KATIA LIN
ESTUDO ANÁTOMO-FUNCIONAL POR RESSONÂNCIA
MAGNÉTICA EM PACIENTES COM EPILEPSIA
MIOCLÔNICA JUVENIL
Tese apresentada à Universidade Federal de São Paulo – Escola Paulista de Medicina, para obtenção do Título de Doutor em Ciências pelo programa de pós-graduação em Neurologia e Neurociências.
ESTUDO ANÁTOMO-FUNCIONAL POR RESSONÂNCIA
MAGNÉTICA EM PACIENTES COM EPILEPSIA
MIOCLÔNICA JUVENIL
Tese apresentada à Universidade Federal de São Paulo – Escola Paulista de Medicina, para obtenção do Título de Doutor em Ciências pelo programa de pós-graduação em Neurologia e Neurociências.
Orientador: Profa. Dra. Elza Márcia Targas Yacubian Co-orientador: Profo. Dro. Henrique Carrete Júnior
Lin, Katia
Estudo anátomo-funcional por ressonância magnética em pacientes com epilepsia mioclônica juvenil. / Katia Lin. -- São Paulo, 2009.
xi,114f.
Tese (Doutorado) – Universidade Federal de São Paulo. Escola Paulista de Medicina. Programa de Pós-graduação em Neurologia e Neurociências.
Título em inglês: Magnetic resonance anatomic and functional study of juvenile myoclonic epilepsy patients.
ESCOLA PAULISTA DE MEDICINA
DEPARTAMENTO DE NEUROLOGIA E NEUROCIRURGIA
Chefe do Departamento: Profa. Dra. Débora Amado Scerni
Coordenador do Curso de Pós-graduação: Profa. Dra. Maria da Graça Naffah Mazzacoratti
KATIA LIN
ESTUDO ANÁTOMO-FUNCIONAL POR RESSONÂNCIA
MAGNÉTICA EM PACIENTES COM EPILEPSIA MIOCLÔNICA
JUVENIL
Presidente da banca:
Profa. Dra. Elza Márcia Targas Yacubian
BANCA EXAMINADORA
Profo. Dro. Henrique Carrete Junior Profo. Dro. Nitamar Abdala
Profo. Dro. Ricardo Guarnieri Profo. Dro. Ylmar Corrêa Neto
SUPLENTES
Profo. Dro. Roger Walz
Profa. Dra. Laura Maria de Figueiredo Ferreira Guilhoto
Esta tese foi realizada na Disciplina de Neurologia Clínica, Departamento de Neurologia e Neurocirurgia da Universidade Federal de São Paulo – Escola Paulista de Medicina, durante o curso de Pós-graduação em Neurologia e Neurociências. Auxílio financeiro: CAPES, FAPESP e CNPq.
Dedicatória
Aos pacientes com epilepsia.
Aos meus pais, pelo exemplo, sempre.
À Profa. Dra. Elza Márcia Targas Yacubian e ao Profo. Dr. Henrique Carrete Júnior, com quem passei inesquecíveis cinco anos de aprendizado intenso e que me abriram os olhos e as portas para o vasto mundo da epileptologia e neuroimagem.
Aos meus amigos, colegas, funcionários da UNIPETE e do ambulatório de neuro-epilepsia, ao Departamento de Diagnóstico por Imagem e seus funcionários, pela amizade, colaboração e apoio, possibilitando a execução deste trabalho.
Ao Jaime Lin e Mirella Maccarini Peruchi, pela sua amizade, incentivo e apoio.
A Patrícia Guilhem de Almeida Ramos, pela orientação estatística.
A Janet Fu McDevitt e Enelise Arnold pelas revisões em Inglês e Português.
Aos pacientes, pela confiança depositada em nosso trabalho na difícil jornada de seu tratamento.
E ao meu esposo, Fabrício de Souza Neves, meu porto seguro, sempre.
Sumário
Dedicatória ... vi
Agradecimentos ... vii
Lista ... ix
Resumo ... x
1 INTRODUÇÃO ... 01
1.1 Objetivos ... 05
2 ARTIGOS PUBLICADOS ... 07
2.1 Artigo 1 – Proton magnetic resonance spectroscopy study of juvenile myoclonic epilepsy patients suggests involvement of a specific neuronal network ... 08
2.2 Artigo 2 – Magnetic resonance spectroscopy reveals an epileptic network in juvenile myoclonic epilepsy ………..………. 11
2.3 Artigo 3 – Voxel-based morphometry evaluation of patients with photosensitive juvenile myoclonic epilepsy ..………. 22
3 DISCUSSÃO .………... 31
4 CONCLUSÕES ……… 38
5 ANEXOS ……… 40
5.1 Anexo 1 – Parecer do Comitê de Ética Institucional ... 41
5.2 Anexo 2 – Termo de Consentimento Livre e Esclarecido ... 43
5.3 Anexo 3 – Outros artigos publicados em co-autoria durante o programa de pós-graduação ... 46
6 REFERÊNCIAS ……… 106 Abstract
Bibliografia consultada
1H-MRS Espectroscopia de prótons por ressonância magnética
Cr Creatina + fosfocreatina
CTCG Crise tônico-clônico generalizada
EEG Eletrencefalografia
EGI Epilepsia generalizada idiopática
EMJ Epilepsia mioclônica juvenil
GLX Glutamato + glutamina
ILAE International League Against Epilepsy
NAA N-acetilaspartato
PET Positron emission tomography
RM Ressonância magnética
SB Substância branca
SC Substância cinzenta
SPECT Single photon emission computerized tomography
TC Tomografia computadorizada
VBM Morfometria baseada em voxels
Resumo
Objetivo: As bases neuroanatômicas e as anormalidades bioquímicas subjacentes à
epilepsia mioclônica juvenil (EMJ) não são totalmente conhecidas. Apesar de o tálamo atuar na sincronização de diversas regiões do córtex cerebral durante uma crise, há evidências sugerindo que nem todos os neurônios corticais são afetados de forma homogênea. Compreender a participação destas redes neuronais específicas na EMJ pode esclarecer alguns de seus mecanismos fisiopatológicos. O objetivo deste estudo foi investigar as diferenças metabólicas e estruturais cerebrais entre pacientes com EMJ e controles normais.
Métodos: Todos os pacientes possuíam o diagnóstico de EMJ baseado em história e
semiologia das crises, eletrencefalografia (EEG), vídeo-EEG e neuroimagem por ressonância magnética (RM) convencional normal, conforme os critérios da Comissão de Classificação e Terminologia da International League Against Epilepsy, 1989. Sessenta pacientes com EMJ foram submetidos a protocolos de espectroscopia de prótons e morfometria baseada em voxels (VBM) por RM de 1,5 T. O grupo controle foi constituído por 30 voluntários saudáveis, pareados por sexo, idade e dominância manual. Este estudo foi realizado após aprovação do comitê de ética da instituição e obtenção de consentimento informado, por escrito, de todos os participantes.
Resultados: Demonstrou-se redução da razão de N-acetilaspartato/Creatina (NAA/Cr) dos pacientes com EMJ em relação ao grupo controle nos córtices frontal, pré-frontal e no tálamo. Observou-se diferença na razão do complexo glutamato-glutamina (GLX)/Cr nos córtices frontal, pré-frontal, ínsula, corpo estriado e cíngulo posterior entre os dois grupos. Análise por regressão múltipla nos pacientes com EMJ demonstrou maior correlação funcional entre o tálamo e o córtex pré-frontal. Também foi encontrada correlação negativa entre NAA/Cr e a duração da epilepsia. Análise estrutural quantitativa por VBM demonstrou redução do volume da substância cinzenta no tálamo, ínsula e cerebelo bilateralmente e aumento do volume do córtex frontal nos pacientes.
envolvimento de um circuito tálamo-cortical específico na fisiopatologia desta síndrome, considerada “generalizada”. Reduções em NAA podem representar perda ou lesão de neurônios ou axônios bem como disfunções metabólicas, enquanto o GLX é considerado um neurotransmissor excitatório, envolvido na patogênese das crises epilépticas. As anormalidades estruturais encontradas reforçam a existência de uma rede ictogênica anátomo-funcional específica na EMJ e o conceito de ‘system epilepsies’.
Em 1867, Herpin foi o primeiro autor a descrever extensivamente um provável caso de epilepsia mioclônica juvenil (EMJ) em sua obra “Des accès incomplets
de l´épilepsie”, ao relatar o caso de um adolescente que iniciou, aos 13 anos de idade, com abalos (“secousses”) na parte superior do corpo, ao despertar pela manhã, progredindo para crises tônico-clônicas generalizadas (CTCG) três meses depois (Herpin, 1867). Em 1899, Rabot, em sua tese intitulada “La myoclonie épileptique”, descreveu abalos súbitos comparáveis à descarga elétrica, com duração menor do que um segundo e que se repetiam a cada minuto durante várias horas, ressaltando sua predominância ao despertar e sua exacerbação com a proximidade de CTCG, considerando-os como um sinal de epilepsia e denominando estas manifestações de mioclonias epilépticas (Rabot, 1899). Em 1957, Janz e Christian caracterizaram pela primeira vez os sinais e sintomas clínicos, traços de personalidade, características eletrencefalográficas, resposta ao tratamento e o prognóstico de 47 pacientes e utilizaram o termo “pequeno mal impulsivo” para descrever uma síndrome clínica atualmente conhecida como EMJ, termo inicialmente introduzido por Lund e colaboradores em 1976 e reconhecido oficialmente na Proposta para a Classificação das Epilepsias e Síndromes Epilépticas da International League Against Epilepsy (ILAE; Commission, 1985).
3
uma CTCG enquanto a história de mioclonias (de início mais precoce) é obtida apenas retrospectivamente após questionamento específico. Seu prognóstico quanto ao controle das crises é satisfatório, mas é necessário o tratamento medicamentoso contínuo, habitualmente representado pelo valproato de sódio, por toda a vida (Delgado-Escueta & Bacsal, 1984; Wolf, 1992; Genton et al., 1994, Zifkin et al., 2005; Nordli, 2005).
A condição sine qua non para o diagnóstico da EMJ proposta pela ILAE
(Commission, 1989) é a ausência de alterações na ressonância magnética (RM) ou tomografia computadorizada (TC) do encéfalo, sendo estes exames de neuroimagem não recomendados rotineiramente na avaliação destes pacientes (Commission, 1997). O desenvolvimento de novas técnicas por RM, altamente sensíveis, no entanto, permitiu a identificação de diversas anormalidades funcionais e estruturais sutis nos pacientes com diagnóstico de EMJ (Koepp & Hamandi, 2006).
Desde o relato original de Gibbs e Lennox (1937), muitas hipóteses surgiram na tentativa de explicar os mecanismos das crises de ausência e das descargas de complexos de espícula-onda em indivíduos com EGI. Penfield e Jasper (1954) elaboraram a “teoria centrencefálica”, correlacionando as EGI às alterações patológicas no tronco encefálico e diencéfalo; Bancaud e colaboradores (1974) sustentaram uma origem cortical; enquanto Gloor (1968) postulou a “teoria corticorreticular”, integrando as teorias cortical e centrencefálica no conceito de que crises generalizadas resultariam de uma interação anormal entre estruturas corticais e subcorticais, representando uma resposta do córtex hiperexcitável às projeções tálamo-corticais – sendo esta última a teoria mais aceita atualmente (Gloor, 1978; Gloor, 1979; Velasco et al., 1989; Avoli et al., 2001).
ausência típica (Prevett et al., 1995) até distúrbios de migração neuronal nos lobos frontais (Janz & Neimanis, 1961; Meencke & Janz, 1984; Meencke & Janz, 1985; Meencke, 1985; Meencke & Veith, 1992).
No esclarecimento das bases neuroanatômicas e fisiopatológicas subjacentes às EGI, novas técnicas de RM têm sido de fundamental importância, já que têm sido capazes de demonstrar alterações cerebrais focais, não evidenciadas pela RM convencional. Além de ter um custo menor que exames como PET e SPECT, a RM, por ser não invasiva e não envolver radiação ionizante, é um exame que oferece riscos mínimos aos pacientes. Dentre as novas técnicas, destacam-se a espectroscopia de prótons por RM (1H-MRS) e a morfometria baseada em voxels (VBM) (Duncan, 2002).
Diferentemente da RM convencional, a 1H-MRS, aprovada pelo FDA em 1995, possibilita a obtenção de informações in vivo de forma não-invasiva acerca da composição metabólica e bioquímica do parênquima encefálico, permitindo uma melhor caracterização tecidual, já que em muitos casos as alterações metabólicas precedem as anormalidades estruturais (Duncan, 2002; Simister et al., 2003a). Desde a última década, um número considerável de estudos tem demonstrado, através da 1H-MRS, disfunções metabólicas associadas às EGI com elevada sensibilidade, principalmente no tálamo e córtices pré-frontais (Savic et al., 2000; Bernasconi et al., 2003; Simister et al., 2003b; Pan
et al., 2008).
Em adição, estudos de VBM, técnica originalmente desenvolvida por Ashburner e Friston (2000) com o propósito de avaliar a morfologia encefálica que possibilita a comparação de diferenças na concentração da substância cinzenta (SC) e da substância branca (SB) em diversas regiões cerebrais de forma simultânea e automatizada, têm demonstrado resultados conflitantes na concentração da SC no tálamo, na região pré-frontal, nas regiões frontobasais e frontais mesiais (Woermann et al., 1999; Betting et al., 2006a; Tae et al., 2006; Kim et al., 2007).
5
achados e determinar sua especificidade para subssíndromes específicas dentre as EGI (Koepp & Hamandi, 2006).
Desse modo, embora os dados eletrencefalográficos sugiram uma disfunção tálamo-cortical como mecanismo principal das EGI (Gloor, 1968), as bases neuroanatômicas e as anormalidades neuroquímicas responsáveis pelas EGI ainda não estão bem definidas. Embora o tálamo exerça um papel central na sincronização de diversas regiões corticais durante a crise epiléptica, evidências recentes sugerem que os neurônios corticais podem não estar homogeneamente envolvidos (Gloor, 1979; Avoli et
al., 2001). Pode haver a ativação de grupos de regiões cerebrais corticais e subcorticais conectadas anatômica e funcionalmente que, em virtude de alguma instabilidade funcional, provavelmente geneticamente determinada, respondem com descargas epileptiformes (Wolf, 2005). A compreensão da interação entre estas redes neuronais específicas pode trazer esclarecimentos sobre a fisiopatologia da EMJ. O objetivo desde estudo é, portanto, analisar as diferenças metabólicas e estruturais encefálicas entre pacientes com EMJ e indivíduos normais e avaliar em qual extensão essas alterações refletem o envolvimento de redes neuronais específicas envolvidas na ictogênese da EMJ.
1.1 Objetivos
O objetivo deste estudo foi avaliar:
1. As características demográficas e clínicas de um grupo homogêneo de 60 pacientes com diagnóstico bem estabelecido de EMJ, em acompanhamento regular e de longa data em um serviço terciário de saúde;
2. A presença de anormalidades estruturais e metabólicas no encéfalo de pacientes com EMJ não demonstráveis na RM convencional através de estudo de 1H-MRSI e VBM na tentativa de delimitar a rede anatômico-funcional envolvida na geração e propagação das crises epilépticas nestes pacientes;
4. A existência de disfunção cerebral progressiva através da correlação dos achados acima com parâmetros clínicos como tempo de doença, frequência de crises e tratamento medicamentoso;
7
ARTIGO 1
Proton magnetic resonance spectroscopy study of juvenile myoclonic epilepsy
patients suggests involvement of a specific neuronal network (Prêmio Cesare Lombroso)
Katia Lin, Henrique Carrete Junior, Jaime Lin, Mirella Maccarini Peruchi, Gerardo Maria Araújo Filho, Tatiana Frascareli Pascalicchio, Mirian Bittar Guaranha, Laura Maria
Figueiredo Ferreira Guilhoto, Elza Márcia Targas Yacubian
99
Awards Works: Expanded Abstract
J
ournal of Epilepsy and ClinicalNeurophysiology
J Epilepsy Clin Neurophysiol 2008; 14(3):99-100
Proton Magnetic Resonance Spectroscopy Study of
Juvenile Myoclonic Epilepsy Patients Suggests Involvement
of a Specific Neuronal Network*
K. Lin, H. Carrete Junior, J. Lin, M.M. Peruchi, G.M. Araújo Filho, T.F. Pascalicchio, M.B. Guaranha, L.M.F.F. Guilhoto, E.M.T. Yacubian
UNIPETE – Hospital São Paulo – UNIFESP/EPM
* Trabalho premiado com o Prêmio Cesare Lombroso durante o XXXII Congresso Brasileiro de Epilepsia 2008. Received June 16, 2008; accepted July 18, 2008.
ABSTRACT
Objectives: The neuroanatomical basis and the neurochemical abnormalities that underlay juvenile myoclonic epilepsy (JME) are not fully defined. While the thalamus plays a central role in synchronization of widespread regions of the cerebral cortex during a seizure, emerging evidence suggests that all cortical neurons may not be homogeneously involved. The purpose of this study was to investigate the cerebral metabolic differences between patients with JME and normal controls. Methods: All patients had a JME diagnosis based on seizure history and semiology, EEG recording, normal magnetic resonance neuroimaging (MRI) and video-EEG. Forty JME patients (JME-P) were submitted to 1.5 T MRI proton spectroscopy (1H-MRS), multi-voxel with PRESS sequence (TR/TE = 1500/30 ms) over the following locations: prefrontal cortex (PC), frontal cortex (FC), thalamus, basal nuclei, posterior cingulate gyrus (PCG), insular, parietal and occipital cortices. We determined ratios for integral values of N-acetyl aspartate (NAA) and glutamine-glutamate (GLX) over creatine-phosphocreatine (Cr). The control group (CTL) consisted of 20 age and sex-matched healthy volunteers. Results: Group analysis demonstrated a tendency for lower NAA/Cr ratio of JME-P compared to CTL predominantly on FC, PC, thalamus and occipital cortex. When compared to CTL, JME-P had a statistically significant difference in GLX/Cr on FC, PC, insula, basal nuclei, PCG and on thalamus. When evaluating the relationship among the various components of this epileptic network among JME-P, the strongest correlation occurred between thalamus and PC. Also, we found a significant negative correlation between NAA/Cr and duration of epilepsy. Conclusion: Reductions in NAA may represent loss or injury of neurons and/or axons, as well as metabolic dysfunction while glutamate is considered to be an excitatory neurotransmitter in the brain which is involved in the pathogenesis of epileptogenic seizures.
Key words: Juvenile myoclonic epilepsy, magnetic resonance spectroscopy, magnetic resonance imaging.
Espectroscopia por Ressonância Magnética de Prótons em Epilepsia Mioclônica Juvenil Sugere o Comprometimento de uma
Rede Neuronal Específica
OBJETIVOS
As bases neuroanatômicas e as anormalidades bioquími-cas subjacentes à epilepsia mioclônica juvenil (EMJ) não são totalmente conhecidas.1 Apesar de o tálamo atuar na
sincro-nização de múltiplas regiões do córtex cerebral durante uma
crise, há evidências sugerindo que nem todos os neurônios corticais são afetados de forma homogênea.2 Pode haver
insta-100
REFERENCES
1. Avoli M, Rogawski MA, Avanzini G. Generalized epileptic disorders: an update. Epilepsia 2001; 42:445-57.
2. Hirsch E, Andermann F, Chauvel P, Engel Jr J, Lopes da Silva F, Lüders H. Foreword. In: Hirsch E, Andermann F, Chauvel P, Engel Jr J, Lopes da Silva F, Lüders H (eds.). Generalized seizures: from clinical phenomenology to underlying systems and networks 2006, John Libbey Eurotext, Montrouge, p. XIX.
3. Norden AD, Blumenfeld H. The role of subcortical structures in human epilepsy. Epilepsy Behav 2002;3:219-231.
4. Wolf P. Juvenile myoclonic epilepsy. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P (eds.). Epileptic syndromes in infancy, childhood and adolescence 1992; John Libbey & Company Ltd., Montrouge, p. 313-27.
5. Commission on classification and terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989; 30:389-99. 6. Bernasconi A, Bernasconi N, Natsume J, Antel SB, Andermann
F, Arnold DL. Magnetic resonance spectroscopy and imaging of the thalamus in idiopathic generalized epilepsy. Brain 2003; 126:2447-54.
7. Benjamin Z, Andermann E, Andermann F. Mechanisms, genetics, and pathogenesis of juvenile myoclonic epilepsy: review. Curr Opin Neurol 2005; 18:147-53.
8. Gotman J. Epileptic networks studied with EEG-fMRI. Epilepsia 2008; 49:42-51.
Endereço para correspondência:
Katia Lin
Rod. Amaro Antônio Vieira, 2740 - A - 502 CEP 88034-102, Florianópolis, SC, Brasil E-mail: linkatia@uol.com.br bilidades funcionais nas regiões afetadas, provavelmente
determinadas geneticamente.4 Compreender a participação
destas redes específicas na EMJ pode esclarecer alguns de seus mecanismos fisiopatológicos. O objetivo deste estudo foi investigar as diferenças metabólicas cerebrais entre pa-cientes com EMJ e controles normais.
MÉTODOS
Todos os pacientes possuíam o diagnóstico de EMJ base-ado em história e semiologia das crises, eletroencefalografia (EEG), vídeo-EEG e neuroimagem por ressonância magnéti-ca (RM) normal, conforme os critérios da Comissão de Clas-sificação e Terminologia da International League Against Epilepsy (ILAE), 1989.5 Quarenta pacientes com EMJ, livres
de crises por, no mínimo, 48 horas foram submetidos à espectroscopia por RM de prótons de 1,5 T (1H-MRS). Após a aquisição nos planos axial, sagital e coronal (para a localiza-ção das estruturas de interesse), estas imagens de RM con-vencional foram utilizadas para executar uma análise multi-voxel de 1H-MRS em seqüência PRESS (TR/TE = 1500/30 ms) nos seguintes locais: córtex pré-frontal, córtex frontal, tálamo, núcleos da base, cíngulo posterior, ínsula, córtex parietal e córtex occipital. Foram calculadas as razões dos valores integrais de N-acetil aspartato (NAA) e glutamina-glutamato (GLX) sobre creatina-fosfocreatina (Cr). O grupo controle foi constituído por 20 voluntários saudáveis, pareados por sexo e idade. Este estudo foi realizado após aprovação do comitê de ética da instituição e obtenção de consentimento informado, por escrito, de todos os participantes.
RESULTADOS
Demonstrou-se uma tendência à redução da razão NAA/Cr dos pacientes com EMJ em relação ao grupo con-trole em todas as regiões estudadas, sendo mais proeminente nos córtices frontal, pré-frontal, occipital e tálamo. Observou-se diferença na razão GLX/Cr nos córtices frontal, pré-frontal, ínsula, núcleos da base, cíngulo posterior e tálamo entre os dois grupos estudados. Avaliando as relações entre os componen-tes desta rede de estruturas nos paciencomponen-tes com EMJ, a corre-lação mais forte foi demonstrada entre o tálamo e o córtex pré-frontal. Estas correlações não foram observadas no grupo controle. Também foi encontrada correlação negativa entre NAA/Cr e duração da epilepsia nos pacientes com EMJ.
CONCLUSÃO
O comprometimento de algumas regiões cerebrais nes-tes paciennes-tes sugere que uma circuitaria tálamo-cortical específica (com comprometimento de algumas estruturas
corticais e subcorticais) pode ser mais importante na fisiopato-logia desta síndrome, ainda considerada “generalizada”.3,6 O
tálamo é a principal fonte de entrada para o córtex, mas estas conexões são superadas, em número, pelas conexões recípro-cas do córtex de volta para o tálamo, envolvendo os núcleos da base, o sistema límbico e projeções difusas do lobo frontal (córtex pré-motor, córtex motor primário e córtex motor su-plementar).3 Reduções em NAA podem representar perda
ou lesão de neurônios ou axônios bem como disfunções meta-bólicas, enquanto o GLX é considerado um neurotransmissor excitatório, envolvido na patogênese das crises epilépticas.6 A
atividade elétrica anormal do tálamo na EMJ pode ser expli-cada pela hiperexcitabilidade devido ao aumento no GLX neuronal.6 No entanto, enquanto a atividade epiléptica
trans-corre com hiperexcitabilidade neuronal em algumas regiões, em outras regiões há evidências de atividade neuronal dimi-nuída. Isso ocorre porque o tálamo tem importantes conexões inibitórias que podem, inclusive, levar a inibição funcional aguda de algumas regiões, como o lobo frontal, o que pode estar associado aos baixos níveis de GLX/Cr encontrados.7,8
11
ARTIGO 2
Magnetic resonance spectroscopy reveals an epileptic network in juvenile
myoclonic epilepsy
Katia Lin, Henrique Carrete Junior, Jaime Lin, Mirella Maccarini Peruchi, Gerardo Maria Araújo Filho, Mirian Salvadori Bittar Guaranha, Laura Maria Figueiredo Ferreira Guilhoto,
Américo Ceiki Sakamoto, Elza Márcia Targas Yacubian
Magnetic resonance spectroscopy reveals an epileptic
network in juvenile myoclonic epilepsy
Katia Lin, Henrique Carrete Jr, Jaime Lin, Mirella Maccarini Peruchi, Gerardo Maria de Arau´jo Filho, Mirian Salvadori Bittar Guaranha, Laura Maria Figueiredo Ferreira
Guilhoto, Ame´rico Ceiki Sakamoto, and Elza Ma´rcia Targas Yacubian
Unidade de Pesquisa e Tratamento das Epilepsias (UNIPETE), Universidade Federal de Sa˜o Paulo (UNIFESP/EPM), Sa˜o Paulo, SP, Brazil
SUMMARY
Purpose: To investigate the cerebral metabolic
differences between patients with juvenile myo-clonic epilepsy (JME) and normal controls and to evaluate to what extent these metabolic altera-tions reflect involvement of an epileptic network.
Methods: Sixty patients with JME were submitted
to multi-voxel proton spectroscopy (1H-MRS) at 1.5 T over medial prefrontal cortex (MPC), pri-mary motor cortex (PMC), thalamus, striatum, posterior cingulate gyrus (PCG), and insular, pari-etal, and occipital cortices. We determined ratios
for integral values of N-acetyl-aspartate (NAA)
and glutamate-glutamine (GLX) over creatine-phosphocreatine (Cr). The control group (CTL) consisted of 30 age- and sex-matched healthy vol-unteers.
Results: The NAA/Cr ratio, a measure of
neuro-nal injury, was reduced in PMC, MPC, and thala-mus among patients. In addition, they had an altered GLX/Cr ratio, which is involved in excit-atory activity, on PMC, MPC, and PCG, where it
was reduced, whereas it was increased on insula
and striatum. Multiple regression analysis
revealed the strongest correlation between
thalamus and MPC, but the thalamus was also correlated with insula, occipital cortex, and
stria-tum among patients. Lower NAA/Cr was
observed with advancing age and duration of epilepsy, regardless of frequency of seizures and antiepileptic drug therapy in thalamus and frontal cortex.
Discussion: The identification of a specific
net-work of neurochemical dysfunction in patients with JME, with diverse involvement of particular structures within the thalamocortical circuitry, suggests that cortical hyperexcitability in JME is not necessarily diffuse, supporting the knowledge that the focal/generalized distinction of epilepto-genesis should be reconsidered. Furthermore, evi-dence is provided toward progressive neuronal dysfunction in JME.
KEY WORDS: Glutamate, Idiopathic generalized
epilepsy, Magnetic resonance imaging, N-acetyl
aspartate, Thalamus.
Juvenile myoclonic epilepsy (JME), a well-defined clinical syndrome among idiopathic generalized epilep-sies (IGEs), is the most common IGE in adults, comprising 5–11% of patients with epilepsy (Janz & Christian, 1957; Wolf, 1992; Panayiotopoulos et al., 1994). It is character-ized by myoclonia, generalcharacter-ized tonic–clonic seizures (GTCS), and, less frequently, absences. Its seizures occur
predominantly at awakening and are associated with clear precipitating factors, such as sleep deprivation, fatigue, and alcohol intake. Its electroencephalography (EEG) findings show generalized 4–6 Hz irregular spike- or polyspike-wave activity, with a maximum in frontocentral regions (Delgado-Escueta & Enrile-Bacsal, 1984; Genton et al., 1994).
The diagnosis of JME proposed by the International League Against Epilepsy (ILAE) includes the absence of magnetic resonance imaging (MRI) and computed tomo-graphy (CT) abnormalities (Commission, 1989), and, therefore, neuroimaging is not routinely recommended (Commission, 1997). However, the development of new
Accepted October 6, 2008; Early View publication February 12, 2009. Address correspondence to Katia Lin, Rod. Amaro Antnio Vieira, 2740, Bl. A, Apto. 502, Itacorubi, Florianpolis 88034-102, SC, Brazil. E-mail: linkatia@uol.com.br
Wiley Periodicals, Inc.
ª2009 International League Against Epilepsy
FULL-LENGTH ORIGINAL RESEARCH
MRI techniques has allowed the identification of diverse functional and structural alterations in these patients (Koepp & Hamandi, 2006).
Magnetic resonance spectroscopy (MRS) provides non-invasive in vivo information about the chemical composi-tion of the brain (Simister et al., 2003a), and proton MRS (1H-MRS) has revealed metabolic dysfunction related to epilepsy with high sensitivity (70–100%) (Pan et al., 2008). These abnormalities are characterized mainly by lower thalamic and prefrontal N-acetyl-aspartate (NAA)
concentrations in IGE patients than in controls (Savic et al., 2000; Bernasconi et al., 2003; Mory et al., 2003; Simister et al., 2003b; Savic et al., 2004; Haki et al., 2007). Despite this, a major concern in these studies is that various IGE syndromes were often grouped together and larger studies with more homogeneous patient populations are needed to determine the robustness of these findings (Koepp & Hamandi, 2006).
Although the major mechanism of IGE is related to tha-lamocortical dysfunction, its neuroanatomic basis and its neurochemical abnormalities are not fully understood (Gloor, 1968; Avoli et al., 2001). The purpose of this study is to investigate the cerebral metabolic differences between JME patients and healthy controls and to evaluate to what extent these metabolic alterations reflect involve-ment of a specific network of dysfunction.
Methods
Subjects
We studied a cohort of 60 long-term followed consecu-tive patients with JME from the epilepsy clinic at Hospital S¼o Paulo, Universidade Federal de S¼o Paulo (UNIFESP/EPM), S¼o Paulo, SP, Brazil. Demographic and clinical data were obtained through interviews with the patients and their relatives, and by reviewing hospital charts. JME diagnosis was based on seizure history and semiology, EEG and interictal € ictal video-EEG in all patients, according to the Commission on Classification and Terminology of ILAE (Commission, 1989). The video-EEG protocol consisted of 6 h of continuous moni-toring with 32-channel EEG, with electrodes placed according to the 10–20 International Electrode System, and included hyperventilation and photic stimulation.
Age of onset of epilepsy was defined as the age at which the patient developed habitual and recurrent seizures, whereas the duration included the interval between age of onset and time of neuroimaging. Estimation of the fre-quency of seizures was based on review of seizure calen-dars and specific questioning of the patient and family members. The interval since the most recent GTCS was also recorded. The criteria for classification of seizure control were: GTCS—good (<1 seizure per year), moder-ate (1–4 seizures per year), or poor (>4 seizures per year); myoclonus—good (<5 seizures or clusters per month),
moderate (5–14 seizures or clusters per month), or poor (‡15 seizures or clusters per month or daily seizures); and
absence—good (<5 seizures per month), moderate (5–14 seizures per month), or poor (‡15 seizures per month or
daily seizures) (Prasad et al., 2003).
All patients were healthy apart from having seizures. None had a history of status epilepticus, drug intoxication, or drug-related encephalopathy, and routine 1.5 T MRI of the brain was normal in all subjects. All patients were seizure-free for at least 48 h before neuroimaging.
The control group consisted of 30 unmedicated, right-handed (except one), healthy volunteers [14 males; mean age € standard deviation (SD) = 30.1 € 9.0 years; range = 20–56 years].
The Ethics Committee of the UNIFESP/EPM approved the study, and informed consent was obtained from all participants.
Proton MRS
Multivoxel MRS was obtained by two-dimensional (2D) hybrid chemical shift imaging (CSI) technique, with two-phase encoding gradients before gradient readout, using a 1.5 T scanner (Magnetom Sonata–Siemens AG, Erlangen, Germany). A spin-echo point-resolved spectro-scopy (PRESS) sequence was performed (TR = 1500 ms, TE = 30 ms, four averages, 512 data points, delta fre-quency =)2.7 ppm, bandwidth = 1,000 Hz) with two
spectroscopic volumes of interest (VOIs) of 80.0·80.0·
15.0 mm each; one centered on thalami, parallel to the anterior commissure–posterior commissure (AC-PC) line plus 10; and the other centered on frontal lobes, parallel to
the AC-PC line. The caudomesial portion of frontal lobes was avoided owing to susceptibility artifacts generated from this region. Three T2-weighted MRI sequences, cov-ering the whole brain, were acquired in sagittal, axial, and coronal orientations to identify the appropriate positioning of the VOIs under similar anatomic position in all subjects. Before recording the spectrum, the homogeneity of the magnetic field over the VOIs was optimized (shimming) automatically, observing the water signal intensity, using double measurements with gradient inversion and one iteration step. Water suppression was achieved using chemical shift selective (CHESS) radiofrequency pulses before PRESS excitation. Spectra showing abnormal baseline because of major artifact resonances, poor signal-to-noise ratio (SNR), broad peaks, excessive line width, or poor separation of individual peaks were excluded from the statistical analysis, totaling <10.0% of data.
All data were transferred offline and processed on a workstation by using the software developed for multi-voxel 1H-MRS (Syngo MR Spectroscopy Evaluation Programme–Syngo v.2004A, Siemens Medical Solutions, Erlangen, Germany). Spectral data sets were zero filled from 512 to 1,024 points, multiplied by a Hanning function (center 0 ms, width 512 ms) and Fourier transformation 1192
K. Lin et al.
was performed in time domain signal to a frequency domain signal, as well as the spatial axes to a spectral matrix. A polynomial baseline correction was applied (sixth order), followed by an automatic phase correction [using the reference peaks of choline (Cho) and creatine plus phosphocreatine-contained compounds (Cr) to deter-mine the best fit in the spectrum] and after that, an auto-matic curve fitting was performed. The corrected spectra were analyzed to calculate ratios of integral values of NAA at 2.0 parts per million (ppm) and glutamate-gluta-mine complex (GLX) at 2.2 ppm to Cr at 3.0 ppm (Fig. 1). A single operator unaware of the subject’s diagnosis, selected voxels on the right and the left hemispheres for the following regions of the brain: thalamus, striatum, insular cortex, posterior cingulate gyrus, occipital cortex, medial prefrontal cortex, primary motor cortex, and parietal cortex. Each voxel, with a dimension of 10.0 ·10.0·15.0 mm, was selected under identical
con-ditions from one subject to another and was located entirely within the studied structures to minimize any metabolite contamination from adjacent cerebrospinal fluid or gray and white matter outside the studied struc-tures. The number of spectra averaged for each structure in each cerebral hemisphere was: thalamus = 4; insular cortex, medial prefrontal cortex, primary motor cortex, and parietal cortex = 2; and striatum, cingulate gyrus, and occipital cortex = 1 (Fig. 1).
Statistical analysis
All results were presented as mean € SD to define dispersion. Statistical analysis was performed with SPSS for Windows 10.01 Standard Version (SPSS Inc., Chicago,
IL, U.S.A.), STATISTICA for Windows 4.2 (StatSoft Inc., Tulsa, OK, U.S.A.) and STATA Statistics Data Anal-ysis 5 (Stata Corporation, College Station, TX, U.S.A.). Differences between patients and controls with respect to age were assessed using Student’st-test. The gender distri-bution and manual dominance were examined by Fisher’s exact test. Given that concentrations of measured metabo-lites varied among different regions as assessed by paired analysis of variance (ANOVA); and between right and left sides, as assessed using paired Student’st-test, which were
consistent with previous studies (Grachev & Apkarian, 2000; Kadota et al., 2001), only values from the same region and side were compared. Differences for each spec-troscopic ratio between JME and controls within a given structure were compared using Student’st-test. The results were corrected for multiple comparisons using a false dis-covery rate (FDR) of 15%, which may be reasonable in most problems (Genovese et al., 2002).
Furthermore, a multiple regression analysis with stepwise method was performed to evaluate how each thalamus and the other structures were metabolically linked, as reflected by the degree of neuronal dysfunction as measured by NAA/Cr.
Analysis of variance was performed for NAA/Cr against frequency of seizures and AED treatment, Pearson’s correlation coefficient (r) was computed against the interval (months) since the most recent GTCS, and multiple regression analysis was performed between NAA/Cr and age and duration of epilepsy.
Abnormality of the regional concentration of a given metabolite in individual patients was defined as a value outside the 2 SD of the mean of normal controls.
Figure 1.
Axial T2-weighted MRI demonstrating the position of both volumes of interest and the averaged voxels; and a
representative spectrum. Cho, choline-containing compounds; Cr, creatine plus phosphocreatine; GLX, glutamate plus glutamine; I, integral; NAA,N-acetyl-aspartate; MRI, magnetic resonance imaging; ppm, chemical shift in parts per million.
Epilepsia ILAE
MRS in Juvenile Myoclonic Epilepsy
Results
Patients
Patients and controls were similar in age (p = 0.09), gender distribution (p = 1.00), and hand preference (p = 0.80). As expected, most of our patients had age of onset of seizures around puberty and good control of sei-zures. A higher prevalence of absence seizures in our patient population might be because of the use of video-EEG in the clinical characterization of JME. Their major clinical data are presented in Table 1.
Group analysis
We found a significant reduction of NAA/Cr among JME patients compared with controls in the primary motor cortex, medial prefrontal cortex, and thalamus. No signifi-cant intergroup variation was observed in the other studied regions, although they showed a trend to lower values among patients (Table 2).
When compared with controls, JME patients had a sig-nificant difference in GLX/Cr in the primary motor cortex, medial prefrontal cortex, insula, striatum, and cingulum. However, although this ratio was reduced among patients in the primary motor cortex, medial prefrontal cortex and cingulum, it was increased in the insula and striatum (Table 3).
Individual results
Forty-three percent of patients with JME had abnormal left insular NAA/Cr levels, whereas more than 20% of them presented altered NAA/Cr ratio in bilateral medial prefrontal cortices, thalami, and right parietal cortex. Regarding GLX/Cr, 51% of patients presented abnormal levels in right occipital cortex, whereas more than 20% of patients had altered levels in left insula, thalamus, and stri-atum. No individual among controls presented any metab-olite ratio outside the 2 SD of the mean of normal controls (Table 4).
Correlations within the epileptic network
The strongest correlation based on Beta % value among patients occurred between the thalamus and medial pre-frontal cortex, but the thalamus was also correlated with insula, occipital cortex and striatum (Table 5). In contrast, no correlation was observed in controls between thalami and other structures, except only for right and left thalami [Beta % = 59.3%; regression coefficient = 0.502; confidence interval (CI) = 0.22–0.79; p = 0.001].
Relation to age and duration of epilepsy
There was no correlation between NAA/Cr of any structure and age in controls. In contrast, among patients, we found lower NAA/Cr ratio with advancing age and duration of epilepsy on primary motor cortex, medial prefrontal cortex and thalamus (Fig. 2).
Relation to seizure frequency at the time of 1H-MRS and time since the most recent GTCS
The patients were divided into three groups: good, mod-erate, and poor control of seizures. A trend toward lower values of NAA/Cr was observed among those patients with moderate and poor control of myoclonia in all studied regions, when compared to those with good control, being significantly lower in the left thalamus (2.00 € 0.49 vs. 2.37 € 1.08; p < 0.001) and left parietal cortex (2.28 € 0.42 vs. 2.33 € 0.41; p = 0.018). When patients with moderate and poor control of GTCS were compared to those with good control, lower values of NAA/Cr were observed among those with worse control as well, being significantly lower in the left thalamus (2.12 € 0.70 vs. 2.48 € 1.26; p < 0.001). However, neither the frequency of absence seizures nor the duration since the most recent GTCS was significantly correlated with this metabolite ratio.
Relation to concurrent AED therapy
The AED taken by the JME group was as follows: val-proate (41 patients), lamotrigine (6), carbamazepine (5), topiramate (14), clobazam (1), clonazepam (12), phenyt-oin (2), and phenobarbital (12). Thirty-five patients were on monotherapy, whereas 17 patients were taking two AEDs and eight were receiving three AEDs.
Table 1. Clinical data of the patients
Clinical data JME (n = 60)
Gender male/female 29 (48.3%)/31 (51.7%) Age (years) 26.6 ± 8.85
Hand preference 52 (86.7%) right-handed Age of onset of sz (years) 12.6 ± 4.56
Duration of epilepsy (years)
14.2 ± 9.83
Duration in months since last GTCS
21.5 ± 32.45
Type of sz 60 (100.0%) myoclonia; 58 (96.7%) GTCS; 39 (65.0%) absences Frequency of sz
Myocloniaa 38 (63.3%) good; 11 (18.3%) moderate; 11 (18.3%) poor GTCSb 41 (70.7%) good; 10 (17.2%) moderate; 7 (12.1%) poor Absencesc 28 (71.8%) good; 4 (10.3%) moderate; 7 (17.9%) poor
GTCS, generalized tonic–clonic seizure; JME, juvenile myoclonic epilepsy; SD, standard deviation; sz, seizure.
aGood (<5 sz or clusters per month), moderate (5–14 sz or
clusters per month), or poor (‡15 sz or clusters per month or
daily sz).
bGood (<1 sz per year), moderate (1–4 sz per year), or poor
(>4 sz per year).
cGood (<5 sz per month), moderate (5–14 sz per month), or
poor (‡15 sz per month or daily seizures). 1194
K. Lin et al.
We did not find any difference in NAA/Cr ratio in any region when comparing patients with different treatments. Comparisons between patients with and without valproate were also conducted, and no differences were found.
Furthermore, among those patients taking carbamazepine and phenytoin, which may potentially aggravate seizures in patients with JME (Genton et al., 2000), only one patient was not classified as having good control of seizures.
Table 2. NAA/Cr in patients and controls
Right Left
Controls JME p-value Controls JME p-value
Primary motor cortex 2.43 ± 0.35 2.20 ± 0.40 0.011* 2.41 ± 0.28 2.26 ± 0.43 0.097 Medial prefrontal cortex 2.26 ± 0.23 2.08 ± 0.38 0.012* 2.12 ± 0.16 2.02 ± 0.34 0.072 Parietal cortex 2.23 ± 0.39 2.05 ± 0.56 0.154 2.45 ± 0.40 2.34 ± 0.44 0.299 Occipital cortex 2.97 ± 0.67 2.68 ± 0.71 0.161 3.03 ± 0.83 2.59 ± 0.60 0.074 Insula 1.68 ± 0.25 1.77 ± 0.34 0.244 1.42 ± 0.16 1.48 ± 0.37 0.343 Thalamus 1.99 ± 0.47 1.88 ± 0.24 0.147 2.24 ± 0.90 1.91 ± 0.22 0.012* Striatum 1.59 ± 0.59 1.66 ± 0.72 0.659 1.36 ± 0.39 1.45 ± 0.33 0.269 Cingulum 2.14 ± 0.87 1.91 ± 0.61 0.242 2.25 ± 0.84 2.01 ± 0.70 0.196
Cr, creatine plus phosphocreatine; JME, juvenile myoclonic epilepsy; NAA,N-acetyl-aspartate. *Significant; while corrected for multiple comparisons using false discovery rate [q = 0.15; c(V) = 1].
Table 3. GLX/Cr in patients and controls
Right Left
Controls JME p-value Controls JME p-value
Primary motor cortex 0.52 ± 0.30 0.38 ± 0.30 0.049* 0.53 ± 0.27 0.37 ± 0.26 0.013* Medial prefrontal cortex 0.59 ± 0.33 0.45 ± 0.29 0.041* 0.49 ± 0.28 0.35 ± 0.29 0.028* Parietal cortex 0.46 ± 0.30 0.44 ± 0.36 0.835 0.54 ± 0.33 0.48 ± 0.32 0.452 Occipital cortex 0.78 ± 0.25 0.63 ± 0.49 0.135 0.59 ± 0.36 0.71 ± 0.57 0.359 Insula 0.47 ± 0.30 0.52 ± 0.27 0.421 0.32 ± 0.24 0.48 ± 0.29 0.012* Thalamus 0.28 ± 0.17 0.26 ± 0.17 0.652 0.22 ± 0.14 0.33 ± 0.33 0.084 Striatum 0.31 ± 0.19 0.33 ± 0.24 0.587 0.16 ± 0.15 0.31 ± 0.27 0.002* Cingulum 0.67 ± 0.32 0.50 ± 0.34 0.042* 0.71 ± 0.40 0.60 ± 0.34 0.232
Cr, creatine plus phosphocreatine; GLX, glutamate-glutamine complex; JME, juvenile myoclonic epilepsy. *Significant; while corrected for multiple comparisons using false discovery rate [q = 0.15; c(V) = 1].
Table 4. Number and percentage of patients with abnormal (higher or lower) metabolic values using a two standard deviation cutoff from the mean of normal controls
NAA GLX
Right Left Right Left
Lower Higher Lower Higher Lower Higher Lower Higher
Primary motor cortex 2/58 (3.4%) 1/58 (1.7%) 7/58 (12.1%) 2/58 (3.4%) 0 (0.0%) 0 (0.0%) 5/58 (8.6%) 0 (0.0%) Medial prefrontal cortex 8/58 (13.8%) 6/58 (10.3%) 10/58 (17.2%) 5/58 (8.6%) 1/58 (1.7%) 0 (0.0%) 1/58 (1.7%) 0 (0.0%) Parietal cortex 5/54 (9.3%) 7/54 (13.0%) 5/54 (9.3%) 2/54 (3.7%) 5/54 (9.3%) 0 (0.0%) 3/54 (5.6%) 0 (0.0%) Occipital cortex 4/55 (7.3%) 0 (0.0%) 10/55 (18.2%) 0 (0.0%) 11/55 (20.0%) 17/55 (30.9%) 11/55 (20.0%) 0 (0.0%) Insula 9/60 (15.0%) 3/60 (5.0%) 16/58 (27.6%) 9/58 (15.5%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 12/58 (20.7%) Thalamus 10/60 (16.7%) 3/60 (5.0%) 13/59 (22.0%) 2/59 (3.4%) 5/60 (8.3%) 0 (0.0%) 0 (0.0%) 17/59 (28.8%) Striatum 5/58 (8.6%) 0/58 (0.0%) 5/54 (9.3%) 0 (0.0%) 5/58 (8.6%) 0 (0.0%) 0 (0.0%) 19/54 (35.2%) Cingulum 3/55 (5.5%) 1/55 (1.8%) 2/54 (3.7%) 2/54 (3.7%) 4/55 (7.3%) 4/55 (7.3%) 2/54 (3.7%) 0 (0.0%)
Cr, creatine plus phosphocreatine; GLX, glutamate-glutamine complex; JME, juvenile myoclonic epilepsy; NAA, N -acetyl-aspartate.
MRS in Juvenile Myoclonic Epilepsy
Discussion
The pathogenic mechanism of JME in humans remains under investigation. Because the main pathophysiologic theories of generalized seizures were based on electro-physiologic studies, a neuroanatomic substrate was hypothesized, but such substrate has been unraveled with the advent of better neuroimaging techniques, which pro-vided evidence that it is a multifocal rather than a truly generalized syndrome (Binnie, 2004; Benjamin et al., 2005).
Over the last decade 1H-MRS has contributed toward understanding the changes in brain metabolism associated with IGE. Nevertheless, there is relative inhomogeneity in studied IGE groups, sometimes including IGE sub-syndromes that were not classifiable. In addition, few studies included a sufficiently large JME patient popula-tion and thus the results have been highly variable (Savic et al., 2000, 2004; Bernasconi et al., 2003; Mory et al., 2003; Simister et al., 2003b; Haki et al., 2007).
N-Acetyl-aspartate
As demonstrated in previous studies (Savic et al., 2000, 2004; Bernasconi et al., 2003; Mory et al., 2003; Simister et al., 2003b; Haki et al., 2007), we found lower concentra-tions of primary motor cortex, medial prefrontal cortex, and thalamic NAA/Cr of JME patients compared with controls. Furthermore, when we performed individual analysis, we found many patients with abnormal NAA/Cr also in the parietal and insular cortex.
N-Acetyl-aspartate is located exclusively in neurons and, therefore, it is considered a marker of neuronal and axonal integrity (Goldstein, 1976). Its reduced concentra-tion may reflect general neuronal dysfuncconcentra-tion, specific mitochondrial dysfunction, neuronal lesion leading to release of aminohydrolase and degradation of NAA, or neuronal loss (Savic et al., 2000). However, a decreased
density of neurons seems less probable in our patients when considering that neuropathologic studies did not demonstrate neuronal loss in the thalamus and revealed cortical and subcortical ‘‘microdysgenesis’’ associated with IGE (Meencke & Janz, 1984; Meencke, 1985). More-over, some volumetric-based MRI studies have identified widespread cerebral structural changes in patients with IGE, in particular increased frontal and thalamic gray-matter content (Woermann et al., 1998, 1999a; Betting et al., 2006).
These structural and neurochemical findings may be the epiphenomenon of ‘‘minor’’ regional cortical dysplasia, which may be related to mutations described in the gene
EFHC1(Suzuki et al., 2004). In fact, reduced concentra-tion of NAA has been found even when the cell density is normal or elevated in cortical dysplasia (Kuzniecky et al., 1997; Li et al., 1998), where neurometabolism may be dysfunctional, which would be compatible with our find-ings of reduced NAA/Cr along with a normal or altered GLX/Cr, or vice versa.
Glutamate-glutamine complex
Glutamate is the principal excitatory neurotransmitter in the mammalian brain, and its elevated concentrations have been identified in human epileptic tissue (Woermann et al., 1999b; Simister et al., 2003a). Although 1H-MRS may reliably detect the complex signal GLX at short echo-times (Woermann et al., 1999b), further studies are needed to determine whether GLX elevation directly represents elevation of glutamate or glutamine or an increase in glutamatergic neurons and thus, GLX values should be interpreted with caution (Simister et al., 2003b).
Even so, the reduced GLX/Cr found in the primary motor cortex, medial prefrontal cortex, and cingulum among our patients is in accordance with neuropsycholo-gical (Devinsky et al., 1997; Savic et al., 2004) and positron emission tomography (PET) with 18-fluorodeoxyglucose
Table 5. Significant correlations between thalamic NAA/Cr and other brain structures of the
epileptic network among patientsa
Other structures Regression coef. CI 95% Beta % Correlation coef. p-value
R-thalamus
R-medial prefrontal cortex 0.52 0.25 to 0.84 45.9 0.444 <0.001* L-occipital cortex 0.21 )0.33 to)0.03 29.0 )0.309 0.018*
L-medial prefrontal cortex 0.05 0.01 to 0.86 25.2 0.262 0.047* L-thalamus
L-insula 0.89 0.37 to 1.41 29.1 0.418 0.001*
R-striatum 0.44 0.17 to 0.35 27.7 0.407 0.002*
R-medial prefrontal cortex 0.97 0.32 to 1.63 25.0 0.371 0.004*
R-insula )0.61 )1.18 to)0.04 18.2 )0.279 0.036*
Beta %, standardized regression coefficient percentage; CI, confidence interval; Correlation coef, partial correlation coefficient; Cr, creatine plus phosphocreatine; L, left; NAA,N-acetyl-aspartate; R, right; Regression coef, regression coefficient.
*Significant at p < 0.05.
a
Table presents final models obtained from multiple regression analysis with stepwise method.
1196
K. Lin et al.
Figure 2.
Correlations for age and duration of epilepsy against thalamic and frontal NAA/Cr in patients with JME. Cr, creatine plus phosphocreatine; JME, juvenile myoclonic epilepsy; NAA,N-acetyl-aspartate.
Epilepsia ILAE
MRS in Juvenile Myoclonic Epilepsy
(Swartz et al., 1996) studies suggesting that JME subjects have deficient frontal lobe functions.
Conversely, the increased GLX/Cr observed in the thal-amus (and adjacent regions, e.g., striatum and insula) may be the result of hyperexcitability. Although opposite corti-cal and subcorticorti-cal responses may be intriguing, it is possi-ble that, while seizure activity is ongoing, there may be regions of the brain with inhibited activity (Norden & Blumenfeld, 2002). Combined EEG and functional MRI have demonstrated thalamic activation with widespread cortical deactivation with a frontal maximum during generalized seizures. These hypoactivations may reflect a relative deafferentation of the cortex during generalized spike-wave, mediated by a widespread hyperpolarization of the thalamus, which could also explain the thalamic activation (Gotman et al., 2005; Gotman, 2008). In other words, abnormally increased activity in some regions of the brain during seizures may disrupt function in certain pathways or cause other changes leading secondarily to inhibition or inactivation of other brain regions, for exam-ple, frontal and parietal association cortex, causing loss of consciousness, as observed in the ‘‘network inhibition hypothesis’’(Norden & Blumenfeld, 2002).
The epileptic network
A network is a functionally and anatomically con-nected set of cortical and subcortical brain regions where activity in any one part affects the activity in all others (Hirsch et al., 2006). MRS evidence for such a network has been suggested in IGE. However, to the best of our knowledge, other than Bernasconi et al. (2003), who found one single correlation between thalamic and insu-lar NAA/Cr, no other previous report has addressed this issue directly using MRS. Therefore, it was unclear whether the metabolic perturbations described in other studies (Savic et al., 2000, 2004; Bernasconi et al., 2003; Mory et al., 2003; Simister et al., 2003b; Haki et al., 2007) reflected involvement of a metabolic network or were uncorrelated.
The thalamus is the major source of input to the cortex, yet these connections are exceeded in number by recipro-cal connections from the cortex back to the thalamus. It has widespread interconnections to the frontal lobe including premotor, primary motor, and supplementary motor cortices, as well as to the primary sensory modali-ties, basal ganglia, cerebellum, and the limbic system, serving as an integrative function for a variety of complex tasks. In addition, its reticular nucleus is a shell of GABA-ergic neurons, with strong inhibitory connections with the relay nuclei, and may induce a burst-firing mode that leads to cortical slow wave, serving to regulate cortical excitability and, therefore, seizure threshold (Norden & Blumenfeld, 2002; Blumenfeld et al., 2003).
The analysis of the thalami in 30 controls revealed a single significant relationship between the right and left
thalami, probably reflecting their general healthy meta-bolic state. Hence, the associations found between the thalamus and particular subcortical and cortical structures in patients reflect a specific effect most likely caused by the seizures/epilepsy as opposed to any intrinsic relation-ship in metabolism between these structures. In fact, the specificity of the alterations and their association within the known thalamocortical pathway—since the strongest correlation found among patients occurred between thala-mus and medial prefrontal cortex—suggest that this is a result of a functional interconnection, corroborating the ‘‘corticoreticular’’theory (Gloor, 1968) for the pathophys-iology of JME.
Relation of metabolites with other clinical data
We found a negative correlation between NAA/Cr and age and duration of epilepsy in patients, suggesting that progressive neuronal dysfunction may evolve in patients with JME. Except for Bernasconi et al. (2003), no other previous study succeeded in finding this association. Because duration is a composite variable, consisting of both age of onset of seizures and the patient’s age, any change in that variable might be related to age of onset or increasing age, but there was no such relationship in our healthy controls, as neurochemical changes are not expected in the young age range of our subjects (Grachev & Apkarian, 2000; Kadota et al., 2001).
Our results also demonstrated that poorly controlled seizures may be associated with more altered brain metab-olites, although it was observed only in some regions. Conversely, the lack of difference of NAA/Cr between JME patients with adequate seizure control and those with persistent seizures as observed in most analyzed regions, along with the lack of association observed with time since the most recent GTCS, supports the idea that neuro-nal dysfunction in JME could be related primarily to the underlying epileptogenic process rather than to the effect of the seizures themselves or to interictal activity (Bernasconi et al., 2003).
The effects of medication on NAA/Cr were not signifi-cant. It appears unlikely that AEDs were responsible for the altered concentrations of NAA/Cr in our patients, as no effects of therapy were found in most other studies (Savic et al., 2000; Bernasconi et al., 2003; Mory et al., 2003; Simister et al., 2003b; Savic et al., 2004; Haki et al., 2007).
Methodological considerations
Direct comparison among studies is difficult because of differences in acquisition techniques and tissue composi-tions, as well as to the variable number of patients and methods used for clinical distinction of IGE sub-syndromes (Simister et al., 2003b). Moreover, JME may represent a heterogeneous disorder, since it is related to several independent genetic abnormalities (Benjamin et al., 2005).
1198
K. Lin et al.
A limitation of this study is that we did not perform absolute quantification of metabolite concentrations, which would require overly long acquisition times and possibly sedation of patients. The intensity ratios NAA/Cr and GLX/Cr were used to simplify quantification across patients. Nevertheless, Cr is a ubiquitous compound found in both neurons and glial cells, and its level remains rela-tively stable even in most pathologies; therefore, it is used as an internal standard, and MRS results are often expressed as a ratio of one given metabolite to Cr (Cendes et al., 2002).
Although variation in voxel gray matter composition can affect any metabolite concentration, previous studies have failed to demonstrate differences in voxel fractional gray matter content when these voxels were positioned under identical approach conditions (Simister et al., 2003b). Therefore, we did not perform gray–white matter segmentation and we used smaller voxels to avoid partial volume averaging.
Reasonable reliability can be achieved with MRS at 1.5 T (Simister et al., 2003a; Rosen & Lenkinski, 2007), and most studies were performed with this magnetic field strength (Savic et al., 2000; Bernasconi et al., 2003; Mory et al., 2003; Simister et al., 2003b; Savic et al., 2004; Haki et al., 2007). Although several investigators have shown that the SNR gains at 3 and 4 T are relatively modest, the ability to acquire spectra at higher magnetic field strengths may improve spectral resolution and the quantification of chemical compounds such as glutamate (Rosen & Len-kinski, 2007). MRS is noninvasive and requires no isotopes or contrast agents, so that repeated measurements can be made, enabling longitudinal studies in a single sub-ject, giving insights into disease evolution and progres-sion, even in cases in which MRI reveals no distinct lesions (Cendes et al., 2002; Rosen & Lenkinski, 2007).
In conclusion, previous studies about the site of initial epileptiform discharges in patients with IGE (Velasco et al., 1989; Meeren et al., 2002) remain controversial. However, experimental models of absence seizures dem-onstrated evidences of an underlying focal abnormality that is responsible for the onset of the generalized spike-wave discharges, whereas the thalamus would be second-arily affected (Meeren et al., 2002, 2005). Although investigating the existence of an epileptic disturbed network has been difficult because of the lack of a tech-nique for exploring brain functional connectivity in vivo, the recognition of a multifocal thalamocortical network mechanism in the pathophysiology of JME has been increasingly evidenced, with a consistent body of data with the development of novel neuroimaging techniques.
Acknowledgments
This work was supported by a grant from CAPES/CNPq, FAPESP, S¼o Paulo, SP, Brazil. We are grateful for Janet Fu McDevitt, who
reviewed the manuscript as a native English speaker, and for Patrcia Guilhem de Almeida Ramos, for her expert assistance in statistical analyses.
Conflict of interest: We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is con-sistent with those guidelines. None of the authors has any conflict of interest to disclose.
References
Avoli M, Rogawski MA, Avanzini G. (2001) Generalized epileptic disorders: an update.Epilepsia42:445–457.
Benjamin Z, Andermann E, Andermann F. (2005) Mechanisms, genetics, and pathogenesis of juvenile myoclonic epilepsy: review.Curr Opin Neurol18:147–153.
Bernasconi A, Bernasconi N, Natsume J, Antel SB, Andermann F, Arnold DL. (2003) Magnetic resonance spectroscopy and imaging of the thalamus in idiopathic generalized epilepsy.Brain 126:2447– 2454.
Betting LE, Mory SB, Li LM, Lopes-Cendes I, Guerreiro MM, Guerreiro CAM, Cendes F. (2006) Voxel-based morphometry in patients with idiopathic generalized epilepsies.Neuroimage32:498–502. Binnie CD. (2004) Evidence of reflex epilepsy on functional systems in
the brain and‘‘generalized’’epilepsy. In Wolf P, Inoue Y, Zifkin B (Eds) Reflex epilepsies: progress in understanding. John Libbey Eurotext, Montrouge, pp. 7–14.
Blumenfeld H, Westerveld M, Ostroff RB, Vanderhill SD, Freeman J, Necochea A, Uranga P, Tanhehco T, Smith A, Seibyl JP, Stokking R, Studholme C, Spencer SS, Zubal IG. (2003) Selective frontal, parie-tal, and temporal networks in generalized seizures. Neuroimage 19:1556–1566.
Cendes F, Knowlton RC, Novotny E, Li LM, Antel S, Sawrie S, Laxer KD, Arnold D. (2002) Magnetic resonance spectroscopy in epilepsy: clinical issues.Epilepsia43:32–39.
Commission on Classification and Terminology of the Interna-tional League Against Epilepsy. (1989) Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30:389–399.
Commission on Neuroimaging of the International League Against Epilepsy. (1997) Recommendations for neuroimaging of patients with epilepsy.Epilepsia38:1255–1256.
Delgado-Escueta AV, Enrile-Bacsal F. (1984) Juvenile myoclonic epilepsy of Janz.Neurology34:285–294.
Devinsky O, Gershengorn J, Brown E, Perrine K, Vazquez B, Luciano D. (1997) Frontal functions in juvenile myoclonic epilepsy.J Neurol Neurosurg Psychiatr10:243–246.
Genovese CR, Lazar NA, Nichols T. (2002) Thresholding of statistical maps in functional neuroimaging using the false discovery rate. Neuroimage15:870–878.
Genton P, Salas-Puig X, Tunon A, Lahoz C, Sanchez MDSG. (1994) Juvenile myoclonic epilepsy and related syndromes: clinical and neurophysiological aspects. In Malafosse A, Genton P, Hirsch E, Marescaux C, Broglin D, Bernasconi R (Eds)Idiopathic generalized epilepsies: clinical, experimental and genetic aspects. John Libbey Eurotext, Montrouge, pp. 253–265.
Genton P, Gelisse P, Thomas P, Dravet C. (2000) Do carbamazepine and phenytoin aggravate juvenile myoclonic epilepsy? Neurology 55:1106–1109.
Gloor P. (1968) Generalized cortico-reticular epilepsies: some consider-ations on the pathophysiology of generalized bilaterally synchronous spike and wave discharge.Epilepsia9:249–263.
Goldstein FB. (1976) Aminohydrolases of brain: enzymatic hydrolysis of N-acetyl-L-aspartate and other N-acetyl-L-amino acids. J Neuro-chem26:45–49.
Gotman J, Grova C, Bagshaw A, Kobayashi E, Aghakhani Y, Dubeau F. (2005) Generalized epileptic discharges show thalamocortical activa-tion and suspension of the default state of the brain.Proc Natl Acad Sci USA102:15236–15240.
Gotman J. (2008) Epileptic networks studied with EEG-fMRI.Epilepsia 49:42–51.
MRS in Juvenile Myoclonic Epilepsy
Grachev ID, Apkarian AV. (2000) Chemical heterogeneity of the living human brain: a proton MR spectroscopy study on the effects of sex, age, and brain region.Neuroimage11:554–563.
Haki C, Gms¸tas¸ OG, Bora I, Gms¸tas¸ AU, Parlak M. (2007) Proton magnetic resonance spectroscopy study of bilateral thalamus in juve-nile myoclonic epilepsy.Seizure16:287–295.
Hirsch E, Andermann F, Chauvel P, Engel J Jr, Lopes da Silva F, Lders H, Eds. (2006) Foreword. InGeneralized seizures: from clinical phenomenology to underlying systems and networks. John Libbey Eurotext, Montrouge, XIX pp.
Janz D, Christian W. (1957) Impulsiv petit mal.Dtsch Z Nervenheilkd 176:346–386.
Kadota T, Horinouchi T, Kuroda C. (2001) Development and aging of the cerebrum: assessment with proton MR spectroscopy.AJNR Am J Neuroradiol22:128–135.
Koepp MJ, Hamandi K. (2006) Systems and networks in myoclonic sei-zures and epilepsies. In Hirsch E, Andermann F, Chauvel P, Engel J Jr, Lopes da Silva F, Lders H (Eds)Generalized seizures: from clini-cal phenomenology to underlying systems and networks. John Libbey Eurotext, Montrouge, pp. 163–182.
Kuzniecky R, Hetherington H, Pan J, Hugg J, Palmer C, Gilliam F, Faught E, Morawetz R. (1997) Proton spectroscopic imaging at 4.1 tesla in patients with malformations of cortical development and epilepsy.Neurology48:1018–1024.
Li LM, Cendes F, Bastos AC, Andermann F, Dubeau F, Arnold DL. (1998) Neuronal metabolic dysfunction in patients with cortical developmental malformations: a proton magnetic resonance spectro-scopic imaging study.Neurology50:755–759.
Meencke HJ, Janz D. (1984) Neuropathological findings in primary generalized epilepsy: a study of eight cases.Epilepsia25:8–21. Meencke HJ. (1985) Neuron density in the molecular layer of the frontal
cortex in primary generalized epilepsy.Epilepsia26:450–454. Meeren HK, Pijn JP, van Luijtelaar EL, Coenen AML, Lopes da Silva
FH. (2002) Cortical focus drives widespread corticothalamic net-works during spontaneous absence seizures in rats. J Neurosci 22:1480–1495.
Meeren HK, van Luijtelaar EL, Lopes da Silva FH, Coenen AM. (2005) Evolving concepts on the pathophysiology of absence seizures: the cortical focus theory.Arch Neurol62:371–376.
Mory SB, Li LM, Guerreiro CAM, Cendes F. (2003) Thalamic dysfunc-tion in juvenile myoclonic epilepsy: a proton MRS study.Epilepsia 44:1402–1405.
Norden AD, Blumenfeld H. (2002) The role of subcortical structures in human epilepsy.Epilepsy Behav3:219–231.
Pan JW, Williamson A, Cavus I, Hetherington HP, Zaveri H, Petroff OAC, Spencer DD. (2008) Neurometabolism in human epilepsy. Epilepsia49:31–41.
Panayiotopoulos CP, Obeid T, Tahan AR. (1994) Juvenile myoclonic epilepsy: a 5-year prospective study.Epilepsia35:285–296.
Prasad A, Kuzniecky RI, Knowlton RC, Welty TE, Martin RC, Mendez M, Faught RE. (2003) Evolving antiepileptic drug treat-ment in juvenile myoclonic epilepsy. Arch Neurol 60:1100– 1105.
Rosen Y, Lenkinski RE. (2007) Recent advances in magnetic resonance neurospectroscopy.Neurotherapeutics4:330–345.
Savic I, Lekvall A, Greitz D, Helms G. (2000) MR spectroscopy shows reduced frontal lobe concentrations of N-acetyl aspartate in patients with juvenile myoclonic epilepsy.Epilepsia41:290–296.
Savic I, sterman Y, Helms G. (2004) MRS shows syndrome differentiated metabolite changes in human-generalized epilepsies. Neuroimage21:163–172.
Simister RJ, McLean MA, Barker GJ, Duncan JS. (2003a) A proton mag-netic resonance spectroscopy study of metabolites in the occipital lobes in epilepsy.Epilepsia44:550–558.
Simister RJ, McLean MA, Barker GJ, Duncan JS. (2003b) Proton MRS reveals frontal lobe metabolite abnormalities in idiopathic general-ized epilepsy.Neurology61:897–902.
Suzuki T, Delgado-Escueta AV, Aguan K, Alonso ME, Shi J, Hara Y, Nishida M, Numata T, Medina MT, Takeuchi T, Morita R, Bai D, Ganesh S, Sugimoto Y, Inazawa J, Bailey JN, Ochoa A, Jara-Prado A, Rasmussen A, Ramos-Peek J, Cordova S, Rubio-Donnadieu F, Inoue Y, Osawa M, Kaneko S, Oguni H, Mori Y, Yamakawa K. (2004) Mutations in EFHC1 cause juvenile myoclonic epilepsy.Nat Genet36:842–849.
Swartz BE, Simpkins F, Halgren E, Mandelkern M, Brown C, Krisdakumtorn T, Gee M. (1996) Visual working memory in primary generalized epilepsy: an 18FDG-PET study.Neurology 47:1203– 1212.
Velasco M, Velasco F, Velasco AL, Lujan M, Vazquez del Mercado J. (1989) Epileptiform EEG activities of the centromedial thalamic nuclei in patients with intractable partial motor, complex partial, and generalized epilepsies.Epilepsia30:295–306.
Woermann FG, Sisodiya SM, Free SL, Duncan JS. (1998) Quantitative MRI in patients with idiopathic generalized epilepsy. Evidence of widespread cerebral structural changes.Brain121:1161–1167. Woermann FG, Free SL, Koepp MJ, Sisodiya SM, Duncan JS. (1999a)
Abnormal cerebral structure in juvenile myoclonic epilepsy demonstrated with voxel-based analysis of MRI.Brain122:2101– 2108.
Woermann FG, McLean MA, Bartlett PA, Parker GJ, Barker GJ, Duncan JS. (1999b) Short echo time single-voxel 1H magnetic resonance spectroscopy in magnetic resonance imaging-negative temporal lobe epilepsy: different biochemical profile compared with hippocampal sclerosis.Ann Neurol45:369–376.
Wolf P. (1992) Juvenile myoclonic epilepsy. In Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P (Eds)Epileptic syndromes in infancy, childhood and adolescence. John Libbey & Company Ltd, Montrouge, pp. 313–327.
1200
K. Lin et al.
ARTIGO 3
Voxel-based morphometry evaluation of patients with photosensitive juvenile myoclonic epilepsy
Katia Lin, Andrea Parolin Jackowski, Henrique Carrete Junior, Gerardo Maria Araújo Filho, Henrique Hattori Silva, Mirian Salvadori Bittar Guaranha, Laura Maria Figueiredo Ferreira
Guilhoto, Rodrigo Affonseca Bressan, Elza Márcia Targas Yacubian
