|Year : 2008 | Volume
| Issue : 1 | Page : 117-120
Predictive factors of seizure control in childhood onset epilepsy
Eli Shahar, Jacob Genizi
Child Neurology Unit and Epilepsy Service, Meyer Children Hospital, Rambam Medical Center, Rappaport School of Medicine, Haifa, Israel
Child Neurology Unit and Epilepsy Service, Meyer Children Hospital, Rambam Medical Center, Haifa 31096
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Prediction of the long-term outcomes of childhood-onset epilepsy remains crucial for the future well-being of the affected children and their families and for planning proper therapeutic and educational programs. Objective: To identify and analyze the early predictive factors of seizure control in childhood-onset epilepsies referred at the age of 1 month up to the age of 18 years to the Epilepsy Service at the Meyer Children Hospital, Rambam Medical Center, Haifa, Israel. Materials and Methods: In this study, children who were newly diagnosed with epileptic disorders and treated with antiepileptic drug therapy - who became either completely controlled for at least twelve months or those remaining intractable - were included. Partially responding children were excluded from the analysis. The etiology was segregated into either symptomatic or nonsymptomatic epilepsy, referred to as idiopathic epilepsy. Results: Overall, 74 children (mean age: 4.27 ± 4 years at the first seizure) were found eligible for analysis followed for a mean period of 4.5 years. Fifty-three (72%) children became seizure-free for a mean period of 20 months on antiepileptic drug (AED) therapy and 21 (28%) remained uncontrolled. Sixty out of 74 children (81%) had idiopathic epilepsy and 14 (19%) had symptomatic epilepsy. In those with idiopathic epilepsy, 46 (77%) children gained complete seizure control in comparison to 7 out of 14 (50%) children in the symptomatic group ( P < 0.01). Thirty-nine out of 47 (83%) children who had normal cognition became seizure-free, and 14 (52%) out of 27 mentally retarded children also became seizure-free ( P < 0.01). The outcome of seizure control was not affected by age at onset and seizure type itself during the presentation. Conclusion: The present study reveals that the single most predictive factor of a favorable seizure control is preserved cognitive function in accordance with idiopathic epilepsy. However, a fairly high number of children with impaired cognition may also achieve satisfactory seizure control.
Keywords: Childhood-onset epilepsy, cognition, seizure control, symptomatic and idiopathic epilepsy
|How to cite this article:|
Shahar E, Genizi J. Predictive factors of seizure control in childhood onset epilepsy. J Pediatr Neurosci 2008;3:117-20
| Introduction|| |
Prediction of the long-term outcomes of childhood-onset epilepsy remains cardinal for the future psychological and emotional well-being of the affected children and their families. Awareness of the outcome profile in different risk groups among children with epilepsy may be extremely important for the individual patient who is attempting to design his personal treatment protocol as well as for the future planning of educational programs. A few recent studies attempted to sort out the denominators of outcome prediction in childhood-onset epilepsy taking into account the vast complexity of the practically endless epilepsy presentations at different ages throughout infancy and childhood. ,, Braathen and Melander  found that the strong predictive factors of seizure control involve the age at onset, seizure type and abnormal electroencephalography (EEG) findings. Recent wide-scale studies of outcome prediction in children with epilepsy include the Nova Scotia Study of Childhood Epilepsy  and the Dutch Study of Epilepsy in Childhood  that involved 1055 children. Both studies had prospectively followed-up children with newly diagnosed epilepsy and attempted to design models of outcome prediction using a multivariate analysis. Following an exhaustive process of analysis of many variable attributes, both the models of prediction managed to successfully predict the outcome in approximately 65% of patients. Outcome predictors were also different in both studies as was the case in the other ones: The research from Nova Scotia , outlines the following significant predictive attributes: age at onset, intelligence, neonatal seizures and number of seizures before treatment. The Dutch study  revealed that the underlying etiology, a history of febrile seizures and age at onset may serve as the strong outcome predictive factors. In fact, both the studies as well as others ,,,,, have outlined multiple, complex and perplexing system for seizure outcome prection and as such render to become impractical for the practicing pediatricians and nursing staff who take care of such children. A common simple query by parents refers to the actual control of seizures and using complex figures and statistics, as expressed in those outcome predictive models, may increase confusion and anxiety between the patients and their families. Therefore, a simple method of outcome prediction may be plausible in order to relieve the anxiety of parents and care-givers and plan an individualized treatment strategy on the basis of such simple legible outcome prediction measures for each child.
The present study is an attempt to delineate simple and legible predictive outcome attributes in childhood-onset epilepsy by considering different seizure presentations, the underlying etiology of either idiopathic or symptomatic epilepsy and also the overall response to AEDs.
| Materials and Methods|| |
Definitions of etiology of the epilepsy
The underlying etiology of the different epilepsies was classified according to the classification of the epilepsies and epileptic syndromes according to the International League against Epilepsy.  Remote symptomatic group: defined as seizure occurring with an identifiable previous brain injury or in the presence of a static encephalopathy known to be associated with increased risk of seizures. Symptomatic epilepsy: secondary to an underlying pathological pathology of the cortex, including cerebrovascular disorders, traumatic brain injury and congenital structural abnormalities of the cortex (such as cerebral dysgenesis, cerebral tumors and previous central nervous system infections). Both remote symptomatic and symptomatic epilepsies were clumped together. Idiopathic epilepsy: defined as unprovoked seizures occurring without a disorder or abnormality of the brain that may be associated with an increased risk of epilepsy.
Eligible for analysis
Group 1: Children completely controlled on antiepileptic medications for a period of at least 12 months or longer. Group 2: Children who remained intractable within a period of 12 months when subjected to AED therapy. Children who remained partially controlled within a period 12 months were excluded.
Included in the study
Children in the age group of 1 month-18 years with new onset epilepsy and who were treated with anticonvulsant treatment at the Epilepsy Service of the Meyer children's hospital, Rambam Medical Center, Haifa, Israel, between 1995 and 2004. The thorough history was taken for very patient and it included the following: 1. Seizure characterization: age at onset, frequency and duration, clinical manifestation including secondary generalization. 2. Course of pregnancy, labor and delivery and the postnatal course. 3. Response to antiepileptic drugs and final outcome. 4. Developmental Milestones and cognitive status.
The perinatal history was considered normal if the child was delivered at term without complications, requiring any neonatal intensive care unit observation. The comorbid conditions present at the onset included the presence of any developmental delay (global, isolated motor or speech delay), mental retardation, attention-deficit hyperactivity disorder and learning disabilities. Global developmental delay was defined as a significant delay in two or more of the following developmental domains: gross or fine motor, speech and language, social and interpersonal and activities of daily living.
Repeat surface recordings were performed applying the accepted International 10-20 System using bipolar and referential montages during wakefulness, drowsiness and if feasible during natural sleep following sleep deprivation, if required. A child was considered to have developed epilepsy if he experienced two or more consecutive seizures
Antiepileptic drug therapy
Selected as the "drug of choice" in accordance with the classification of the unique epileptic disorder taking into account the various clinical presentations as well as EEG abnormalities.
The correlations between seizure control and variants including gender, cognition, seizure description and epilepsy types were all tested by the chi-square test for trend in proportions (χ2 test) and with the Cramer correlation. T-test was used to investigate the difference between controlled and uncontrolled patients, analyzing the following variants: age, number of AEDs used, duration of disease and the ratio between number of AEDs and duration of disease. A logistic regression analysis was attempted to analyze the following possible predictive factors for the duration of seizure control: the seizure type, cognition, epilepsy type and number of AEDs.
| Results|| |
Seventy-four (74) children were found eligible for the study diagnosed at the age of 1 month to 8 years (mean age: 4.27 ± 4 years) and followed-up for a mean period of 4.5 years. Overall, 47 (63.5%) children were found to have cognition within the normal range and 27 (36.5%) children were diagnosed with mental retardation.
[Table - 1] delineates the association between the epilepsy type, either idiopathic or symptomatic. Within the overall group, 53 (72%) children became seizure-free for a mean period of 20 months on antiepileptic medications and 21 (28%) remained uncontrolled. In the idiopathic type, 46 (77%) children gained complete seizure control in comparison to 7 out of 14 (50%) children in the symptomatic group ( P < 0.01).
[Table - 2] delineates the correlation between complete seizure control accomplished and cognition: 39 (83%) of normally cognitive children gained complete seizure control in comparison to 14 out of 27 (52%) children with mental retardation ( P < 0.01). Thirty-eight (52%) children developed generalized seizure and 36 (48%) developed partial epilepsy. However, the seizure type at presentation imposed no statistically significant difference upon the outcome.
| Discussion|| |
Childhood epilepsy is a disorder that remits in as many as 60% of all the affected children allowing the discontinuation od antiepileptic medications. ,,,,,,, Sorting out those children who will grow out of their epileptic disorder with lack of cognitive impairment and vise versa, those who may remain intractable suffering from developmental handicaps is crucial for planning an individualized diagnostic and therapeutic program as well as relieving the anxiety of the parents and patients. Various complex models of outcome prediction are presently based on the clinical manifestations of possible epilepsy accompanied with EEG abnormalities. ,,, Such models however differ in various atrributes: The basic definitions of epilepsy, the semiology of different epilepsy syndromes and the period of seizure control considered remission. As such, a unified analysis of outcome predictors in pediatric epilepsy is presently non-existant. As such, complex and perplexing models of outcome prediction presenting different favorable prognostic factors may create confusion in between medical personnel and enhance the anxiety of the parents. Two wide-scale outcome prediction studies analyzing 1055 children are characteristic examples of such differences in the delineation of prediction attributes: The Dutch study  classified patients on the basis of epilepsy types, syndromes and etiology according to the International League against Epilepsy (ILAE) definitions and that the etiology, a history of febrile seizures and age at presentation serve as most important predictive factors of outcome. The Nova Scotia study,  being an earlier study, lacked the delineations into unique epileptic syndromes and also excluded and children with absence epilepsy and symptomatic generalized epilepsies; this study sorted out that the age at onset, intelligence, neonatal seizures and the numbers of seizures prior to treatment served as the best predictive attributes of outcome. Geelhoed et al .  have recently assessed the accuracy of outcome prediction models of childhood-onset epilepsy by merging both wide-scale Dutch  and Nova Scotia  studies for the purpose of analysis had found that the most consistent predictors of outcome were the epilepsy type, age at onset (older than 10-12 years) and the presence of neurological or cognitive deficits. Even by following such an exhaustive and time-consuming survey of outcome prediction, successful prediction was eventually accomplished in 66-69% of all the children in these wide-scale data survey. The authors also added a cumulative survey of more than 2000 children of six wide-scale extensively analyzed pediatric cohorts of epilepsy outcome prediction, further denoting the marked variability of attributes of seizure outcomes across the various studies and eventually allowing a correct prediction of favorable outcomes in totally 60% of children.
The present study was therefore delineated in order to design a simplified practical model that aimed to identify the early predictive factors of seizure control in children with new-onset epilepsy requiring anticonvulsant therapy being free of seizures for 12 months or longer. Therefore, we can avoid exhaustive, time-consuming, expensive and frustrating complex prediction models that would not allow a better prediction of seizure control. Seventy-four children were analyzed for a mean period of 4.5 years; the overall seizure control was satisfactory and 75% of the children were seizure-free for at least 12 months. These figures of immediate seizure control are in concert with the previous studies, reporting approximately 70% remission rate in all newly diagnosed children. ,,,, Smaller figures were reported by Oskoui et al. ,  who found that only 53% of the epileptic children became seizure-free. As for early seizure remission and the overall outcome, different authors are in agreement that the chance of future remission is significantly reduced if the seizures are not controlled within the first year of treatment. ,, In a recent study, Sillanpaa and Schmidt  state that early remission does not necessarily guarantee a complete remission of seizures in the future and on the other hand the initial failure of drug therapy does not necessarily predict a poor outcome.
We examined the correlations between seizure control and the following variants: epilepsy type, cognition, seizure type and age at onset. Tang-Wai et al .  have proposed that although cryptogenic epilepsy is presumably very similar to symptomatic epilepsy without a genetic or other identifiable cause, its outcome is favorable as idiopathic epilepsy in comparison to children diagnosed with "pure" symptomatic epilepsy. Following this line of evidence, we have segregated our patients into two groups: the first group included children diagnosed as having idiopathic/cryptogenic epilepsy and the second group included children with "pure" symptomatic epilepsy. In our experience, 80% of children in the nonsymptomatic, I.E. idiopathic/cryptogenic group achieved seizure control in comparison to 50% in the symptomatic group. Camfield et al.  and later Sillanpaa and Schmidt  noted that the underlying etiology, either symptomatic or nonsymptomatic epilepsy is the single most important predictive factor for outcome and our results are indeed in agreement with their observations. As for cognition, in our study, 83% of normally cognitive children became seizure-free in comparison to 52% of the mentally retarded children who became seizure-free. Similar predictive associations between the overall cognitive state and seizure control were also previously reported. ,, Berg et al .  analyzed a cohort of 594 children and reported a remission rate of 74% children; in this group, children with idiopathic etiology and relatively higher cognition had the highest rate of remission. Therefore, assessment of the epilepsy type, either symptomatic or non-symptomatic, along with the cognitive state of the affected child may both serve as simple to assess guidelines of seizure control prediction to be conveyed to the parents as well as medical personnel.
In summary, the models of seizure control prediction are complex and the analysis results differ across various studies. In our study, we attempted to delineate easily understandable simple measures of outcome prediction. Both the best predictive measures in the present study included the cognitive state of the affected child and concordantly the type of epilepsy. As such, normally developed children with nonsymptomatic epilepsy may have a very high chance of more than 80% to become seizure-free on AED therapy. Although children with symptomatic epilepsies may have a lower chance to become seizure-free, a fairly high percentage of them, around 50%, may also become seizure-free and such an occurrence may serve as a comforting message for the future for mentally retarded children. Our study strongly suggests that the outcome of childhood epilepsy is overall favorable as seizure control will be achieved mainly in children with preserved cognition, as observed by others. Since the current complex and perplexing models of outcome prediction fail to predict the outcome in one out of every three children,  further longitudinal multi-center prospective studies using identical attributes of data collection, seizure-type delineation and seizure control analysis may contribute to the overall accurate prediction of seizure control in pediatric-onset epilepsy.
| References|| |
|1.||Braathen G, Melander H. Early discontinuation of treatment in children with uncomplicated epilepsy: A prospective study with a model for prediction of outcome. Epilepsia 1997;38:561-9. [PUBMED] |
|2.||Camfield CS, Camfield PR, Gordon K, Wirrell E, Dooley JM. Incidence of epilepsy in childhood and adolescence: A population-based study in Nova Scotia from 1977 to 1985. Epilepsia 1996;37:19-23. [PUBMED] |
|3.||Geerts AT, Arts WF, Brouwer OF, Peters AC, Peeters EA, Stroink H, et al . Validation of two prognostic models predicting outcome at two years after diagnosis in a new cohort of children with epilepsy: the Dutch Study of Epilepsy in Childhood. Epilepsia 2006;47:960-5. [PUBMED] [FULLTEXT]|
|4.||Arts WF, Brouwer OF, Peters AC, Stroink H, Peeters EA, Schmitz PI, et al . Course and prognosis of childhood epilepsy: 5-year follow-up of the Dutch study of epilepsy in childhood. Brain 2004;127:1774-84. [PUBMED] [FULLTEXT]|
|5.||Camfield C, Camfield P, Gordon K, Smith B, Dooley J. Outcome of childhood epilepsy: A population-based study with a simple predictive scoring system for those treated with medication. J Pediatr 1993;122:861-8. [PUBMED] |
|6.||Oskoui M, Webster RI, Zhang X, Shevell MI. Factors predictive of outcome in childhood epilepsy. J Child Neurol 2005;20:898-904. [PUBMED] |
|7.||Chadwick D, Taylor J, Johnson T. Outcomes after seizure recurrence in people with well-controlled epilepsy and the factors that influence it. Epilepsia 1996;37:1043-50. [PUBMED] |
|8.||Sillanpaa M, Schmidt D. Prognosis of seizure recurrence after stopping antiepileptic drugs in seizure-free patients: A long-term population-based study of childhood-onset epilepsy. Epilepsy Behav 2006;8:713-9. |
|9.||Tang-Wai R, Oskoui M, Webster R, Shevell M. Outcomes in pediatric epilepsy: Seeing through the fog. Pediatr Neurol 2005;33:244-50. [PUBMED] [FULLTEXT]|
|10.||Brorson LO, Wranne L. Long-term prognosis in childhood epilepsy: survival and seizure prognosis. Epilepsia 1987;28:324-30. [PUBMED] |
|11.||Proposal for revised classification of epilepsies and epileptic syndromes: Commission on Classification and Terminology of the International League Against Epilepsy. Epilepsia 1989;30:389-99. |
|12.||Berg AT, Shinnar S, Levy SR, Testa FM, Smith-Rapaport S, Beckerman B, et al . Two-year remission and subsequent relapse in children with newly diagnosed epilepsy. Epilepsia 2001;42:1553-62. [PUBMED] [FULLTEXT]|
|13.||Geelhoed M, Boerrigter AO, Camfield P, Geerts AT, Arts W, Smith B, et al . The accuracy of outcome prediction models for childhood-onset epilepsy. Epilepsia 2005;46:1526-32. [PUBMED] [FULLTEXT]|
|14.||Hauser E, Freilinger M, Seidl R, Groh C. Prognosis of childhood epilepsy in newly referred patients. J Child Neurol 1996;11:201-4. [PUBMED] |
|15.||Annegers JF, Hauser WA, Elveback LR. Remission of seizures and relapse in patients with epilepsy. Epilepsia 1979;20:729-37. [PUBMED] |
|16.||Reynolds EH. The early treatment and prognosis of epilepsy. Jpn J Psychiatry Neurol 1988;42:429-35. [PUBMED] |
|17.||Wakamoto H, Nagao H, Hayashi M, Morimoto T. Long-term medical, educational and social prognoses of childhood-onset epilepsy: A population-based study in a rural district of Japan. Brain Dev 2000;22:246-55. [PUBMED] [FULLTEXT]|
[Table - 1], [Table - 2]