Brain & Development 31 (2009) 294299 www.elsevier.

com/locate/braindev

Original article

Seizures in childhood ischemic stroke in Taiwan

Division of Pediatric Neurology, Department of Pediatrics, Chang Gung Childrens Hospital, #5 Fu-Shin Street, Kwei-Shan, Taoyuan 333, Taiwan b Division of Neuroradiology, Chang Gung Childrens Hospital and Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Taoyuan, Taiwan Received 5 December 2007; received in revised form 14 May 2008; accepted 14 May 2008

Abstract In this retrospective study, we collected clinical and radiographic data on children (age range, 1 month to 18 years) with symptoms and radiographic conrmation of seizure after ischemic stroke for the period of January 1996 to July 2006. Thirty-nine out of 94 children with ischemic stroke had poststroke seizures. Thirty-three out of 39 children with poststroke seizures had new onset seizures but only data of 28 were available. Infection was the most common etiology in the early poststroke seizure group (52.4%) but not in the late poststroke seizure group (0%). Infarction involving arterial ischemic stroke of anterior circulation were the most common in both the early poststroke seizure (61.9%) and the late poststroke seizure group (57.1%). Epilepsy was the most common sequelae in both the early poststroke seizure (38.1%) and late poststroke seizure group (100%). Children who had initial focal neurological sign (100% vs. 38.1%; P = 0.007) or the focal cortical dysfunction on EEG (85.7% vs. 33.3%; P = 0.029) were prone to develop late poststroke seizures. Late poststroke seizures had a high risk of developing poststroke epilepsy (100% vs. 38.1%; P = 0.007). We conclude that seizures commonly occur in childhood ischemic stroke. Most poststroke seizures developed at an early stage. Infection was the most common etiology that caused early poststroke seizures in childhood ischemic stroke. Initial focal neurological signs and focal cortical dysfunction on EEG are risk factors for developing epilepsy. Poststroke seizures did not aect mortality, but there was a signicant dierence in normal outcome and epilepsy between those with or without poststroke seizures. 2008 Elsevier B.V. All rights reserved.
Keywords: Poststroke seizure; Epilepsy; Childhood ischemic stroke

1. Introduction Jackson was the rst one to observe a relationship between seizures and cerebrovascular disease in 1864 [1]. Stroke is the most common cause of seizures in the elderly, and poststroke epilepsy is the most common neurologic sequelae of stroke. Childhood stroke is dened as a sudden onset neurological decit due to a cerebrovascular
Corresponding author. Tel.: +886 3 328 1200x8200; fax: +886 3 328 8957. E-mail address: lincgh@adm.cgmh.org.tw (K.-L. Lin). 0387-7604/$ - see front matter 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2008.05.006
*

event occurring between 30 days and 18 years of age [2]. Many studies about the incidence of seizures after stroke in adults have been reported, but there are few reports about seizures in childhood ischemic stroke. Therefore, we performed this retrospective study to determine the incidence of initial seizures and poststroke epilepsy after childhood ischemic stroke occurring in a tertiary medical center in Taiwan. In addition, we also investigated the clinical presentations, radiological ndings, etiologies, interictal electroencephalogram (EEG) patterns, seizure types and outcomes of initial seizure depending on the time of seizure after ischemic stroke.

J.-C. Lee et al. / Brain & Development 31 (2009) 294299

295

2. Materials and methods 2.1. Patients

nephrotic syndrome, vasculopathy, mitochondria or others. 2.5. Radiologic ndings

cardiovascular,

We searched for consecutive cohort of children (1 month to 18 years) with seizure attack after ischemic stroke in the medical records database of Chung Gung Childrens Hospital, a tertiary medical center in Taiwan, from January 1996 to July 2006. Admissions for ischemic stroke were identied using the following International Classication of Disease, 9th revision (ICD-9) codes for ischemic stroke, in any discharge coding: 433 (occlusion and stenosis of precerebral arteries), 434 (occlusion of cerebral arteries), 436 (acute, but ill-dened, cerebrovascular disease), 437 (other and ill-dened cerebral vascular disease), and 437.6 (nonpyogenic thrombosis of intracranial venous sinus). Transient ischemic attack, hypoxiaischemic encephalopathy, and neonatal stroke were not included in this study because they exhibited dierent etiologic patterns. 2.2. Seizures We classied seizures according to the International League Against Epilepsy (ILAE) criteria [3]. Seizures did not include transient amnesia or isolated, transient change in level of consciousness. According to the guidelines of ILAE, early poststroke seizures were dened as those occurring within 7 days and late poststroke seizures were dened as unprovoked seizures developing beyond 1 week after stroke [4]. In all cases, the diagnosis was conrmed by the neurologist in charge of the patients. Early poststroke seizures occurrence was ascertained by chart review and late poststroke seizures were systematically recorded at each follow-up visit. All available information about seizures (date, type, and number of seizures), interictal EEG and current antiepileptic treatment were all reviewed. Interictal EEG was performed 12 days after initial seizure attack and considered abnormal when focal, lateralized, or generalized slowing or epileptiform discharges were present. 2.3. Clinical features Basic demographic information was recorded. Clinical presentations were classied as fever, diuse neurological signs, and focal neurological signs. The diagnostic evaluations included neuroimaging, cardiac assessment, prothrombotic assays, immunologic assays, infection studies, and metabolic studies. 2.4. Etiologies According to the clinical manifestation, radiologic imaging, and laboratory ndings, etiologies were classied into infection, trauma, metabolism, moyamoya,

All patients underwent brain computed tomography (CT) or magnetic resonance imaging (MRI) with or without conventional angiography, magnetic resonance angiography (MRA), or computed tomography angiography (CTA). All imaging studies were reviewed by the pediatric neuroradiologist. According to the imaging results, ischemic stroke was further classied as arterial ischemic stroke or sinovenous thrombosis. The vascular patterns of arterial ischemic stroke were classied as anterior circulation (anterior and middle cerebral arteries) or posterior circulation (vertebrobasilar territory). 2.6. Outcomes Based on the assessment at the last follow-up visit, the outcome was classied as normal, neurologic decits, epilepsy, or death. The SPSS 12.0 software was used for statistical analysis. The data were analyzed using Chi-square test for independence and Fishers Exact Test when the expected number in any cell was less than 5. P-value <0.05 was considered statistically signicant. 3. Results 3.1. Patients and clinical characteristics There were total 94 children with ischemic stroke in our 10 year medical records database. Thirty-nine children with seizure attack after ischemic stroke were noted during this study. We excluded 11 patients for the following reasons: previous seizure history before stroke (6), death within initial hospitalization (1), or insucient data (4). Therefore, 28 children whose age of rst episode varied between 40 days and 16 years were enrolled in this study. There were 15 boys and 13 girls. According to the ILAE guidelines for epidemiologic studies on epilepsy [4], 21 patients (13 males, 8 females) had early poststroke seizures and seven patients (2 males, 5 females) had late poststroke seizures. The mean ages are 4.3 and 4.2 years old in the early poststroke seizure and late poststroke seizure group, respectively. Gender and age dierences between the early poststroke seizure and late poststroke seizure groups were not signicant. 3.2. Clinical presentations Focal neurological signs were the most common presentation in both early poststroke seizure group (n = 8, 38.1%) and late poststroke seizure group (n = 7, 100%). This is more frequent in the late poststroke sei-

296

J.-C. Lee et al. / Brain & Development 31 (2009) 294299 Table 1 Clinical presentations, etiologies, radiologic ndings of 28 children with poststroke seizures Early seizures n = 21 Clinical presentations Fever Diuse neurological signs Consciousness disturbance Headache Focal neurological signs* Hemiplegia Cranial nerve palsy Sensory impairment Visual impairment Etiologies Infection** Trauma Metabolism Moyamoya Nephrotic syndrome Vasculopathy Cardiovascular Mitochondria Unknown Radiologic ndings Side of involved Both Left Right Arterial territory Anterior circulation Posterior circulation Both Venous thrombosis 6 (%) (28.6) Late seizures n=7 0 (%)

zure group (P-value = 0.007). In the focal neurological signs, hemiplegia was the most common focal neurological sign in both the groups but no signicant dierence between these two groups. In comparison with the late poststroke seizure group, there was a higher frequency of fever (n = 6, 28.6%), consciousness disturbance (n = 4, 19.0%), headache (n = 4, 19.0%) in the early-seizure group. But these dierences were not signicant between these two groups. 3.3. Etiologies The most common etiology in the early poststroke seizure group was infection (52.4%), followed by trauma (19%), metabolic disorders (14.3%), cardiovascular disorders (9.5%), moyamoya disease (4.8%), and nephrotic syndrome (4.8%). The etiologies in the late poststroke seizure group are trauma (14.3%), metabolic disorders (14.3%), moyamoya disease (14.3%), vasculopathy (28.6%), and mitochondria disease (28.6%). The etiologies of two patients remained unidentied. Therefore, infection was the most common etiology in the early poststroke seizure (n = 11, 52.4%) but not in the late poststroke seizure group (n = 0, 0%) (P-value = 0.023). Infection included meningitis (n = 6), encephalitis (n = 1), meningoencephalitis (n = 3), and sepsis (n = 5). Four patients had sepsis and meningoencephalitis or meningitis at the same time. The pathogens including Streptococcus pneumoniae (n = 3), Haemophilus inuenzae (n = 1), Pseudomonas aeruginosa (n = 1), and Mycobacterium tuberculosis (n = 2) were identied in seven patients. 3.4. Radiology ndings All the patients received brain CT and/or MRI. Ischemic infarction was noted in all patients. Radiological evaluations of the infarction are presented in Table 1. In 10 patients, infarction involving more than one side was found. Infarction involving arterial ischemic stroke of anterior circulation (anterior and middle cerebral arteries) were the most common presentation in both the early poststroke seizure group (n = 13, 61.9%) and late poststroke seizure group (n = 4, 57.1%). Infarction of hemisphere and arterial territory involved were all insignicant in both the early poststroke seizure and the late poststroke seizure group. 3.5. Seizure patterns and EEG In 28 patients with seizure after ischemic stroke, there were 11 patients with focal seizures in early poststroke seizure group and three in late poststroke seizure group (Table 2). Eleven patients had recurrence of afebrile seizure in the early poststroke seizure group. Eight patients had recurrent seizure within 24 h. Other three patients

4 4 8 6 3 3 0 11 4 3 1 1 0 2 0 1

(19.0) (19.0) (38.1) (28.6) (14.3) (14.3)

0 1 7 5 2 0 2 0 1 1 1 0 2 0 2 1

(14.3) (100.0) (71.4) (28.6) (28.6)

(52.4) (19.0) (14.3) (4.8) (4.8) (9.5) (4.8)

(14.3) (14.3) (14.3) (28.6) (28.6) (14.3)

8 7 6

(38.1) (33.3) (28.6)

2 3 2

(28.6) (42.9) (28.6)

13 2 3 3

(61.9) (9.5) (14.3) (14.3)

4 1 1 1

(57.1) (14.3) (14.3) (14.3)

Three patients (10.7%) had two etiologies. * P = 0.007 by Chi-square test. ** P = 0.023 by Chi-square test. Table 2 Seizure type and EEG pattern of 28 children with poststroke seizures Early seizures n = 21 Seizure type Generalized Focal EEG pattern Focal cortical dysfunction* Diuse cortical dysfunction Focal epileptiform Multifocal epileptiform Focal epileptiform with generalization Negative 10 11 7 7 8 2 1 1 (%) (47.6) (52.4) (33.3) (33.3) (38.1) (9.5) (4.8) (4.8) Late seizures n=7 4 3 6 2 4 0 0 0 (%) (57.1) (42.9) (85.7) (28.6) (57.1) (0.0) (0.0) (0.0)

Nine patients had two patterns of EEG ndings. * P = 0.029 by Chi-square test.

had recurrent seizure on 3, 15, and 180 days, respectively. In the late poststroke seizure group, all of seven patients had recurrent seizures within 24 h.

J.-C. Lee et al. / Brain & Development 31 (2009) 294299

297

Nine patients had two patterns of EEG abnormalities. Children with late poststroke seizures tend to have a focal cortical dysfunction on EEG (85.7% vs. 33.3%; P-value = 0.029). However, seizure type and other EEG abnormalities, such as diuse cortical dysfunction, focal epileptiform, multifocal epileptiform, and focal epileptiform with generalization were all insignicant in both the early poststroke seizure and late poststroke seizure groups (Table 2). 3.6. Outcomes The mean interval of follow-up was 2 years (range: 0.510 years). No neurological decits or epilepsy were noted in three patients (14.3%) of early poststroke seizure group. Epilepsy is the most common sequela in both the early poststroke seizure (n = 8, 38.1%) and the late poststroke seizure group (n = 7, 100%). Epilepsy is more frequent in the late poststroke seizure group (Pvalue = 0.007). Other outcomes, such as motor impairment, speech impairment, hearing impairment, visual impairment, cognitive impairment, developmental delay, and death, were all insignicant in both the early poststroke seizure and late poststroke seizure group (Table 3). Furthermore, we compared the outcomes of those with poststroke seizures to those without poststroke seizures. Normal outcomes (n = 24, 51.1%) in the group of the patients without poststroke seizure was signicant dierence (P = 0.000). Poststroke epilepsy (n = 7, 100%) in the group of the patients with late poststroke seizures was signicant dierence (P = 0.000) (Table 3). 4. Discussion Most previous clinical studies make a distinction between early and poststroke seizures based on dier-

ences in their presumed pathophysiology. Many authors dened early poststroke seizures as those occurring within 14 days after stroke [57]. However, Hauser et al. and Lamy et al. dened early poststroke seizures as those occurring within 1 week after stroke [8,9]. According to the guidelines of the ILAE [4], seizures occurring more than 1 week after clinical identied cerebral infarction or intracerebral hemorrhage or subarachnoid hemorrhage are dened as late seizures after cerebrovascular diseases. Therefore, we dened early poststroke seizures as those occurring within 7 days of stroke and unprovoked seizures developing beyond 1 week after stroke were termed late poststroke seizures. The incidence of poststroke seizure in ve children studies ranged from 14% to 73% [1014] reecting the dierent populations in each study. In our study, 39 out of 94 ischemic stroke patients (41.4%) had seizures after ischemic stroke. From the available data and our study, the overall incidence of poststroke seizures appears to be higher in children than in adult populations (2.27.1%) [15]. Twenty-one of 94 cases (22.3%) had early poststroke seizures. Previous study by Honer et al. reported 24% of 72 patients with post-ischemic stroke had early onset seizures [16]. The incidence of the early poststroke seizures was similar to our study. Regarding initial symptoms and signs, children with focal neurological signs tend to have late poststroke seizures. Lesser et al. addressed that acute transient cytotoxic metabolic alterations and structural changes are responsible for stroke-related seizures [17]. Furthermore, De Carolis et al. reported that distinction is important between early and late poststroke seizures because early poststroke seizures are probably due to biochemical reactions provoked by ischemia, whereas late poststroke seizures are due to chronic epileptic foci [18]. This may explain why the late poststroke seizure

Table 3 Outcomes of 75 children with early poststroke seizures or late poststroke seizures vs. without poststroke seizures after ischemic stroke Outcomes
*

Early seizures n = 21 (%) (9.5) (4.8) (4.8) (38.1) (47.6) (47.6) (4.8) (4.8) (14.3) (9.5) (4.8)

Late seizures n=7 0 1 0 1 7 3 2 0 2 1 0 0 0 (%) (14.3) (14.3) (100.0) (42.9) (28.6) (28.6) (14.3)

No seizures n = 47 24 9 7 2 0 21 12 1 1 2 1 0 2 (%) (51.1) (19.1) (14.9) (4.3) (44.7) (25.5) (2.1) (2.1) (4.3) (2.1) (4.3)

Normal Death <30 days >6 months Poststroke epilepsy** Neurological decits Motor impairment Speech impairment Visual impairment Cognitive impairment Hearing impairment Developmental delay Others
* **

2 1 0 1 8 10 10 1 1 0 3 2 1

P = 0.000 by Chi-square test. P = 0.000 by Chi-square test.

298

J.-C. Lee et al. / Brain & Development 31 (2009) 294299

group can present as focal neurological sign in stead of diuse neurological sign after ischemic stroke. About seizure types, early poststroke seizures were more likely to be generalized in series of Horner et al [16]. On the contrary, Gupta et al. reported that early poststroke seizures were more likely to be partial (57%), whereas late poststroke seizures were more likely to be generalized (65%) [19]. However, focal seizures with or without secondary generalization, and generalized seizure were not signicant dierence between the early poststroke seizure and late poststroke seizure group in our study. Infection has been described as an etiology for stroke in earlier studies [20,21]. Acute and exacerbating chronic infection may act by activating coagulation and chronic infections and may contribute to atherogenesis. Genetic predisposition of the inammatory host response may be an important codeterminant for atherogenesis and stroke risk [22]. In our study, infection was the most common etiology of the early poststroke seizure, but none of the late poststroke seizure group was due to infection. This may explain that infection could induce acute biochemical reactions, but few of those become chronic epileptic foci. Some previous studies agree that late poststroke seizure was a risk factor for poststroke epilepsy [7,14,23]. The occurrence of early and late poststroke seizures parallels that of post-traumatic epilepsy. These similar pathophysiologic mechanisms are cellular biochemical dysfunction for early poststroke seizures and epileptogenic gliotic scarring for late poststroke seizures [24,25]. In our study, 100% of seven patients with late poststroke seizures had poststroke epilepsy. This high risk of poststroke epilepsy is similar to that observed in the late poststroke seizures of other studies [24,26,27] and furthermore is higher than that reported for the general population experiencing a rst unprovoked seizure [28]. Therefore, Dhanuka et al. also made a conclusion that early poststroke seizures were rather common, did not recur and did not be treated with antiepileptics. Late poststroke seizures were less common but were associated with poststroke epilepsy [29]. In addition to neuroimaging, EEG may be helpful in the early evaluation of poorly dened poststroke focal neurologic symptoms. Patients with frontal intermittent rhythmic delta activities (FIRDAs) and diuse slowing on the poststroke EEG have a high risk to develop the late poststroke seizure, while the chance is reduced in those with normal EEG ndings [30]. In our study, the patients with focal cortical dysfunction on interictal EEG were prone to have late poststroke seizures. Focal cortical dysfunction on interictal EEG might be more related to structural changes even epileptics foci than simply wide region of ischemic or infarcted tissue involving the cerebral cortex or subcortical territory. Furthermore, in our study, the patients with focal cortical

dysfunction had high rate of poststroke epilepsy. Therefore, we concluded that children with focal cortical dysfunction on interictal EEG after ischemic stroke and late poststroke seizure are prone to have poststroke epilepsy. Regarding the outcomes of those with or without poststroke seizures, we found that poststroke seizures did not aect mortality. However, the patients with poststroke seizures were prone to develop poststroke epilepsy and those without poststroke seizures were apt to have normal outcomes without any other neurological sequelae. Nevertheless, we were unable to gather enough data to make a detailed comparison between the patients who had seizures and those that did not, which is a great limitation of this study. 5. Conclusion Seizures commonly occur in childhood ischemic stroke. Most poststroke seizures developed at an early stage. Infection was the most common etiology that caused early poststroke seizures in childhood ischemic stroke. Children who have initial focal neurological signs or the focal cortical dysfunction on EEG are prone to develop late poststroke seizures. Late poststroke seizures had a high risk of developing epilepsy. Poststroke seizures did not aect mortality, but there was a significant dierence in normal outcome and epilepsy between those with or without poststroke seizures. References
[1] Jackson JJH. Epileptiform convulsions from cerebral disease. In: Taylor J, Homes G, Walshe FMR, editors. Selected writings of John Hughlings Jackson on epilepsy and epileptiform convulsion. London: Hodder and Stoughton, Ltd.; 1931. p. 3040. [2] Lynch JK, Hirtz DG, DeVeber G, Nelson KB. Report of the National Institute of Neurological Disorders and Stroke workshop on perinatal and childhood stroke. Pediatrics 2002;109:11623. [3] Proposal for revised clinical and electroencephalographic classication of epileptic seizures. From the Commission on Classication and Terminology of the International League Against Epilepsy. Epilepsia 1981; 22:489501. [4] Guidelines for epidemiologic studies on epilepsy. Commission on Epidemiology and Prognosis, International League Against Epilepsy. Epilepsia 1993;34:5926. [5] Berger AR, Lipton RB, Lesser ML, Lantos G, Portenoy RK. Early seizures following intracerebral hemorrhage: implications for therapy. Neurology 1988;38:13635. [6] Faught E, Peters D, Bartolucci A, Moore L, Miller PC. Seizures after primary intracerebral hemorrhage. Neurology 1989;39:108993. [7] Fish DR, Miller DH, Roberts RC, Blackie JD, Gilliatt RW. The natural history of late-onset epilepsy secondary to vascular disease. Acta Neurol Scand 1989;80:5246. [8] Hauser WA R-LM, Rosenstein R. Risk for seizures and epilepsy following cerebrovascular insults. Epilepsia 1984;25:666. [9] Lamy C, Domigo V, Semah F, Arquizan C, Trystram D, Coste J, et al. Early and late seizures after cryptogenic ischemic stroke in young adults. Neurology 2003;60:4004.

J.-C. Lee et al. / Brain & Development 31 (2009) 294299 [10] Aicardi J, Amsili J, Chevrie JJ. Acute hemiplegia in infancy and childhood. Dev Med Child Neurol 1969;11:16273. [11] Solomon GE, Hilal SK, Gold AP, Carter S. Natural history of acute hemiplegia of childhood. Brain 1970;93:10720. [12] Schoenberg BS, Mellinger JF, Schoenberg DG. Cerebrovascular disease in infants and children: a study of incidence, clinical features, and survival. Neurology 1978;28:7638. [13] Lanska MJ, Lanska DJ, Horwitz SJ, Aram DM. Presentation, clinical course, and outcome of childhood stroke. Pediatr Neurol 1991;7:33341. [14] Yang JS, Park YD, Hartlage PL. Seizures associated with stroke in childhood. Pediatr Neurol 1995;12:1368. [15] Camilo O, Goldstein LB. Seizures and epilepsy after ischemic stroke. Stroke 2004;35:176975, [A journal of cerebral circulation]. [16] Horner S, Ni XS, Duft M, Niederkorn K, Lechner H. EEG, CT and neurosonographic ndings in patients with postischemic seizures. J Neurol Sci 1995;132:5760. [17] Lesser RP, Lu ders H, Dinner DS, Morris HH. Epileptic seizures due to thrombotic and embolic cerebrovascular disease in older patients. Epilepsia 1985;26:62230. [18] De Carolis P, DAlessandro R, Ferrara R, Andreoli A, Sacquegna T, Lugaresi E. Late seizures in patients with internal carotid and middle cerebral artery occlusive disease following ischemic events. J Neurol Neurosurg Psychiatry 1984;47:13457. [19] Gupta SR, Naheedy MH, Elias D, Rubino FA. Postinfarction seizures. A clinical study. Stroke 1988;19:147781. [20] Kirkham FJ, Prengler M, Hewes DK, Ganesan V. Risk factors for arterial ischemic stroke in children. J Child Neurol 2000;15:299307.

299

[21] Lee YY, Lin KL, Wang HS, Chou ML, Hung PC, Hsieh MY, et al. Risk factors and outcomes of childhood ischemic stroke in Taiwan. Brain Dev 2008;30:149. [22] Lindsberg PJ, Grau AJ. Inammation and infections as risk factors for ischemic stroke. Stroke 2003;34:251832. [23] Daniele O, Mattaliano A, Tassinari CA, Natale E. Epileptic seizures and cerebrovascular disease. Acta Neurol Scand 1989;80:1722. [24] Bladin CF, Alexandrov AV, Bellavance A, Bornstein N, Chambers B, Cote R, et al. Seizures after stroke: a prospective multicenter study. Arch Neurol 2000;57:161722. [25] Silverman IE, Restrepo L, Mathews GC. Poststroke seizures. Arch Neurol 2002;59:195201. [26] So EL, Annegers JF, Hauser WA, OBrien PC, Whisnant JP. Population-based study of seizure disorders after cerebral infarction. Neurology 1996;46:3505. [27] Burn J, Dennis M, Bamford J, Sandercock P, Wade D, Warlow C. Epileptic seizures after a rst stroke: the Oxfordshire Community Stroke Project. BMJ 1997;315:15827. [28] Hauser WA, Rich SS, Lee JR, Annegers JF, Anderson VE. Risk of recurrent seizures after two unprovoked seizures. N Engl J Med 1998;338:42934. [29] Dhanuka AK, Misra UK, Kalita J. Seizures after stroke: a prospective clinical study. Neurol India 2001;49: 336. [30] De Reuck J, Goethals M, Claeys I, Van Maele G, De Clerck M. EEG ndings after a cerebral territorial infarct in patients who develop early-and late-onset seizures. Eur Neurol 2006;55:20913.