For children with intractable epilepsy, surgery provides significant reduction in seizure frequency, with no significant declines in intellectual or behavioral functioning (Datta, et al., 2011). Prior to surgery, children must undergo a thorough assessment to determine the location of the seizure-focus and to evaluate risks of post-operative impairment (Lee, 2010). Currently, fMRI offers one of the most reliable and least invasive means of localizing language prior to surgery (McDonald, Saykin, William & Assaf, 2006). Dichotic listening, a behavioral task in which subjects are asked to listen to two competing stimuli simultaneously, offers a possible alternative for children who cannot complete fMRI studies. Previous studies have relied on research-based listening tasks and the type of quantitative analysis of the fMRI rarely available in the clinical setting. Instead, this study examined how well dichotic listening results predict language lateralization from fMRI within a clinical setting. Data were gathered through a records review of 13 children with intractable epilepsy referred to Austin Neuropsychology through the epilepsy treatment team at Dell Children’s Medical Center. Overall, children classified as atypical language dominance on the fMRI studies showed lower levels of right ear advantage on the dichotic listening measure. Despite this trend, a discriminant analysis using the dichotic listening results to predict fMRI classification showed no significant improvement over chance classification. A secondary analysis examined factors related to a child’s ability to complete an fMRI language study, comparing 12 children from the original sample with 6 children referred through the same process and over the same time period who could not obtain a successful fMRI determination of language lateralization. Overall, children who successfully completed the fMRI language studies showed a trend of lower levels of difficulty with behavioral regulation and higher levels of intelligence. Although the non-significant results highlight the limitations of dichotic listening as a clinical tool, the failure rate within the total sample, along with the information about the roles of intelligence and behavioral regulation, may help spur the development of alternative methods of language lateralization. / text

Introdução: A lateralização atípica de linguagem (LAL) é mais frequentemente observada em indivíduos com epilepsia no hemisfério esquerdo. Pacientes com epilepsia secundária a esclerose mesial temporal (EMT) apresentam alta frequência de LAL. A presença de EMTE, diretamente ou por efeitos à distância, perturbaria a determinação típica de lateralidade de linguagem (LL). Estudos de LL com RMf, em populações homogêneas de pacientes com EMT são escassos. Os padrões regionais de reorganização, nesta população, assim como os fatores clínicos e funcionais associados à LAL são pouco conhecidos. Os objetivos deste estudo foram (1) Validar paradigma de linguagem em RMf capaz de ativar áreas frontais e temporais de linguagem (2) Avaliar LL regional, em população homogênea de pacientes com EMT e verificar o papel de variáveis clínicas e da atividade epileptiforme na determinação de LAL. Foram avaliados 46 pacientes com EMT unilateral (24 EMTE e 22 EMTD) e 24 controles saudáveis, destros. Todos os participantes foram submetidos a RMf de linguagem com paradigma de nomeação responsiva à leitura (NRL) desenvolvido em língua portuguesa. O paradigma foi eficaz em ativar áreas de linguagem. A correlação entre IL-RMf e IL-Wada foi estatisticamente significativa no GFM (0,387; p < 0,01); GFI (0,594, p < 0,001); GTS (0,357, p < 0,05); GTM (0,509, p < 0,001) e GTI (0,489, p < 0,001). Os pacientes foram classificados em discordantes e concordantes (EEG/RM), após monitorização prolongada por vídeo-EEG. As classificações individuais de LL do teste de Wada e RMf foram concordantes em 84%(GFM), 86% (GFI), 74%(GTS), 83%(GTM) e 86%(GTI) dos casos. A frequência de LAL variou de 17 a 33%, para EMTE; 14 a 17%, para EMTD e de 0 a 10%, no grupo controle. Os grupos EMTEc e EMTEd apresentaram menores valores de IL-RMf, em relação ao grupo controle no GFM e GFI (p <= 0.05). Adicionalmente, o grupo EMTEd apresentou menor IL-RMf no GTM (p <= 0.01) e GTI (p <= 0.05). O grupo EMTDd apresentou menor IL-RMf no GFM; GFI; GTM e GTI (p <= 0.05). A ocorrência de mais de 5 descargas/hora no período interictal associou-se a LAL no GFM (p <= 0.01). Início da epilepsia, maior tempo de epilepsia e presença de IPI não foram associadas a LAL. Concluiu-se que a presença de EMT associou-se a LAL nas regiões frontais e temporais, sugerindo reorganização de linguagem envolvendo amplas regiões cerebrais, distantes da zona epileptogênica. O achado de LAL observada no grupo EMTDd sugere que a atividade epileptiforme temporal esquerda pode influenciar a reorganização de linguagem, independentemente da presença de lesão estrutural à esquerda. A maior frequência de descargas à esquerda associou-se a LAL / Atypical language lateralization (ALL) is associated with left hemisphere epilepsy, especially in epilepsy secondary to left mesial temporal sclerosis (MTS). Recent findings indicate that MTS, directly or by an at-distance effect may disrupt typical language lateralization (LL). Language-fMRI investigations in homogeneous MTS populations are scarce. Frontal and temporal language reorganization patterns in this population remain incompletely understood, as well as clinical factors associated with ALL. Our objectives were (1) To validate an fMRI paradigm effective to activate frontal and temporal language areas (2) To evaluate regional LL patterns in a homogeneous population of MTS patients and to investigate clinical factors associated with ALL. We studied forty-six right-handed patients (24 left MTS and 22 right MTS) and 24 controls. Patients and controls were similar in age and gender. Left and right MTS groups were comparable in terms of clinical variables. All participants underwent a language fMRI with reading response-naming paradigm (RRN) task developed in Brazilian Portuguese. The task effectively activated frontal and temporal language areas, with a high concordance with the Wada test. All patients were classified as concordant and discordant after video-EEG monitoring. Correlation between Wada test and fMRI LIs was significant in MFG (0.387; p < 0.01); IFG (0.594, p < 0.001); STG (0.357, p < 0.05); MTG (0.509, p < 0.001) e ITG (0.489, p < 0.001). Individual LI classifications with Wada test and fMRI were concordant in 84%(MFG), 86% (IFG), 74%(STG), 83%(MTG) e 86%(ITG). The frequency of ALL was 17-33%, in left MTS; 14-17%, in right MTS and 0-10%, in controls. Both concordant and discordant-left MTS patients presented lower LI than controls in MFG and IFG (p<=0.05). Additionally, discordant left MTS group presented lower LI in MTG (p<=0.01) and ITG (p<=0.05). Discordant right MTS group presented lower LI in MFG; IFG; MTG and ITG (p <= 0.05). Interictal left-sided spike frequency (< 5/h) was associated with lower LI (p <= 0.01) in MTG. In conclusion, MTS was associated with frontal and temporal language reorganization. Presence of ALL in discordant RMTS group suggests that left temporal epileptic activity may influence language reorganization, even in the absence of a left structural lesion. Higher left-sided spike frequency in MTS was associated with a left-right shift of speech representation

Introdução: A lateralização atípica de linguagem (LAL) é mais frequentemente observada em indivíduos com epilepsia no hemisfério esquerdo. Pacientes com epilepsia secundária a esclerose mesial temporal (EMT) apresentam alta frequência de LAL. A presença de EMTE, diretamente ou por efeitos à distância, perturbaria a determinação típica de lateralidade de linguagem (LL). Estudos de LL com RMf, em populações homogêneas de pacientes com EMT são escassos. Os padrões regionais de reorganização, nesta população, assim como os fatores clínicos e funcionais associados à LAL são pouco conhecidos. Os objetivos deste estudo foram (1) Validar paradigma de linguagem em RMf capaz de ativar áreas frontais e temporais de linguagem (2) Avaliar LL regional, em população homogênea de pacientes com EMT e verificar o papel de variáveis clínicas e da atividade epileptiforme na determinação de LAL. Foram avaliados 46 pacientes com EMT unilateral (24 EMTE e 22 EMTD) e 24 controles saudáveis, destros. Todos os participantes foram submetidos a RMf de linguagem com paradigma de nomeação responsiva à leitura (NRL) desenvolvido em língua portuguesa. O paradigma foi eficaz em ativar áreas de linguagem. A correlação entre IL-RMf e IL-Wada foi estatisticamente significativa no GFM (0,387; p < 0,01); GFI (0,594, p < 0,001); GTS (0,357, p < 0,05); GTM (0,509, p < 0,001) e GTI (0,489, p < 0,001). Os pacientes foram classificados em discordantes e concordantes (EEG/RM), após monitorização prolongada por vídeo-EEG. As classificações individuais de LL do teste de Wada e RMf foram concordantes em 84%(GFM), 86% (GFI), 74%(GTS), 83%(GTM) e 86%(GTI) dos casos. A frequência de LAL variou de 17 a 33%, para EMTE; 14 a 17%, para EMTD e de 0 a 10%, no grupo controle. Os grupos EMTEc e EMTEd apresentaram menores valores de IL-RMf, em relação ao grupo controle no GFM e GFI (p <= 0.05). Adicionalmente, o grupo EMTEd apresentou menor IL-RMf no GTM (p <= 0.01) e GTI (p <= 0.05). O grupo EMTDd apresentou menor IL-RMf no GFM; GFI; GTM e GTI (p <= 0.05). A ocorrência de mais de 5 descargas/hora no período interictal associou-se a LAL no GFM (p <= 0.01). Início da epilepsia, maior tempo de epilepsia e presença de IPI não foram associadas a LAL. Concluiu-se que a presença de EMT associou-se a LAL nas regiões frontais e temporais, sugerindo reorganização de linguagem envolvendo amplas regiões cerebrais, distantes da zona epileptogênica. O achado de LAL observada no grupo EMTDd sugere que a atividade epileptiforme temporal esquerda pode influenciar a reorganização de linguagem, independentemente da presença de lesão estrutural à esquerda. A maior frequência de descargas à esquerda associou-se a LAL / Atypical language lateralization (ALL) is associated with left hemisphere epilepsy, especially in epilepsy secondary to left mesial temporal sclerosis (MTS). Recent findings indicate that MTS, directly or by an at-distance effect may disrupt typical language lateralization (LL). Language-fMRI investigations in homogeneous MTS populations are scarce. Frontal and temporal language reorganization patterns in this population remain incompletely understood, as well as clinical factors associated with ALL. Our objectives were (1) To validate an fMRI paradigm effective to activate frontal and temporal language areas (2) To evaluate regional LL patterns in a homogeneous population of MTS patients and to investigate clinical factors associated with ALL. We studied forty-six right-handed patients (24 left MTS and 22 right MTS) and 24 controls. Patients and controls were similar in age and gender. Left and right MTS groups were comparable in terms of clinical variables. All participants underwent a language fMRI with reading response-naming paradigm (RRN) task developed in Brazilian Portuguese. The task effectively activated frontal and temporal language areas, with a high concordance with the Wada test. All patients were classified as concordant and discordant after video-EEG monitoring. Correlation between Wada test and fMRI LIs was significant in MFG (0.387; p < 0.01); IFG (0.594, p < 0.001); STG (0.357, p < 0.05); MTG (0.509, p < 0.001) e ITG (0.489, p < 0.001). Individual LI classifications with Wada test and fMRI were concordant in 84%(MFG), 86% (IFG), 74%(STG), 83%(MTG) e 86%(ITG). The frequency of ALL was 17-33%, in left MTS; 14-17%, in right MTS and 0-10%, in controls. Both concordant and discordant-left MTS patients presented lower LI than controls in MFG and IFG (p<=0.05). Additionally, discordant left MTS group presented lower LI in MTG (p<=0.01) and ITG (p<=0.05). Discordant right MTS group presented lower LI in MFG; IFG; MTG and ITG (p <= 0.05). Interictal left-sided spike frequency (< 5/h) was associated with lower LI (p <= 0.01) in MTG. In conclusion, MTS was associated with frontal and temporal language reorganization. Presence of ALL in discordant RMTS group suggests that left temporal epileptic activity may influence language reorganization, even in the absence of a left structural lesion. Higher left-sided spike frequency in MTS was associated with a left-right shift of speech representation