Cognitive development in Dravet syndrome: A retrospective, multicenter study of 26 patients

Transcription

1 FULL-LENGTH ORIGINAL RESEARCH Cognitive development in Dravet syndrome: A retrospective, multicenter study of 26 patients *Francesca Ragona, *Tiziana Granata, ybernardo Dalla Bernardina, yfrancesca Offredi, yfrancesca Darra, zdomenica Battaglia, *Monica Morbi, xdaniela Brazzo, {Simona Cappelletti, zdaniela Chieffo, *Ilaria De Giorgi, yelena Fontana, *Elena Freri, **Carla Marini, yyalessio Toraldo, zznicola Specchio, xpierangelo Veggiotti, zzfederico Vigevano, **Renzo Guerrini, zfrancesco Guzzetta, and xxcharlotte Dravet *Department of Pediatric Neuroscience, IRCCS Foundation Neurological Institute C. Besta, Milano, Italy; ychild Neuropsychiatry Unit, University of Verona, Verona, Italy; zdepartment of Child Neurology and Psychiatry, Catholic University, Roma, Italy; xdepartment of Child Neuropsychiatry, Foundation Neurological Institute C. Mondino, Pavia, Italy; {Clinical Neuropsychology Unit, Bambino Gesu Children s Hospital, Roma, Italy; **Child Neurology Unit, Pediatric Hospital A. Meyer, University of Florence, Firenze and IRCCS Stella Maris, Pisa, Italy; yydepartment of Psychology, University of Pavia, Pavia, Italy; zzneurology Department Bambino Gesu Children s Hospital, Roma, Italy; and xxdepartment of Child Neurology and Psychiatry, Honorary Consultant Catholic University, Roma, Italy SUMMARY Purpose: To clarify the role of epilepsy and genetic background in determining the cognitive outcome of patients with Dravet syndrome. Methods: In this retrospective study, we reviewed the clinical history and cognitive development of 26 patients who had been followed with standardized evaluations since seizure onset. The cognitive outcome was quantified as differential general quotient (dgq) between ages 12 and 60 months. Statistical analysis correlated the dgq with genotype and epilepsy course. Key Findings: Epilepsy started at the mean age of 5.6 months. All patients experienced prolonged convulsive seizures, whereas absences and myoclonus were reported in 17. Cognitive outcome was poor in almost all patients; the mean dgq was 33 points, varying from 6 77 points. The analysis of individual cognitive profiles identified seven patients in whom the dgq was <20 points; the main clinical characteristic in this subset of patients was lack of early absences and myoclonus. The statistical analysis of the whole series failed to reveal significant differences in cognitive outcome with regard to the presence of SCN1A mutations and their type. In particular, mutation-carrier patients with the best cognitive outcome harbored either missense or truncating mutations. Significance: Dravet syndrome encompasses different epileptic and cognitive phenotypes that probably result from both genetic and epigenetic factors. In this series, early appearance of myoclonus and absences was associated with the worst cognitive outcome. KEY WORDS: Severe myoclonic epilepsy in infancy, Cognitive development, Epileptic status, SCN1A, Myoclonus. Dravet syndrome (DS), also known as severe myoclonic epilepsy in infancy (SMEI), is an epileptic encephalopathy with onset in the first year of life, in previously healthy children, and characterized by febrile and afebrile, generalized seizures, focal seizures, or both. Family history of epilepsy and/or febrile seizures is frequent. The course is characterized by recurrent polymorphic seizures (myoclonic, focal, atypical absences, and generalized and unilateral clonic and tonic clonic seizures) slowing of psychomotor Accepted October 29, 2010; Early View publication January 26, Address correspondence to: Tiziana Granata, Department of Pediatric Neuroscience, IRCCS Foundation Neurological Institute C. Besta, Via Celoria 11, Milano, Italy. granata@istituto-besta.it Wiley Periodicals, Inc. ª 2011 International League Against Epilepsy development, and appearance of neurologic signs (Dravet, 1978). Stagnation of psychomotor development becomes obvious during the second year of life so that all patients are delayed after the age of third years (Dravet et al., 1982; CassØ-Perrot et al., 2001). Electroencephalography (EEG), which is usually normal at onset, will subsequently show generalized and focal discharges, photosensitivity, and slowing of background activity (Dalla Bernardina et al., 1982; Dravet et al., 2002). Prognosis is unfavorable in most cases: seizures become drug-resistant; all patients exhibit motor and cognitive impairment. Mortality rate is high (Dravet et al., 2005). In most patients the syndrome results from de novo mutations of the SCN1A gene, which encodes the voltage-gated sodium channel Nav1.1 (Claes et al., 2001). Many studies have failed to find a correlation 386

2 387 Cognitive Development in Dravet Syndrome between type of mutation and phenotype severity (Ohmori et al., 2002; Wallace et al., 2003; Fukuma et al., 2004; Oguni et al., 2005). Although cognitive and behavioral disorders are prominent, only a few studies have addressed these disorders. CassØ-Perrot and coworkers studied a series of 20 children and found slowing of cognitive development in all of them; in most, severe mental retardation was apparent before the age of 6, and there was a correlation trend between frequency of convulsive seizures and degree of mental impairment (CassØ-Perrot et al., 2001; Wolff et al., 2006). To clarify the respective contribution of epilepsy and of the underlying genetic abnormality in determining cognitive outcome, we retrospectively examined the clinical histories of 26 patients. In particular, we aimed at: (1) describing the course of epilepsy in the first years of life, (2) evaluating cognitive development, (3) analyzing the correlation between the course of epilepsy and cognitive development, and (4) finding out any possible correlations between causative mutations and cognitive outcome. Patients and Methods We studied 26 patients (13 male), now aged 5 19 years, observed in five Italian centers that formed a study group led by one of the authors (Ch.D.). Inclusion criteria were: (1) clinical diagnosis of Dravet syndrome (Engel, 2001); (2) screening for molecular analysis of the SCN1A gene; (3) clinical course clearly traceable on the basis of clinical charts and family diaries, particularly with regard to frequency and semiology of seizures; and (4) at least two standardized cognitive evaluations by the age of 5, of which the first performed within 1 year from the onset of seizures. The upper limit of assessment was set based on previous studies indicating that arrest of cognitive development occurs within the first years after seizure onset (CassØ- Perrot et al., 2001; Ragona et al., 2010). The average delay from the first seizure to the first evaluation was 5.9 months (range 0 13 months); the mean number of evaluations per patient was 4.6 (range 2 7). Other investigations included EEG recordings while awake and asleep, with video and polygraphic monitoring and brain MRI. Mutation analysis of the SCN1A gene was performed for all patients, including mutational screening by denaturing high performance liquid chromatography (DHPLC) followed, whenever necessary, by multiplex ligation probe amplification (MLPA). To face the intrinsic limits of a retrospective study, preliminary meetings of the group were held to elaborate a database for collecting clinical information and defining the potential predictors of cognitive outcome. Agreement was reached on the definition of type of seizures, particularly with regard to their duration and to the definition of epileptic status (see subsequent text). Finally, the group chose to assess cognitive development using the Griffiths Mental Development Scale and Brunet Lezine scales: these scales were selected, among several other tests performed in the different centers, because they had been serially performed in all patients until the age of 5 years, and because of their reliability in providing a figure of the general quotient. The epilepsy histories and EEG samples of each patient were thoroughly reviewed by one of the authors (Ch.D.) during further ad hoc meetings. Particular attention was paid to age at onset and type of initial seizures, subsequent seizure types, and mean frequency of convulsive seizures, absences, myoclonic seizures, as well as to the count of prolonged convulsive seizures (lasting between 10 and 30 min) and of episodes of status (lasting >30 min). Statistical analysis was carried out by general linear models and included the following clinical variables: age at seizure onset, number of episodes of early status epilepticus (before age 18 months), proportion of months with absences and/or myoclonic seizures before age 36 months, number of prolonged convulsive seizures per year and number of seizure-free months per year (in the period between onset and the last cognitive assessment). These variables were used as predictors of cognitive development. For each patient the serial general quotients (GQs) at different epochs during the course were interpolated by linear regression; the differential general quotient (dgq 12 60) was obtained by comparing the GQ at 12 months with the GQ obtained at 60 months as resulting from the linear interpolation. The statistical analysis correlated the clinical variables and the dgq 12 60, as well as genotype and dgq Results Epilepsy course The age at first seizure ranged between 3 and 10 months (mean months); in 17 patients (65% of cases) the onset was within the sixth month of life. All patients experienced convulsive seizures lasting for longer than 10 min, and 21 experienced at least one episode of status epilepticus during the first 5 years of life. Prolonged convulsive seizures and status epilepticus were particularly frequent between the age of 6 and 18 months; the mean number of episodes of status before 18 months was (range 0 7). Twenty-three children had at least one focal seizure during the first 5 years of life; 17 experienced focal or massive myoclonic jerks and/or atypical absences that appeared within 36 months in 13 of them and in the fourth year of life in the remaining 4 (Table 1). Given the high frequency of prolonged convulsive seizures in the first year after onset and the appearance of absences and myoclonic jerks during the second and third years, seizure-free periods were very short in all cases. Fever sensitivity was reported in 21 patients, and photosensitivity in 8. Interictal paroxysmal EEG abnormalities appeared in all patients during the second year of life; a photo paroxysmal EEG response was

3 388 F. Ragona et al. Table 1. Clinical data of the series Case, gender, present age (years) AEDs during the first 18 months of life Differential GQ (GQ 12 months GQ 60 months) GQ at 12 GQ at 60 months a months a Age at onset (months) Number of epileptic status <18 months Mean number of prolonged seizures per year b Number of months with absences and/or myoclonus during the first 3 years of life Group 1 1, M, 19 PB c Truncating 2, F, 6.5 VPA Truncating 3, F, 8 VPA, VPA + BDZ Truncating 4, M, 17 PB, VPA Negative 5, F, 11 PB, VPA + BDZ Negative 6, F, 9 VPA, VPA + BDZ Truncating 7, M, 8.5 VPA, PB + BDZ Missense 8, F, 12 PB, PB + VPA, VPA + BDZ Genetic analysis Negative 9, M, 11 VPA Truncating 10, M, 14 PB Truncating 11, F, 7.5 VPA Truncating 12, F, 15 VPA, VPA + PB Truncating 13, F, 10 PB Truncating 14, M, 15 VPA + PB Truncating 15, M, 13 VPA + TPM c Truncating 16, M, 18 VPA, VPA + BDZ Truncating 17, F, 8.5 VPA + PB Negative 18, F, 19 PB, VPA + BDZ Truncating 19, M, 9.5 PB + VPA, VPA Truncating Group 2 20, M, 8.5 VPA + PB c Missense 21, F, 7 PB Truncating 22, F, 7.5 VPA, VPA + BDZ Truncating 23, M, 13 VPA c Negative 24, M, 15 VPA Negative 25, M, 7 VPA Missense 26, M, 5.5 PB Truncating M, male; F, female; AEDs, antiepileptic drugs; GQ, general quotient; PB, Phenobarbital; VPA, valproate; BDZ, benzodiazepine; TPM, topiramate; GQ, general quotient. a GQ as obtained from a linear interpolation. b Computed between onset and last cognitive assessment. c Myoclonia appeared in the fourth year of life.

4 389 Cognitive Development in Dravet Syndrome detected in 13 patients (50% of cases) within 42 months of life. The first-line treatment was valproate (16 patients) or phenobarbital monotherapy (10 patients), but given the recurrence of seizures, polytherapy was used in 23 of 26 cases during the first three years of life (Table 1). Five patients received, during the course of the disease, inappropriate treatments for short periods: carbamazepine (four cases), phenytoin (one case), and lamotrigine (one case). These drugs were rapidly withdrawn after worsening of seizures or occurrence of status epilepticus. One patient died at the age of 7 years after drowning caused by a generalized convulsive seizure. Neuroimaging MRI was judged to be normal in 23 (88%) of 26 patients. In the three remaining patients, mild abnormalities were found, including mild supratentorial atrophy, ventricular enlargement, and arachnoid cyst in the middle cranial fossa. life. They were mainly classified as attention deficit (18 cases), hyperactivity (15 cases), and autistic-like behaviors (four cases). Cognitive slowing resulted in a decline of differential general quotient (dgq 12 60) of variable severity: the mean dgq was 33 points, ranging from 6 77 (Table 1). The analysis of individual cognitive profiles showed that 19 patients (cases 1 19) exhibited a steep decrease in GQ (group 1, Fig. 2), whereas in seven patients (cases 20 26), who had been assessed at comparable ages, the decrease was less steep, and the dgq was lower than 20 points (group 2, Fig. 3). On the basis of this arbitrary cut-off of 20 points of dgq, we observed that, although the difference was statistically nonsignificant, absences and myoclonus during the first 3 years of life were present in 12 of 19 patients of group 1 (63%) and in only one patient of group 2 (14%). By contrast, the two groups did not differ in age at onset (group 1: months; group 2: months) or presence of episodes of status epilepticus Genetic analysis The genetic analysis detected de novo mutations of the SCN1A gene in 20 patients (77% of cases), of which truncating mutations were observed in 17 and missense mutations in 3. MLPA did not identify deletions in the six mutationnegative patients (Table 1). Cognitive and neurologic course At the first evaluation, performed at a mean age of 11 months (range 4 26 months), neurologic examination was normal in all children, and the general quotient was within the normal range in all but two (cases 8 and 19, Table 1). At the age of 5, twelve patients exhibited mild neurologic deficits (ataxia or pyramidal signs). In all children, slowing of cognitive achievements was evident from the second year of life (Fig. 1), and in 20 patients behavioral disorders appeared during the first 4 years of Figure 2. Cognitive development of patients of group 1 (cases 1 19, Table 1), mean decrease of GQ is 39 points. Epilepsia ILAE Figure 1. Cognitive development of individual patients. Mean decrease of GQ is 33 points. Epilepsia ILAE Figure 3. Cognitive development of patients of group 2 (cases 20 26, Table 1), mean decrease of GQ is 12 points. Epilepsia ILAE

5 390 F. Ragona et al. during the first 18 months of life (group 1: 13 of 19 patients; group 2: 7 of 7 patients). Statistical analysis: epilepsy course and cognitive development Statistical analysis failed to detect any significant correlation between severity of cognitive impairment and the following clinical variables: age at seizure onset, months with absences and myoclonus during the first 3 years of life, mean number of prolonged convulsive seizures and of episodes of convulsive status epilepticus from the onset to the last cognitive assessment, and mean number of seizure-free months. A positive correlation, by contrast, was found between the number of episodes of convulsive status in the first 18 months of life and GQ at 60 months (F 1,15 = 4.569, p = 0.049): the higher the number of episodes, the higher the GQ at 60 months. Statistical analysis: genotype and cognitive development The statistical analysis failed to reveal significant differences between patients carrying mutations and mutationnegative cases, or between patients with missense mutations and those with truncating mutations. Discussion In the patients included in this study, the course of epilepsy was typical for DS, with recurrent, drug-resistant, polymorphic seizures and episodes of status epilepticus. Nevertheless, different epilepsy phenotypes were observed: In some patients the epilepsy course was characterized by early recurrence of status (patients 10, 14, 17, 20, 23, and 25), whereas in the remaining the most prominent seizures were early myoclonus and absences (cases 2, 3, 6, 7, 10, 11, 18, and 19). Finally, nine patients, never experiencing myoclonic seizures or atypical absences (cases 4, 5, 9, 12, 17, 21, 22, 25, and 26), were classified as SMEI borderline (SMEB). Slowing of cognitive achievements became evident during the course in all patients, and resulted in decline of GQ. Nonetheless a wide range of severity was still apparent. In most patients (1 19, group 1) there was severe decline, whereas in a minority, at the end of follow-up the cognitive deficit was mild, with the GQ being higher than 69 in 7 patients (cases 20 26, group 2). Cognitive slowing and appearance of ataxia and pyramidal signs were described as traits of the syndrome since the original report by Dravet (Dravet, 1978). In 2001 the syndrome was included among the epileptic encephalopathies in the Proposed Diagnostic scheme for epilepsy (Engel, 2001), assuming that the epileptic activity might contribute to the mental and motor impairment. In the same year, however, the demonstration of de novo mutation of the SCN1A gene (Claes et al., 2001) in DS opened new perspectives in the understanding of the pathogenesis of both epilepsy and associated neurologic manifestations, raising the issue of the relative role of genetic background and epilepsy course in influencing mental outcome. The analysis of the whole series of patients did not provide statistical evidence of a correlation between the presence and type of mutation and cognitive outcome. In particular, mutated patients with the best cognitive outcome carried both missense and truncating mutations. However, a more reliable correlation between genotype and phenotype will hopefully be clarified by prospective studies on larger populations. The correlation between epilepsy course and cognitive outcome has been investigated only in a few, small series of patients and with conflicting results. Wolff and coworkers studied 12 patients (Wolff et al., 2006) with a mean age of 5 years (range 11 months to 12 years) and observed a correlation trend between frequency of convulsive seizures and mental deterioration. In that series, the three children with less than five seizures per month performed better than those with higher frequency. By contrast, more recently, Akiyama and coworkers failed to find a significant correlation between seizure frequency, number of episodes of convulsive status epilepticus, and mental abilities in 31 patients with DS followed up to until the age of at least 18 years (Akiyama et al., 2010). Our data suggest that the epilepsy phenotype actually bears a prognostic value and that, in addition to convulsive seizures, other variables are likely to concur in influencing mental outcome. The analysis of individual cognitive profiles indeed demonstrates that the early appearance of myoclonus and absences is associated with the worst cognitive outcome, whereas convulsive seizures and convulsive status do not represent, per se, a bad prognostic factor for mental outcome. The main clinical characteristic in patients with the less severe cognitive outcome (dgq <20 points) was in fact the lack of early absences and myoclonus, irrespectively, from the presence and number of prolonged seizures or convulsive status, which in some patients had recurred at high frequency during infancy (see Table 1). The unexpected positive correlation between the number of episodes of convulsive status during the first 18 months of life and the GQ at 60 months remains difficult to explain. We ruled out a number of possible artifacts, for example, differences in the time course of assessments or spurious correlations between predictors, but we could find no obvious alternative explanations. It is also possible that differences in seizure semiology (e.g., generalized versus hemigeneralized), duration, and treatment of convulsive status might have misled the results, but this bias is inevitable in a retrospective study. A parsimonious explanation of our data may be that two different epilepsy phenotypes, one dominated by convulsive seizures and one dominated by non convulsive seizures, may be associated with a different cognitive outcome. The possibility of a variant form of Dravet syndrome lacking absences and myoclonus has been described under

6 391 Cognitive Development in Dravet Syndrome the heading of SMEB since 1988 (Ogino et al., 1988) and commonly defined as SMEI borderline (SMEB). This borderline form, generally speaking, shares with SMEI the long-term prognosis, including seizure frequency and mental outcome (Ohtsuka et al., 1991; Oguni et al., 2001; Akiyama et al., 2010), as well as the genetic background, and may, therefore, be considered a variant phenotype within the spectrum of Dravet syndrome. Some authors reported that in a minority of patients with SMEB, mental abilities may, at least in part, be preserved (Dalla Bernardina et al., 1987; Dulac et al., 1990; Oguni et al., 2005). In our series, nine patients were classified as SMEB: among these, four had a relatively good mental outcome, thereby confirming that SMEB does not have an invariably gloom prognosis as SMEI. The correlation between epilepsy course and cognitive outcome should consider at least two additional issues, in addition to seizures: the characteristics of EEG and treatment. These two variables were not included in the statistical analysis because EEG studies had been performed with different timing in the various patients and each patient had been treated with different drug associations at different times. Although we never observed severe interictal epileptiform EEG activity, the potential impact of interictal discharges on cognitive development cannot be disregarded, and should be the focus of prospective ad hoc studies. As far as treatment is concerned, most of our patients received appropriate therapy, and inappropriate drugs, when given, had been rapidly withdrawn because of worsening of seizure frequency. However, it is known that antiepileptic drugs, and particularly polytherapies, play a detrimental role on cognitive development (Mula & Trimble, 2009). It will be of interest to compare cognitive development in elder and in newly diagnosed patients to verify if early and appropriate treatment with less aggressive treatment schedules and the use of new drugs (e.g., stiripentol) imply a better cognitive outcome. On the whole, our data suggest that the epilepsy phenotype contributes in determining cognitive development, and that the early appearance of myoclonus and/or absences might have a negative prognostic impact. However, the wide range of cognitive impairment suggests that, in addition to seizures, other variables, including antiepileptic drug treatment, rehabilitation, and familial environment, greatly impact cognitive development, as in other epilepsy syndromes (Noeker et al., 2005; Austin & Caplan, 2007). The variability of outcome, and the appearance of neurologic deficits hardly ascribable to epileptic activity only (e.g., ataxia, extrapyramidal signs) (Ragona et al., 2010), suggest that the channelopathy itself is probably crucial in determining the phenotype. The role of channelopathies in inducing symptoms other than seizures is well documented in the mouse genetic model of SMEI in which a selective failure of excitability of c-aminobutyric acid (GABA)ergic neurons, resulting from SCN1A mutations, underlies both epilepsy and motor deficits (Kalume et al., 2007). It has been demonstrated that reduced excitability of cerebellar Purkinje neurons, observed in the mutant mice, is by its own sufficient to cause ataxia (Grusser-Cornehls & Baurle, 2001). It is conceivable that the genetic determinants (type of mutations, and modulating and epigenetic factors) may underlie different epilepsy and mental phenotypes, and that epilepsy is just one of the variables that concur in determining the overall outcome. Acknowledgments This work has been supported by a grant from Mariani Foundation. We thank Mrs Marina De Negri for her editorial assistance. Disclosure We confirm that we have read the Journal s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. None of the authors has any conflict of interest to disclose. References Akiyama M, Kobayashi K, Yoshinaga H, Ohtsuka Y. (2010) A long-term follow-up study of Dravet syndrome up to adulthood. Epilepsia 51: Austin JK, Caplan R. (2007) Behavioral and psychiatric comorbidities in pediatric epilepsy: toward an integrative model. Epilepsia 48: CassØ-Perrot C, Wolff M, Dravet C. (2001) Neuropsychological aspects of severe myoclonic epilepsy in infancy. In Jambaque I, Lassonde M, Dulac O (Eds) Neuropsychology of childhood epilepsy. Kluwer Academic/ Plenum Publishers, New York, pp Claes L, Del-Favero J, Ceulemans B, Lagae L, Van Broeckhoven C, De Jonghe P. (2001) De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsy of infancy. Am J Hum Genet 68: Dalla Bernardina B, Capovilla G, Gattoni MB. (1982) Epilepsie myoclonique grave de la premi re annøe. Rev EEG Neurophysiol 12: Dalla Bernardina B, Capovilla G, Chiamenti C. (1987) Myoclonic epilepsies of infancy and early childhood: nosological and prognostic approach. In Wolf P, Darn M, Janz D, Dreifuss FE (Eds) Advances in epileptology. Raven Press, New York, pp Dravet C. (1978) Les Øpilepsies graves de l enfant. Vie MØd 8: Dravet C, Roger J, Bureau M, Dalla Bernardina B. (1982) Myoclonic epilepsies in childhood. In Akimoto H, Kazamatsuri H, Seino M, Ward A (Eds) Advances in epileptology: the 13 Epilepsy International Symposium. Raven Press, New York, pp Dravet C, Bureau M, Oguni H. (2002) Severe myoclonic epilepsy in infancy (Dravet Syndrome). In Roger J, Bureau M, Dravet C, Jenton P, Tassinari CA, Wolff P (Eds) Epileptic syndromes in infancy, childhood and adolescence. John Libbey, London, pp Dravet C, Bureau M, Oguni H, Fukuyama Y, Cokar O. (2005) Severe myoclonic epilepsy in infancy: Dravet syndrome. Adv Neurol 95: Dulac O, Plouin P, Chiron C. (1990) Benign form of myoclonic epilepsy in children. Neurophysiol Clin 20: Engel J Jr; International League Against Epilepsy (ILAE). (2001) A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE task force on classification and terminology. Epilepsia 42: Fukuma G, Oguni H, Shirasaka Y, Watanabe K, Miyajima T, Yasumoto S, Ohfu M, Inoue T, Watanachai A, Kira R, Matsuo M, Muranaka H, Sofue F, Zhang B, Kaneko S, Mitsudome A, Hirose S. (2004) Mutations of neuronal voltage-gated na+ channel alpha 1 subunit gene SCN1A in

7 392 F. Ragona et al. core severe myoclonic epilepsy in infancy (SMEI) and in borderline SMEI (SMEB). Epilepsia 45: Grusser-Cornehls U, Baurle J. (2001) Mutant mice as a model for cerebellar ataxia. Prog Neurobiol 63: Kalume F, Yu FH, Westenbroek RE, Scheuer T, Catterall WA. (2007) Reduced sodium current in Purkinje neurons from Nav1.1 mutant mice: implications for ataxia in severe myoclonic epilepsy in infancy. J Neurosci 27: Mula M, Trimble MR. (2009) Antiepileptic drug-induced cognitive adverse effects: potential mechanisms and contributing factors. CNS Drugs 23: Noeker M, Haverkamp-Krois A, Haverkamp F. (2005) Development of mental health dysfunction in childhood epilepsy. Brain Dev 27:5 16. Ogino T, Ohtsuka Y, Amano R, Yamatogi Y, Ohtahara S. (1988) An investigation on the borderland of severe myoclonic epilepsy in infancy. Jpn J Psychiatry Neurol 42: Oguni H, Hayashi K, Awaya Y, Fukuyama Y, Osawa M. (2001) Severe myoclonic epilepsy in infants a review based on the Tokyo women s medical university series of 84 cases. Brain Dev 23: Oguni H, Hayashi K, Osawa M, Awaya Y, Fukuyama Y, Fukuma G, Hirose S, Mitsudome A, Kaneko S. (2005) Severe myoclonic epilepsy in infancy: clinical analysis and relation to SCN1A mutations in a Japanese cohort. Adv Neurol 95: Ohmori I, Ouchida M, Ohtsuka Y, Oka E, Shimizu K. (2002) Significant correlation of the SCN1A mutations and severe myoclonic epilepsy in infancy. Biochem Biophys Res Commun 295: Ohtsuka Y, Maniwa S, Ogino T, Yamatogi Y, Ohtahara S. (1991) Severe myoclonic epilepsy in infancy: a long-term follow-up study. Jpn J Psychiatry Neurol 45: Ragona F, Brazzo D, De Giorgi I, Morbi M, Freri E, Teutonico F, Gennaro E, Zara F, Binelli S, Veggiotti P, Granata T. (2010) Dravet syndrome: early clinical manifestations and cognitive outcome in 37 italian patients. Brain Dev 32: Wallace RH, Hodgson BL, Grinton BE, Gardiner RM, Robinson R, Rodriguez-Casero V, Sadleir L, Morgan J, Harkin LA, Dibbens LM, Yamamoto T, Andermann E, Mulley JC, Berkovic SF, Scheffer IE. (2003) Sodium channel alpha1-subunit mutations in severe myoclonic epilepsy of infancy and infantile spasms. Neurology 61: Wolff M, CassØ-Perrot C, Dravet C. (2006) Severe myoclonic epilepsy of infants (Dravet syndrome): natural history and neuropsychological findings. Epilepsia 47(suppl 2):45 48.