- Email: [email protected]
West syndrome – The University of Hong Kong experience (1970–2000)
Brain & Development 23 (2001) 609–615 www.elsevier.com/locate/braindev
Original article
West syndrome – The University of Hong Kong experience (1970–2000) Virginia Wong* Department of Paediatrics, The University of Hong Kong, Queen Mary Hospital, Pokfulam, Hong Kong Received 21 April 2001; received in revised form 30 July 2001; accepted 2 August 2001
Abstract Objective: To study the clinical pattern of West syndrome (WS) in a university based hospital. Methodology: The database of children seen in the Epilepsy Clinic of Queen Mary Hospital and Duchess of Kent Children’s Hospital during a 30-year period (1970–2000) was reviewed. Results: A total of 105 cases had WS (1.9%). The number of new cases of WS admitted per year ranged from one to eight. The range of annual incidence of WS to newly diagnosed epilepsy was 0.8–4.8%. The etiology included idiopathic (N ¼ 19, 18%), cryptogenic (N ¼ 23; 22%), symptomatic (N ¼ 56; 53.3%) and unknown (N ¼ 7; 5.7%). Adrenocorticotropic hormone (ACTH) and/or prednisone were given to 42 children (40%). Most were effective in controlling WS on an all-or-none fashion. Seizure outcome included 12 with remission, persistent in the same form in two and persisting but changed to another form in the rest. Neurological outcome at the last follow up in 2000 December included multiple disabilities ðN ¼ 16Þ, cerebral palsy ðN ¼ 22Þ, mental retardation ðN ¼ 94Þ and Lennox–Gastaut syndrome ðN ¼ 13Þ. We analysed the following risk factors in relation to poor outcome: age of onset, age of presentation, time lag before treatment, etiology, family history of epilepsy and hormonal treatment. Only etiology of WS has a positive correlation with poor outcome ðP , 0:0005Þ. Conclusions: WS is an uncommon epileptic syndrome. The majority had poor outcome, especially those with causes identified. Infantile spasm is a specific epileptic phenomenon in a maturational stage of a child when heterogeneous disorders can present with the same clinical epileptic and electroencephalographic phenomenon. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Infantile spasm; West syndrome; Children; Seizure; Epilepsy; Vigabatrin; Risk factors; Prognosis; Annual incidence
1. Introduction West syndrome (WS) is a rare epileptic syndrome consisting of infantile spasm (IS), hypsarrhythmia and mental retardation. The etiology, seizure classification and drug treatment of WS had been under evolutionary change over the past decade. The ultimate prognosis, in general, is poor despite treatment [1–12]. The segregation of etiology into idiopathic, cryptogenic and symptomatic depends on the facilities available locally, as neuroradiologic techniques like computerized tomography (CT), magnetic resonance imaging (MRI) and positron emission tomography (PET) contributed to the rising trend of symptomatic cases [13–15]. Rare neurometabolic or neurodegenerative diseases presenting with IS contributed to a minority of symptomatic cases, and thorough workup depends on the availability of sophisticated laboratories and professionals which were unavailable in most clinical child neurology settings. Adrenocorticotropic hormone (ACTH)/steroid has been
useful in the suppression of IS. However, the use of low dose versus high dose ACTH/steroid was still under intense debate [16–24]. Although vigabatrin (VGA) had been advocated as the first line of treatment recently, its usefulness had been limited by the occurrence of visual field defect in some [25–29]. VGA had been found to be especially useful in children with WS due to tuberous sclerosis (TS) [30–32]. There has been long debate of the first antiepileptic drug (AED) of choice in WS, ACTH/steroid or VGA [33,34]. Other newer AED like topiramate was also studied in new WS cases, and reported to be useful [35]. The Japanese reported usefulness of pyridoxine in a small proportion of IS patients [36,37]. The objective is to study the clinical pattern of WS in a university based hospital during a 30-year period (1970– 2000) and to identify the risk factors relating to poor outcome.
2. Methodology * Tel: 1852-2855-4485; fax: 1852-2855-1523. E-mail address: [email protected] (V. Wong).
The database of children (,16 years on admission) seen
0387-7604/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0387-760 4(01)00296-0
610
V. Wong / Brain & Development 23 (2001) 609–615
in the Epilepsy Clinic of Queen Mary Hospital and Duchess of Kent Children’s Hospital during a 30-year period (1970– 2000) was reviewed. Both hospitals are teaching hospitals of the University of Hong Kong. Children admitted for nonepileptic convulsive disorders or febrile convulsions were excluded. Most cases had been actively followed up by the author since 1980 except for those who were lost on follow up in the 1970s. A retrospective (1970–1979) and prospective medical record (1980–2000) review was conducted. Children were included if they satisfied the diagnostic criteria of IS with flexor or extensor spasms, and WS which consisted of IS, hypsarrhythmia pattern in electroencephalography (EEG) and associated mental retardation. 2.1. Statistical analysis Pearson chi-square was used to determine six risk factors (age of onset, age of presentation, time lag before treatment, etiology, family history of epilepsy and hormonal treatment) and abnormal outcome. Logistic regression analysis was carried out after obtaining the significant factors from Pearson chi-square test. 3. Results 3.1. Epilepsy database Altogether around 5500 children had epilepsy. A total of 105 cases had WS (1.9%). The number of new cases of WS admitted per year was one to eight (Fig. 1). The increase in the number of cases in 1980s to 2000s as compared to 1970s was due to the newly developed integrated child neurology and developmental pediatrics program in 1980s by the author within the University of Hong Kong as the only centre available locally. The annual range of newly diagnosed WS in newly diagnosed epilepsy during this period was 0.8–4.8%. The sex ratio of boys:girls was 55:50 (or 1.1:1). Nineteen children were lost on follow up.
being referred to a child neurologist. Video EEG was available for seizure monitoring in 1998 in our centre. 3.4. Family history None of our WS cases had history of IS or epilepsy in the immediate family members. There were only two sisters in a family with TS. 3.5. Workup for underlying etiology The etiological workup included: renal and liver biochemistries, glucose, lactate and pyruvate (since 1985), anion gap, ammonia, urine for simple amino acid and metabolic screening for most cases. Chromosomal study was evaluated depending on the clinical features. Cerebrospinal fluid analysis was performed if clinically indicated. Blood and urine were also taken for congenital infections like rubella and cytomegalovirus (CMV) if features were suggestive such as microcephaly, developmental delay, failure to thrive and visual or hearing impairment. EEG was performed for all cases after admission. Video EEG was only available in our centre in 1998 for diagnostic evaluation and monitoring of electrophysiological response to anticonvulsants. Interictal EEG showed typical features of hypsarrhythmia in most cases except the case with hemimegalencephaly and another two children with Aicardi syndrome. Intravenous pyridoxine (vitamin B6) test with 50–100 mg pyridoxine and simultaneous EEG monitoring were performed routinely in all cases since 1980s but none showed EEG response to intravenous pyridoxine. CAT scan of the brain was available locally in the government hospitals on a free basis since 1985 and MRI of the brain since 1995. PET scan was only available locally in the private setting in 2000. Thus, the role played by neuroimaging in identifying etiology in our study depended on the availability of these tools during the initial or subsequent follow up of these cases. The number of children who underwent MRI or PET scan of the brain were 87 and 3, respectively.
3.2. Age of onset and age of presentation (Fig. 2) 3.6. Etiology (Fig. 4) The peak ages of onset and presentation both were 0–3 months. If one categorized the age intervals by 6-monthly, then the peak age of onset is centred around 6 months. The mean age at last consultation was 5.6 years (based on 2000 December follow up), with age range from 12 months to 30 years. 3.3. Clinical seizure pattern (Fig. 3) Clear-cut description of seizure pattern by chief caretaker or through the nursing record was obtained in 84 cases. Clinical features were typical in most with the majority presenting with flexor spasms. Only a few had spasms misinterpreted as normal by primary care doctors before
The etiology included idiopathic ðN ¼ 19Þ, cryptogenic ðN ¼ 23Þ, symptomatic ðN ¼ 56Þ and unknown ðN ¼ 7Þ. ‘Unknown’ means investigations like CT/MRI of the brain were not available in the 1970s for complete evaluation. Idiopathic is defined as no causes identified despite available investigations and the children had normal development prior to development of WS. Symptomatic is defined as definite causes identified with existing investigations as mentioned above. Cryptogenic is defined as negative findings with existing investigative tools available locally and yet these children had delayed development or abnormal neurological findings. One child developed IS within 1 week after receiving pertussis vaccination, and thus, the
V. Wong / Brain & Development 23 (2001) 609–615
611
Fig. 1. Yearly distribution of new cases of WS and new cases of epilepsy (1970–2000).
relationship is speculative and we categorized this as a cyptogenic case. For symptomatic cases, prenatal causes included neuronal migration disorder (N ¼ 11), congenital CMV infection (N ¼ 1), Rett syndrome (RS) (N ¼ 1), Down syndrome (DS) (N ¼ 2), Aicardi syndrome (N ¼ 2), Angelman syndrome (AS) ðN ¼ 1Þ, TS (N ¼ 10), hyperphenylalaninemia (N ¼ 1) and organic aciduria (argininosuccinase deficiency) (N ¼ 1). Perinatal included hypoxic-ischemic encephalopathy (N ¼ 3) and persistent neonatal hypoglycemia (N ¼ 1). Postnatal included meningitis (three; one with Salmonella;
two with Haemophilus influenzae) and herpes simplex encephalitis (N ¼ 1). Urine for metabolic screening was performed in all cases. Only two were positive. One had phenylketonuria and the other had organic aciduria (argininosuccinase deficiency). 3.7. Neuroimaging (Fig. 5) MRI scan of the brain was performed for 87 children. MRI features of neuronal migration disorder (N ¼ 13) included lissencephaly, pachygyria/agyria, hemimegalence-
612
V. Wong / Brain & Development 23 (2001) 609–615
Fig. 2. Ages of onset and presentation of WS.
Fig. 4. Etiology of WS.
phaly, schizencephaly and agenesis of corpus callosum. Some cases had a mixture of these radiological features. PET scan was only performed for three cases as PET is only available on a private setting. In a 12-month-old girl with intractable seizures with good initial response to VGB and having a relapse 6 months later, PET scan of the brain showed focal hypometabolism in her left temporal lobe. Epilepsy surgery may be indicated in this case. However, epilepsy surgery program has not been launched for children in Hong Kong yet. A 3.5-year-old boy with TS had hypometabolic regions corresponding to the site of tubers. He subsequently had left temporal lobectomy and lesionectomy of left parieto-occipital tubers performed in Montreal Neurological Institute under Professor F. Andermann. Another 4-year-old boy with severe mental retardation, cortical blindness and cerebral atrophy shown on MRI had global cerebral hypometabolism in PET scan.
WS. ACTH in the form of Synacthen was injected intramuscularly on a daily basis. In the 1980s, high dose ACTH (2 IU/kg/day) was given for at least an 8–12 weeks course. Low dose regime (0.2 IU/kg/day) ACTH was used for 4–6 weeks course in the 1990s. Due to the failure of production of ACTH in the market in 1997 and with the launching of VGA in Hong Kong in 1997, VGA was started as the first choice for any new case of IS. VGA led to complete cessation of IS in all cases tried within 24 h as the first line treatment. Visual field monitoring was regularly conducted for these children by the ophthalmologist and optometrist with behavioral visual field testing if cooperative, funduscopic examination and visual evoked potential evaluation. No reported visual problems occurred so far. Complete remission occurred with VGA monotherapy in etiologies like idiopathic ðN ¼ 3Þ, congenital CMV infection ðN ¼ 1Þ and organic aciduria ðN ¼ 1Þ. As for TS cases with IS, they already had the spasms long before they were seen by us and none of them responded to VGA monotherapy.
3.8. Treatment Altogether 42 children received hormonal treatment. ACTH and prednisone were given to 38 and four children, respectively. ACTH was effective in the majority of cases within the first week of injection. The response was actually seen as an all-or-none one. ACTH was given on an emergency basis in 1980–1997 for any newly admitted cases of
Fig. 3. Seizure types at presentation.
Fig. 5. MRI findings in WS ðN ¼ 87Þ.
V. Wong / Brain & Development 23 (2001) 609–615
613
We analysed the following risk factors in relation to poor outcome: age of onset, age of presentation, time lag before treatment, etiology, family history of epilepsy and hormonal treatment. Only a symptomatic etiology has a positive correlation with poor outcome ðP , 0:0005Þ. 3.10. Mortality Nine (8.57%) children died, of which seven died due to steroid related complications. One was due to disseminated intravascular coagulation and fulminant sepsis related to the use of high dose ACTH. 3.11. Risk factors and poor outcome (Table 1)
Fig. 6. Major anticonvulsants used during active seizure phase in 1980– 2000.
Only two factors were statistically significantly related to poor outcome: etiology ðP , 0:0005Þ and family history of epilepsy ðP ¼ 0:007Þ. When both risk factors were put into the regression, only etiology is statistically significant ðP ¼ 0:001Þ with coefficient of 3.845.
In 1970s to early 1980s, the main anticonvulsants used were phenobarbitone and phenytoin. Carbamazepine was not used for WS. For those who failed to respond to ACTH/steroid or having relapse after ACTH/steroid withdrawal, other anticonvulsants were used. This included valproic acid, clonazapam, nitrazepam and clobazam. Newer anticonvulsants like VGA, lamotrigine and topiramate were tried for intractable cases as they were licensed locally during the recent 5 years (Fig. 6).
4. Discussion
3.9. Outcome
4.1. Etiology in relation to development of IS
The outcome was analysed for those with active follow up until 2000 December. Altogether 19 cases were lost for follow up. For the new WS cases seen in 1999–2000, the minimum interval for follow up since diagnosis was 9 months. Seizure outcome included 12 with remission, persistent in the same form in two (cases admitted in 2000) and persisting but changed to another form in the rest. Neurological outcome included mental retardation (N ¼ 94; 89.5%), cerebral palsy (N ¼ 22; 21%), multiple disabilities (N ¼ 16; 15.2%), visual impairment (N ¼ 31; 29.5%), hearing impairment (N ¼ 2; 1.9%) and Lennox– Gastaut syndrome (N ¼ 13; 12.4%).
Idiopathic type of WS has the best prognosis. Most had normal neurological and developmental outcome. We noted that despite the multiple and heterogeneous causes of WS, if one looked at the specific etiology per se, one could find that only occasionally IS is part of the presentation. Thus, there might be some yet unexplained triggering factor in generating the WS symptomatology. In our own TS database of 70 children, 10% presented with IS. In a few cases, the depigmented spots were not very obvious such that the diagnosis was delayed until MRI scan showed the typical tubers in the brain. One girl suffering from TS with intractable IS had a focal resection of a massive tuber in her left parietal region and complete remission of her seizures occurred which lasted for 12 months in her last follow up. Both neurofibromatosis (NF) and incontinenti pigmenti (IP) can present with IS. It is interesting that in our NF database of nearly 100 cases, none had IS. Both of our IP cases present with focal or generalized seizures rather than IS. In our DS database of around 400 children only two had WS. One case was related to a severe hypoxic-ischemic event post-cardiac surgery, and another had acute myeloid leukemia as well. Thus, we failed to find the increased prevalence of WS in DS.
Table 1 Analysis of risk factors for poor outcome in WS Pearson chisquare P-value
Age of onset Age of presentation Time lag before treatment Etiology Family history of epilepsy Hormonal treatment
0.83 0.742 0.447 ,0.0005 0.007 0.856
Logistic regression P value
Coefficient
0.001 0.967
3.845 –
Our university based tertiary hospital WS database showed that WS is indeed a rare epileptic disorder. The best outcome occurred in those children with idiopathic etiology. IS is a specific epileptic phenomenon in a maturational stage of a child when heterogeneous disorders like neuronal migration disorders, TS and other syndromes can present with the same clinical epileptic and EEG signatures.
614
V. Wong / Brain & Development 23 (2001) 609–615
One of our RS cases in our RS database ðN ¼ 12Þ initially presented with IS. She subsequently regressed with autistic features and typical hand wringing stereotypic movement. Thus one needs to be alert to the possible association of WS and RS. Another girl with AS presented with WS during the clinical course. This is an unusual presentation as in our AS database of 20 cases, only one had IS. Thus, one should be alert to unusual jerky movement or features of unexplained laughter in children with WS and the possible association of WS and AS. In our database of around 50 cases of congenital CMV infection, only one had IS. None of our cases with other congenital infections like rubella presented with IS. In our neurometabolic disease database ðN ¼ 50Þ, two cases presented with IS. One girl with organic aciduria (argininosuccinase deficiency) presented with intractable IS which responded immediately to VGA. Another girl had hyperphenylalaninemia and moderate degree of mental retardation with IS subsequently changed to other seizure types. In our neurodegenerative disease database, only one girl with infantile type of neuronal ceroid lipofuscinosis diagnosed by skin biopsy presented with IS. She subsequently died of her own disease. We performed intravenous pyridoxine infusion (50– 100 mg) during EEG monitoring since 1980s. However, none of our IS cases had EEG response. Oral pyridoxine at 200–400 mg/day was continued for some WS cases intractable to all conventional AEDs, but none had any positive response. 4.2. EEG Most of our WS cases had hypsarrhythmia pattern in the inerictal EEG. One boy with hemimegalencephaly had hemihypsarrhythmia. Other atypical hypsarrhythmic features included burst suppression pattern in the child with hemimegalencephaly and unilateral hypsarrhythmia in Aicardi syndrome. Video EEG monitoring was available in our centre in 1998 and was mainly used for diagnostic evaluation and monitoring of response of seizure to anticonvulsants. For the cases treated with VGA, the hysparrhythmic pattern still persisted even when the clinical seizures were well controlled. This was unlike those with ACTH treatment where EEG and clinical remission correlated. For those with VGA treatment, the EEG pattern might evolve to generalized slow sharp and spike wave discharges during the clinical course. 4.3. Neuroimaging In the days with purely computerized axial tomography (CAT) scan, almost all cases showed pseudo-cerebral atrophy after the use of ACTH/steroid. With the availability of MRI scan locally in 1995, more children with cortical dysplasia were identified, thus shifting the cryptogenic cases to symptomatic etiology. The PET scan was useful
in identifying focal hypometabolic lesions not detected by MRI of the brain. One 13-month-old girl with intractable WS and normal MRI had a focal hypometabolic area in her left parieto-temporal region. These children might be good candidates for epilepsy surgery. 4.4. Treatment In the era before the availability of VGA locally, the first line treatment for WS was ACTH injection, and the response rate was high and immediate (usually within a week). However, the relapse rate was relatively high as well. In 1980s for parents of patients who were reluctant to allow their children to try ACTH or oral prednisone as the patients were too young to give consent, the commonest first line AED used was sodium valproate, followed by benzodiazepines. Phenobarbitone or phenytoin was used in 1970s to 1980s. The availability of VGA locally in 1997 replaced valproic acid as the AED of first choice. The local availability of VGA reached a turning point when the pharmaceutical company stopped manufacturing ACTH gel (Synacthen). Altogether, 38 children with WS had received VGA as the initial treatment or during the active seizure phase. VGA was successful in inducing seizure cessation almost within 24 h in those given as initial monotherapy. It had been an all-or-none effect. Even children with symptomatic etiology responded. One 4-month-old child with organic aciduria (argininosuccinase deficiency) who failed to respond to sodium valproate (VPA) or clonazepam responded after the first 500 mg dose of VGA. She remained seizure free for at least 24 months at the last follow up. Another 7-month-old child with congenital CMV infection also achieved complete and long lasting seizure control with VGA monotherapy within 12 h of a 250 mg dose. Another 6-month-old girl with idiopathic WS who failed to respond to VPA or clonazepam for 4 weeks also achieved complete cessation of seizures with a single dose of 500 mg VGA. However, hypsarrhythmia pattern failed to respond to treatment with VGA even with clinical remission in most children with IS. Other newer AEDs were used during the active course of seizure which were unresponsive to VPA or benzodiazepines. However, the response to lamotrigine and topiramate was not as satisfactory as reported. Ketogenic diet was tried in two children during the intractable seizure course, with one having complete remission for 2 years with only one anticonvulsant as maintenance. As WS is such a rare disorder with a heterogeneous etiological background, the collaboration of multicentre clinical trials is urgently needed to improve the ultimate prognosis of this catastrophic epileptic syndrome. Acknowledgements We thank our paediatric colleagues who participated in medical care of these children during the period 1970–2000.
V. Wong / Brain & Development 23 (2001) 609–615
This paper was presented in the 4th International Symposium of the West Syndrome and Other Infantile Epileptic Encephalopathies (ISWS) in Tokyo, Japan in 2001, February 9–11.
[19]
[20]
References [1] Appleton RE. Infantile spasms. Arch Dis Child 1993;69:614–618. [2] Cowan LD, Hudson LS. The epidemiology and natural history of infantile spasms. J Child Neurol 1991;6:355–364. [3] Hrachovy RA, Glaze DG, Frost JD. A retrospective study of spontaneous remission and long-term outcome in patients with infantile spasms. Epilepsia 1991;32(2):212–214. [4] Meencke HJ, Gerhard C. Morphological aspects of etiology and the course of infantile spasms (West-syndrome). Neuropediatrics 1985;16:59–66. [5] Bobele GB, Bodensteiner JB. Infantile spasms. Neurol Clin 1990;8:633–645. [6] Bellman MH, Ross EM, Miller DL. Infantile spasms and pertussis immunisation. Lancet 1983;1:1031–1034. [7] Lombroso CT. A prospective study of infantile spasms: clinical and therapeutic correlations. Epilepsia 1983;24:135–158. [8] Riikonen R. A long-term follow-up study of 214 children with the syndrome of infantile spasms. Neuropediatrics 1982;13:14–23. [9] Matsumoto A, Watanabe K, Negoro T, Sugiura M, Iwase K, Hara K, et al. Long-term prognosis after infantile spasms: a statistical study of prognostic factors in 200 cases. Dev Med Child Neurol 1981;23:51– 65. [10] Kurokawa T, Goya N, Fukuyama Y, Suzuki M, Seki T, Ohtahara S. West syndrome and Lennox–Gastaut syndrome: a survey of natural history. Pediatrics 1980;65:81–88. [11] Riikonen R, Donner M. Incidence and aetiology of infantile spasms from 1960 to 1976: a population study in Finland. Dev Med Child Neurol 1979;21:333–343. [12] Koo B, Hwang PA, Logan WJ. Infantile spasms: outcome and prognostic factors of cryptogenic and symptomatic groups. Neurology 1993;43:2322–2327. [13] Aydinli N, Caliskan M, Ozmen M, Tonguc E. Neuroradiologic aspects of West syndrome. Pediatr Neurol 1998;19:211–216. [14] Okumura A, Watanabe K, Negoro T, Aso K, Natsume J, Kubota T, et al. Evolutional changes and outcome of West syndrome: correlation with magnetic resonance imaging findings. Epilepsia 1998;39(Suppl 5):46–49. [15] Chugani HT, Conti JR. Etiologic classification of infantile spasms in 140 cases: role of positron emission tomography. J Child Neurol 1996;11:44–48. [16] Snead OC. Treatment of infantile spasms. Pediatr Neurol 1990;6:147–150. [17] Bobele GB, Bodensteiner JB. The treatment of infantile spasms by child neurologists. J Child Neurol 1994;9:432–435. [18] Shields WD, Shewmon DA, Chugani HT, Peacock WJ. Treatment of
[21]
[22]
[23] [24]
[25]
[26] [27]
[28]
[29] [30]
[31]
[32]
[33] [34]
[35] [36]
[37]
615
infantile spasms: medical or surgical? Epilepsia 1992;33(Suppl 4):S26–S31. Hrachovy RA, Frost JD, Glaze DG. High-dose, long-duration versus low-dose, short-duration corticotropin therapy for infantile spasms. J Pediatr 1994;124:803–806. Snead OC, Benton JW, Myers GJ. ACTH and prednisone in childhood seizure disorders. Neurology 1983;33:966–970. Snead OC, Benton JW, Hosey LC, Swann JW, Spink D, Martin D, et al. Treatment of infantile spasms with high-dose ACTH: efficacy and plasma levels of ACTH and cortisol. Neurology 1989;39:1027–1031. Glaze DG, Hrachovy RA, Frost JD, Kellaway P, Zion TE. Prospective study of outcome of infants with infantile spasms treated during controlled studies of ACTH and prednisone. J Pediatr 1998;112:389–396. Riikonen R, Donner M. ACTH therapy in infantile spasms: side effects. Arch Dis Child 1980;55:664–672. Yanagaki S, Oguni H, Hayashi K, Imai K, Funatuka M, Tanaka T, et al. A comparative study of high-dose and low-dose ACTH therapy for West syndrome. Brain Dev 1999;21:461–467. Appleton RE, Peters AC, Mumford JP, Shaw DE. Randomised, placebo-controlled study of vigabatrin as first-line treatment of infantile spasms. Epilepsia 1999;40:1627–1633. Appleton RE. Guideline for prescribing vigabatrin. BMJ 1998;317:1322. Aicardi J, Mumford JP, Dumas C, Wood S. Vigabatrin as initial therapy for infantile spasms: a European retrospective survey. Epilepsia 1996;37:638–642. Fejerman N, Cersosimo R, Caraballo R, et al. Vigabatrin as a firstchoice drug in the treatment of West syndrome. J Child Neurol 2000;15:161–165. Vigabatrin Paediatric Advisory Group. Guideline for prescribing vigabatrin in children has been revised. BMJ 2000;320:1404–1405. Fukushima K, Inoue Y, Fujiwara T, Yagi K. Long-term course of West syndrome associated with tuberous sclerosis. Epilepsia 1998;39(Suppl 5):50–54. Hancock E, Osborne JP. Vigabatrin in the treatment of infantile spasms in tuberous sclerosis: literature review. J Child Neurol 1999;14:71–74. Chiron C, Dumas C, Jambaque I, Mumford J, Dulac O. Randomized trial comparing vigabatrin and hydrocortisone in infantile spasms due to tuberous sclerosis. Epilepsy Res 1997;26:389–395. Riikonen RS. Steroids or vigabatrin in the treatment of infantile spasms? Pediatr Neurol 2000;23:403–408. Cossette P, Riviello JJ, Carmant L. ACTH versus vigabatrin therapy in infantile spasms: a retrospective study. Neurology 1999;52:1691– 1694. Glauser TA, Clark PO, Strawsburg R. A pilot study of topiramate in the treatment of infantile spasms. Epilepsia 1998;39(12):1324–1328. Ohtsuka Y, Ogino T, Asano T, Hattori J, Ohta H, Oka E. Long-term follow-up of vitamin B6-responsive West syndrome. Pediatr Neurol 2000;23:202–206. Blennow G, Starck L. High dose B6 treatment in infantile spasms. Neuropediatrics 1986;17:7–10.
Original article
West syndrome – The University of Hong Kong experience (1970–2000) Virginia Wong* Department of Paediatrics, The University of Hong Kong, Queen Mary Hospital, Pokfulam, Hong Kong Received 21 April 2001; received in revised form 30 July 2001; accepted 2 August 2001
Abstract Objective: To study the clinical pattern of West syndrome (WS) in a university based hospital. Methodology: The database of children seen in the Epilepsy Clinic of Queen Mary Hospital and Duchess of Kent Children’s Hospital during a 30-year period (1970–2000) was reviewed. Results: A total of 105 cases had WS (1.9%). The number of new cases of WS admitted per year ranged from one to eight. The range of annual incidence of WS to newly diagnosed epilepsy was 0.8–4.8%. The etiology included idiopathic (N ¼ 19, 18%), cryptogenic (N ¼ 23; 22%), symptomatic (N ¼ 56; 53.3%) and unknown (N ¼ 7; 5.7%). Adrenocorticotropic hormone (ACTH) and/or prednisone were given to 42 children (40%). Most were effective in controlling WS on an all-or-none fashion. Seizure outcome included 12 with remission, persistent in the same form in two and persisting but changed to another form in the rest. Neurological outcome at the last follow up in 2000 December included multiple disabilities ðN ¼ 16Þ, cerebral palsy ðN ¼ 22Þ, mental retardation ðN ¼ 94Þ and Lennox–Gastaut syndrome ðN ¼ 13Þ. We analysed the following risk factors in relation to poor outcome: age of onset, age of presentation, time lag before treatment, etiology, family history of epilepsy and hormonal treatment. Only etiology of WS has a positive correlation with poor outcome ðP , 0:0005Þ. Conclusions: WS is an uncommon epileptic syndrome. The majority had poor outcome, especially those with causes identified. Infantile spasm is a specific epileptic phenomenon in a maturational stage of a child when heterogeneous disorders can present with the same clinical epileptic and electroencephalographic phenomenon. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Infantile spasm; West syndrome; Children; Seizure; Epilepsy; Vigabatrin; Risk factors; Prognosis; Annual incidence
1. Introduction West syndrome (WS) is a rare epileptic syndrome consisting of infantile spasm (IS), hypsarrhythmia and mental retardation. The etiology, seizure classification and drug treatment of WS had been under evolutionary change over the past decade. The ultimate prognosis, in general, is poor despite treatment [1–12]. The segregation of etiology into idiopathic, cryptogenic and symptomatic depends on the facilities available locally, as neuroradiologic techniques like computerized tomography (CT), magnetic resonance imaging (MRI) and positron emission tomography (PET) contributed to the rising trend of symptomatic cases [13–15]. Rare neurometabolic or neurodegenerative diseases presenting with IS contributed to a minority of symptomatic cases, and thorough workup depends on the availability of sophisticated laboratories and professionals which were unavailable in most clinical child neurology settings. Adrenocorticotropic hormone (ACTH)/steroid has been
useful in the suppression of IS. However, the use of low dose versus high dose ACTH/steroid was still under intense debate [16–24]. Although vigabatrin (VGA) had been advocated as the first line of treatment recently, its usefulness had been limited by the occurrence of visual field defect in some [25–29]. VGA had been found to be especially useful in children with WS due to tuberous sclerosis (TS) [30–32]. There has been long debate of the first antiepileptic drug (AED) of choice in WS, ACTH/steroid or VGA [33,34]. Other newer AED like topiramate was also studied in new WS cases, and reported to be useful [35]. The Japanese reported usefulness of pyridoxine in a small proportion of IS patients [36,37]. The objective is to study the clinical pattern of WS in a university based hospital during a 30-year period (1970– 2000) and to identify the risk factors relating to poor outcome.
2. Methodology * Tel: 1852-2855-4485; fax: 1852-2855-1523. E-mail address: [email protected] (V. Wong).
The database of children (,16 years on admission) seen
0387-7604/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0387-760 4(01)00296-0
610
V. Wong / Brain & Development 23 (2001) 609–615
in the Epilepsy Clinic of Queen Mary Hospital and Duchess of Kent Children’s Hospital during a 30-year period (1970– 2000) was reviewed. Both hospitals are teaching hospitals of the University of Hong Kong. Children admitted for nonepileptic convulsive disorders or febrile convulsions were excluded. Most cases had been actively followed up by the author since 1980 except for those who were lost on follow up in the 1970s. A retrospective (1970–1979) and prospective medical record (1980–2000) review was conducted. Children were included if they satisfied the diagnostic criteria of IS with flexor or extensor spasms, and WS which consisted of IS, hypsarrhythmia pattern in electroencephalography (EEG) and associated mental retardation. 2.1. Statistical analysis Pearson chi-square was used to determine six risk factors (age of onset, age of presentation, time lag before treatment, etiology, family history of epilepsy and hormonal treatment) and abnormal outcome. Logistic regression analysis was carried out after obtaining the significant factors from Pearson chi-square test. 3. Results 3.1. Epilepsy database Altogether around 5500 children had epilepsy. A total of 105 cases had WS (1.9%). The number of new cases of WS admitted per year was one to eight (Fig. 1). The increase in the number of cases in 1980s to 2000s as compared to 1970s was due to the newly developed integrated child neurology and developmental pediatrics program in 1980s by the author within the University of Hong Kong as the only centre available locally. The annual range of newly diagnosed WS in newly diagnosed epilepsy during this period was 0.8–4.8%. The sex ratio of boys:girls was 55:50 (or 1.1:1). Nineteen children were lost on follow up.
being referred to a child neurologist. Video EEG was available for seizure monitoring in 1998 in our centre. 3.4. Family history None of our WS cases had history of IS or epilepsy in the immediate family members. There were only two sisters in a family with TS. 3.5. Workup for underlying etiology The etiological workup included: renal and liver biochemistries, glucose, lactate and pyruvate (since 1985), anion gap, ammonia, urine for simple amino acid and metabolic screening for most cases. Chromosomal study was evaluated depending on the clinical features. Cerebrospinal fluid analysis was performed if clinically indicated. Blood and urine were also taken for congenital infections like rubella and cytomegalovirus (CMV) if features were suggestive such as microcephaly, developmental delay, failure to thrive and visual or hearing impairment. EEG was performed for all cases after admission. Video EEG was only available in our centre in 1998 for diagnostic evaluation and monitoring of electrophysiological response to anticonvulsants. Interictal EEG showed typical features of hypsarrhythmia in most cases except the case with hemimegalencephaly and another two children with Aicardi syndrome. Intravenous pyridoxine (vitamin B6) test with 50–100 mg pyridoxine and simultaneous EEG monitoring were performed routinely in all cases since 1980s but none showed EEG response to intravenous pyridoxine. CAT scan of the brain was available locally in the government hospitals on a free basis since 1985 and MRI of the brain since 1995. PET scan was only available locally in the private setting in 2000. Thus, the role played by neuroimaging in identifying etiology in our study depended on the availability of these tools during the initial or subsequent follow up of these cases. The number of children who underwent MRI or PET scan of the brain were 87 and 3, respectively.
3.2. Age of onset and age of presentation (Fig. 2) 3.6. Etiology (Fig. 4) The peak ages of onset and presentation both were 0–3 months. If one categorized the age intervals by 6-monthly, then the peak age of onset is centred around 6 months. The mean age at last consultation was 5.6 years (based on 2000 December follow up), with age range from 12 months to 30 years. 3.3. Clinical seizure pattern (Fig. 3) Clear-cut description of seizure pattern by chief caretaker or through the nursing record was obtained in 84 cases. Clinical features were typical in most with the majority presenting with flexor spasms. Only a few had spasms misinterpreted as normal by primary care doctors before
The etiology included idiopathic ðN ¼ 19Þ, cryptogenic ðN ¼ 23Þ, symptomatic ðN ¼ 56Þ and unknown ðN ¼ 7Þ. ‘Unknown’ means investigations like CT/MRI of the brain were not available in the 1970s for complete evaluation. Idiopathic is defined as no causes identified despite available investigations and the children had normal development prior to development of WS. Symptomatic is defined as definite causes identified with existing investigations as mentioned above. Cryptogenic is defined as negative findings with existing investigative tools available locally and yet these children had delayed development or abnormal neurological findings. One child developed IS within 1 week after receiving pertussis vaccination, and thus, the
V. Wong / Brain & Development 23 (2001) 609–615
611
Fig. 1. Yearly distribution of new cases of WS and new cases of epilepsy (1970–2000).
relationship is speculative and we categorized this as a cyptogenic case. For symptomatic cases, prenatal causes included neuronal migration disorder (N ¼ 11), congenital CMV infection (N ¼ 1), Rett syndrome (RS) (N ¼ 1), Down syndrome (DS) (N ¼ 2), Aicardi syndrome (N ¼ 2), Angelman syndrome (AS) ðN ¼ 1Þ, TS (N ¼ 10), hyperphenylalaninemia (N ¼ 1) and organic aciduria (argininosuccinase deficiency) (N ¼ 1). Perinatal included hypoxic-ischemic encephalopathy (N ¼ 3) and persistent neonatal hypoglycemia (N ¼ 1). Postnatal included meningitis (three; one with Salmonella;
two with Haemophilus influenzae) and herpes simplex encephalitis (N ¼ 1). Urine for metabolic screening was performed in all cases. Only two were positive. One had phenylketonuria and the other had organic aciduria (argininosuccinase deficiency). 3.7. Neuroimaging (Fig. 5) MRI scan of the brain was performed for 87 children. MRI features of neuronal migration disorder (N ¼ 13) included lissencephaly, pachygyria/agyria, hemimegalence-
612
V. Wong / Brain & Development 23 (2001) 609–615
Fig. 2. Ages of onset and presentation of WS.
Fig. 4. Etiology of WS.
phaly, schizencephaly and agenesis of corpus callosum. Some cases had a mixture of these radiological features. PET scan was only performed for three cases as PET is only available on a private setting. In a 12-month-old girl with intractable seizures with good initial response to VGB and having a relapse 6 months later, PET scan of the brain showed focal hypometabolism in her left temporal lobe. Epilepsy surgery may be indicated in this case. However, epilepsy surgery program has not been launched for children in Hong Kong yet. A 3.5-year-old boy with TS had hypometabolic regions corresponding to the site of tubers. He subsequently had left temporal lobectomy and lesionectomy of left parieto-occipital tubers performed in Montreal Neurological Institute under Professor F. Andermann. Another 4-year-old boy with severe mental retardation, cortical blindness and cerebral atrophy shown on MRI had global cerebral hypometabolism in PET scan.
WS. ACTH in the form of Synacthen was injected intramuscularly on a daily basis. In the 1980s, high dose ACTH (2 IU/kg/day) was given for at least an 8–12 weeks course. Low dose regime (0.2 IU/kg/day) ACTH was used for 4–6 weeks course in the 1990s. Due to the failure of production of ACTH in the market in 1997 and with the launching of VGA in Hong Kong in 1997, VGA was started as the first choice for any new case of IS. VGA led to complete cessation of IS in all cases tried within 24 h as the first line treatment. Visual field monitoring was regularly conducted for these children by the ophthalmologist and optometrist with behavioral visual field testing if cooperative, funduscopic examination and visual evoked potential evaluation. No reported visual problems occurred so far. Complete remission occurred with VGA monotherapy in etiologies like idiopathic ðN ¼ 3Þ, congenital CMV infection ðN ¼ 1Þ and organic aciduria ðN ¼ 1Þ. As for TS cases with IS, they already had the spasms long before they were seen by us and none of them responded to VGA monotherapy.
3.8. Treatment Altogether 42 children received hormonal treatment. ACTH and prednisone were given to 38 and four children, respectively. ACTH was effective in the majority of cases within the first week of injection. The response was actually seen as an all-or-none one. ACTH was given on an emergency basis in 1980–1997 for any newly admitted cases of
Fig. 3. Seizure types at presentation.
Fig. 5. MRI findings in WS ðN ¼ 87Þ.
V. Wong / Brain & Development 23 (2001) 609–615
613
We analysed the following risk factors in relation to poor outcome: age of onset, age of presentation, time lag before treatment, etiology, family history of epilepsy and hormonal treatment. Only a symptomatic etiology has a positive correlation with poor outcome ðP , 0:0005Þ. 3.10. Mortality Nine (8.57%) children died, of which seven died due to steroid related complications. One was due to disseminated intravascular coagulation and fulminant sepsis related to the use of high dose ACTH. 3.11. Risk factors and poor outcome (Table 1)
Fig. 6. Major anticonvulsants used during active seizure phase in 1980– 2000.
Only two factors were statistically significantly related to poor outcome: etiology ðP , 0:0005Þ and family history of epilepsy ðP ¼ 0:007Þ. When both risk factors were put into the regression, only etiology is statistically significant ðP ¼ 0:001Þ with coefficient of 3.845.
In 1970s to early 1980s, the main anticonvulsants used were phenobarbitone and phenytoin. Carbamazepine was not used for WS. For those who failed to respond to ACTH/steroid or having relapse after ACTH/steroid withdrawal, other anticonvulsants were used. This included valproic acid, clonazapam, nitrazepam and clobazam. Newer anticonvulsants like VGA, lamotrigine and topiramate were tried for intractable cases as they were licensed locally during the recent 5 years (Fig. 6).
4. Discussion
3.9. Outcome
4.1. Etiology in relation to development of IS
The outcome was analysed for those with active follow up until 2000 December. Altogether 19 cases were lost for follow up. For the new WS cases seen in 1999–2000, the minimum interval for follow up since diagnosis was 9 months. Seizure outcome included 12 with remission, persistent in the same form in two (cases admitted in 2000) and persisting but changed to another form in the rest. Neurological outcome included mental retardation (N ¼ 94; 89.5%), cerebral palsy (N ¼ 22; 21%), multiple disabilities (N ¼ 16; 15.2%), visual impairment (N ¼ 31; 29.5%), hearing impairment (N ¼ 2; 1.9%) and Lennox– Gastaut syndrome (N ¼ 13; 12.4%).
Idiopathic type of WS has the best prognosis. Most had normal neurological and developmental outcome. We noted that despite the multiple and heterogeneous causes of WS, if one looked at the specific etiology per se, one could find that only occasionally IS is part of the presentation. Thus, there might be some yet unexplained triggering factor in generating the WS symptomatology. In our own TS database of 70 children, 10% presented with IS. In a few cases, the depigmented spots were not very obvious such that the diagnosis was delayed until MRI scan showed the typical tubers in the brain. One girl suffering from TS with intractable IS had a focal resection of a massive tuber in her left parietal region and complete remission of her seizures occurred which lasted for 12 months in her last follow up. Both neurofibromatosis (NF) and incontinenti pigmenti (IP) can present with IS. It is interesting that in our NF database of nearly 100 cases, none had IS. Both of our IP cases present with focal or generalized seizures rather than IS. In our DS database of around 400 children only two had WS. One case was related to a severe hypoxic-ischemic event post-cardiac surgery, and another had acute myeloid leukemia as well. Thus, we failed to find the increased prevalence of WS in DS.
Table 1 Analysis of risk factors for poor outcome in WS Pearson chisquare P-value
Age of onset Age of presentation Time lag before treatment Etiology Family history of epilepsy Hormonal treatment
0.83 0.742 0.447 ,0.0005 0.007 0.856
Logistic regression P value
Coefficient
0.001 0.967
3.845 –
Our university based tertiary hospital WS database showed that WS is indeed a rare epileptic disorder. The best outcome occurred in those children with idiopathic etiology. IS is a specific epileptic phenomenon in a maturational stage of a child when heterogeneous disorders like neuronal migration disorders, TS and other syndromes can present with the same clinical epileptic and EEG signatures.
614
V. Wong / Brain & Development 23 (2001) 609–615
One of our RS cases in our RS database ðN ¼ 12Þ initially presented with IS. She subsequently regressed with autistic features and typical hand wringing stereotypic movement. Thus one needs to be alert to the possible association of WS and RS. Another girl with AS presented with WS during the clinical course. This is an unusual presentation as in our AS database of 20 cases, only one had IS. Thus, one should be alert to unusual jerky movement or features of unexplained laughter in children with WS and the possible association of WS and AS. In our database of around 50 cases of congenital CMV infection, only one had IS. None of our cases with other congenital infections like rubella presented with IS. In our neurometabolic disease database ðN ¼ 50Þ, two cases presented with IS. One girl with organic aciduria (argininosuccinase deficiency) presented with intractable IS which responded immediately to VGA. Another girl had hyperphenylalaninemia and moderate degree of mental retardation with IS subsequently changed to other seizure types. In our neurodegenerative disease database, only one girl with infantile type of neuronal ceroid lipofuscinosis diagnosed by skin biopsy presented with IS. She subsequently died of her own disease. We performed intravenous pyridoxine infusion (50– 100 mg) during EEG monitoring since 1980s. However, none of our IS cases had EEG response. Oral pyridoxine at 200–400 mg/day was continued for some WS cases intractable to all conventional AEDs, but none had any positive response. 4.2. EEG Most of our WS cases had hypsarrhythmia pattern in the inerictal EEG. One boy with hemimegalencephaly had hemihypsarrhythmia. Other atypical hypsarrhythmic features included burst suppression pattern in the child with hemimegalencephaly and unilateral hypsarrhythmia in Aicardi syndrome. Video EEG monitoring was available in our centre in 1998 and was mainly used for diagnostic evaluation and monitoring of response of seizure to anticonvulsants. For the cases treated with VGA, the hysparrhythmic pattern still persisted even when the clinical seizures were well controlled. This was unlike those with ACTH treatment where EEG and clinical remission correlated. For those with VGA treatment, the EEG pattern might evolve to generalized slow sharp and spike wave discharges during the clinical course. 4.3. Neuroimaging In the days with purely computerized axial tomography (CAT) scan, almost all cases showed pseudo-cerebral atrophy after the use of ACTH/steroid. With the availability of MRI scan locally in 1995, more children with cortical dysplasia were identified, thus shifting the cryptogenic cases to symptomatic etiology. The PET scan was useful
in identifying focal hypometabolic lesions not detected by MRI of the brain. One 13-month-old girl with intractable WS and normal MRI had a focal hypometabolic area in her left parieto-temporal region. These children might be good candidates for epilepsy surgery. 4.4. Treatment In the era before the availability of VGA locally, the first line treatment for WS was ACTH injection, and the response rate was high and immediate (usually within a week). However, the relapse rate was relatively high as well. In 1980s for parents of patients who were reluctant to allow their children to try ACTH or oral prednisone as the patients were too young to give consent, the commonest first line AED used was sodium valproate, followed by benzodiazepines. Phenobarbitone or phenytoin was used in 1970s to 1980s. The availability of VGA locally in 1997 replaced valproic acid as the AED of first choice. The local availability of VGA reached a turning point when the pharmaceutical company stopped manufacturing ACTH gel (Synacthen). Altogether, 38 children with WS had received VGA as the initial treatment or during the active seizure phase. VGA was successful in inducing seizure cessation almost within 24 h in those given as initial monotherapy. It had been an all-or-none effect. Even children with symptomatic etiology responded. One 4-month-old child with organic aciduria (argininosuccinase deficiency) who failed to respond to sodium valproate (VPA) or clonazepam responded after the first 500 mg dose of VGA. She remained seizure free for at least 24 months at the last follow up. Another 7-month-old child with congenital CMV infection also achieved complete and long lasting seizure control with VGA monotherapy within 12 h of a 250 mg dose. Another 6-month-old girl with idiopathic WS who failed to respond to VPA or clonazepam for 4 weeks also achieved complete cessation of seizures with a single dose of 500 mg VGA. However, hypsarrhythmia pattern failed to respond to treatment with VGA even with clinical remission in most children with IS. Other newer AEDs were used during the active course of seizure which were unresponsive to VPA or benzodiazepines. However, the response to lamotrigine and topiramate was not as satisfactory as reported. Ketogenic diet was tried in two children during the intractable seizure course, with one having complete remission for 2 years with only one anticonvulsant as maintenance. As WS is such a rare disorder with a heterogeneous etiological background, the collaboration of multicentre clinical trials is urgently needed to improve the ultimate prognosis of this catastrophic epileptic syndrome. Acknowledgements We thank our paediatric colleagues who participated in medical care of these children during the period 1970–2000.
V. Wong / Brain & Development 23 (2001) 609–615
This paper was presented in the 4th International Symposium of the West Syndrome and Other Infantile Epileptic Encephalopathies (ISWS) in Tokyo, Japan in 2001, February 9–11.
[19]
[20]
References [1] Appleton RE. Infantile spasms. Arch Dis Child 1993;69:614–618. [2] Cowan LD, Hudson LS. The epidemiology and natural history of infantile spasms. J Child Neurol 1991;6:355–364. [3] Hrachovy RA, Glaze DG, Frost JD. A retrospective study of spontaneous remission and long-term outcome in patients with infantile spasms. Epilepsia 1991;32(2):212–214. [4] Meencke HJ, Gerhard C. Morphological aspects of etiology and the course of infantile spasms (West-syndrome). Neuropediatrics 1985;16:59–66. [5] Bobele GB, Bodensteiner JB. Infantile spasms. Neurol Clin 1990;8:633–645. [6] Bellman MH, Ross EM, Miller DL. Infantile spasms and pertussis immunisation. Lancet 1983;1:1031–1034. [7] Lombroso CT. A prospective study of infantile spasms: clinical and therapeutic correlations. Epilepsia 1983;24:135–158. [8] Riikonen R. A long-term follow-up study of 214 children with the syndrome of infantile spasms. Neuropediatrics 1982;13:14–23. [9] Matsumoto A, Watanabe K, Negoro T, Sugiura M, Iwase K, Hara K, et al. Long-term prognosis after infantile spasms: a statistical study of prognostic factors in 200 cases. Dev Med Child Neurol 1981;23:51– 65. [10] Kurokawa T, Goya N, Fukuyama Y, Suzuki M, Seki T, Ohtahara S. West syndrome and Lennox–Gastaut syndrome: a survey of natural history. Pediatrics 1980;65:81–88. [11] Riikonen R, Donner M. Incidence and aetiology of infantile spasms from 1960 to 1976: a population study in Finland. Dev Med Child Neurol 1979;21:333–343. [12] Koo B, Hwang PA, Logan WJ. Infantile spasms: outcome and prognostic factors of cryptogenic and symptomatic groups. Neurology 1993;43:2322–2327. [13] Aydinli N, Caliskan M, Ozmen M, Tonguc E. Neuroradiologic aspects of West syndrome. Pediatr Neurol 1998;19:211–216. [14] Okumura A, Watanabe K, Negoro T, Aso K, Natsume J, Kubota T, et al. Evolutional changes and outcome of West syndrome: correlation with magnetic resonance imaging findings. Epilepsia 1998;39(Suppl 5):46–49. [15] Chugani HT, Conti JR. Etiologic classification of infantile spasms in 140 cases: role of positron emission tomography. J Child Neurol 1996;11:44–48. [16] Snead OC. Treatment of infantile spasms. Pediatr Neurol 1990;6:147–150. [17] Bobele GB, Bodensteiner JB. The treatment of infantile spasms by child neurologists. J Child Neurol 1994;9:432–435. [18] Shields WD, Shewmon DA, Chugani HT, Peacock WJ. Treatment of
[21]
[22]
[23] [24]
[25]
[26] [27]
[28]
[29] [30]
[31]
[32]
[33] [34]
[35] [36]
[37]
615
infantile spasms: medical or surgical? Epilepsia 1992;33(Suppl 4):S26–S31. Hrachovy RA, Frost JD, Glaze DG. High-dose, long-duration versus low-dose, short-duration corticotropin therapy for infantile spasms. J Pediatr 1994;124:803–806. Snead OC, Benton JW, Myers GJ. ACTH and prednisone in childhood seizure disorders. Neurology 1983;33:966–970. Snead OC, Benton JW, Hosey LC, Swann JW, Spink D, Martin D, et al. Treatment of infantile spasms with high-dose ACTH: efficacy and plasma levels of ACTH and cortisol. Neurology 1989;39:1027–1031. Glaze DG, Hrachovy RA, Frost JD, Kellaway P, Zion TE. Prospective study of outcome of infants with infantile spasms treated during controlled studies of ACTH and prednisone. J Pediatr 1998;112:389–396. Riikonen R, Donner M. ACTH therapy in infantile spasms: side effects. Arch Dis Child 1980;55:664–672. Yanagaki S, Oguni H, Hayashi K, Imai K, Funatuka M, Tanaka T, et al. A comparative study of high-dose and low-dose ACTH therapy for West syndrome. Brain Dev 1999;21:461–467. Appleton RE, Peters AC, Mumford JP, Shaw DE. Randomised, placebo-controlled study of vigabatrin as first-line treatment of infantile spasms. Epilepsia 1999;40:1627–1633. Appleton RE. Guideline for prescribing vigabatrin. BMJ 1998;317:1322. Aicardi J, Mumford JP, Dumas C, Wood S. Vigabatrin as initial therapy for infantile spasms: a European retrospective survey. Epilepsia 1996;37:638–642. Fejerman N, Cersosimo R, Caraballo R, et al. Vigabatrin as a firstchoice drug in the treatment of West syndrome. J Child Neurol 2000;15:161–165. Vigabatrin Paediatric Advisory Group. Guideline for prescribing vigabatrin in children has been revised. BMJ 2000;320:1404–1405. Fukushima K, Inoue Y, Fujiwara T, Yagi K. Long-term course of West syndrome associated with tuberous sclerosis. Epilepsia 1998;39(Suppl 5):50–54. Hancock E, Osborne JP. Vigabatrin in the treatment of infantile spasms in tuberous sclerosis: literature review. J Child Neurol 1999;14:71–74. Chiron C, Dumas C, Jambaque I, Mumford J, Dulac O. Randomized trial comparing vigabatrin and hydrocortisone in infantile spasms due to tuberous sclerosis. Epilepsy Res 1997;26:389–395. Riikonen RS. Steroids or vigabatrin in the treatment of infantile spasms? Pediatr Neurol 2000;23:403–408. Cossette P, Riviello JJ, Carmant L. ACTH versus vigabatrin therapy in infantile spasms: a retrospective study. Neurology 1999;52:1691– 1694. Glauser TA, Clark PO, Strawsburg R. A pilot study of topiramate in the treatment of infantile spasms. Epilepsia 1998;39(12):1324–1328. Ohtsuka Y, Ogino T, Asano T, Hattori J, Ohta H, Oka E. Long-term follow-up of vitamin B6-responsive West syndrome. Pediatr Neurol 2000;23:202–206. Blennow G, Starck L. High dose B6 treatment in infantile spasms. Neuropediatrics 1986;17:7–10.