Brain & Development xxx (2014) xxx–xxx www.elsevier.com/locate/braindev
Original article
Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy Mahmoud Mohammadi a,c, Tohru Okanishi a, Kazuo Okanari a, Shiro Baba a, Hironobu Sumiyoshi a, Satoru Sakuma a, Ayako Ochi a, Elysa Widjaja b, Cristina Y. Go a, O. Carter Snead III a, Hiroshi Otsubo a,⇑ a
Division of Neurology, The Hospital for Sick Children, Toronto, ON, Canada Diagnostic Imaging, The Hospital for Sick Children, Toronto, ON, Canada c Department of Pediatric Neurology, Children’s Medical Center, Tehran, Iran b
Received 14 August 2013; received in revised form 12 March 2014; accepted 13 March 2014
Abstract Background: Generalized paroxysmal fast activity (GPFA) consists of burst of generalized rhythmic discharges; 100–200 lV; 1–9 s; 8–26 Hz; with frontal predominance; appearing during NREM sleep. GPFA was originally described as an electrographic feature of Lennox–Gastaut Syndrome (LGS). We analyzed GPFA on scalp video EEG (VEEG) in children to evaluate that GPFA presents in patients with intractable localization-related epilepsy. Methods: We collected cases with GPFA with intractable localization-related epilepsy who underwent scalp VEEG, MRI, and magnetoencephalography (MEG) prior to intracranial video EEG (IVEEG) and surgical resection. We collected 50 epochs of GPFA per patient during the first night during scalp VEEG. We analyzed amplitude, duration and frequency of GPFA over the bilateral frontal region between surgical resection side with grid placement and non-resection side. Results: We identified 14 (14%) patients with GPFA on scalp VEEG. The mean amplitude ranged from 145 to 589 lV (mean 293 lV). The mean duration ranged from 1.18 to 2.31 s (mean 1.6 s). The mean frequencies ranged from 9.3 to 14.7 Hz (mean 11.1 Hz). The amplitude (307 ± 156 lV) and duration (1.62 ± 0.8 s) of GPFAs in all the patients over the resection side were significantly higher than those (279 ± 141 lV, 1.58 ± 0.8 s) of the non-resection side (p < 0.001). All nine patients who showed significant duration differences between two hemispheres (p < 0.05) had longer duration of GPFA over the resection side. Eight of 12 patients who showed significant amplitude differences between two hemispheres (p < 0.05) had higher amplitude of GPFA over the resection side. Four of six patients who showed significant frequency differences between two hemispheres (p < 0.05) had higher frequency of GPFA over the resection side. Nine (64%) patients became seizure free after surgical resection including multilobar resections in eight patients. Conclusions: GPFA can exist in localization-related epilepsy with secondary bilateral synchrony. Although EEG shows GPFA on scalp VEEG, the precise localization of the epileptogenic zone using IVEEG could achieve the successful surgical resection. Ó 2014 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved.
Keywords: Generalized paroxysmal fast activity (GPFA); Localization-related epilepsy; Intracranial video EEG; Children; Cortico-thalamic epileptic network; Asymmetry; Amplitude; Duration; Epilepsy surgery; Epileptogenic hemisphere
⇑ Corresponding author. Address: Division of Neurology, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada. Tel.: +1 416 813 6660; fax: +1 416 813 6334. E-mail address: [email protected] (H. Otsubo).
http://dx.doi.org/10.1016/j.braindev.2014.03.006 0387-7604/Ó 2014 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
2
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
1. Introduction Generalized paroxysmal fast activity (GPFA) is a pattern of; (1), 100–200 lV; (2), 1–9 s; (3), 8–26 Hz; (4), bursts of generalized rhythmic discharges with frontal predominance; (5), appearing most frequently during NREM sleep [1]. Jasper and Kershman, first recognized these runs of rapid spikes in sleep as “paroxysmal fast rhythm” [2]. Gastaut detected paroxysmal fast rhythm with tonic seizures in patients with Lennox–Gastaut syndrome (LGS) [3]. Brenner and Atkinson described EEG findings in patients with multiple type of seizures and mental retardation as “generalized paroxysmal fast activity” [4]. Other terms such as grand mal discharges, fast paroxysmal rhythms, rhythmic spikes, and runs of rapid spikes were also used to describe GPFA [1]. Since GPFA was associated with tonic seizures, clonic movements, and subtle jerks in LGS, it has been one of the criteria used in diagnosing LGS [5] as well as a late variant of LGS [6]. GPFA, however, also can be seen in patients with frontal and temporal lobe epilepsy [1]. The paroxysmal fast activities (PFA) of LGS are widespread in distribution and bilaterally synchronous over both hemispheres, with highest amplitude over the frontal and central head regions. Shifting asymmetries of PFA are common, but rarely the pattern may show persistent amplitude asymmetry, unilateral or even focal [7]. EEG and clinical differentiation between LGS and focal epilepsy with secondary bilateral synchrony (SBS) may be difficult or even impossible [8]. Halasz hypothesized that the mechanism of GPFA is the permanent or momentary breakdown of the GABAergic inhibitory process operating during the generation of spike-wave discharge [9]. The usual spike and wave complex is thought to consists of a balance of excitatory (spike components) and inhibitory (slow components) components [10]. Therefore, the lack of slow waves during GPFA suggests an absence of inhibitory processes. As well, Halasz et al., reported GPFA in three patients with non-malignant seizures to prove that GPFA can be a possible electrographic variant in certain generalized epilepsies showing atypical features [11]. These authors concluded that GPFA could be seen in patients with treatable epilepsy with a better seizure outcome than that seen in patients with LGS. Sueda et al. differentiated the localized origin and propagation of PFA in patients with epileptic spasms, in contrast to bilaterally generated PFA in patients with LGS, using time–frequency analyses of magnetoencephalography (MEG) [12]. To our knowledge, there have been no reports on GPFA in patients undergoing epilepsy surgery, prior to their receiving the surgery. The characteristics of GPFA and its correlation with the epileptogenic zone as defined by intracranial video EEG (IVEEG) and seizure outcome remain unclear. We analyzed GPFA on
scalp recordings performed in pediatric patients with intractable localization-related epilepsies, who subsequently underwent IVEEG for epilepsy surgery. We hypothesized that asymmetry of GPFA would be present in patients with intractable localization-related epilepsy secondary to the resectable epileptogenic zone, and further that there are characteristic features of GPFA in the intractable localization-related epilepsy. 2. Patients and methods We selected patients with GPFA among the patients who underwent intracranial video EEG between 2004 and 2012 at the Hospital for Sick Children. All patients were admitted to the epilepsy monitoring unit (EMU) for scalp video EEG (VEEG) between 1 and 5 days. In addition, MRI, neuropsychological assessment and magnetoencephalography (MEG) were performed. Thereafter, because the data were concordant for localization of an epileptogenic zone, the patients underwent intracranial video EEG (IVEEG) to delineate the epileptogenic zone pursuant to resective surgery. Parents or guardians gave informed consent for all procedures. This study received prior approval from the Research Ethics Board at The Hospital for Sick Children. We recorded scalp video-EEG (HARMONIE 5.4, Stellate, Montreal, PQ, Canada) using 19 or 25 scalp electrodes placed according to the International 10–20 system (subtemporal electrodes; F9, F10, T9, T10, P9 and P10 or midline electrodes; F1, F2, C1, C2, P1 and P2). A single reference was placed at Oz, Pz’ (located 1 cm behind Pz), or FCz, whichever was the most inactive electrode. Sampling rate was 200 or 500 Hz. We selected the first 50 GPFAs for each patient during NREM sleep, at the first and/or second night of admission. We followed three criteria to select GPFAs (1) amplitude of 100–200 lV; (2) duration of 1–9 s; (3) frequency of 8–26 Hz, for bursts of generalized rhythmic discharges with frontal predominance [1]. We applied a band pass filter of 5–70 Hz without notch filter for the selection and analysis of GPFA. We used a program “Signal statistics” (HARMONIE 5.4, Stellate, Montreal, PQ, Canada) to analyze the amplitude, duration and frequency. We analyzed the amplitude of GPFAs using each of electrodes on referential montage EEG (ex, Fp1-Ref, Fp2-Ref, F3-Ref, F4-Ref). Single Oz electrode was used for the reference. If a patient has the lesions and/or epileptic zone around the occipital lobe, Pz electrode is used for the reference. During the period of GPFA, we measured the amplitude difference between minimum and maximum amplitudes of each GPFA. We compared the amplitude for paired electrodes, Fp1-Ref v.s. Fp2-Ref or F3-Ref v.s. F4-Ref. First, we found the electrode with the maximum amplitude among four electrodes. Second, we compared the paired electrodes, one of which had maximum
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
amplitude. For example, if the maximum amplitude was recorded on F3-Ref, F3-Ref and F4-Ref were analyzed to evaluate hemispheric difference. We used one pair of electrodes, Fp1-F3 and Fp2-F4 for the analysis of the duration and frequency. We applied a paired t-test with Bonferroni correction to compare amplitudes, durations and frequency of GPFA between resection and non-resection sides. Patients underwent 1.5 or 3 tesla (T) MRI. We used a whole-head gradiometer-based Omega system (151 channels, VSM MedTech Ltd., Coquitlam, BC, Canada) for the MEG. Our procedures and analysis of MEG for surgical evaluation were described somewhere [13]. We placed the IVEEG electrodes based on seizure semiology, MRI, MEG, and neuropsychological data to cover eloquent cortex and epileptic zones [14]. We determined the resection area based on seizure onset zone and active interictal zone, location of eloquent cortex, the location of MRI lesion and MEG spike dipoles [15].
3
3. Results Table 1 describes the clinical profiles and features of GPFA. We found 28 patients with interictal generalized spike and waves in 103 patients who underwent intracranial video EEG. We found 14 patients with GPFA among the 28 patients. Eight patients (57%) were female, and six were male (43%). Age of seizure onset ranged between 1 month and 12 years with a mean age of 4.2 years. The scalp VEEG for analysis of GPFA was performed between 3 and 17 years old with a mean age of 11.4 years. Following the scalp VEEG, the children underwent IVEEG between 5 and 18 years of age with a mean age of 12.9 years. Nine (64%) of the 14 patients who had GPFA documented prior to IVEEG and respective surgery presented with partial seizures, including two (14%) patients with secondary generalization. Three (21%) patients had generalized tonic-clonic seizures without
Table 1 Amplitude, duration and frequency of GPFA.
SVEEG: scapl video-EEG; PS: partial seizure; GTCS: generalized tonic clonic seizure; ES: epileptic spasms; PS2G: partial seizure with secondary generalization; *: significantly higher with p < 0.05.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
4
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
focal signs. Seven (50%) patients presented with epileptic spasms. Four patients had symmetric epileptic spasms. Eleven patients (79%) had normal neurologic exam. One (7%) patient had decreased visual field at nasal side but otherwise normal. One patient demonstrated bilateral dysmetria and hand tremor. The other one demonstrated decreased peripheral vision on both sides and mild weakness of right hand fingers.
Six (43%) patients were developmentally delayed. Four patients (29%) were developmentally normal. The other four patients showed attention deficit, hyperactivity disorders, learning difficulties and academic achievement problems. We collected 50 GPFA in each of the 14 patients (Fig. 1). The mean amplitude of GPFA ranged from 145 to 589 lV with a mean of 293 lV. The mean
A FP1-F7 F7-T3 T3-T5 T5-O1 FP2-F8
*
F8-T4 T4-T6 T6-O2 FP1-F3 F3-C3 C3-P3 P3-O1 FP2-F4
*
F4-C4 C4-P4 P4-O2
100 uV 1 sec
FZ-CZ CZ-PZ
B Fp1-REF
F3-REF
Fp2-REF 200 uV 1 sec F4-REF Fig. 1. (A) AP-bipolar montage EEG (500 Hz sampling rate; low frequency filter, 5 Hz; high frequency filter 70 Hz; notch filter on) showing generalized paroxysmal fast activities (GPFA) in case #9 with generalized tonic clonic seizures. GPFA appears with bilateral frontal predominance. The fast activity starts slightly earlier on the right side (*Fp2-F8, Fp2-F4) than the left side. The duration and frequency of GPFA showed 15.1 Hz, 2.61 s at Fp1-F3 and 15.9 Hz, 2.68 s at Fp2-F4. (B) Selected referential montage channels (Oz reference; low frequency filter, 5 Hz; high frequency filter, 70 Hz; notch filter on) of Fp1-Ref, Fp2-Ref, F3-Ref, F4-Ref from GPFA of (A). Fp2-Ref channel shows GPFA with maximum amplitude among four electrodes. Therefore we compared amplitude of Fp2-Ref (398 lV) with Fp1-Ref (292 lV) to evaluate hemispheric difference. The asymmetric amplitude correlated with the epileptogenic hemisphere. The patient underwent resective surgery over the right frontal and temporal regions. The pathology was microdysgenesis. She has been seizure free for four years with medications.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
duration of GPFA ranged from 1.18 to 2.31 s with a mean of 1.6 s. The mean frequency of GPFA ranged from 9.3 to 14.7 Hz with a mean of 11.1 Hz. Table 2 describes the MRI, MEG, resective areas, histopathological findings and seizure outcomes. More than 50% of MEG spike dipoles were located in one hemisphere in all 14 patients. Thirteen patients showed 83–100% of MEG spike dipoles in one hemisphere. We placed subdural grid over eight (57%) left hemispheres and six (43%) right hemispheres. The seizure onset zone and active interictal zone were localized in the multiple lobes in 12 (86%), consisting of three lobes in five (36%) patients and two lobes in three (21%) patients. Ten patients of the 12 patients with multiple lobe epileptic foci involved frontal lobe. Only two (14%) patients had seizure onset and active interictal zone involving the frontal lobe alone. Thirteen of 14 patients underwent resective surgery following IVEEG. In one patient (case 2), MRI showed the abnormal signal in the left frontal lobe in spite of 94% of MEG spike dipoles lateralized in the right hemisphere. The patient initially underwent IVEEG using bilateral subdural strip electrodes over the bilateral frontal lobes and a depth electrode into the left frontal MRI lesion to determine the seizure onset. After we confirmed the right hemispheric seizure onset, the patient had a second surgery to place the subdural grid over the right hemisphere. The seizure onset zone was localized in the right primary motor cortex on IVEEG. The parents
5
refused the proposed resection, but consented to have a biopsy performed in the right frontal lobe instead. Surgical pathology revealed focal cortical dysplasia in four (29%) patients, microdysgenesis in four (29%), subpial gliosis in two (14%), tuberous sclerosis complex, filaminopathy, oligodendrogliosis, and normal in one each. One patient presented both dysembryoplastic neruoepitherial tumor and focal cortical dysplasia. Eleven patients had more than one year follow-up after the resective surgery. Six patients were seizure free with anti-epileptic medications (Engel classification I). Two patients were classified as Engel II, and three patients were Engel III. Two patients had less than one year follow-up. One patient did not undergo resective surgery. We compared the three characteristic features of GPFA: amplitude, duration and frequency between the side of resection after grid recording and the nonresection side. The amplitude of GPFA in all the patients over the resection side (307 ± 156 lV) was significantly higher than that over the non-resection side (279 ± 141 lV) (p < 0.001). The duration of GPFA in all the patients over the resection side (1.62 ± 0.8 s) was significantly higher than that over the non-resection side (1.58 ± 0.8 s) (p < 0.001). The frequency of GPFA in all the patients was no significant difference between the resection side (11.06 ± 2.1 Hz) and the non-resection side (11.08 ± 2.2 Hz).
Table 2 Clinical profiles, MRI, MEG, resective surgery and surgical outcomes.
PS: partial seizure; GTCS: generalized tonic clonic seizure; ES: epileptic spasms; PS2G: partial seizure with secondary generalization; F: frontal lobe; T: Temporal lobe; P: Parietal lobe; O: occipital lobe; MT: mesial temporal; FP: frontoparietal; DNET: dysembryoplastic neuroepithelial tumor; FCD: focal cortical dysplasia. a The epileptogenic zone overlapped the eloqunet cortex. Only biopsy was performed in the right frontal lobe.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
6
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
All nine patients who showed significant duration differences between two sides (p < 0.05) had the longer duration of GPFA over the resection side. Eight of 12 patients who showed significant amplitude differences between two sides (p < 0.05) had the higher amplitude of GPFA over the resection side. Four of six patients who showed significant frequency differences between two sides (p < 0.05) had the higher frequency of GPFA over the resection side. We compared three characteristic features of GPFA between resection and non-resection sides among ten patients (Table 1. Number of patients 1,3,5,6,8,9,10,12,13,14) with good seizure outcome (Engel I and II). Seven (70%) of them showed higher amplitude over the resection side than the non-resection side including six patients with a statistical significance. In three patients (Table 1. Number of patients 4, 7, 11) with relatively poor seizure outcome (Engel III), one of them showed significant higher amplitude of GPFA in the non-resection side than in the resection side. 4. Discussion GPFA is considered a sign of cortico-thalamic discharges to produce generalized, frontal predominance, rhythmic, and lack of complex of spike and slow wave. Thus, GPFA is not only seen in LGS, but can be seen in localization-related epilepsy with secondary generalization and/or secondary bilateral synchrony [1,11]. In our series, three patients presented with generalized tonic clonic seizures and two patients partial seizures with secondary generalization, and seven patients epileptic spasms. They had focal onset proven by IVEEG. However, their seizure onset and interictal zones were widespread over the multiple lobes. The previous reports of GPFA were frequently recognized in LGS, furthermore GPFA might present brief tonic fits or clonic movements not infrequently [5]. When GPFA became longer, tonic seizures were more associated than shorter GPFA [1]. GPFA has a potential of ictal symptoms or relating with ictal symptomatic network in a subset of intractable epilepsy patients. Children with LGS as known as epileptic encephalopathy involving both cortex and subcortical structures, thus GPFA is most frequently captured and seems to be a sign of LGS. GPFA consist of high amplitude alpha range fast activities running up to six seconds without isolated slow waves. When GPFA last longer, they can produce clinical seizures; however, even brief GPFA can produce subtle clinical seizures through the cortico-subcortical network. We compared GPFA at bilateral frontal electrodes, because GPFA appeared at bilateral frontal electrodes, GPFA appeared predominantly over the bilateral frontal regions regardless of the localization of either seizure onset zones or epileptogenic zones. GPFA can be a part of secondary bilateral synchrony
or generalized discharges provoked by the epileptic cortico-thalamic network in the localization-related epilepsy. Nine (62%) patients became seizure free after surgical resection over the multiple lobe resection. The precise lateralization of epileptogenic hemisphere, IVEEG to localize the epileptogenic zone and surgical treatment can be a reasonable option to control intractable epilepsy in children with GPFA. Lennox–Gastaut syndrome has been reported surgically treated with successful seizure outcome [16,17]. The most of seizure free patients following focal resection for LGS had intractable localization-related epilepsy. Their results indicated secondary bilateral synchrony to project generalized epileptiform discharges. Wyllie et al. described epilepsy surgery was successful for selected children and adolescents with a congenital or early-acquired brain lesion, despite abundant generalized or bilateral epileptiform discharges on EEG [18]. Our series showed that the patients with GPFA, yet presenting a sign of lateralization for the epileptogenic zone on the scalp VEEG, precise localization of the epileptogenic zone using various modalities using MRI, MEG, PET and IVEEG would be highly recommended. The detection of GPFA must not discourage from considering pre-surgical evaluation. If the GPFA can show lateralized signs of the amplitude and duration asymmetry, the other techniques for definition of the epileptogenic zone are highly recommended. Then surgical treatment can achieve the successful seizure outcome in patients with intractable localization-related epilepsy and GPFA. In our series, the amplitude and duration of GPFA over the resection side were significantly higher than that over the non-resection side. The frequency of GPFA was no significant difference between the resection side and the non-resection side. Six of 10 good-seizure-outcome patients showed significantly higher amplitude of GPFA over the resection side than the non-resection side. Eight of 10 good-seizure-outcome patients showed significantly longer duration of GPFA over the resection side than the non-resection side. In three patients with poor seizure outcome, the GPFA amplitude was significantly higher in the non-resection side than the resection side for one patients. The amplitude and duration of GPFA in the localization-related epilepsy with SBS and generalized seizures may contribute to the excitatory cortex in epileptogenic hemisphere. The frequency of GPFA, however, had no differences between two hemispheres. The thalamo-cortical network may conduct the identical rhythm of both hemispheres. The predominant high amplitude of GPFA in the contralateral hemisphere may indicate potential independent epileptogenesis in the non-resection side. Further collection of patients with GPFA will be performed to analyze the correlation GPFA with the epileptogenic zone/hemisphere and epileptic disorders.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
5. Conclusion GPFA can exist in intractable localization-related epilepsy with SBS and generalized seizures. GPFA is not always a sign of malignant epileptic encephalopathy represented by Lennox–Gastaut syndrome. In a subset of the intractable localization-related epilepsy with SBS and generalized seizures, the amplitude and duration of GPFA may contribute to the excitatory cortex in epileptogenic hemisphere, furthermore, the bilateral identical rhythm may be conducted by thalamo-cortical network. Although EEG shows GPFA on scalp VEEG, the precise localization of the epileptogenic zone using IVEEG could achieve the successful surgical resection for a subset of patients with GPFA.
Acknowledgements We thank Dr. Mahdi Razzaghi from Bloomsburg University (PA, USA), for his invaluable assistance in statistical tests selection. Drs. Tohru Okanishi and Satoru Sakuma were supported by Ontario Brain Institute (OBI). References [1] Laoprasert P. Atlas of pediatric EEG. New York: McGraw Hill; 2011. [2] Jasper H, Kershman J. Electroencephalograph classification of the epilepsies. AMA Arch Neurol Psychiatry 1941;45: 903–43. [3] Gastaut H, Roger J, Ouahchi S, Timsit M, Broughton R. An electro-clinical study of generalized epileptic seizures of tonic expression. Epilepsia 1963;4:15–44. [4] Brenner RP, Atkinson R. Generalized paroxysmal fast activity: electroencephalographic and clinical features. Ann Neurol 1982; 11:386–90.
7
[5] Beaumanoir A. The Lennox–Gastaut syndrome. In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, editors. Epileptic syndromes in infancy, childhood and adolescence. London: John Libbey Eurotext; 1985. p. 89–99. [6] Bauer G, Aichner F, Saltuari L. Epilepsies with diffuse slow spikes and waves of late onset. Eur Neurol 1983;22:344–50. [7] Markand ON. Lennox–Gastaut syndrome (childhood epileptic encephalopathy). J Clin Neurophysiol 2003;20:426–41. [8] Niedermeyer E. Lennox–Gastaut syndrome. Clinical description and diagnosis. Adv Exp Med Biol 2002;497:61–75. [9] Hala´sz P. Runs of rapid spikes in sleep – a characteristic EEG expression of generalized malignant epileptic encephalopathies. A conceptual review with new pharmacological data. In: Degen R, Rodin EA, editors. Epilepsy, sleep, and sleep deprivation. New York: Elsevier; 1991. p. 49–78. [10] Lothman EW. The neurobiology of epileptiform discharges. Am J EEG Technol 1993;33:93–112. [11] Hala´sz P, Janszky J, Barcs G, Szucs A. Generalised paroxysmal fast activity (GPFA) is not always a sign of malignant epileptic encephalopathy. Seizure 2004;13:270–6. [12] Sueda K, Takeuchi F, Shiraishi H, Nakane S, Sakurai K, Yagyu K, et al. Magnetoencephalographic analysis of paroxysmal fast activity in patients with epileptic spasms. Epilepsy Res 2013;104: 68–77. [13] Ochi A, Go CY, Otsubo H. Clinical MEG analyses for children with intractable epilepsy. In: Pang EW, editor. Magnetoencephalography. Tech North, America: Elseiver; 2011 ch. 9. [14] Snead 3rd OC. Surgical treatment of medically refractory epilepsy in childhood. Brain Dev 2001;23:199–207. [15] 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. [16] Jung YJ, Kang HC, Choi KO, Lee JS, Kim DS, Cho JH, et al. Localization of ictal onset zones in Lennox–Gastaut syndrome using directional connectivity analysis of intracranial electroencephalography. Seizure 2011;20:449–57. [17] Lee YJ, Kang HC, Lee JS, Kim SH, Kim DS, Shim KW, et al. Resective pediatric epilepsy surgery in Lennox–Gastaut syndrome. Pediatrics 2010;125:e58–66. [18] Wyllie E, Lachhwani DK, Gupta A, Chirla A, Cosmo G, Worley S, et al. Successful surgery for epilepsy due to early brain lesions despite generalized EEG findings. Neurology 2007;69:389–97.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
Original article
Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy Mahmoud Mohammadi a,c, Tohru Okanishi a, Kazuo Okanari a, Shiro Baba a, Hironobu Sumiyoshi a, Satoru Sakuma a, Ayako Ochi a, Elysa Widjaja b, Cristina Y. Go a, O. Carter Snead III a, Hiroshi Otsubo a,⇑ a
Division of Neurology, The Hospital for Sick Children, Toronto, ON, Canada Diagnostic Imaging, The Hospital for Sick Children, Toronto, ON, Canada c Department of Pediatric Neurology, Children’s Medical Center, Tehran, Iran b
Received 14 August 2013; received in revised form 12 March 2014; accepted 13 March 2014
Abstract Background: Generalized paroxysmal fast activity (GPFA) consists of burst of generalized rhythmic discharges; 100–200 lV; 1–9 s; 8–26 Hz; with frontal predominance; appearing during NREM sleep. GPFA was originally described as an electrographic feature of Lennox–Gastaut Syndrome (LGS). We analyzed GPFA on scalp video EEG (VEEG) in children to evaluate that GPFA presents in patients with intractable localization-related epilepsy. Methods: We collected cases with GPFA with intractable localization-related epilepsy who underwent scalp VEEG, MRI, and magnetoencephalography (MEG) prior to intracranial video EEG (IVEEG) and surgical resection. We collected 50 epochs of GPFA per patient during the first night during scalp VEEG. We analyzed amplitude, duration and frequency of GPFA over the bilateral frontal region between surgical resection side with grid placement and non-resection side. Results: We identified 14 (14%) patients with GPFA on scalp VEEG. The mean amplitude ranged from 145 to 589 lV (mean 293 lV). The mean duration ranged from 1.18 to 2.31 s (mean 1.6 s). The mean frequencies ranged from 9.3 to 14.7 Hz (mean 11.1 Hz). The amplitude (307 ± 156 lV) and duration (1.62 ± 0.8 s) of GPFAs in all the patients over the resection side were significantly higher than those (279 ± 141 lV, 1.58 ± 0.8 s) of the non-resection side (p < 0.001). All nine patients who showed significant duration differences between two hemispheres (p < 0.05) had longer duration of GPFA over the resection side. Eight of 12 patients who showed significant amplitude differences between two hemispheres (p < 0.05) had higher amplitude of GPFA over the resection side. Four of six patients who showed significant frequency differences between two hemispheres (p < 0.05) had higher frequency of GPFA over the resection side. Nine (64%) patients became seizure free after surgical resection including multilobar resections in eight patients. Conclusions: GPFA can exist in localization-related epilepsy with secondary bilateral synchrony. Although EEG shows GPFA on scalp VEEG, the precise localization of the epileptogenic zone using IVEEG could achieve the successful surgical resection. Ó 2014 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved.
Keywords: Generalized paroxysmal fast activity (GPFA); Localization-related epilepsy; Intracranial video EEG; Children; Cortico-thalamic epileptic network; Asymmetry; Amplitude; Duration; Epilepsy surgery; Epileptogenic hemisphere
⇑ Corresponding author. Address: Division of Neurology, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada. Tel.: +1 416 813 6660; fax: +1 416 813 6334. E-mail address: [email protected] (H. Otsubo).
http://dx.doi.org/10.1016/j.braindev.2014.03.006 0387-7604/Ó 2014 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
2
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
1. Introduction Generalized paroxysmal fast activity (GPFA) is a pattern of; (1), 100–200 lV; (2), 1–9 s; (3), 8–26 Hz; (4), bursts of generalized rhythmic discharges with frontal predominance; (5), appearing most frequently during NREM sleep [1]. Jasper and Kershman, first recognized these runs of rapid spikes in sleep as “paroxysmal fast rhythm” [2]. Gastaut detected paroxysmal fast rhythm with tonic seizures in patients with Lennox–Gastaut syndrome (LGS) [3]. Brenner and Atkinson described EEG findings in patients with multiple type of seizures and mental retardation as “generalized paroxysmal fast activity” [4]. Other terms such as grand mal discharges, fast paroxysmal rhythms, rhythmic spikes, and runs of rapid spikes were also used to describe GPFA [1]. Since GPFA was associated with tonic seizures, clonic movements, and subtle jerks in LGS, it has been one of the criteria used in diagnosing LGS [5] as well as a late variant of LGS [6]. GPFA, however, also can be seen in patients with frontal and temporal lobe epilepsy [1]. The paroxysmal fast activities (PFA) of LGS are widespread in distribution and bilaterally synchronous over both hemispheres, with highest amplitude over the frontal and central head regions. Shifting asymmetries of PFA are common, but rarely the pattern may show persistent amplitude asymmetry, unilateral or even focal [7]. EEG and clinical differentiation between LGS and focal epilepsy with secondary bilateral synchrony (SBS) may be difficult or even impossible [8]. Halasz hypothesized that the mechanism of GPFA is the permanent or momentary breakdown of the GABAergic inhibitory process operating during the generation of spike-wave discharge [9]. The usual spike and wave complex is thought to consists of a balance of excitatory (spike components) and inhibitory (slow components) components [10]. Therefore, the lack of slow waves during GPFA suggests an absence of inhibitory processes. As well, Halasz et al., reported GPFA in three patients with non-malignant seizures to prove that GPFA can be a possible electrographic variant in certain generalized epilepsies showing atypical features [11]. These authors concluded that GPFA could be seen in patients with treatable epilepsy with a better seizure outcome than that seen in patients with LGS. Sueda et al. differentiated the localized origin and propagation of PFA in patients with epileptic spasms, in contrast to bilaterally generated PFA in patients with LGS, using time–frequency analyses of magnetoencephalography (MEG) [12]. To our knowledge, there have been no reports on GPFA in patients undergoing epilepsy surgery, prior to their receiving the surgery. The characteristics of GPFA and its correlation with the epileptogenic zone as defined by intracranial video EEG (IVEEG) and seizure outcome remain unclear. We analyzed GPFA on
scalp recordings performed in pediatric patients with intractable localization-related epilepsies, who subsequently underwent IVEEG for epilepsy surgery. We hypothesized that asymmetry of GPFA would be present in patients with intractable localization-related epilepsy secondary to the resectable epileptogenic zone, and further that there are characteristic features of GPFA in the intractable localization-related epilepsy. 2. Patients and methods We selected patients with GPFA among the patients who underwent intracranial video EEG between 2004 and 2012 at the Hospital for Sick Children. All patients were admitted to the epilepsy monitoring unit (EMU) for scalp video EEG (VEEG) between 1 and 5 days. In addition, MRI, neuropsychological assessment and magnetoencephalography (MEG) were performed. Thereafter, because the data were concordant for localization of an epileptogenic zone, the patients underwent intracranial video EEG (IVEEG) to delineate the epileptogenic zone pursuant to resective surgery. Parents or guardians gave informed consent for all procedures. This study received prior approval from the Research Ethics Board at The Hospital for Sick Children. We recorded scalp video-EEG (HARMONIE 5.4, Stellate, Montreal, PQ, Canada) using 19 or 25 scalp electrodes placed according to the International 10–20 system (subtemporal electrodes; F9, F10, T9, T10, P9 and P10 or midline electrodes; F1, F2, C1, C2, P1 and P2). A single reference was placed at Oz, Pz’ (located 1 cm behind Pz), or FCz, whichever was the most inactive electrode. Sampling rate was 200 or 500 Hz. We selected the first 50 GPFAs for each patient during NREM sleep, at the first and/or second night of admission. We followed three criteria to select GPFAs (1) amplitude of 100–200 lV; (2) duration of 1–9 s; (3) frequency of 8–26 Hz, for bursts of generalized rhythmic discharges with frontal predominance [1]. We applied a band pass filter of 5–70 Hz without notch filter for the selection and analysis of GPFA. We used a program “Signal statistics” (HARMONIE 5.4, Stellate, Montreal, PQ, Canada) to analyze the amplitude, duration and frequency. We analyzed the amplitude of GPFAs using each of electrodes on referential montage EEG (ex, Fp1-Ref, Fp2-Ref, F3-Ref, F4-Ref). Single Oz electrode was used for the reference. If a patient has the lesions and/or epileptic zone around the occipital lobe, Pz electrode is used for the reference. During the period of GPFA, we measured the amplitude difference between minimum and maximum amplitudes of each GPFA. We compared the amplitude for paired electrodes, Fp1-Ref v.s. Fp2-Ref or F3-Ref v.s. F4-Ref. First, we found the electrode with the maximum amplitude among four electrodes. Second, we compared the paired electrodes, one of which had maximum
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
amplitude. For example, if the maximum amplitude was recorded on F3-Ref, F3-Ref and F4-Ref were analyzed to evaluate hemispheric difference. We used one pair of electrodes, Fp1-F3 and Fp2-F4 for the analysis of the duration and frequency. We applied a paired t-test with Bonferroni correction to compare amplitudes, durations and frequency of GPFA between resection and non-resection sides. Patients underwent 1.5 or 3 tesla (T) MRI. We used a whole-head gradiometer-based Omega system (151 channels, VSM MedTech Ltd., Coquitlam, BC, Canada) for the MEG. Our procedures and analysis of MEG for surgical evaluation were described somewhere [13]. We placed the IVEEG electrodes based on seizure semiology, MRI, MEG, and neuropsychological data to cover eloquent cortex and epileptic zones [14]. We determined the resection area based on seizure onset zone and active interictal zone, location of eloquent cortex, the location of MRI lesion and MEG spike dipoles [15].
3
3. Results Table 1 describes the clinical profiles and features of GPFA. We found 28 patients with interictal generalized spike and waves in 103 patients who underwent intracranial video EEG. We found 14 patients with GPFA among the 28 patients. Eight patients (57%) were female, and six were male (43%). Age of seizure onset ranged between 1 month and 12 years with a mean age of 4.2 years. The scalp VEEG for analysis of GPFA was performed between 3 and 17 years old with a mean age of 11.4 years. Following the scalp VEEG, the children underwent IVEEG between 5 and 18 years of age with a mean age of 12.9 years. Nine (64%) of the 14 patients who had GPFA documented prior to IVEEG and respective surgery presented with partial seizures, including two (14%) patients with secondary generalization. Three (21%) patients had generalized tonic-clonic seizures without
Table 1 Amplitude, duration and frequency of GPFA.
SVEEG: scapl video-EEG; PS: partial seizure; GTCS: generalized tonic clonic seizure; ES: epileptic spasms; PS2G: partial seizure with secondary generalization; *: significantly higher with p < 0.05.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
4
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
focal signs. Seven (50%) patients presented with epileptic spasms. Four patients had symmetric epileptic spasms. Eleven patients (79%) had normal neurologic exam. One (7%) patient had decreased visual field at nasal side but otherwise normal. One patient demonstrated bilateral dysmetria and hand tremor. The other one demonstrated decreased peripheral vision on both sides and mild weakness of right hand fingers.
Six (43%) patients were developmentally delayed. Four patients (29%) were developmentally normal. The other four patients showed attention deficit, hyperactivity disorders, learning difficulties and academic achievement problems. We collected 50 GPFA in each of the 14 patients (Fig. 1). The mean amplitude of GPFA ranged from 145 to 589 lV with a mean of 293 lV. The mean
A FP1-F7 F7-T3 T3-T5 T5-O1 FP2-F8
*
F8-T4 T4-T6 T6-O2 FP1-F3 F3-C3 C3-P3 P3-O1 FP2-F4
*
F4-C4 C4-P4 P4-O2
100 uV 1 sec
FZ-CZ CZ-PZ
B Fp1-REF
F3-REF
Fp2-REF 200 uV 1 sec F4-REF Fig. 1. (A) AP-bipolar montage EEG (500 Hz sampling rate; low frequency filter, 5 Hz; high frequency filter 70 Hz; notch filter on) showing generalized paroxysmal fast activities (GPFA) in case #9 with generalized tonic clonic seizures. GPFA appears with bilateral frontal predominance. The fast activity starts slightly earlier on the right side (*Fp2-F8, Fp2-F4) than the left side. The duration and frequency of GPFA showed 15.1 Hz, 2.61 s at Fp1-F3 and 15.9 Hz, 2.68 s at Fp2-F4. (B) Selected referential montage channels (Oz reference; low frequency filter, 5 Hz; high frequency filter, 70 Hz; notch filter on) of Fp1-Ref, Fp2-Ref, F3-Ref, F4-Ref from GPFA of (A). Fp2-Ref channel shows GPFA with maximum amplitude among four electrodes. Therefore we compared amplitude of Fp2-Ref (398 lV) with Fp1-Ref (292 lV) to evaluate hemispheric difference. The asymmetric amplitude correlated with the epileptogenic hemisphere. The patient underwent resective surgery over the right frontal and temporal regions. The pathology was microdysgenesis. She has been seizure free for four years with medications.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
duration of GPFA ranged from 1.18 to 2.31 s with a mean of 1.6 s. The mean frequency of GPFA ranged from 9.3 to 14.7 Hz with a mean of 11.1 Hz. Table 2 describes the MRI, MEG, resective areas, histopathological findings and seizure outcomes. More than 50% of MEG spike dipoles were located in one hemisphere in all 14 patients. Thirteen patients showed 83–100% of MEG spike dipoles in one hemisphere. We placed subdural grid over eight (57%) left hemispheres and six (43%) right hemispheres. The seizure onset zone and active interictal zone were localized in the multiple lobes in 12 (86%), consisting of three lobes in five (36%) patients and two lobes in three (21%) patients. Ten patients of the 12 patients with multiple lobe epileptic foci involved frontal lobe. Only two (14%) patients had seizure onset and active interictal zone involving the frontal lobe alone. Thirteen of 14 patients underwent resective surgery following IVEEG. In one patient (case 2), MRI showed the abnormal signal in the left frontal lobe in spite of 94% of MEG spike dipoles lateralized in the right hemisphere. The patient initially underwent IVEEG using bilateral subdural strip electrodes over the bilateral frontal lobes and a depth electrode into the left frontal MRI lesion to determine the seizure onset. After we confirmed the right hemispheric seizure onset, the patient had a second surgery to place the subdural grid over the right hemisphere. The seizure onset zone was localized in the right primary motor cortex on IVEEG. The parents
5
refused the proposed resection, but consented to have a biopsy performed in the right frontal lobe instead. Surgical pathology revealed focal cortical dysplasia in four (29%) patients, microdysgenesis in four (29%), subpial gliosis in two (14%), tuberous sclerosis complex, filaminopathy, oligodendrogliosis, and normal in one each. One patient presented both dysembryoplastic neruoepitherial tumor and focal cortical dysplasia. Eleven patients had more than one year follow-up after the resective surgery. Six patients were seizure free with anti-epileptic medications (Engel classification I). Two patients were classified as Engel II, and three patients were Engel III. Two patients had less than one year follow-up. One patient did not undergo resective surgery. We compared the three characteristic features of GPFA: amplitude, duration and frequency between the side of resection after grid recording and the nonresection side. The amplitude of GPFA in all the patients over the resection side (307 ± 156 lV) was significantly higher than that over the non-resection side (279 ± 141 lV) (p < 0.001). The duration of GPFA in all the patients over the resection side (1.62 ± 0.8 s) was significantly higher than that over the non-resection side (1.58 ± 0.8 s) (p < 0.001). The frequency of GPFA in all the patients was no significant difference between the resection side (11.06 ± 2.1 Hz) and the non-resection side (11.08 ± 2.2 Hz).
Table 2 Clinical profiles, MRI, MEG, resective surgery and surgical outcomes.
PS: partial seizure; GTCS: generalized tonic clonic seizure; ES: epileptic spasms; PS2G: partial seizure with secondary generalization; F: frontal lobe; T: Temporal lobe; P: Parietal lobe; O: occipital lobe; MT: mesial temporal; FP: frontoparietal; DNET: dysembryoplastic neuroepithelial tumor; FCD: focal cortical dysplasia. a The epileptogenic zone overlapped the eloqunet cortex. Only biopsy was performed in the right frontal lobe.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
6
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
All nine patients who showed significant duration differences between two sides (p < 0.05) had the longer duration of GPFA over the resection side. Eight of 12 patients who showed significant amplitude differences between two sides (p < 0.05) had the higher amplitude of GPFA over the resection side. Four of six patients who showed significant frequency differences between two sides (p < 0.05) had the higher frequency of GPFA over the resection side. We compared three characteristic features of GPFA between resection and non-resection sides among ten patients (Table 1. Number of patients 1,3,5,6,8,9,10,12,13,14) with good seizure outcome (Engel I and II). Seven (70%) of them showed higher amplitude over the resection side than the non-resection side including six patients with a statistical significance. In three patients (Table 1. Number of patients 4, 7, 11) with relatively poor seizure outcome (Engel III), one of them showed significant higher amplitude of GPFA in the non-resection side than in the resection side. 4. Discussion GPFA is considered a sign of cortico-thalamic discharges to produce generalized, frontal predominance, rhythmic, and lack of complex of spike and slow wave. Thus, GPFA is not only seen in LGS, but can be seen in localization-related epilepsy with secondary generalization and/or secondary bilateral synchrony [1,11]. In our series, three patients presented with generalized tonic clonic seizures and two patients partial seizures with secondary generalization, and seven patients epileptic spasms. They had focal onset proven by IVEEG. However, their seizure onset and interictal zones were widespread over the multiple lobes. The previous reports of GPFA were frequently recognized in LGS, furthermore GPFA might present brief tonic fits or clonic movements not infrequently [5]. When GPFA became longer, tonic seizures were more associated than shorter GPFA [1]. GPFA has a potential of ictal symptoms or relating with ictal symptomatic network in a subset of intractable epilepsy patients. Children with LGS as known as epileptic encephalopathy involving both cortex and subcortical structures, thus GPFA is most frequently captured and seems to be a sign of LGS. GPFA consist of high amplitude alpha range fast activities running up to six seconds without isolated slow waves. When GPFA last longer, they can produce clinical seizures; however, even brief GPFA can produce subtle clinical seizures through the cortico-subcortical network. We compared GPFA at bilateral frontal electrodes, because GPFA appeared at bilateral frontal electrodes, GPFA appeared predominantly over the bilateral frontal regions regardless of the localization of either seizure onset zones or epileptogenic zones. GPFA can be a part of secondary bilateral synchrony
or generalized discharges provoked by the epileptic cortico-thalamic network in the localization-related epilepsy. Nine (62%) patients became seizure free after surgical resection over the multiple lobe resection. The precise lateralization of epileptogenic hemisphere, IVEEG to localize the epileptogenic zone and surgical treatment can be a reasonable option to control intractable epilepsy in children with GPFA. Lennox–Gastaut syndrome has been reported surgically treated with successful seizure outcome [16,17]. The most of seizure free patients following focal resection for LGS had intractable localization-related epilepsy. Their results indicated secondary bilateral synchrony to project generalized epileptiform discharges. Wyllie et al. described epilepsy surgery was successful for selected children and adolescents with a congenital or early-acquired brain lesion, despite abundant generalized or bilateral epileptiform discharges on EEG [18]. Our series showed that the patients with GPFA, yet presenting a sign of lateralization for the epileptogenic zone on the scalp VEEG, precise localization of the epileptogenic zone using various modalities using MRI, MEG, PET and IVEEG would be highly recommended. The detection of GPFA must not discourage from considering pre-surgical evaluation. If the GPFA can show lateralized signs of the amplitude and duration asymmetry, the other techniques for definition of the epileptogenic zone are highly recommended. Then surgical treatment can achieve the successful seizure outcome in patients with intractable localization-related epilepsy and GPFA. In our series, the amplitude and duration of GPFA over the resection side were significantly higher than that over the non-resection side. The frequency of GPFA was no significant difference between the resection side and the non-resection side. Six of 10 good-seizure-outcome patients showed significantly higher amplitude of GPFA over the resection side than the non-resection side. Eight of 10 good-seizure-outcome patients showed significantly longer duration of GPFA over the resection side than the non-resection side. In three patients with poor seizure outcome, the GPFA amplitude was significantly higher in the non-resection side than the resection side for one patients. The amplitude and duration of GPFA in the localization-related epilepsy with SBS and generalized seizures may contribute to the excitatory cortex in epileptogenic hemisphere. The frequency of GPFA, however, had no differences between two hemispheres. The thalamo-cortical network may conduct the identical rhythm of both hemispheres. The predominant high amplitude of GPFA in the contralateral hemisphere may indicate potential independent epileptogenesis in the non-resection side. Further collection of patients with GPFA will be performed to analyze the correlation GPFA with the epileptogenic zone/hemisphere and epileptic disorders.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
M. Mohammadi et al. / Brain & Development xxx (2014) xxx–xxx
5. Conclusion GPFA can exist in intractable localization-related epilepsy with SBS and generalized seizures. GPFA is not always a sign of malignant epileptic encephalopathy represented by Lennox–Gastaut syndrome. In a subset of the intractable localization-related epilepsy with SBS and generalized seizures, the amplitude and duration of GPFA may contribute to the excitatory cortex in epileptogenic hemisphere, furthermore, the bilateral identical rhythm may be conducted by thalamo-cortical network. Although EEG shows GPFA on scalp VEEG, the precise localization of the epileptogenic zone using IVEEG could achieve the successful surgical resection for a subset of patients with GPFA.
Acknowledgements We thank Dr. Mahdi Razzaghi from Bloomsburg University (PA, USA), for his invaluable assistance in statistical tests selection. Drs. Tohru Okanishi and Satoru Sakuma were supported by Ontario Brain Institute (OBI). References [1] Laoprasert P. Atlas of pediatric EEG. New York: McGraw Hill; 2011. [2] Jasper H, Kershman J. Electroencephalograph classification of the epilepsies. AMA Arch Neurol Psychiatry 1941;45: 903–43. [3] Gastaut H, Roger J, Ouahchi S, Timsit M, Broughton R. An electro-clinical study of generalized epileptic seizures of tonic expression. Epilepsia 1963;4:15–44. [4] Brenner RP, Atkinson R. Generalized paroxysmal fast activity: electroencephalographic and clinical features. Ann Neurol 1982; 11:386–90.
7
[5] Beaumanoir A. The Lennox–Gastaut syndrome. In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, editors. Epileptic syndromes in infancy, childhood and adolescence. London: John Libbey Eurotext; 1985. p. 89–99. [6] Bauer G, Aichner F, Saltuari L. Epilepsies with diffuse slow spikes and waves of late onset. Eur Neurol 1983;22:344–50. [7] Markand ON. Lennox–Gastaut syndrome (childhood epileptic encephalopathy). J Clin Neurophysiol 2003;20:426–41. [8] Niedermeyer E. Lennox–Gastaut syndrome. Clinical description and diagnosis. Adv Exp Med Biol 2002;497:61–75. [9] Hala´sz P. Runs of rapid spikes in sleep – a characteristic EEG expression of generalized malignant epileptic encephalopathies. A conceptual review with new pharmacological data. In: Degen R, Rodin EA, editors. Epilepsy, sleep, and sleep deprivation. New York: Elsevier; 1991. p. 49–78. [10] Lothman EW. The neurobiology of epileptiform discharges. Am J EEG Technol 1993;33:93–112. [11] Hala´sz P, Janszky J, Barcs G, Szucs A. Generalised paroxysmal fast activity (GPFA) is not always a sign of malignant epileptic encephalopathy. Seizure 2004;13:270–6. [12] Sueda K, Takeuchi F, Shiraishi H, Nakane S, Sakurai K, Yagyu K, et al. Magnetoencephalographic analysis of paroxysmal fast activity in patients with epileptic spasms. Epilepsy Res 2013;104: 68–77. [13] Ochi A, Go CY, Otsubo H. Clinical MEG analyses for children with intractable epilepsy. In: Pang EW, editor. Magnetoencephalography. Tech North, America: Elseiver; 2011 ch. 9. [14] Snead 3rd OC. Surgical treatment of medically refractory epilepsy in childhood. Brain Dev 2001;23:199–207. [15] 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. [16] Jung YJ, Kang HC, Choi KO, Lee JS, Kim DS, Cho JH, et al. Localization of ictal onset zones in Lennox–Gastaut syndrome using directional connectivity analysis of intracranial electroencephalography. Seizure 2011;20:449–57. [17] Lee YJ, Kang HC, Lee JS, Kim SH, Kim DS, Shim KW, et al. Resective pediatric epilepsy surgery in Lennox–Gastaut syndrome. Pediatrics 2010;125:e58–66. [18] Wyllie E, Lachhwani DK, Gupta A, Chirla A, Cosmo G, Worley S, et al. Successful surgery for epilepsy due to early brain lesions despite generalized EEG findings. Neurology 2007;69:389–97.
Please cite this article in press as: Mohammadi M et al. Asymmetrical generalized paroxysmal fast activities in children with intractable localization-related epilepsy. Brain Dev (2014), http://dx.doi.org/10.1016/j.braindev.2014.03.006
