Characteristics and Prognosis of Epilepsy in Children With Cerebral Palsy Dimitrios I.

Zafeiriou, MD, PhD; Eleftherios E. Kontopoulos, MD, PhD; Ioannis Tsikoulas, MD, PhD

ABSTRACT The aims of the study were to describe the prevalence and characteristics of epilepsy in a population of patients with cerebral palsy in a university referral center and to determine the rate of relapse caused by discontinuation of antiepileptic drugs after a 3-year seizure-free period. A total of 178 consecutive patients with cerebral palsy and epilepsy were prospectively followed for 9.2 ± 2.4 years after onset of seizures and compared to a control group of 150 epileptic patients without cerebral palsy (median follow-up period, 10.5 years). The overall prevalence of epilepsy was 36.1%. Patients with atonicdiplegic, dystonic, tetraplegic, and hemiplegic cerebral palsy had a higher incidence of epilepsy (87.5%, 87.1%, 56.5%, and 42%, respectively). In all, 134 (75.3%) patients were seizure free for more than 3 years and could discontinue therapy, whereas 44 patients (24.7%) were still on antiepileptic drugs. Eighteen of 134 patients relapsed after a 3-year seizure-free period and subsequent discontinuation of antiepileptic drugs, thus giving a relapse rate of 13.4%. First seizures occurred during the first year of life in 69.7% of the patients with epilepsy and cerebral palsy. Complete control of seizures could be achieved in 65.2% of the patients with cerebral palsy and epilepsy; however, regardless of the prognosis of seizures, epilepsy was a major prognostic factor regarding both the presence of mental retardation and the motor development of children with cerebral palsy. (J Child Neurol 1999;14:289-294).

Epilepsy is one of the most prevalent neuroimpairments in childhood, occurring in approximately 5 in 1000 children,.’1 The rate of epilepsy in children with cerebral palsy is strikingly higher, ranging from 15% to more than 60%, depending on the type of cerebral palsy and the origin of the series.’ In these children, epilepsy is an index of the severity of cerebral palsy’; associated disabilities such as mental retardation are much more common in patients with cerebral palsy with epilepsy than in those without seizures.3--7 The presence of epilepsy seems to be a more predictive factor of mental development than the extent of the brain damage. Epilepsies associated with cerebral palsy are difficult to control,8 although remission, even in the presence of brain damage, can occur.9 However, there is still controversy concerning the optimal seizure-free period needed

Received March 25, 1998. Received revised Sept 18, 1998. Accepted for publication Oct 1, 1998. From the First Pediatric Greece.

Clinic,

Aristotle

University

of

Thessaloniki,

Thessaloniki,

Address correspondence to Dr Dimitrios I. Zafeiriou, Child Neurologist, Egnatia St. 106, 54622 Thessaloniki, Greece. Tel./fax: 30-31-241845; e-mail:

[email protected].

before

discontinuing antiepileptic drugs, since a high rate

relapse was noted in two previous studies, in which antiepileptic drugs were discontinued after a seizure-free period of 2 years.10,11 In the present retrospective study, the aims were to describe the characteristics of epilepsy in patients with cerebral palsy and to determine the relapse rate due to discontinuation of antiepileptic drugs after a 3-year seizure-free period. of

DEFINITIONS

Epilepsy was defined as two or more epileptic seizures, unprovoked by an immediate, identifiable cause and excluding neonatal seizures (ie, seizures occurring in the first month of life) and febrile seizures. 12 The diagnosis of epilepsy was based on historical records, clinical description, and electroencephalogram (EEG) fmdings. Seizure types were classified according to the criteria of the International League Against Epilepsy. 13 The partial seizure group was accordingly subdivided into simple, complex, and secondary generalized seizure types, and the generalized seizure group into typical and atypical absence, myoclonic and/or clonic, tonic-clonic, and atonic seizure types.&dquo;3 Although some patients with generalized tonicclonic seizures actually might have had secondary generalized 289

290

tonic-clonic them

on

seizures, there

clinical

was no

possibility of distinguishing

EEG grounds. Infantile spasms and stadefined according to the criteria of the Inter-

or even

epilepticus were League Against Epilepsy. 14 Status epilepticus was considered from age 1 month until the prevalence day and did not include infantile spasms. Frequent seizures were defined as seizures occurring every day or week, infrequent seizures as seizures occurring less frequently. Cerebral palsy was defmed as a nonprogressive disorder of tus

national

movement and posture caused

by a defect or lesion of the immaaccording to Bax, with a slight modification. 15 Impairments and disabilities were defmed according to World Health Organization criteria in which impairment repreture brain and

was

classified

disability the conseimpairments for function and activities. Mental retardation in the present study corresponded to an IQ of 70 or less. Prenatal was defined as the period before the start of labor resulting in delivery, perinatal as the period from birth through the first week of life, and postnatal as the period after the first week sents disturbances at the organ level and

quence of

of life. METHODS AND SUBJECTS The series studied consisted of 178 consecutive children with

epilepsy and cerebral palsy,

who

prospectively followed for (9.2 -!- 2.4 years) at the Department of Developmental Medicine and Child Neurology. The patients were derived from a total population of 493 children with cerebral palsy, thus giving an overall prevalence of epilepsy in cerebral palsy patients of 36.1%. A total of 150 consecutive epileptic patients without cerebral palsy seen at the Epilepsy Outpatient Department and followed prospectively for at least 5 years (10.5 ± 2.1 years) formed the control group. These patients had been diagnosed as having epilepsy using the same criteria and classification as for the patients with cerebral palsy. Excluded from the control group were patients with neonatal seizures only, febrile seizures, or benign epilepsy with centrotemporal spikes. In our department, the policy regarding antiepileptic drug discontinuation requires at least 3 seizure-free years for almost all types of epileptic syndromes (excluding neonatal seizures and benign epilepsy with centrotemporal spikes, for which the seizure-free period needed prior to antiepileptic drug discontinuation can be shorter). For this reason, we felt that although not fully representative, the control group of epileptic children could be considered large enough to provide a meaningful comparison with epileptic children with cerebral palsy. Epilepsy outcome was assessed according to the children’s status at their last follow-up visit. Both groups were compared regarding sex, age at onset of seizures, epilepsy and seizure types, presence of neonatal seizures or status epilepticus, frequency of seizures, need for mono- or polytherapy, EEG fmdings, computed tomography (CT) or magnetic resonance imaging (MRI) findings, presence of other impairments and disabilities (including the ability for independent walking), and recurrence rate after antiepileptic drug discontinuation, to determine the role of cerebral palsy in a given epileptic population. Furthermore, etiology and type of cerebral palsy, as well as associated impairments and disabilities, were were

at least 5 years after onset of seizures

separately analyzed for epileptic versus nonepileptic patients with cerebral palsy. Statistical analysis was performed using the Xz test and Fisher’s exact test; a regression analysis model was used to determine prognostic risk factors for relapse of seizures in children with cerebral palsy and epilepsy. There was no difference in gender distribution between the two groups (73 boys and 105 girls, versus 66 boys and 84 girls, in the group of patients with cerebral palsy and the control group, respectively). Regarding the onset of seizures, 71.3% of the patients with epilepsy and cerebral palsy had their first seizure in the first year of life, which was significant compared to the control group (5.3%; P < .001; Figure 1). The etiology of cerebral palsy was prenatal in 69 (38.8%) patients, perinatal in 59 (33.1%), postnatal in 19 (10.7%), and unknown in 31 (17.4%) patients. Classification according to seizure type (Table 1) demonstrated that simple partial and generalized tonic-clonic seizures dominated in epileptic patients with cerebral palsy. In the control group, children less frequently had simple partial, tonic, and secondary generalized tonic-clonic seizures (P < .001) and more frequently had generalized tonic-clonic seizures and absence seizures (P < .001). Classification according to epilepsy type in children with cerebral palsy (Table 2) disclosed that 85.4% had symptomatic localization-related epilepsies and 14.6% had cryptogenic or symptomatic generalized epilepsies (P < .001 and P < .01, respectively, compared to the control group). In the control group, cryptogenic localization-related epilepsies, idiopathic generalized epilepsies, and special syndromes were present more frequently compared to the whole group of epileptic patients with cerebral palsy (P < .Ol, P < .001, and P < .001, respectively). Neonatal seizures preceded epilepsy in 35 (1J.7°1o) patients with cerebral palsy and in 11 (7.3%) patients of the control group (P < .001). In contrast, status epilepticus was present in 6.7% and 4.7% of the cerebral palsy group and the control group, respectively (results not significant).

Figure 1. Age at onset of seizures in epileptic children with cerebral palsy (Group A) and in the control group (Group B).

291

Table 1. SeizureTypes in Epileptic Children With Cerebral Palsy and Controls

Table 3. Electroencephalogram Findings in Epileptic Children With Cerebral Palsy and Controls

Some patients exhibited * P < .05. Some patients exhibited *p < .001.

more

than

one

seizure type.

Polytherapy was needed for 94 (52.8%) patients with cerebral palsy and 44 (29.3%) patients in the control group (P < .001). Ninety-three (52.2%) patients with epilepsy and cerebral palsy and 47 (31.3%) patients in the control group had frequent seizures, without any significance in favor of either group. As expected, patients receiving polytherapy in both groups had, in the vast majority, frequent seizures (72.3% and 90.9%, respectively). Regarding the EEG findings (Table 3), children in the control group had more frequent normal (P < .O1) or focal epileptiform EEG findings (P < .05) and generalized activity (P < .01), and less frequent focal slowing (P < .001) or generalized slowing (P < .001), compared to the group of epileptic patients with cerebral palsy. Both epileptic and nonepileptic children with cerebral palsy demonstrated more frequently all impairments and disabilities analyzed, compared to the control group (Table 4). The presence of epilepsy in children with cerebral palsy was correlated statistically with an increased frequency of speech problems (P < .O1), mental retardation (P < .001), myoskeletal problems (P < .001), and an inability to walk (P < .001). Comparing children with epilepsy, cerebral palsy, and mental retardation (97 patients) to children with cerebral palsy and mental retardation only (102 patients), the results were highly significant regarding the inability to walk (48 versus 34 patients; P < .001) in favor of the first group. The above

more

than

one

finding.

’P < .01. $P< 001.

results could have been influenced somewhat by different degrees of mental retardation in the two groups (42 of 48 patients with epilepsy, cerebral palsy, and mental retardation versus 17 of 34

patients with cerebral palsy and mental retardation only, respectively, had IQs below 50, which was statistically significant; P < .01). Cortical, central, or combined atrophy and an infarct picture were the most frequently demonstrated neuroimaging findings in children with epilepsy and cerebral palsy (Table 5). Control patients demonstrated cortical or combined atrophy and infarct less frequently and normal findings more frequently, than epileptic patients with cerebral palsy (P < .001). RESULTS

The overall prevalence of epilepsy in the population of children with cerebral palsy was 36.1%. Patients with atonicdiplegic, dystonic, tetraplegic, and hemiplegic cerebral palsy had higher prevalences of epilepsy (87.5%, 87.1%, 56.5%, and 42%, respectively; Figure 2). A total of 134 of 178 patients with cerebral palsy could discontinue antiepileptic drugs successfully after a 3-year seizure-free period (overall remission rate, 75.3%); 116 (86.6%) of these 134 patients discontinued antiepileptic drugs without relapsing in a follow-up period of 5.8 ± 1.2 years, while the remaining 18 patients relapsed in the months following antiepileptic drug discontinuation,

Table 2. EpilepsyTypes in Children With Cerebral Palsy and Controls Table 4.

*p< .001. TP < .01.

Impairments and Disabilities

Some patients exhibited * P < .001.

more

than

one

impairment.

in Three

Groups

292

Table 5.

Findings

ComputedTomagraphic/Magnetic in

Resonance

imaging

Epileptic Children With Cerebral Palsy and Controls

*P < .001.

thus giving a relapse rate of 13.4%. Forty-four (24.7%) patients with cerebral palsy were still having seizures or were seizure free for less than 3 years at the end of the study. From the control group, 121 (80.7%) children were seizure free after 3 years (P < .01), while the relapse rate was 4.7% (P < .05). Eleven (61.1%) of 18 patients who relapsed after antiepileptic drug discontinuation had hemiplegic cerebral palsy, 4 (22.2%) had dystonic cerebral palsy, and 3 (16.7%) had diplegic cerebral palsy. Seventeen (94.4%) patients relapsed during the first year after antiepileptic drug discontinuation and one (5.6%) after 17 months; all those who relapsed had localization-related symptomatic epilepsy with focal seizures (n 15; 83.3%), except for two (11.1%) patients with generalized seizures and one (5.6%) patient with infantile spasms. Regarding neuroimaging findings, 13 patients demonstrated an infarct picture, 4 cortical atrophy, and 1 a congenital malformation. Prognostic factors for relapse included hemiplegic cerebral palsy (P < .001), localizationrelated symptomatic epilepsy (P < .001), focal seizures (P < .001), and an infarct picture in neuroimaging (P < .001). =

DISCUSSION

The

frequency of epilepsy varies with the form of cerebral

palsy. In patients with congenital hemiplegia, incidence fig-

Figure 2. Epilepsy in the various types of cerebral palsy. Group A, epileptic children with cerebral palsy; Group B, nonepileptic children with cerebral palsy. Numbers in bars indicate number of patients in each group.

of 27%, 17 37.4%, 18 and 44%5 are commonly quoted, although the frequency tends to be lower in children with spastic diplegia (especially those bom at term )19 ranging from 16%19 to 27%.2° The epilepsy incidence is definitely higher in patients with tetraplegic cerebral palsy (50% to 90%),17 whereas in patients with dystonic or athetoid cerebral palsy it is estimated to be 32% and 11%, respectively 2l As well, in patients with atonic diplegia (a form of cerebral palsy that has been incorporated into the hypotonic cerebral palsy group in all recent cerebral palsy series), the epilepsy incidence reaches 64.3%. 22 The high incidence of epilepsy in this form of cerebral palsy is the reason it is given separate ures

mention in the cerebral palsy classification used in the present study. In this study, atonic diplegia, tetraplegic cerebral palsy, and dystonic cerebral palsy had higher prevalences of epilepsy, as expected; patients with hemiplegic cerebral palsy had the highest relapse rate of all cerebral palsy forms (61.1%), as in the study of Delgado et al.llI Epilepsy in patients with cerebral palsy usually begins in the first years of life, with 36.7%1° to 46.6%~ of patients having their first seizure in their first 12 months. In our study population, 69.7% of the children had their first seizure within the first year of life. It remains difficult to determine a cut-off point beyond which it would be unlikely for epilepsy to

occur.

Most

epileptic patients with cerebral palsy suffer, as from localization-related epilepsies2; focal and genexpected, eralized seizures dominate the clinical picture.2 On these points our study was fully in agreement with the literature. The interictal EEG in these patients has been studied and documented extensively, 23-11 showing almost exclusively focal findings or generalized slowing. In the present study, apart from the abovementioned EEG findings, a considerable proportion of patients demonstrated primary interictal generalized epileptiform activity. The latter phenomenon can be attributed either to a genetic predisposition, or to a quick secondary bilateral synchronization such as the one sometimes induced by a frontal focus. On rare occasions, deep subcortical cerebral lesions also can generate this kind of epileptiform actiVity.31 Skatvedt3l found a remission of epilepsy in 43.5% of 46 children with cerebral palsy after 1 year of follow-up and Sillanpaa.,3~ also after 1 year of follow-up, noticed a 75% to 100% reduction of seizure frequency in 25% of children with cerebral palsy and epilepsy (51 patients). Brorson and Wranne 33 estimated that only half of patients with brain damage and epilepsy can be treated succesfully with antiepileptic drugs. Sixty-three percent of the children in their study with epilepsy and mental retardation or cerebral palsy were seizure free after 12 years of follow-up; the previous figure declined to 37% for epileptic children who had both cerebral palsy and mental retardation. After a 2-year follow-up period, Aksul° concluded that antiepileptic drugs could be discontinued in only 8 (14.0%) of 57 patients with cerebral palsy and epilepsy, with 5 of the 8 having relapses in the first 12 months after withdrawal of antiepileptic drugs. In a study on epilepsy and severe cerebral palsy, Traverse et al~~ con-

293

cluded that two thirds of children achieved either complete or acceptable control of their seizures; however, only half of the children achieved seizure control 12 years after the onset of the epilepsy, 60% of them without antiepileptic drugs. Delgado et alll proposed that discontinuation of antiepileptic drugs in children with cerebral palsy should be practiced when possible after patients have been seizure free for at least 2 years; however, 41.5% of their patients had seizure relapses, with patients with spastic hemiplegia having the highest relapse rate (61.5%). In our study the relapse rate was only 13.4%, a figure that can be attributed to the longer seizure-free period required before discontinuing antiepileptic drugs, and that provides evidence that the symptomatic localization-related epilepsies of cerebral palsy need at least 3 seizure-free years prior to antiepileptic drug discontinuation. Indeed, the fact that 94.4% of the relapses occurred during the first year after antiepileptic drug discontinuation allows us to conclude that a 4-year seizure-free period before antiepileptic drug discontinuation is necessary for children with cerebral palsy and epilepsy, especially for those with the hemiplegic form of cerebral palsy. As previously

mentioned, epilepsy per se is an indica-

palsy. The deleterious

compared to those without.9,35-37 In the (1) epilepsy was significantly more common in children with cerebral palsy and mental retardation than in children with cerebral palsy without mental retardation, (2) there were significantly more patients with cerebral palsy and epilepsy unable to walk than in the rest of the cerebral palsy group (without seizures), and (3) inability to walk was found to be significantly more frequent in children with cerebral palsy, mental retardation, and epilepsy, compared to children with cerebral palsy and mental retardation only. It has been postulated that the presence of specific risk factors predisposes to the co-occurrence of mental retardation, cerebral palsy, and epilepsy. From the results of this study we conclude that epilepsy is a major prognostic factor regarding both the presence of mental retardation and the motor development of children with cerebral palsy.

mental retardation, current study,

8.

Aicardi J:

1-71.

Epilepsy in brain-injured children. Dev Med Child Neurol 1990;32:191-202. 9. Goulden KJ, Shinnar S, Koller H, et al: Epilepsy in children with mental retardation: A cohort study. EpiLepsia 1991;32:690-697. 10. Aksu F: Nature and prognosis of seizures in patients with cerebral palsy. Dev Med Child Neurol 1990;32:661-668. 11. Delgado MR, Riela AR, Mills J, Pitt A: Discontinuation of antiepileptic drug treatment after two seizure-free years in children with cerebral palsy. Pediatrics 1996;97:192-197. 12. Commission on Epidemiology and Prognosis of the International League Against Epilepsy: Guidelines for epidemiologic studies on epilepsy. Epilepsia 1993;34:592-596. 13. Commission on Classification and Terminology of the International League Against Epilepsy: Proposal for revised clinical and electroencephalographic classification of epileptic seizures. EpiLepsia 1981;22:489-501. 14. Commission on Classification and Terminology of the International League Against Epilepsy: Proposal for revised clinical classification of epilepsies and epileptic syndromes. Epilepsia 1989;30: 389-399. 15.

Bax MCO: Terminology and classification of cerebral Med Child Neurol 1964;6:295-297.

epilepsies

in children and adolescents.

World Health Organization: International Classification of Impairments, Disabilities and Handicaps. Geneva, World Health Organization, 1980. 17. Aicardi J: Diseases of the Nervous System in Childhood. London, Mac Keith Press, 1992. 18. Zafeiriou DI, Kontopoulos E, Tsikoulas I, Anastasiou A: Epilepsy and EEG findings in congenital hemiplegia: Long-term outcome, abstract. Epilepsia 1996;37(Suppl 4):110. 19. Veelken N, Hagberg B, Hagberg G, Olow I: Diplegic cerebral palsy in Swedish term and preterm children—differences in reduced optimality, relations to neurology and pathogenetic factors. Neu-

20. 21. 22.

ropediatrics 1983;14:20-28. Ingram TTS: The neurology of cerebral palsy.

2.

Epilepsy and cerebral palsy, in Aicardi J (ed): Epilepsy Children, 2nd ed. New York, Raven Press, 1994, pp 350-351. Trevathan E, Yeargin-Allsopp M, Murphy CC, Ding G: Epilepsy

25.

Gibbs

26.

Perlstein MA: The electroencephalogram in infantile cerebral palsy. Proc Res Nerv Ment Dis 1946;26:377-384.

27.

Perlstein MA: The electroencephalogram in infantile cerebral palsy. Am J Phys Med 1955;34:4 77-496. Skatvedt M: The electroencephalographic findings and the problem of epilepsy in a cerebral palsy material of 370 patients. Acta Paediatr 1955;43(Suppl 103):22-23. Zafeiriou DI: EEG patterns of 103 children with cerebral palsy and epilepsy: Correlation with the prognosis of seizures, abstract. Epilepsia 1992;32(Suppl 1):21.

Aicardi J:

among children with mental retardation. Ann Neurol

1988;24:321.

4.

Uvebrandt P: Hemiplegic cerebral palsy. Aetiology and outcome. Acta Paediatr Scand 1988;77(Suppl 345):5-100.

5.

Goutieres F, Challamel MJ, Aicardi J,

Vargha-Khadem F, Isaacs EB, Van der Werf S, et al: Development of intelligence and memory in children with hemiplegic cerebral

EL, Gibbs FA: Electroencephalographic prediction of epilepsy as complication of cerebral palsy. Neurology 1963b;13: 143-145.

28.

Gilly R: Les hémiplegies congenitales : semiologie, etiologie et prognostic. Arch Pediatr 1972;29: 839-851.

6.

1950;37:448-454.

Gibbs FA, Gibbs EL: Electroencephalographic and clinical aspects of cerebral palsy. Pediatrics 1963a;32:73-84.

in 3.

Arch Dis Child

1966;41:337-357. Kyllerman M: Dyskinetic Cerebral Palsy. Doctoral diss, Dept of Pediatrics, University of Göteborg, Göteborg, Sweden, 1981. Zafeiriou DI: Course and Prognosis of Epilepsy in Children With Infantile Cerebral Palsy. Doctoral diss, Institute of Social Pediatrics and Adolescent Medicine, Ludwig-Maximilian University of Munich, Munich, Germany, 1993. Aird RB, Cohen P: Electroencephalography in cerebral palsy. J

24.

Epilepsia 1989;30:

94-106.

palsy. Dev

16.

Pediatr

the

seizures. Brain

Lagergren J: Children with motor handicaps: Epidemiological, medical and sociopaediatric aspects of motor handicapped children in a Swedish country. Acta Paediatr Scand 1981;70(Suppl 289):

23.

References 1. Cowan LD, Bodensteiner LB, Leviton A, Doherty L: Prevalence of

early

7.

tor of the

severity of cerebral palsy. From prospective and epidemiologic studies of cohorts of patients with mental retardation, it is also well documented that the risk of epilepsy is much higher in patients with cerebral palsy and

consequences of

1992;115:315-329.

29.

294

Niedermeyer E: Abnormal EEG patterns (epileptic and paroxysmal), in Niedermeyer E, DaSilva FL (eds): Electroencephalography: BasicPrinciples, Clinical Applications and Related Fields. Baltimore, Urban & Schwarzenberg, 1987, pp 405-510. 31. Skatvedt M: Cerebral palsy: A clinical study of 370 cases. Acta Pae-

30.

35.

Corbett JA, Harris RG, Robinson R: Epilepsy, in Wortis J (ed): Mental Retardation and Developmental Disabilities. New York, Bruner Mazel, 1975, pp 79-111.

36.

Forsgren L, Edvinsson SO, Blomquist HK, et al: Epilepsy in a population of mentally retarded children and adults. Epilepsy Res 1990;6:234-248.

37.

Steffenburg U, Hagberg G, Kyllerman M: Characteristics of seizures in a population-based series of mentally retarded children with active epilepsy. Epilepsia 1996;37:850-856.

38.

Curatolo

diatr

32. 33.

34.

1958;46(Suppl 111):72-83. Sillanpaa M: The significance of motor handicap in the prognosis of childhood epilepsy. Dev Med Child Neurol 1975; 17:52-57. Brorson LO, Wranne L: Long-term prognosis in childhood epilepsy: Survival and seizure prognosis. Epilepsia 1987;28:324-330. Traverse L, Dravet C, Roger J, et al: Epilepsy in children with severe cerebral palsy: Outcome and evolution. Dev Med Child Neurol 1994; 36(Suppl 70):32-33.

P, Arpino C, Stazi MA, Medda E: Risk factors for the coof partial epilepsy, cerebral palsy and mental retardation. Dev Med Child Neurol 1995;37:776-782. occurrence