Secondary Logo

Share this article on:

Dynamic Preictal Discharges in Patients With Mesial Temporal Lobe Epilepsy

Chen, Jia*; Li, Liping*; Wu, Dongyan; Li, Xiaoxuan; Xue, Qing*; Wang, Liying; Du, Jialin*; Wang, Di*; Hu, Minjing*; Ren, Liankun*; Wang, Yuping*

Journal of Clinical Neurophysiology: September 2018 - Volume 35 - Issue 5 - p 381–387
doi: 10.1097/WNP.0000000000000486
Original Research

Purpose: It has been challenging to detect early changes preceding seizure onset in patients with epilepsy. This study investigated the preictal discharges (PIDs) by intracranial electroencephalogram of 11 seizures from 7 patients with mesial temporal lobe epilepsy.

Methods: The EEG segments consisting of 30 seconds before ictal onset and 5 seconds after ictal onset were selected for analysis. After PID detection, the amplitude and interval were measured. According to the timing of PID onset, the 30-second period preceding seizure onset was divided into two stages: before PID stage and PID stage. The autocorrelation coefficients during the two stages were calculated and compared.

Results: Preictal discharge amplitude progressively increased, while PID interval gradually decreased toward seizure onset. The autocorrelation coefficients of PID channels were significantly higher during PID stage than before PID stage. There was an overlap between channels with PIDs and seizure onset channels (80.77%).

Conclusions: Preictal discharges emerge prior to ictal event, with a dynamic change and a spatial correlation with seizure onset zone. These findings deepen our understanding of seizure generation and help early prediction and localization of seizure onset zone.

*Department of Neurology, Beijing Key Laboratory of Neuromodulation, Xuanwu Hospital, Capital Medical University, Beijing, China;

Department of Neurology, China-Japan Friendship Hospital, Beijing, China; and

Department of Neurology, Jiaozhou Central Hospital of Qingdao, Shandong, China.

Address correspondence and reprint requests to Liankun Ren, MD, Department of Neurology, Beijing Key Laboratory of Neuromodulation, Xuanwu Hospital, Capital Medical University, No. 45 Changchun Street, Xicheng District, Beijing, 100053, China; e-mail: everest20@163.com.

The authors have no funding or conflicts of interest to disclose.

J. Chen and L. Li contributed equally to this work.

Ethical Publication Statement: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.

An epileptic seizure is a transient and intrinsic dynamic phenomenon. Despite considerable studies on the ictogenesis, our knowledge on the initiation of seizures is still limited. Epileptic seizures are assumed to be an abrupt onset resulting from hypersynchrony of the populations of neurons. The existence of preictal period, however, has been supported by previous observations, including the increase in cerebral blood flow,1,2 the alterations in heart rate,3,4 and the changes in blood oxygen level prior to seizure onset.5 In particular, the firing of neurons and multiunit activities preceding seizures indicate that preictal EEG changes are detectable.6–8 Various biomarkers have been proposed to identify preictal stage. For example, abnormally intensified gamma rhythms,2,9,10 high-frequency oscillations,11 and the power imbalance between spike-related high-frequency oscillations and postspike slow waves12–14 correlated with the transition to seizures.

Recently, preictal discharges (PIDs) were described to be characterized by higher amplitude and faster propagation speed than interictal discharges (IIDs) and prior to ictal events,15–18 suggesting an imbalance of the interaction between excitement and inhibition. In this study, we aimed to provide new insights into seizure prediction by investigating the dynamics of PIDs with intracranial EEG (iEEG) recordings in drug-resistant epileptic patients who underwent the presurgical evaluation.

Back to Top | Article Outline

MATERIAL AND METHODS

Patient Selection

This study was approved by the institutional review board. Total 334 patients with focal drug-resistant epilepsy who were implanted with intracranial electrodes for presurgical evaluation were recruited at the Epilepsy Surgery Center in Xuanwu Hospital between December 2012 and January 2015. Before invasive EEG recording, all patients underwent noninvasive evaluations including clinical history, neurological examination, neuroimaging, and long-term scalp EEG recording. Electrode grids and strips were implanted subdurally, and the sites of electrode placement were individually determined according to noninvasive data in each patient. The diameter of the subdural electrode was 2 mm, and the center-to-center electrode distance was 10 mm. The EEG signals were sampled at 200 Hz or 1,000 Hz, low-pass filtered at 70 Hz or 300 Hz, and recorded using NIHON KOHDEN EEG-1200C (NIHON KOHDEN Inc., Tokyo, Japan). Among them, 112 patients (33.5%) were diagnosed as having mesial temporal lobe epilepsy (MTLE). The decision to proceed with iEEG recording before surgical resection was made in 21 patients (18.8%). The surgical resection was performed after removal of intracranial electrodes. Postoperatively, 12 of these 21 patients (57.1%) attained seizure free (International League Against Epilepsy class I) in the first year according to the International League Against Epilepsy classification.19 The iEEG recordings of these 12 patients were reviewed by two electroencephalographers (EEGs) to identify the ictal onset and PIDs. Only if the two observers got an agreement, the activities would be identified as PIDs. There were 30 seizures recorded for the 12 patients, however, PIDs were noted in 11 seizures (36.7%) from 7 patients.

Back to Top | Article Outline

Segmentation of PIDs

Preictal discharges were observed preceding 11 seizure events in iEEG recordings of 7 patients (Table 1). Ictal onset was defined as the presence of typical iEEG changes prior to clinical seizure, including low-voltage fast activities, the flattening of background activity, or a slow potential or sustained, rhythmic bursts.20–22 The time and zone of seizure onset were identified with traditional visual inspection by experienced neurophysiologists. The PIDs occurred 6 to 25 seconds before seizure onset in our series. To include all the PIDs, the EEG segment comprising 30 seconds before ictal onset and 5 seconds after ictal onset was extracted from unfiltered iEEG of all implanted electrodes.

TABLE 1

TABLE 1

Back to Top | Article Outline

Data Analysis

All the EEG data were processed with EEGLAB and custom MATLAB (Mathworks, Natick, MA) code. Selected EEG segments of PIDs were band-pass filtered (Butterworth filter) from 1 to 70 Hz. Then, PIDs were detected automatically and the amplitude and timing of each detected PID were measured. The time from the first discharge to the last discharge of PIDs was defined as PID duration, and the interval between two adjacent discharges was calculated. To evaluate the variation tendency of the amplitude and interval of PIDs, linear fitting was performed. For the patients who have more than one seizure with PIDs, linear fitting coefficients of different seizures were averaged.

Furthermore, we computed the normalized autocorrelation coefficients (ACCs) of each channel for selected EEG segments with a sliding window (window, 1 second; overlap, 0.25 seconds; lag, 0.1 seconds). All the channels were classified as PID channels or non-PID channels based on whether PIDs emerged. The 30-second period preceding seizure onset was divided into two stages: before PID stage and PID stage, according to the timing of PID onset. The ACCs of PID channels and non-PID channels during the two stages were individually averaged for each patient.

Back to Top | Article Outline

Statistics

All the numerical results are reported as mean ± SD unless specified. Analysis of samples was performed with either Student t-test (if normal distribution condition was satisfied) or Wilcoxon rank-sum test. A two-sided P value of <0.05 was considered statistically significant. Exact P values are given unless P < 0.001. All statistical tests were conducted with SPSS 17.0 (IBM Corp, Armonk, NY).

Back to Top | Article Outline

RESULTS

Dynamics of PIDs

The duration, number, and rate of PIDs were measured for each patient. For the patients who had more than one seizure with PIDs, these measurements were defined as the average value of multiple seizure events. The mean number of PIDs per patient was 11.90 ± 4.01 (95% confidence interval, 8.20–15.61), and the mean PID duration per patient was 15.39 ± 6.00 seconds (95% confidence interval, 9.84–20.93). The PID rate ranged from 0.60 to 1.40 per second, with the mean of 0.84 ± 0.29 per second (95% confidence interval, 0.57–1.11; Fig. 1G).

FIG. 1

FIG. 1

Next, we measured the amplitudes and intervals of PIDs. As illustrated in the representative patient, the amplitude increased while the interval decreased gradually (Figs. 1C and 1D). The linear fitting regression confirmed the observed changes of amplitude and interval in the representative patient (Figs. 1E and 1F). The similar increasing tendency of PID amplitude over time was shown in most patients (n = 6), and PID interval represented a decreasing trend in all the patients (n = 7; Figs. 1H and 1I).

Back to Top | Article Outline

Autocorrelation Analysis During the Transition to Ictal Activity

Next, we performed a continuous analysis of autocorrelation with a sliding window as mentioned above. As shown in a representative patient, the ACCs in PID channels increased immediately upon PID onset, whereas the ACCs in non-PID channels remained unchanged (Fig. 2E). The distinction between two groups of channels persisted even after ictal initiation.

FIG. 2

FIG. 2

The mean ACCs of PID channels were higher during PID stage than that before PID stage in all the seven patients individually (Fig. 2F). For non-PID channels, the mean ACC showed no obvious change between before PID stage and PID stage. In addition, the average of mean ACCs of PID channels in seven patients was significantly higher during PID stage than that before PID stage (p = 0.04), whereas there was no significant change between the mean ACCs of non-PID channels during the two stages (p = 0.98).

Back to Top | Article Outline

Spatial Distribution of PID Region and Seizure Onset Zone

Finally, we examined whether the region with PIDs matched seizure onset zone (SOZ). The PID channels and channels within SOZ were taken as a whole, and the channels with both PIDs and seizure onset discharges accounted for 80.77% (Fig. 3). The rest 19.23% were divided into two groups: channels in SOZ without PIDs (11.54%) and PID channels beyond SOZ (7.69%).

FIG. 3

FIG. 3

Back to Top | Article Outline

DISCUSSION

The unpredictability of seizure occurrence causes the enhanced risk of injury and sudden death in epilepsy patients. Increasing evidence suggests the existence of a preictal state, which can be detected by different measures.2,23,24 Identifying definite preictal EEG changes is important to provide new opportunities for predicting impending seizures and improving the life quality of patients.

In this study, we recorded the occurrence of PIDs prior to seizure onset in MTLE patients with iEEG recordings, which were observed previously by Huberfeld et al.16 Then, we analyzed the dynamic changes of PIDs toward seizure initiation: (1) the PID amplitude progressively increased, whereas the interval gradually decreased approaching ictal onset, (2) the ACC in PID channels increased simultaneously with PID occurrence, while the ACC in non-PID channels remained unchanged, and (3) the region where PIDs generated from had a big overlap with SOZ. Our data demonstrate that PIDs are premonitory electrophysiological activities of seizure onset with the characteristic changing pattern, revealing the dynamic tempospatial processes that lead to seizures.

Several studies reported that during the preictal period, there were markedly increased glutamatergic excitatory post synaptic potentials, accompanied by the decrease of inhibitory GABAergic inhibitory post synaptic potentials.16,18,25 However, other studies found the excitation of GABAergic inputs preceding ictal initiation, indicating that inhibitory networks contribute to the interictal–ictal transition.6,7,26 Recently, a study on epileptic rats revealed that different hippocampal subregions exhibited different preictal changes.8 These inconsistent findings may be due to the differences in experimental conditions, seizure onset region, and methods.

Previous studies have shown that PIDs are fundamentally different from IIDs. Both glutamatergic and GABAergic networks are involved in IIDs,27–34 and interneurons firing induces IIDs.16,35 Interictal discharges frequency does not increase before the onset of an ictal event in human epileptic patients and in models of focal epilepsy36–38; thus, it is proposed that the after inhibition produced by IIDs protects against the emergence of ictal discharges.39 In contrast, PIDs occur just before ictal events and have different pharmacologic responsiveness from IIDs.16 Preictal discharges seem to depend on principal cells firing rather than interneurons firing and are mediated by glutamatergic activity, which is suggestive of excitatory networks activation.16,40

Based on the classical theory that spike is considered as the excitatory component and the slow wave is considered as the inhibitory component,41 the changes of PID amplitude and interval observed in our study provide support for a prominent role of excitatory effect during the transition to ictal activity. The shape of the autocorrelation function is determined by the frequency composition and the degree of predictability in the signal.42 High ACC reveals a high degree of interdependence between adjacent signals, representing high predictability.43,44 The elevated ACC on PID channels during PID stage indicates that the preictal activities are self-dependent, and PIDs tend to cluster together. Therefore, we hypothesize that during the preictal period, the excitatory effect gradually prevails over inhibitory effect, suggestive of an extension of epileptogenicity until seizure onset.

A series of studies have confirmed that cortical region that generates IIDs (irritative area) is not coincident and is usually larger than SOZ.17,21,39 Compared with IIDs, PIDs have more restricted spatial distribution, and the spatial distribution in situ is in line with SOZ.16 Our data suggest that the region of PIDs largely overlaps with SOZ, which may aid in the localization of SOZ in the presurgical evaluation. However, why PIDs and ictal discharges activate the same region of neurons remains to be unclear. An alternative hypothesis proposed that PIDs are an integral part of ictal event considering their consistent and reproducible emergence at the seizure onset.17 Preictal discharges may actually be the earliest electrographic change of seizure onset, and thus, the sites with the emergence of PIDs should be classified as SOZ. Future investigations on whether PIDs have the same cellular and network mechanism with ictal discharges could help determine these discharges as preictal or ictal.

All seven of the individuals in our study are patients with MTLE and PIDs emerged from the unilateral mesial temporal formation, consistent with previous studies.16,45 A study on the onset of limbic seizures in patients with MTLE reported that PIDs-like events always emerged in seizures derived from the hippocampus but barely in seizures with entorhinal cortex onset.46 Further studies are needed to determine the causal significance of PIDs in seizures originating from different regions.

As stated above, distinct PIDs could be used as a potential biomarker predicting seizure onset. Although the number of patients in this study is limited, the dynamic changes during PID period have been manifested. These characters could help detect preictal period and provide insights into how seizures generate. Given the correlation of spatial distribution between PIDs and ictal onset discharges, identifying PIDs may improve the localization of SOZ. Nevertheless, further large-scale studies are necessary to confirm our conclusions.

Back to Top | Article Outline

CONCLUSIONS

In summary, our study characterized the PIDs based on iEEG recordings from patients with MTLE and demonstrated the dynamic electrophysiological process of PIDs toward seizure onset. Preictal discharges emerge preceding the ictal event, with a progressive increase in the amplitude and decrease in the interval. The autocorrelation coefficients of PID channels are significantly higher during PID stage than those before PID stage. The PID onset region has a correlation with SOZ. These findings deepen our understanding of seizure onset and provide help for seizure prediction and the localization of epileptogenic zone.

Back to Top | Article Outline

ACKNOWLEDGMENTS

This work was supported by National Natural Science Foundation of China (No. 81271447 and 81571271) and Brain Science & Brain Program of Beijing Municipal Commission of Science and Technology (Z161100002616001).

Back to Top | Article Outline

REFERENCES

1. Baumgartner C, Serles W, Leutmezer F, et al. Preictal SPECT in temporal lobe epilepsy: regional cerebral blood flow is increased prior to electroencephalography-seizure onset. J Nucl Med 1998;39:978–982.
2. Zhang T, Zhou J, Jiang R, Yang H, Carney PR, Jiang H. Pre-seizure state identified by diffuse optical tomography. Sci Rep 2014;4:3798.
3. Kerem DH, Geva AB. Forecasting epilepsy from the heart rate signal. Med Biol Eng Comput 2005;43:230–239.
4. Delamont RS, Julu PO, Jamal GA. Changes in a measure of cardiac vagal activity before and after epileptic seizures. Epilepsy Res 1999;35:87–94.
5. Federico P, Abbott DF, Briellmann RS, Harvey AS, Jackson GD. Functional MRI of the pre-ictal state. Brain 2005;128:1811–1817.
6. Bragin A, Azizyan A, Almajano J, Engel J Jr. The cause of the imbalance in the neuronal network leading to seizure activity can be predicted by the electrographic pattern of the seizure onset. J Neurosci 2009;29:3660–3671.
7. Gnatkovsky V, Librizzi L, Trombin F, de Curtis M. Fast activity at seizure onset is mediated by inhibitory circuits in the entorhinal cortex in vitro. Ann Neurol 2008;64:674–686.
8. Fujita S, Toyoda I, Thamattoor AK, Buckmaster PS. Preictal activity of subicular, CA1, and dentate gyrus principal neurons in the dorsal hippocampus before spontaneous seizures in a rat model of temporal lobe epilepsy. J Neurosci 2014;34:16671–16687.
9. Medvedev AV, Murro AM, Meador KJ. Abnormal interictal gamma activity may manifest a seizure onset zone in temporal lobe epilepsy. Int J Neural Syst 2011;21:103–114.
10. Alvarado-Rojas C, Valderrama M, Fouad-Ahmed A, et al. Slow modulations of high-frequency activity (40–140-Hz) discriminate preictal changes in human focal epilepsy. Sci Rep 2014;4:4545.
11. Worrell GA, Parish L, Cranstoun SD, Jonas R, Baltuch G, Litt B. High-frequency oscillations and seizure generation in neocortical epilepsy. Brain 2004;127:1496–1506.
12. Sato Y, Doesburg SM, Wong SM, Boelman C, Ochi A, Otsubo H. Preictal surrender of post-spike slow waves to spike-related high-frequency oscillations (80-200 Hz) is associated with seizure initiation. Epilepsia 2014;55:1399–1405.
13. Sato Y, Doesburg SM, Wong SM, Ochi A, Otsubo H. Dynamic preictal relations in FCD type II: potential for early seizure detection in focal epilepsy. Epilepsy Res 2015;110:26–31.
14. Sato Y, Doesburg SM, Wong SM, et al. Dynamic changes of interictal post-spike slow waves toward seizure onset in focal cortical dysplasia type II. Clin Neurophysiol 2015;126:1670–1676.
15. Proddutur A, Santhakumar V. Marching towards a seizure: spatio-temporal evolution of preictal activity. Epilepsy Curr 2015;15:267–268.
16. Huberfeld G, Menendez de la Prida L, Pallud J, et al. Glutamatergic pre-ictal discharges emerge at the transition to seizure in human epilepsy. Nat Neurosci 2011;14:627–634.
17. Fisher RS, Scharfman HE, deCurtis M. How can we identify ictal and interictal abnormal activity? Adv Exp Med Biol 2014;813:3–23.
18. Alvarado-Rojas C, Huberfeld G, Baulac M, et al. Different mechanisms of ripple-like oscillations in the human epileptic subiculum. Ann Neurol 2015;77:281–290.
19. Wieser HG, Blume WT, Fish D, et al. ILAE Commission Report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001;42:282–286.
20. Gnatkovsky V, de Curtis M, Pastori C, et al. Biomarkers of epileptogenic zone defined by quantified stereo-EEG analysis. Epilepsia 2014;55:296–305.
21. de Curtis M, Gnatkovsky V. Reevaluating the mechanisms of focal ictogenesis: the role of low-voltage fast activity. Epilepsia 2009;50:2514–2525.
22. Wendling F, Bartolomei F, Bellanger JJ, Bourien J, Chauvel P. Epileptic fast intracerebral EEG activity: evidence for spatial decorrelation at seizure onset. Brain 2003;126:1449–1459.
23. Mormann F, Kreuz T, Rieke C, et al. On the predictability of epileptic seizures. Clin Neurophysiol 2005;116:569–587.
24. Stacey W, Le Van Quyen M, Mormann F, Schulze-Bonhage A. What is the present-day EEG evidence for a preictal state? Epilepsy Res 2011;97:243–251.
25. Zhang ZJ, Koifman J, Shin DS, et al. Transition to seizure: ictal discharge is preceded by exhausted presynaptic GABA release in the hippocampal CA3 region. J Neurosci 2012;32:2499–2512.
26. Lasztoczi B, Nyitrai G, Heja L, Kardos J. Synchronization of GABAergic inputs to CA3 pyramidal cells precedes seizure-like event onset in juvenile rat hippocampal slices. J Neurophysiol 2009;102:2538–2553.
27. Uva L, Avoli M, de Curtis M. Synchronous GABA-receptor-dependent potentials in limbic areas of the in-vitro isolated adult Guinea pig brain. Eur J Neurosci 2009;29:911–920.
28. Uusisaari M, Smirnov S, Voipio J, Kaila K. Spontaneous epileptiform activity mediated by GABA(A) receptors and gap junctions in the rat hippocampal slice following long-term exposure to GABA(B) antagonists. Neuropharmacology 2002;43:563–572.
29. Huberfeld G, Wittner L, Clemenceau S, et al. Perturbed chloride homeostasis and GABAergic signaling in human temporal lobe epilepsy. J Neurosci 2007;27:9866–9873.
30. Avoli M, D'Antuono M, Louvel J, et al. Network and pharmacological mechanisms leading to epileptiform synchronization in the limbic system in vitro. Prog Neurobiol 2002;68:167–207.
31. Wittner L, Huberfeld G, Clemenceau S, et al. The epileptic human hippocampal cornu ammonis 2 region generates spontaneous interictal-like activity in vitro. Brain 2009;132:3032–3046.
32. de Curtis M, Radici C, Forti M. Cellular mechanisms underlying spontaneous interictal spikes in an acute model of focal cortical epileptogenesis. Neuroscience 1999;88:107–117.
33. Uva L, Breschi GL, Gnatkovsky V, Taverna S, de Curtis M. Synchronous inhibitory potentials precede seizure-like events in acute models of focal limbic seizures. J Neurosci 2015;35:3048–3055.
34. Cohen I, Navarro V, Clemenceau S, Baulac M, Miles R. On the origin of interictal activity in human temporal lobe epilepsy in vitro. Science 2002;298:1418–1421.
35. Ziburkus J, Cressman JR, Barreto E, Schiff SJ. Interneuron and pyramidal cell interplay during in vitro seizure-like events. J Neurophysiol 2006;95:3948–3954.
36. Gotman J. Relationships between interictal spiking and seizures: human and experimental evidence. Can J Neurol Sci 1991;18:573–576.
37. Katz A, Marks DA, McCarthy G, Spencer SS. Does interictal spiking change prior to seizures? Electroencephalogr Clin Neurophysiol 1991;79:153–156.
38. Leung LW. Spontaneous hippocampal interictal spikes following local kindling: time-course of change and relation to behavioral seizures. Brain Res 1990:513:308–314.
39. de Curtis M, Avanzini G. Interictal spikes in focal epileptogenesis. Prog Neurobiol 2001;63:541–567.
40. Stafstrom CE. Distinct mechanisms mediate interictal and pre-ictal discharges in human temporal lobe epilepsy. Epilepsy Curr 2011;11:200–202.
41. Blumenfeld H. Cellular and network mechanisms of spike-wave seizures. Epilepsia 2005;46(suppl 9):21–33.
42. Broersen PMT. Automatic Autocorrelation and Spectral Analysis. London, United Kingdom: Springer, 2006.
43. Houle TT, Penzien DB, Rains JC. Time-series features of headache: individual distributions, patterns, and predictability of pain. Headache 2005;45:445–458.
44. Wu Z, Xia X, Wang J. EEG autocorrelation analysis of neuronal population at criticality. Appl Mech Mater 2014;482:363–366.
45. Wasade VS, Gaddam S, Burdette DE, et al. Intracranial electrographic analysis of preictal spiking and ictal onset ill uni- and bitemporal epilepsy. Epileptic Disord 2015;17:156–164.
46. Bartolomei F, Wendling F, Regis J, Gavaret M, Guye M, Chauvel P. Pre-ictal synchronicity in limbic networks of mesial temporal lobe epilepsy. Epilepsy Res 2004;61:89–104.
Keywords:

Seizure prediction; Seizure onset zone; Intracranial EEG; Autocorrelation coefficient; Preictal period

© 2018 by the American Clinical Neurophysiology Society