsumoto H, Marsan CA. Cortical cellular phenomena in experimental epilepsy: interictal manifestations. Exp Neurol 1964;9:286–304. Mormann F, Andrzejak RG, Elger CE, Lehnertz K. Seizure prediction: the long and winding road. Brain 2007;130:314–33. Nayak D, Valentín A, Selway RP, Alarcón G. Can single pulse electrical stimulation provoke responses similar to spontaneous interictal epileptiform discharges? Clin Neurophysiol 2014;125:1306–11. Neckelmann D, Amzica F, Steriade M. Changes in neuronal conductance during different components of cortically generated spike-wave seizures. Neuroscience 2000;96:475–85. 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. Otsubo H, Ochi A, Imai K, Akiyama T, Fujimoto A, Go C, et al. High-frequency oscillations of ictal muscle activity and epileptogenic discharges on intracranial EEG in a temporal lobe epilepsy patient. Clin Neurophysiol 2008;119:862–8. Otsubo H, Sharma R, Elliott I, Holowka S, Rutka JT, Snead 3rd OC. Confirmation of two magnetoencephalographic epileptic foci by invasive monitoring from subdural electrodes in an adolescent with right frontocentral epilepsy. Epilepsia 1999;40:608–13. Palmini A, Gambardella A, Andermann F, Dubeau F, da Costa JC, Olivier A, et al. Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol 1995;37:476–87. Prince DA, Gutnick MJ. Neuronal activities in epileptogenic foci of immature cortex. Trans Am Neurol Assoc 1971;96:88–91. Prince DA. Cortical cellular activities during cyclically occurring inter-ictal epileptiform discharges. Electroencephalogr Clin Neurophysiol 1971;31: 469–84. Sato Y, Fukuda M, Oishi M, Shirasawa A, Fujii Y. Ictal near-infrared spectroscopy and electrocorticography study of supplementary motor area seizures. J Biomed Opt 2013;18:76022. 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. Steriade M, Amzica F, Neckelmann D, Timofeev I. Spike-wave complexes and fast components of cortically generated seizures. II. Extra- and intracellular patterns. J Neurophysiol 1998;80:1456–79. Steriade M, Amzica F. Intracellular study of excitability in the seizure-prone neocortex in vivo. J Neurophysiol 1999;82:3108–22. Taylor DC, Falconer MA, Bruton CJ, Corsellis JA. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry 1971;34:369–87. Timofeev I, Grenier F, Steriade M. Contribution of intrinsic neuronal factors in the generation of cortically driven electrographic seizures. J Neurophysiol 2004;92:1133–43. Timofeev I, Steriade M. Review neocortical seizures: initiation, development and cessation. Neuroscience 2004;123:299–336.
