- Email: [email protected]
Autism and epilepsy: A retrospective follow-up study
Brain & Development 29 (2007) 486–490 www.elsevier.com/locate/braindev
Original article
Autism and epilepsy: A retrospective follow-up study Hitoshi Hara Yokohama Central Area Habilitation Center for Children, Yokohama, Japan Kanagawa Day Treatment & Guidance Center for Children, Japan Received 16 August 2006; received in revised form 21 December 2006; accepted 31 December 2006
Abstract So-called ‘‘idiopathic’’ autism, which exhibited no major complications before diagnosis is well-known as one of the risk factors for epilepsy. This retrospective follow-up study aimed to clarify the characteristics of epilepsy in the autism; onset of seizure, seizure types, EEG findings and epilepsy outcome and the differences as a group between the autism with epilepsy and those without epilepsy. One hundred thirty individuals with autistic disorder or atypical autism diagnosed in childhood were followed up over 10 years and were evaluated almost every year up to 18–35 years of age. Their medical records related to perinatal conditions, IQ, social maturity scores and several factors of epilepsy were reviewed in October 2005. Thirty-three of the follow-up group (25%) exhibited epileptic seizures. The onset of epilepsy was distributed from 8 to 26 years of age. Two types of seizure were observed; partial seizure with secondarily generalized seizure and generalized seizure. Twenty of the epileptics (61%) showed the partial seizure. Although 18% of the non-epileptic group exhibited epileptic discharges on EEG, 68% of the epileptic group revealed epileptiform EEG findings before the onset of epilepsy. No differences were observed concerning the sex ratio, autistic disorder/atypical autism and past history of febrile seizures between the epileptic and non-epileptic groups. Lower IQ, lower social maturity score and higher frequency of prescribed psychotropics were observed in the epileptic group compared to the non-epileptics. Idiopathic autism was confirmed as the high risk factor for epilepsy. Epileptiform EEG findings predict subsequent onset of epileptic seizures in adolescence. Epilepsy is one of negative factors on cognitive, adaptive and behavioral/emotional outcomes for individuals with autism. Ó 2007 Elsevier B.V. All rights reserved. Keywords: Autism; Epilepsy; Follow-up
1. Introduction Many studies [1–8] have pointed out a close relationship between autism and epilepsy. The reported rate of epilepsy in autism has varied from 12.6% to 39.2%. Since autism is defined by a particular behavioral phenotype, not by specific etiology or pathophysiology, all the studies mentioned above included individuals with ‘‘symptomatic’’ or secondary autism, i.e., West syndrome, tuberous sclerosis, fragile-X syndrome that had already exhibited epileptic seizures before the diagnosis of autism. E-mail address: [email protected] 0387-7604/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2006.12.012
So-called idiopathic or primary autism, which exhibited no major complications before the diagnosis, is also well-known as one of the risk factors for epilepsy. This retrospective follow-up study aimed to clarify the characteristics of epilepsy in idiopathic autism, i.e., onset of seizure, seizure types, EEG findings and epilepsy outcome, and the differences as a group between autism with and without epilepsy.
2. Methods The subjects of this study consisted of children referred to the Kanagawa Day Treatment & Guidance
H. Hara / Brain & Development 29 (2007) 486–490
Center for Children for the evaluation of development, diagnosis and treatment of challenging behaviors. The author, one of the child neurologists at the center, checked up all subjects in the ordinary clinical setting and followed up almost every year from their first visit to adulthood. These subjects were selected from those individuals re-diagnosed by the author as having one of pervasive developmental disorders, autistic disorder or atypical autism, according to DSM-IV criteria [9] as of October 2005. The subjects with atypical autism met at least three but less than five symptoms of the DSM-IV criteria in total. These symptoms were derived from all three major domains: (1) social interaction, (2) language, communication and imagination and (3) behavioral flexibility; not just one or two. Several individuals with autistic disorder or atypical autism were excluded because they exhibited some other neurological diseases, i.e., West syndrome, neurofibromatosis, Prader–Willi syndrome or cerebral palsies due to perinatal complications. The medical records related to perinatal conditions (birth weight, birth head circumference), months of age when they walked alone, cognitive level evaluated by Tanaka–Binet test at the age of 18 years, social maturity scores, treatment with psychotropic drugs, routine EEG findings and several factors of epilepsy were obtained for this study. Cognitive level was divided into four categories: normal/mildly retarded (IQ > 50), moderately retarded (IQ 35–49), severely retarded (IQ 20–34) and profoundly retarded (IQ < 19). Social maturity scores (SQ) or the Japanese version of the Vinland social maturity scale (SM-Social Maturity Scale, 1980) [10] were evaluated at around 10 years of age. The most recent SQ was used in this study when several SQs were obtained. After they were diagnosed as having autistic disorder or atypical autism, their parents or caretakers were recommended to visit the clinic regularly, to be checked up the examinations including a routine EEG per a year from their first visits to around 12 years of age. When age of the first visit to our clinic as over 12 years, two or three EEGs were examined. If abnormal findings on EEG were detected during this period, the EEG follow-up continued annually until the findings disappeared. The caregiver could also consult about the daily activities or behavioral problems of their child annually. If necessary, the author prescribed psychotropic drugs or haloperidol, risperidone, fluvoxamine, clomipramine, methylphenidate, lithium, etc. for control of their problems. The diagnosis of epilepsy, epileptic seizures, active epilepsy and remission of epilepsy is essentially defined by the International League Against Epilepsy (ILAE) [11]. In this study, however, the individuals who exhibit-
487
ed one definite seizure and some epileptiform EEG abnormalities (EEA) included epileptics as an exception of ILAE’s definition. The EEA consisted of spikes, sharps, spike/sharpwaves, but not 6–14 Hz positive-spike bursts, 6-Hz spike-wave phantoms nor slow wave bursts without spikes. The EEA were divided into five groups according to the dominant area described by Kawasaki et al. [5]: Fgroup, left and right Fp, F and Cz areas; CP-group, left and right C and P areas; T-group, left and right aT and mT areas; O-group, left and right O areas and Diff.group, diffuse spike-wave discharges bilaterally or unilaterally.
3. Results The subjects of this study consisted of 130 individuals with autistic disorder or atypical autism diagnosed in childhood (males, 104; females, 26). Their first visits ranged from 1 to 14 years of age (Median = 4 years). All but four subjects were followed up over 10 years by the author from their first visit. The median age of the subjects was 21 years (range: 18–35) at the time of this review. Thirty-three out of the 130 patients (25%) developed epilepsy during the follow-up period. The characteristics of the two groups, i.e., epileptics (n = 33) and nonepileptics (n = 97) were compared (Table 1). No statistical differences were observed concerning the gender, diagnosis of autistic disorder/atypical autism, presence/ absence of speech loss in infancy [12], average birth Table 1 Characteristics of the individuals with or without epilepsy
Gender (M/F) Birth weight (g) Head circumference (cm)a Age when walking alone (months) Diagnosis of DSM- IV Autistic disorder Atypical autism Speech loss in infancy (+/ ) Cognitive level*** Normal–mildly retarded Moderately retarded Severely retarded Profoundly retarded Adaptive level (SQ)b,** Psychotropic drugs (+/ )* a b * ** ***
Epileptics (n = 33)
Non-epileptics (n = 97)
25/8 3227 ± 374 33.5 ± 1.0 14.4 ± 7.0
81/16 3172 ± 457 33.5 ± 1.7 13.7 ± 3.6
26 7 7/26
75 22 17/80
1 (3.1%) 6 (20.0%) 8 (22.2%) 18 (56.3%) 42.1 ± 16.8 21/12
31 24 28 14 59.4 ± 20.6 39/58
Six data points were not available. Two data points were not available. p = 0.026 (Fisher, both side). p < 0.05. p < 0.001.
488
H. Hara / Brain & Development 29 (2007) 486–490
weight, average head circumference at birth and average month when they walked alone. Lower cognitive level (v2 = 26.373, df = 3, p < 0.001), lower SQ (t = 1.979, p < 0.05) and a higher frequency of treatment with psychotropic (p = 0.026, Fisher’s both-side) were observed in the epileptics compared to the non-epileptics. More than half of profoundly retarded autistic individuals (56%) developed epilepsy in this series. The comparisons between the two groups concerning seizures and EEG findings are presented in Table 2. There was no statistical difference between the two groups, regarding the history of febrile seizures, (18% vs. 12%). The ages at initial EEG examinations ranged from 3 to 16 years of age (Median age, 4) and the most recent examinations were from 5 to 34 years of age (Median age; 13). In total, 464 EEG records were obtained from the epileptics and 701 from the non-epileptics during the follow-up period. Thus, the median number of EEG records per individual epileptic subject was 13, ranging from 5 to 27, and 6 for the non-epileptics, ranging from 2 to 23. The comparison between the two groups achieved a statistically significant difference (p < 0.001). Twenty-four individuals out of 33 (73%) in the epileptic group exhibited some type of EEA. All but two patients revealed such abnormalities before the onset of epilepsy. Twenty of the non-epileptics (21%) also exhibited epileptic discharges on EEG recordings. A statistically higher proportion of EEA was observed in epileptics compared with non-epileptics (p < 0.001). Comparison of the sites of foci and laterality of the discharges was performed between 24 of the epileptic
Table 2 Comparison between the epileptics and non-epileptics
History of febrile seizures Epileptiform EEG abnormalities (EEA)** Sites of foci F* C-P T O Diff. Right side only Left side only Right and/or left unilateral/bilateral Antiepileptic drugs (AEDs) VPA CBZ ZSM PHT PB CLB * **
p = 0.0682 (Fisher’s one-side). p < 0.001.
Epileptics (n = 33)
Non-epileptics (n = 97)
6 (18%) 24 (73%)
12 (12%) 20 (21%)
11/24 11/24 12/24 7/24 4/24 9/24 5/24 9/24
4/20 8/20 12/20 7/20 3/20 6/20 5/20 6/20
21/33 8/33 8/33 6/33 3/33 1/33
7/97 11/97 – 1/97 1/97 –
individuals and 20 of the non-epileptics with EEA. A higher percentage (46%) of the F-group in the epileptics tended to be observed, compared to 20% in the non-epileptics (p = 0.0682, Fisher’s one-side). The other comparisons exhibited no statistical differences. All epileptics but four were treated with some antiepileptic drugs (AEDs) at the time of this review. Out of 97 subjects, thirteen non-epileptics were also treated with CBZ and/or VPA because of their labile mood or aggressiveness (Table 2). Twenty-four of the epileptics (73%) were considered to be active. As shown in Fig. 1, the age at epilepsy onset was distributed from 8 to 26 years. The median age was 14 years. The four individuals had their first seizure at more than 24 years of age. Their EEA pattern belonged to the F-group, which had been recognized since adolescence. Two types of seizure were observed; partial seizures with secondarily generalized seizures and generalized tonic-clonic seizures. Twenty of the epileptics (61%) exhibited partial seizures. Of these subjects, one exceptional individual exhibited partial facial seizures without secondarily generalized seizures. Nine epileptics exhibited generalized seizures and there were four unclassified seizures in which it was not possible to determine whether they were partial or generalized. There were no other types of seizures; i.e., absence, myoclonic seizures, astatic seizures and so on. Frequency of seizures ranged from 1 to 74 times (Median = 3). Eleven out of the 33 epileptics exhibited only one seizure and had some types of EEA before the onset. In epileptics, 12 out of 84 EEG records (14%) exhibited some types of the EEA at ages ranging from 3 to 7 years. Furthermore, 60 out of 104 (58%) exhibited EEA at ages ranging from 8 to 12 years and 43 out of 99 (43%) at ages ranging from 13 to 17 years.
Fig. 1. EEA and epilepsy onset.
H. Hara / Brain & Development 29 (2007) 486–490
4. Discussion This study has three unique points: first, the study focused on the relationship between ‘‘idiopathic’’ autism and epilepsy, second, this is the longest and largest retrospective follow-up study ever performed, and third, this study based on a huge number of routine EEG records (1165 records) obtained during the follow-up period. The reported incidence of epilepsy in autistic individuals has varied from 13% to 40%. The occurrence of epilepsy in this study is 25%, which appears to be relatively lower. Some reports [2,4,8] have indicated that epilepsy is more common in female autistic adolescents than in males, whereas the occurrence of epilepsy in this study was the same for both genders. These two results likely reflect that the follow-up period reaches early adulthood and that this study excluded ‘‘symptomatic’’ autism, which usually exhibits seizures during infancy. This study did not detect any risk factors for developing epilepsy except for EEA. A previous study from the same center [13] suggested that autistic children with refracted course, or speech loss in infancy, tended to exhibit EEA, but did not exhibit definite epilepsy. The findings in this study re-confirmed that there was no correlation between speech loss in infancy and epilepsy. However, only 11 out of 24 individuals with speech loss in infancy (46%) had EEA, compared to 32 out of 106 who did not (30%). This difference did not reach statistical significance. As a group, the cognitive, adaptive and behavioral/ emotional outcomes of the epileptics are not favorable. Lower cognitive level, lower SQ and a higher percentage of administration of psychotropic drugs were observed in the epileptics compared with the non-epileptics. Thus, autistic individuals with epilepsy require more support throughout their lifetime, both qualitatively and quantitatively. Interestingly, a history of febrile seizures was not a risk factor for developing epilepsy in this study population. The occurrence of febrile seizures in the autistic individuals (14% in total) appears to be slightly higher than in the general population; the prevalence of febrile seizures in Japan estimated 50–100 per 1000. It is likely that the pathogenesis of epilepsy in autism differs from that of epilepsies following febrile seizures. Out of the 33 epileptics, 24 individuals exhibited some types of EEA before the onset of epilepsy. In other words, routine EEG examinations after diagnosis of autism, especially over 8 years of age could be a significant tool in predicting the development of epilepsy. Although three-quarters of the epileptics exhibited some types of EEA, 20 non-epileptics also exhibited EEA. The comparisons between 24 epileptics and 20 non-epileptics concerning sites of foci and laterality revealed no statistically significant differences. However, there was one
489
tendency: epileptics with EEA belonged to the F-group (46%) compared to 20% of the non-epileptics with EEA (p = 0.0682, Fisher’s one-side). Kawasaki et al. [5] speculated that ‘‘paroxysmal at F’’, the EEA of the F-group in this study, is a common abnormality in autism. In their series, 94% of autistic individuals who developed epilepsy over 10 years of age exhibited ‘‘paroxysmal at F’’. Furthermore, these investigators pointed out that in the epileptics, the emergence of ‘‘paroxysmal at F’’ was concordant with the onset of epileptic seizures. The results of this study indicate that ‘‘paroxysmal at F’’ is not common in autistics, but is common in autistic individuals with epilepsy. Concerning the common sites of foci, the percentage of the T-group is larger than that of the F-group in this study. The hypothesis of Kawasaki et al. [5] is partially supported by the results of this study. Only four individuals with epilepsy were applicable for ILAE’s definition of remission. Twenty-four of the subjects (73%) are still active. Thus, epilepsy is not easy to control in autism even when appropriate treatment is administered. Partial seizures with secondarily generalized seizures is probably the predominant type of epileptic seizures in autism, as some studies have reported previously [2,5,8]. However, it remains difficult to classify partial seizures as complex or simple ones in epileptics. As previous reports [2,3,5,7,8] have suggested that the risk for epilepsy was elevated during early adolescence, this study also demonstrated a peak of epilepsy onset during the same period (Fig. 1). This peak likely corresponds with epilepsy onset for idiopathic autism, whereas another peak before early childhood occurs for ‘‘symptomatic’’ autism with some other neurological diseases. In contrast with previous studies [1–8], this series included four individuals who developed epilepsy over 24 years of age. Although no specific characteristics of autism or epilepsy were observed in these four subjects, further follow-up study will need to clarify whether or not there is a ‘‘third’’ peak of epilepsy onset during early adulthood. Approximately, two-third of the subjects in this study did not reach 30 years of age.
Acknowledgment This paper was presented in the 10th International Child Neurology Congress, Montreal, Quebec, in June 2006.
References [1] Deykin EY, MacMahon B. The incidence of seizures among children with autistic symptoms. Am J Psychiatry 1979;136: 1310–2.
490
H. Hara / Brain & Development 29 (2007) 486–490
[2] Olsson I, Steffenburg S, Gillberg C. Epilepsy in autism and autisticlike conditions: a population-based study. Arch Neurol 1988;45:666–8. [3] Volkmar FR, Nelson DS. Seizure disorders in autism. J Am Acad Child Adolesc Psychiatry 1990;29:127–9. [4] Tuchman RF, Rapin I, Shinnar S. Autistic and dysphasic children. II. Epilepsy. Pediatrics 1991;88: 1219–25. [5] Kawasaki Y, Yokota K, Shinomiya M, Shimizu Y, Niwa S. Brief report. Electroenchephalographic paroxysmal activities in the frontal area emerged in middle childhood and during adolescence in a follow-up study of autism. J Autism Dev Disord 1997;27:605–20. [6] Mouridsen SE, Rich B, Isager T. Epilepsy in disintegrative psychosis and infantile autism: a long-term validation study. Dev Med Child Neurol 1999;41:110–4. [7] Giovanardi Rossi PA, Posar A, Parmeggiani A. Epilepsy in adolescents and young adults with autistic disorder. Brain Dev 2000;22:102–6.
[8] Danielsson S, Gillberg IC, Billstedt E, Gillberg C, Olsson I. Epilepsy in young adults with autism: a prospective populationbased follow-up study of 120 individuals diagnosed in childhood. Epilepsia 2005;46:918–23. [9] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th. Washington, DC: American Psychiatric Association; 1994. [10] Miki Y, editor. SM Social Maturity Scale (in Japanese). Tokyo: Nihon-Bunka Kagakusya; 1980. [11] Commission on Epidemiology and Prognosis of the International League against Epilepsy. Guidelines for epidemiologic studies on epilepsy. Epilepsia 1993;34:592–6. [12] Kurita H. Infantile autism with speech loss before the age of thirty months. J Am Acad Child Psychiatry 1985;24:191–6. [13] Hara H, Sasaki M. Autistic syndrome and epilepsy: a comparison between the children with epileptic seizures and only with epileptiform EEG abnormalities. In: Naruse H, Ornitz EM, editors. Proceedings of the international symposium on neurobiology of infantile autism. Tokyo; 1990. p. 201–202.
Original article
Autism and epilepsy: A retrospective follow-up study Hitoshi Hara Yokohama Central Area Habilitation Center for Children, Yokohama, Japan Kanagawa Day Treatment & Guidance Center for Children, Japan Received 16 August 2006; received in revised form 21 December 2006; accepted 31 December 2006
Abstract So-called ‘‘idiopathic’’ autism, which exhibited no major complications before diagnosis is well-known as one of the risk factors for epilepsy. This retrospective follow-up study aimed to clarify the characteristics of epilepsy in the autism; onset of seizure, seizure types, EEG findings and epilepsy outcome and the differences as a group between the autism with epilepsy and those without epilepsy. One hundred thirty individuals with autistic disorder or atypical autism diagnosed in childhood were followed up over 10 years and were evaluated almost every year up to 18–35 years of age. Their medical records related to perinatal conditions, IQ, social maturity scores and several factors of epilepsy were reviewed in October 2005. Thirty-three of the follow-up group (25%) exhibited epileptic seizures. The onset of epilepsy was distributed from 8 to 26 years of age. Two types of seizure were observed; partial seizure with secondarily generalized seizure and generalized seizure. Twenty of the epileptics (61%) showed the partial seizure. Although 18% of the non-epileptic group exhibited epileptic discharges on EEG, 68% of the epileptic group revealed epileptiform EEG findings before the onset of epilepsy. No differences were observed concerning the sex ratio, autistic disorder/atypical autism and past history of febrile seizures between the epileptic and non-epileptic groups. Lower IQ, lower social maturity score and higher frequency of prescribed psychotropics were observed in the epileptic group compared to the non-epileptics. Idiopathic autism was confirmed as the high risk factor for epilepsy. Epileptiform EEG findings predict subsequent onset of epileptic seizures in adolescence. Epilepsy is one of negative factors on cognitive, adaptive and behavioral/emotional outcomes for individuals with autism. Ó 2007 Elsevier B.V. All rights reserved. Keywords: Autism; Epilepsy; Follow-up
1. Introduction Many studies [1–8] have pointed out a close relationship between autism and epilepsy. The reported rate of epilepsy in autism has varied from 12.6% to 39.2%. Since autism is defined by a particular behavioral phenotype, not by specific etiology or pathophysiology, all the studies mentioned above included individuals with ‘‘symptomatic’’ or secondary autism, i.e., West syndrome, tuberous sclerosis, fragile-X syndrome that had already exhibited epileptic seizures before the diagnosis of autism. E-mail address: [email protected] 0387-7604/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2006.12.012
So-called idiopathic or primary autism, which exhibited no major complications before the diagnosis, is also well-known as one of the risk factors for epilepsy. This retrospective follow-up study aimed to clarify the characteristics of epilepsy in idiopathic autism, i.e., onset of seizure, seizure types, EEG findings and epilepsy outcome, and the differences as a group between autism with and without epilepsy.
2. Methods The subjects of this study consisted of children referred to the Kanagawa Day Treatment & Guidance
H. Hara / Brain & Development 29 (2007) 486–490
Center for Children for the evaluation of development, diagnosis and treatment of challenging behaviors. The author, one of the child neurologists at the center, checked up all subjects in the ordinary clinical setting and followed up almost every year from their first visit to adulthood. These subjects were selected from those individuals re-diagnosed by the author as having one of pervasive developmental disorders, autistic disorder or atypical autism, according to DSM-IV criteria [9] as of October 2005. The subjects with atypical autism met at least three but less than five symptoms of the DSM-IV criteria in total. These symptoms were derived from all three major domains: (1) social interaction, (2) language, communication and imagination and (3) behavioral flexibility; not just one or two. Several individuals with autistic disorder or atypical autism were excluded because they exhibited some other neurological diseases, i.e., West syndrome, neurofibromatosis, Prader–Willi syndrome or cerebral palsies due to perinatal complications. The medical records related to perinatal conditions (birth weight, birth head circumference), months of age when they walked alone, cognitive level evaluated by Tanaka–Binet test at the age of 18 years, social maturity scores, treatment with psychotropic drugs, routine EEG findings and several factors of epilepsy were obtained for this study. Cognitive level was divided into four categories: normal/mildly retarded (IQ > 50), moderately retarded (IQ 35–49), severely retarded (IQ 20–34) and profoundly retarded (IQ < 19). Social maturity scores (SQ) or the Japanese version of the Vinland social maturity scale (SM-Social Maturity Scale, 1980) [10] were evaluated at around 10 years of age. The most recent SQ was used in this study when several SQs were obtained. After they were diagnosed as having autistic disorder or atypical autism, their parents or caretakers were recommended to visit the clinic regularly, to be checked up the examinations including a routine EEG per a year from their first visits to around 12 years of age. When age of the first visit to our clinic as over 12 years, two or three EEGs were examined. If abnormal findings on EEG were detected during this period, the EEG follow-up continued annually until the findings disappeared. The caregiver could also consult about the daily activities or behavioral problems of their child annually. If necessary, the author prescribed psychotropic drugs or haloperidol, risperidone, fluvoxamine, clomipramine, methylphenidate, lithium, etc. for control of their problems. The diagnosis of epilepsy, epileptic seizures, active epilepsy and remission of epilepsy is essentially defined by the International League Against Epilepsy (ILAE) [11]. In this study, however, the individuals who exhibit-
487
ed one definite seizure and some epileptiform EEG abnormalities (EEA) included epileptics as an exception of ILAE’s definition. The EEA consisted of spikes, sharps, spike/sharpwaves, but not 6–14 Hz positive-spike bursts, 6-Hz spike-wave phantoms nor slow wave bursts without spikes. The EEA were divided into five groups according to the dominant area described by Kawasaki et al. [5]: Fgroup, left and right Fp, F and Cz areas; CP-group, left and right C and P areas; T-group, left and right aT and mT areas; O-group, left and right O areas and Diff.group, diffuse spike-wave discharges bilaterally or unilaterally.
3. Results The subjects of this study consisted of 130 individuals with autistic disorder or atypical autism diagnosed in childhood (males, 104; females, 26). Their first visits ranged from 1 to 14 years of age (Median = 4 years). All but four subjects were followed up over 10 years by the author from their first visit. The median age of the subjects was 21 years (range: 18–35) at the time of this review. Thirty-three out of the 130 patients (25%) developed epilepsy during the follow-up period. The characteristics of the two groups, i.e., epileptics (n = 33) and nonepileptics (n = 97) were compared (Table 1). No statistical differences were observed concerning the gender, diagnosis of autistic disorder/atypical autism, presence/ absence of speech loss in infancy [12], average birth Table 1 Characteristics of the individuals with or without epilepsy
Gender (M/F) Birth weight (g) Head circumference (cm)a Age when walking alone (months) Diagnosis of DSM- IV Autistic disorder Atypical autism Speech loss in infancy (+/ ) Cognitive level*** Normal–mildly retarded Moderately retarded Severely retarded Profoundly retarded Adaptive level (SQ)b,** Psychotropic drugs (+/ )* a b * ** ***
Epileptics (n = 33)
Non-epileptics (n = 97)
25/8 3227 ± 374 33.5 ± 1.0 14.4 ± 7.0
81/16 3172 ± 457 33.5 ± 1.7 13.7 ± 3.6
26 7 7/26
75 22 17/80
1 (3.1%) 6 (20.0%) 8 (22.2%) 18 (56.3%) 42.1 ± 16.8 21/12
31 24 28 14 59.4 ± 20.6 39/58
Six data points were not available. Two data points were not available. p = 0.026 (Fisher, both side). p < 0.05. p < 0.001.
488
H. Hara / Brain & Development 29 (2007) 486–490
weight, average head circumference at birth and average month when they walked alone. Lower cognitive level (v2 = 26.373, df = 3, p < 0.001), lower SQ (t = 1.979, p < 0.05) and a higher frequency of treatment with psychotropic (p = 0.026, Fisher’s both-side) were observed in the epileptics compared to the non-epileptics. More than half of profoundly retarded autistic individuals (56%) developed epilepsy in this series. The comparisons between the two groups concerning seizures and EEG findings are presented in Table 2. There was no statistical difference between the two groups, regarding the history of febrile seizures, (18% vs. 12%). The ages at initial EEG examinations ranged from 3 to 16 years of age (Median age, 4) and the most recent examinations were from 5 to 34 years of age (Median age; 13). In total, 464 EEG records were obtained from the epileptics and 701 from the non-epileptics during the follow-up period. Thus, the median number of EEG records per individual epileptic subject was 13, ranging from 5 to 27, and 6 for the non-epileptics, ranging from 2 to 23. The comparison between the two groups achieved a statistically significant difference (p < 0.001). Twenty-four individuals out of 33 (73%) in the epileptic group exhibited some type of EEA. All but two patients revealed such abnormalities before the onset of epilepsy. Twenty of the non-epileptics (21%) also exhibited epileptic discharges on EEG recordings. A statistically higher proportion of EEA was observed in epileptics compared with non-epileptics (p < 0.001). Comparison of the sites of foci and laterality of the discharges was performed between 24 of the epileptic
Table 2 Comparison between the epileptics and non-epileptics
History of febrile seizures Epileptiform EEG abnormalities (EEA)** Sites of foci F* C-P T O Diff. Right side only Left side only Right and/or left unilateral/bilateral Antiepileptic drugs (AEDs) VPA CBZ ZSM PHT PB CLB * **
p = 0.0682 (Fisher’s one-side). p < 0.001.
Epileptics (n = 33)
Non-epileptics (n = 97)
6 (18%) 24 (73%)
12 (12%) 20 (21%)
11/24 11/24 12/24 7/24 4/24 9/24 5/24 9/24
4/20 8/20 12/20 7/20 3/20 6/20 5/20 6/20
21/33 8/33 8/33 6/33 3/33 1/33
7/97 11/97 – 1/97 1/97 –
individuals and 20 of the non-epileptics with EEA. A higher percentage (46%) of the F-group in the epileptics tended to be observed, compared to 20% in the non-epileptics (p = 0.0682, Fisher’s one-side). The other comparisons exhibited no statistical differences. All epileptics but four were treated with some antiepileptic drugs (AEDs) at the time of this review. Out of 97 subjects, thirteen non-epileptics were also treated with CBZ and/or VPA because of their labile mood or aggressiveness (Table 2). Twenty-four of the epileptics (73%) were considered to be active. As shown in Fig. 1, the age at epilepsy onset was distributed from 8 to 26 years. The median age was 14 years. The four individuals had their first seizure at more than 24 years of age. Their EEA pattern belonged to the F-group, which had been recognized since adolescence. Two types of seizure were observed; partial seizures with secondarily generalized seizures and generalized tonic-clonic seizures. Twenty of the epileptics (61%) exhibited partial seizures. Of these subjects, one exceptional individual exhibited partial facial seizures without secondarily generalized seizures. Nine epileptics exhibited generalized seizures and there were four unclassified seizures in which it was not possible to determine whether they were partial or generalized. There were no other types of seizures; i.e., absence, myoclonic seizures, astatic seizures and so on. Frequency of seizures ranged from 1 to 74 times (Median = 3). Eleven out of the 33 epileptics exhibited only one seizure and had some types of EEA before the onset. In epileptics, 12 out of 84 EEG records (14%) exhibited some types of the EEA at ages ranging from 3 to 7 years. Furthermore, 60 out of 104 (58%) exhibited EEA at ages ranging from 8 to 12 years and 43 out of 99 (43%) at ages ranging from 13 to 17 years.
Fig. 1. EEA and epilepsy onset.
H. Hara / Brain & Development 29 (2007) 486–490
4. Discussion This study has three unique points: first, the study focused on the relationship between ‘‘idiopathic’’ autism and epilepsy, second, this is the longest and largest retrospective follow-up study ever performed, and third, this study based on a huge number of routine EEG records (1165 records) obtained during the follow-up period. The reported incidence of epilepsy in autistic individuals has varied from 13% to 40%. The occurrence of epilepsy in this study is 25%, which appears to be relatively lower. Some reports [2,4,8] have indicated that epilepsy is more common in female autistic adolescents than in males, whereas the occurrence of epilepsy in this study was the same for both genders. These two results likely reflect that the follow-up period reaches early adulthood and that this study excluded ‘‘symptomatic’’ autism, which usually exhibits seizures during infancy. This study did not detect any risk factors for developing epilepsy except for EEA. A previous study from the same center [13] suggested that autistic children with refracted course, or speech loss in infancy, tended to exhibit EEA, but did not exhibit definite epilepsy. The findings in this study re-confirmed that there was no correlation between speech loss in infancy and epilepsy. However, only 11 out of 24 individuals with speech loss in infancy (46%) had EEA, compared to 32 out of 106 who did not (30%). This difference did not reach statistical significance. As a group, the cognitive, adaptive and behavioral/ emotional outcomes of the epileptics are not favorable. Lower cognitive level, lower SQ and a higher percentage of administration of psychotropic drugs were observed in the epileptics compared with the non-epileptics. Thus, autistic individuals with epilepsy require more support throughout their lifetime, both qualitatively and quantitatively. Interestingly, a history of febrile seizures was not a risk factor for developing epilepsy in this study population. The occurrence of febrile seizures in the autistic individuals (14% in total) appears to be slightly higher than in the general population; the prevalence of febrile seizures in Japan estimated 50–100 per 1000. It is likely that the pathogenesis of epilepsy in autism differs from that of epilepsies following febrile seizures. Out of the 33 epileptics, 24 individuals exhibited some types of EEA before the onset of epilepsy. In other words, routine EEG examinations after diagnosis of autism, especially over 8 years of age could be a significant tool in predicting the development of epilepsy. Although three-quarters of the epileptics exhibited some types of EEA, 20 non-epileptics also exhibited EEA. The comparisons between 24 epileptics and 20 non-epileptics concerning sites of foci and laterality revealed no statistically significant differences. However, there was one
489
tendency: epileptics with EEA belonged to the F-group (46%) compared to 20% of the non-epileptics with EEA (p = 0.0682, Fisher’s one-side). Kawasaki et al. [5] speculated that ‘‘paroxysmal at F’’, the EEA of the F-group in this study, is a common abnormality in autism. In their series, 94% of autistic individuals who developed epilepsy over 10 years of age exhibited ‘‘paroxysmal at F’’. Furthermore, these investigators pointed out that in the epileptics, the emergence of ‘‘paroxysmal at F’’ was concordant with the onset of epileptic seizures. The results of this study indicate that ‘‘paroxysmal at F’’ is not common in autistics, but is common in autistic individuals with epilepsy. Concerning the common sites of foci, the percentage of the T-group is larger than that of the F-group in this study. The hypothesis of Kawasaki et al. [5] is partially supported by the results of this study. Only four individuals with epilepsy were applicable for ILAE’s definition of remission. Twenty-four of the subjects (73%) are still active. Thus, epilepsy is not easy to control in autism even when appropriate treatment is administered. Partial seizures with secondarily generalized seizures is probably the predominant type of epileptic seizures in autism, as some studies have reported previously [2,5,8]. However, it remains difficult to classify partial seizures as complex or simple ones in epileptics. As previous reports [2,3,5,7,8] have suggested that the risk for epilepsy was elevated during early adolescence, this study also demonstrated a peak of epilepsy onset during the same period (Fig. 1). This peak likely corresponds with epilepsy onset for idiopathic autism, whereas another peak before early childhood occurs for ‘‘symptomatic’’ autism with some other neurological diseases. In contrast with previous studies [1–8], this series included four individuals who developed epilepsy over 24 years of age. Although no specific characteristics of autism or epilepsy were observed in these four subjects, further follow-up study will need to clarify whether or not there is a ‘‘third’’ peak of epilepsy onset during early adulthood. Approximately, two-third of the subjects in this study did not reach 30 years of age.
Acknowledgment This paper was presented in the 10th International Child Neurology Congress, Montreal, Quebec, in June 2006.
References [1] Deykin EY, MacMahon B. The incidence of seizures among children with autistic symptoms. Am J Psychiatry 1979;136: 1310–2.
490
H. Hara / Brain & Development 29 (2007) 486–490
[2] Olsson I, Steffenburg S, Gillberg C. Epilepsy in autism and autisticlike conditions: a population-based study. Arch Neurol 1988;45:666–8. [3] Volkmar FR, Nelson DS. Seizure disorders in autism. J Am Acad Child Adolesc Psychiatry 1990;29:127–9. [4] Tuchman RF, Rapin I, Shinnar S. Autistic and dysphasic children. II. Epilepsy. Pediatrics 1991;88: 1219–25. [5] Kawasaki Y, Yokota K, Shinomiya M, Shimizu Y, Niwa S. Brief report. Electroenchephalographic paroxysmal activities in the frontal area emerged in middle childhood and during adolescence in a follow-up study of autism. J Autism Dev Disord 1997;27:605–20. [6] Mouridsen SE, Rich B, Isager T. Epilepsy in disintegrative psychosis and infantile autism: a long-term validation study. Dev Med Child Neurol 1999;41:110–4. [7] Giovanardi Rossi PA, Posar A, Parmeggiani A. Epilepsy in adolescents and young adults with autistic disorder. Brain Dev 2000;22:102–6.
[8] Danielsson S, Gillberg IC, Billstedt E, Gillberg C, Olsson I. Epilepsy in young adults with autism: a prospective populationbased follow-up study of 120 individuals diagnosed in childhood. Epilepsia 2005;46:918–23. [9] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th. Washington, DC: American Psychiatric Association; 1994. [10] Miki Y, editor. SM Social Maturity Scale (in Japanese). Tokyo: Nihon-Bunka Kagakusya; 1980. [11] Commission on Epidemiology and Prognosis of the International League against Epilepsy. Guidelines for epidemiologic studies on epilepsy. Epilepsia 1993;34:592–6. [12] Kurita H. Infantile autism with speech loss before the age of thirty months. J Am Acad Child Psychiatry 1985;24:191–6. [13] Hara H, Sasaki M. Autistic syndrome and epilepsy: a comparison between the children with epileptic seizures and only with epileptiform EEG abnormalities. In: Naruse H, Ornitz EM, editors. Proceedings of the international symposium on neurobiology of infantile autism. Tokyo; 1990. p. 201–202.