Clinical and radiological risk factors for poststroke

Epilepsy & Behavior 88 (2018) 113–116

Contents lists available at ScienceDirect

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

Clinical and radiological risk factors for poststroke epilepsy in childhood Mauricio López-Espejo a,⁎, Marta Hernández-Chávez a, Isidro Huete b a b

Unit of Neurology, Division of Pediatrics, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile Department of Radiology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile

a r t i c l e

i n f o

Article history: Received 11 June 2018 Revised 9 August 2018 Accepted 12 August 2018 Available online xxxx Keywords: Cerebrovascular disease Pediatric stroke Epilepsy Childhood arterial ischemic stroke Brain infarction Stroke outcome

a b s t r a c t Background: There are few studies evaluating risk factors for poststroke epilepsy (PSE) after an arterial ischemic stroke (AIS) in childhood. This study aimed to evaluate clinical and radiological predictors for PSE in a cohort of children with a first-ever AIS. Methods: A retrospective analysis of a single-center prospective consecutive cohort of children beyond neonatal age with a first-ever AIS admitted at the Pontifical Catholic University of Chile's Clinical Hospital between 2003 and 2013. All participants had a brain magnetic resonance imaging at the time of diagnosis. All children underwent follow-up for at least three years with an annual clinical evaluation. We used the current epilepsy definition of the International League Against Epilepsy. Studied variables include demographics, clinical manifestations at onset, stroke risk factors, and radiological characteristics of AIS. Cox proportional hazards regression analysis was used to evaluate PSE risk adjusted for clinical and radiological variables. Results: Among 98 children who met the study criteria, 41 (41.8%) with PSE. Following multivariate analysis, it was determined that the predictors of PSE include young age at AIS (hazard ratio [HR] = 0.91; confidence interval [CI] = 0.84–0.99), the occurrence of acute symptomatic seizures (HR = 3.29; CI = 1.35–8.01), cortical infarction (HR = 5.01; CI = 2.00–12.6), and multifocal infarction (HR = 3.27; CI = 1.01–10.8). Conclusion: Seizures, young age, cortical lesions, and multiple infarction at the time of stroke are independent risk factors for PSE in children following a first-ever AIS. © 2018 Elsevier Inc. All rights reserved.

1. Introduction The annual arterial ischemic stroke (AIS) incidence rate ranges from 1.2 to 7.9 per 100,000 children beyond the first month of life [1,2]. Although AIS has a rather low incidence in children compared with adults, the risk to develop epilepsy is higher in this age group [3–5]. Previous studies have reported several risk factors for poststroke epilepsy (PSE) in childhood, including the occurrence of acute symptomatic seizures (AS) at the time of diagnosis, young age at onset of AIS (notably in infancy), and cortical involvement [5–7]. Also, there are numerous probable associated variables, such as involvement of the middle cerebral artery (MCA) territory and the presence of focal cerebral arteriopathy [3,8]. Although both cortical involvement and younger age at the time of AIS are strongly associated with a higher likelihood of AS, there are still no prospective studies evaluating the interrelation of these variables and the individual effect of these factors on the probability of developing PSE in childhood. Therefore, our aim in the current study was to evaluate the individual effect of ⁎ Corresponding author at: Unit of Neurology, Division of Pediatrics, School of Medicine, Pontificia Universidad Católica de Chile, Diagonal Paraguay 362, Santiago 8330077, Chile. E-mail address: [email protected] (M. López-Espejo).

aforementioned risk factors for PSE in a cohort of children with a firstever AIS. We hypothesized that there are both clinical and radiological independent risk factors for PSE, which can be used to identify high-risk children after an AIS. 2. Materials and methods We conducted a retrospective analysis of a single-center prospective consecutive cohort at the Pontifical Catholic University of Chile's Clinical Hospital. All clinical, laboratory, and radiological information was entered into a database, corrected according to a review of medical and neuroimaging records, and enrolled according to the institutional protocols [9]. We included all children aged 29 days to 18 years with radiologically confirmed AIS (defined as acute neurological symptoms or signs secondary to acute focal cerebral infarction in an arterial distribution on magnetic resonance images [MRIs]) occurring between January 2003 and July 2013. We exclude patients with previous epilepsy, suspected perinatal stroke, previous childhood AIS, moderate to severe hypoxic–ischemic encephalopathy, cerebral sinovenous and veins thrombosis, intracranial hemorrhage (not attributable to hemorrhagic transformation of AIS), prematurity (less than 37 weeks of gestation at birth), cranial surgery, intracranial tumor-related stroke,

https://doi.org/10.1016/j.yebeh.2018.08.012 1525-5050/© 2018 Elsevier Inc. All rights reserved.

Downloaded for Anonymous User (n/a) at University of Chile Catholic ALERTA from ClinicalKey.com by Elsevier on October 13, 2018. For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.

114

M. López-Espejo et al. / Epilepsy & Behavior 88 (2018) 113–116

and moderate to severe traumatic brain injuries. The institution's ethics committee approved this study. A pediatric neurologist obtained imaging and clinical data of acute event during the hospital stay. Demographics (age and sex), clinical manifestations (decreased level of consciousness, defined as Glasgow Coma Score b 12 points; headache in the prior 12 h; AS (defined as any clinical seizure occurring within 7 days of IIA [10]), and focal neurological deficits including motor, visual, or language impairment), underlying conditions (according to the International Pediatric Stroke Study [11]), and radiological stroke characteristics (infarct location, cerebral lobe involved, arterial territory involved, infarct number, and infarct laterality) were analyzed. Sixty-three children had vascular imaging (56 by magnetic resonance angiography). A radiologist evaluated all MRI sequences (diffusion-weighted image, fluid-attenuated inversion recovery, double inversion recovery, and T1 with gadolinium, using 1.5 T with 5-mm-thick slices, and 2.5 mm separation between cuts) and posteriorly were reviewed by two physicians (an experienced neuroradiologist and a pediatric neurologist) together. A pediatric neurologist obtained late seizures and electrophysiology data during at least an annual outpatient appointment or a home visit. Any documented unprovoked seizure that occurred after 1 week of IIA onset was encoded as a late seizure. All children with a late seizure had a standard interictal electroencephalogram (international system 10/20 and activated with hyperventilation and photostimulation) performed within the first two weeks after the seizure. PSE was defined by the occurrence of at least 2 unprovoked seizures occurring N24 h apart within 2 years after the stroke [12]. All statistical analyses were conducted using the IBM SPSS Statistics version 20 (IBM Corp., Somers, NY) software. The clinical presentation, radiological characteristics, sex, and categories of AIS Risk Factors (RFs) were expressed as both absolute and relative frequencies, and the age of onset and the follow-up period, as either means or medians and ranges, as appropriate. Cox proportional hazards regression analysis were used to estimate individual risk. Variables with P b 0.05 in the univariate regression analysis were subjected to a multivariate analysis in which we used the log-rank to compare Cox proportional hazard ratios

(HRs) and evaluate relative risk. A 2-sided P value b0.05 was regarded as statistically significant. 3. Results Among 107 patients, who had met the study criteria, 9 (8.4%) were lost or declined to participate. The study group consisted of 98 patients who suffer a first AIS in childhood, including 41 children (41.8%) who developed PSE and 57 children (58.2%) without PSE. Demographic, underlying conditions and stroke characteristics are summarized in Table 1. The median age of the children at the time of diagnosis was 2.16 years (interquartile range, 0.42–7.82). Time of the follow-up ranged from 0.12 to 8.49 years (mean, 3.92; standard deviation, 2.12). Among 41 patients with PSE, 28 (68.3%) were diagnosed during the first year after AIS, 8 (19.5%) at the second year, 4 (9.8%) at the third year, and 1 (2.4%) at the fourth year of follow-up. All children with PSE were treated with antiseizure medication (ASM). Thirty-one children (75.5%) had been treated with 1 ASM, 7 (17.1%) with 2 ASMs, and 3 (7.4%) with 3 ASMs. Despite the ASM number, six children (14.6%) presented at least one seizure a month. At least 1 AIS risk factor was documented in 90 (91.8%) children, and nearly half (56.2%) had multiple risk factors. The most common detected underlying conditions were acute infections, heart diseases, and other chronic conditions, which account for 38.8%, 37.8%, and 32.7% of the study cohort, respectively. By contrast, central nervous system (CNS) arteriopathies are relatively uncommon. These are documented in one-third of children who underwent a brain vascular imaging (31% of 63). Definable arteriopathies were small vessel disease secondary to CNS infection (50%), transient cerebral arteriopathy (25%), arterial dissection (15%), and moyamoya disease (10%). In the multivariate Cox analysis, age, AS, cortical involvement, and multifocal infarction were still independent factors associated with higher risk to develop epilepsy (Table 2). In contrast, neither arteriopathies nor involvement of the MCA territory is associated with an increased risk of PSE. Also, multivariate Cox regression analyses demonstrated that the adjusted risk of epilepsy decreased 8% for each 1-year increase in age.

Table 1 Baseline clinical and radiological characteristics of children with a first AIS. Frequency Clinical characteristics Demographics Age at onset b1 year Female Male Clinical manifestations Acute symptomatic seizures GCS b 12 at diagnosis Focal deficits Motor deficit Language deficit Visual deficit Cranial nerve deficit Headache Underlying conditions Cardiopathy Congenital Acquired CNS arteriopathy Infections Mild systemic infection CNS infections Shock Chronic conditions Neoplasms Anemia Prothrombotic states

Frequency

Number

Percentage

39 38 60

39.9 38.8 61.2

52 72

53.1 73.5

44 17 9 9 16

44.9 17.3 9.2 9.2 16.3

37 27 10 20 38 28 10 23 32 7 15 20

37.8 27.6 10.2 20.4 38.8 28.6 10.2 23.5 32.7 7.1 15.3 20.4

Stroke characteristics Infarct location Cerebral cortex Internal capsule Corona radiata Corpus callosum Basal ganglia Thalamus Cerebellum Brainstem Frontal Parietal Temporal Occipital Arterial territory involved Anterior cerebral artery Middle cerebral artery Posterior cerebral artery Vertebrobasilar Infarct number Unifocal Multifocal Laterality Bilateral Left hemisphere Right hemisphere Hemorrhagic transformation

Number

Percentage

60 28 67 30 38.8 25 6 8 74 46 22 36

61.2 28.6 68.4 30.6 38 25.5 6.1 8.2 75.5 46.9 22.4 36.7

42 65 37 11

42.9 66.3 37.8 11.2

36 62

36.8 63.3

50 28 20 11

51 28.6 20.4 11.2

Abbreviations: GCS: Glasgow coma scale; CNS: central nervous system; BG: basal ganglia.


M. López-Espejo et al. / Epilepsy & Behavior 88 (2018) 113–116 Table 2 Cox proportional hazard regression models of risk factors for post stroke epilepsy. Multivariate⁎

Univariate Characteristics

HR

95% CI

P value HR

Age (years) Sex (Female) Acute symptomatic seizures GCS b 12 at diagnosis Focal deficits Headache Cardiopathy Congenital Acquired CNS arteriopathy Infections CNS infections Mild systemic infections Shock Chronic conditions Neoplasms Anemia Prothrombotic states

0.90 0.97 4.62 1.94 0.64 0.44 0.59 0.72 0.46 2.32 1.75 2.44 0.51 1.23 0.57 0.58 1.05 0.72

0.83–0.98 0.52–1.82 2.12–10.1 0.89–4.19 0.34–1.19 0.16–1.24 0.29–1.21 0.33–1.55 0.11–1.91 1.23–4.38 0.95–3.24 1.16–5.14 0.21–1.23 0.59–2.57 0.27–1.20 0.14–2.42 0.44–2.51 0.30–1.71

0.01⁎ 0.92 b0.01⁎ 0.09 0.16 0.12 0.15 0.40 0.29 0.01⁎ 0.07 0.02⁎

Stroke characteristics Cortical involvement Internal capsule involvement Corona radiata involvement BG involvement Thalamus involvement Brainstem involvement Frontal lobe infarct Parietal lobe infarct Temporal lobe infarct Occipital lobe infarct ACA territory involvement MCA territory involvement PCA territory involvement VB territory involvement Multifocal infarction Bilateral infarction Left hemisphere infarction Right hemisphere infarction Hemorrhagic transformation

5.58 1.27 1.71 1.37 1.02 1.68 2.01 1.51 1.55 1.14 1.05 2.68 1.06 1.57 2.96 2.51 0.71 0.32 1.51

2.34–13.3 0.62–2.61 0.87–3.37 0.73–2.58 0.49–2.08 0.60–4.73 0.93–4.37 0.82–2.79 0.76–3.16 0.61–2.16 0.57–1.95 1.27–5.65 0.56–1.99 0.62–4.01 1.41–6.22 1.31–4.81 0.35–1.46 0.11–0.89 0.63–3.60

b0.01⁎ 0.51 0.12 0.33 0.96 0.32 0.08 0.19 0.23 0.68 0.87 0.01⁎

95% CI

0.91 0.84–0.99

P value 0.02⁎

3.29 1.35–8.01 b0.01⁎

1.11 0.46–2.71

0.81

1.52 0.53–0.19

0.24

5.01 2.00–12.6

b0.01⁎

3.27 1.01–10.8 0.40 0.13–1.24

0.05⁎ 0.11

0.33 0.10–1.08

0.07

0.13 0.59 0.14 0.46 0.90 0.46

0.87 0.34 b0.01⁎ b0.01⁎ 0.35 0.03⁎ 0.35

Abbreviations: GCS: Glasgow coma scale; CNS: central nervous system; BG: basal ganglia; ACA: anterior cerebral artery; MCA: middle cerebral artery; PCA: posterior cerebral artery; VB: vertebrobasilar. ⁎ Represents the P value ≤ 0.05.

4. Discussion The results of this study show that younger age at the time of the stroke, the occurrence of AS, the ischemic involvement of the cerebral cortex, and the presence of multiple brain infarcts confer an increased risk to develop epilepsy following a first AIS in childhood. This is the first cohort study in South American children exploring this topic. It is well-established that the occurrence of AS is the primary clinical risk factor in developing epilepsy after an AIS in childhood [3,13]. Our results are concordant with this idea. Also, we found a higher frequency of AS than previous studies, likely due to the high proportion of children younger than 2 years at the time of the stroke in our cohort. Even though there is an evident interrelation between AS and cortical infarction, our results show that both factors are independently associated with PSE. Thus, this finding supports the hypothesis that stroke-related variables are affecting the probability of developing PSE independent of the presence of AS. Cortical involvement is the most consistent predictor of poststroke seizures in adults. A recent metaanalysis in the adult population showed that cortical involvement (Odd's ratio [OR] = 3.71; 95% confidence interval [CI] = 2.34–5.90) is strongly associated with an increased risk of PSE [14]. There are studies in children showing the association between cortical infarction and epilepsy. However, only a few reports evaluate this relation using multivariate analyses [7,15].

115

We found that the presence of multifocal infarction is positively associated with PSE independent of the existence of cortical infarctions, or the occurrence of AS. Likely, this may be due to the involvement of different focal neuronal networks, which increased the propensity for evolving recurrent seizures. This finding suggests that the prognosis of children with multiple infarcts could be worse than those with a single lesion. However, it is necessary to consider the infarct size in future studies. The PSE incidence in our AIS children's cohort was 41%. Our data show a higher frequency of PSE compared with that of previous studies with larger number of patients. DeVeber et al. described an incidence of 11% in a prospective, national population-based study that included 691 children beyond neonate age with AIS [16], which is similar to the findings shown in a recent retrospective analysis conducted by Billinghurst et al., with a total of 105 children with AIS and a cumulative incidence of epilepsy at 2 years of 7% [5]. A possible explanation for this elevated incidence of epilepsy is the higher detection of bilateral (51%) and multiple infarcts (63.3%) compared with those of studies mentioned above (ranged 21 to 30% and 41 to 48%, respectively). Although previous studies have been shown that bilateral infarcts are associated with adverse neurological outcomes [8], we did not find a significant association between this variable and increased risk of PSE. About one-third (36.7%) of our children with a first AIS developed PSE during the subsequent two-year period. Moreover, about threethird of those patients evolved PSE during the first year after an AIS. Our results are according to the data from a previous study in which the cumulative incidence of epilepsy trends stabilized after the first 2 years of stroke [5]. Thus, these findings suggest that children have a higher propensity for developing recurrent seizures during the first months following an AIS than after this period. Probably, the selection of a predominantly inhospital cohort and the lower detection of patients with mild symptoms at the time of the AIS are biases that affect the enrollment of children without long-term neurological morbidity; thus, the rate of PSE from this cohort might not accurately reflect the country's statistics. The lack of information about infarct volume and clinical stroke severity is a limitation in our study. The strengths of this study are clear selection criteria controlled for confounding variables (excluding children with prestroke epilepsy and other disorders coursing with epilepsy), prospective recruitment, homogeneous evaluations, and scarce losses to follow-ups. To our knowledge, this is the first report of clinical and radiological predictors of epilepsy after a first AIS in predominantly Hispanic children beyond the neonatal age. In conclusion, young age at onset, AS, cortical infarction, and multiple infarctions are independent risk factors for PSE after a first AIS in childhood. Further studies are needed to explore the impact of these factors on the risk of PSE in different pediatric populations, including ambulatory care setting. Conflict of interest There is no conflict of interest to declare. Acknowledgments The authors thank Pontifical University Catholic of Chile's Clinical Hospital staff and participants for their important contributions to this work. References [1] Agrawal N, Johnston SC, Wu YW, Sidney S, Fullerton HJ. Imaging data reveal a higher pediatric stroke incidence than prior US estimates. Stroke 2009;40(11):3415–21. [2] Giroud M, Lemesle M, Gouyon JB, Nivelon JL, Milan C, Dumas R. Cerebrovascular disease in children under 16 years of age in the city of Dijon, France: a study of incidence and clinical features from 1985 to 1993. J Clin Epidemiol 1995;48(11): 1343–8.


116

M. López-Espejo et al. / Epilepsy & Behavior 88 (2018) 113–116

[3] Fox CK, Glass HC, Sidney S, Lowenstein DH, Fullerton HJ. Acute seizures predict epilepsy after childhood stroke. Ann Neurol 2013;74(2):249–56. [4] Hsu CJ, Weng WC, Peng SS, Lee WT. Early-onset seizures are correlated with late-onset seizures in children with arterial ischemic stroke. Stroke 2014;45(4):1161–3. [5] Billinghurst LL, Beslow LA, Abend NS, Uohara M, Jastrzab L, Licht DJ, et al. Incidence and predictors of epilepsy after pediatric arterial ischemic stroke. Neurology 2017; 88(7):630–7. [6] deVeber GA, MacGregor D, Curtis R, Mayank S. Neurologic outcome in survivors of childhood arterial ischemic stroke and sinovenous thrombosis. J Child Neurol 2000;15(5):316–24. [7] Yang JS, Park YD, Hartlage PL. Seizures associated with stroke in childhood. Pediatr Neurol 1995;12(2):136–8. [8] Greenham M, Gordon A, Anderson V, MacKay MT. Outcome in childhood stroke. Stroke 2016;47(4):1159–64. [9] Lopez-Espejo M, Hernandez-Chavez M. Prevalence and predictors of long-term functional impairment, epilepsy, mortality, and stroke recurrence after childhood stroke: a prospective study of a Chilean cohort. J Stroke Cerebrovasc Dis 2017;26 (7):1646–52.

[10] Beghi E, Carpio A, Forsgren L, Hesdorffer DC, Malmgren K, Sander JW, et al. Recommendation for a definition of acute symptomatic seizure. Epilepsia 2010;51(4):671–5. [11] MacKay MT, Wiznitzer M, Benedict SL, Lee KJ, Deveber GA, Ganesan V, et al. Arterial ischemic stroke risk factors: the International Pediatric Stroke Study. Ann Neurol 2011;69(1):130–40. [12] Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. Epilepsia 2014;55(4):475–82. [13] Mallick AA, Ganesan V, Kirkham FJ, Fallon P, Hedderly T, McShane T, et al. Outcome and recurrence one year after paediatric arterial ischaemic stroke in a populationbased cohort. Ann Neurol 2016;79(5):784–93. [14] Ferlazzo E, Gasparini S, Beghi E, Sueri C, Russo E, Leo A, et al. Epilepsy in cerebrovascular diseases: review of experimental and clinical data with meta-analysis of risk factors. Epilepsia 2016;57(8):1205–14. [15] Morais NMM, Ranzan J, Riesgo RS. Predictors of epilepsy in children with cerebrovascular disease. J Child Neurol 2013;28(11):1387–91. [16] Deveber GA, Kirton A, Booth FA, Yager JY, Wirrell EC, Wood E, et al. Epidemiology and outcomes of arterial ischemic stroke in children: the Canadian Pediatric Ischemic Stroke Registry. Pediatr Neurol 2017;69:58–70.
