Acta Neurol Scand 2015: 132: 401–409 DOI: 10.1111/ane.12409
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA
Lateralizing value of semiology in medial temporal lobe epilepsy Dupont S, Samson Y, Nguyen-Michel V-H, Zavanone C, Navarro V, Baulac M, Adam C. Lateralizing value of semiology in medial temporal lobe epilepsy. Acta Neurol Scand 2015: 132: 401–409. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. Objectives – Analysing the clinical characteristics of seizures constitutes a fundamental aspect of the presurgical evaluation of patients with medial temporal lobe epilepsy and unilateral hippocampal sclerosis (MTLE-HS), the most frequent form of focal epilepsy accessible to surgery. We sought to retrospectively determine whether objective manifestations could have a reliable lateralizing value in a large population of MTLE-HS patients and if their presence could help to identify those patients who would be seizure free after surgery. Material and methods – We analysed the frequency and predictive lateralizing value of objective ictal and postictal signs in 391 patients with MTLE-HS (183 left/208 right). Data were derived from chart review and not from blinded videoEEG analysis. Correlation between the presence of reliable lateralizing signs and postoperative outcome was performed in a subgroup of 302 patients who underwent surgery. Results – Contralateral dystonic posturing was the most frequent and reliable lateralizing sign that correctly lateralized the focus in 96% of patients. Unilateral head/eye deviation was noted in 42% of the patients and predicted unilateral focus in 67%. Ipsilateral postictal nose wiping, contralateral clonus and hypokinesia correctly lateralized the focus in 75%, 81%, respectively, and 100 of patients but were less frequently depicted. Postictal aphasia was a strong lateralizing sign for left MLE-HS. The presence of reliable lateralizing signs was not a predictor of seizure freedom. Conclusion – Seizure semiology is a simple tool that may permit reliable lateralization of the seizure focus in MTLE-HS. The presence of reliable lateralizing signs is not associated with a better postoperative outcome.
Introduction
Surgery is often an effective therapy for refractory medial temporal lobe epilepsy associated with hippocampal sclerosis (MTLE-HS) (1), the most common form of refractory focal epilepsy in adults. However, only two-thirds of patients are seizure free in the 2–3 years after surgery (2). It is not clear what factors govern freedom from seizure after surgery. Obviously, the success of surgical intervention depends on a thorough preoperative evaluation in which evidence from multiple diagnostic tools is used to lateralize and localize the epileptogenic zone (3). The first
S. Dupont1,2,3,4, Y. Samson3,4,5, V.-H. Nguyen-Michel1,6, C. Zavanone2, V. Navarro1,3,4,7, M. Baulac1,3,4, C. Adam1,3 Epilepsy Unit, H^opital de la Pitie-Salp^etriere, APHP, Paris, France; 2Rehabilitation Unit, H^opital de la PitieSalp^etriere, APHP, Paris, France; 3Centre de Recherche de l’Institut du Cerveau et de la Mo€elle Epiniere (ICM), UMPC-UMR 7225 CNRS-UMRS 975 INSERM, Paris, France; 4Universite Pierre et Marie Curie, Paris 6, France; 5Stroke Unit, H^opital de la Pitie-Salp^etriere, APHP, Paris, France; 6Neurophysiology Unit of the Charles Foix Hospital, APHP, Paris, France; 7 Neurophysiology Unit, H^opital de la Pitie-Salp^etriere, APHP, Paris, France 1
Key words: epilepsy; outcome; semiology; video-EEG S. Dupont, Epilepsy Unit and Rehabilitation Unit, H^opital de la Salp^etriere 47, boulevard de l’H^opital, 75651 Paris Cedex 13, France Tel.: +33 1 42 16 41 15 Fax: +33 1 42 16 03 03 e-mail: [email protected] Accepted for publication March 20, 2015
non-invasive phase of the presurgical evaluation combines a gathering of the clinical history of the epilepsy, and the analysis of ictal semiology and ictal discharges thanks to video-EEG recordings (4, 5) and structural brain MRI. At the end of this first phase, when all the data converge towards a single focal and well-lateralized epileptogenic zone in the medial anterior temporal structures, the surgery procedure can be proposed to the patient (5). Lateralization task is mainly performed by visual inspection of interictal EEG and ictal EEG (6–8) and by the inspection of MRI (side of the hippocampal sclerosis). When lateralizing data are discordant, a 401
Dupont et al. second phase must be planed including brain hemodynamic and metabolic analysis using single emission photon tomography (SPECT) and positron emission tomography (PET) (9) and sometimes stereo-electroencephalographic (SEEG) recordings (10). With the advent of such sophisticated methods to lateralize the seizure onset zone, what is the place of ictal and postictal semiology? Objective clinical manifestations of seizures originating in mesiotemporal structures commonly consist of non-lateralized manifestations such as oroalimentary or gestural automatisms. A number of studies have shown the potential lateralizing value of some ictal and postictal manifestations, but to date no significant series of cases have been published. The purpose of this study was to examine ictal and postictal semiology in a large series of welldefined and lateralized patients with MTLE-HS to assess the lateralizing value and the frequency of clinical seizure symptoms. We also wished to determine whether the presence of reliable lateralizing clinical signs might provide a diagnostic complement to help identify patients who will be seizure free after surgery. Material and methods Patient material
The study group was identified from the Epilepsy Data Base of the Epilepsy Unit of the Pitie-Salp^etriere Hospital in Paris. It consisted of all patients suffering from refractory medial temporal lobe epilepsy (MTLE) associated with hippocampal sclerosis who underwent Video-EEG between 1991 and 2013. Criteria for inclusion were as follows: (i) welldocumented refractory MTLE, (ii) unilateral hippocampal sclerosis on structural MRI, (iii) at least one seizure recorded and documented by Video-EEG. All the patients gave their informed consent to the use and publication of data related to their epilepsy at the time of video-EEG.
1. Data related to epilepsy: side of the hippocampal sclerosis, age at seizure onset, age and number of antiepileptic drugs at the time of video- EEG recording, mean and median monthly number of seizures, mean and median duration of video-EEG recording (days), mean and median number of seizures recorded during video-EEG. 2. Seizure features: considered present if associated with at least one, not necessarily all, of the patient’s seizures: a. Objective non-motor ictal signs included gestural automatisms (including walking around), oroalimentary automatisms (chewing–swallowing–smacking–licking), mimetic automatisms (expressing a positive or negative affect such as grimacing–smiling– laughing–crying) and verbal automatisms (vocalization, ictal speech, dysphasia, dysarthria). For these signs, we only noted the presence or absence of any automatism. b. Objective motor ictal signs included dystonic posturing of the upper and lower limbs, eye blinking, tonic posturing of the upper and lower limbs, hypokinesia (= ictal paresis, partial or complete loss of muscle movement), sagging (leaning or falling to one side), clonus, head/eye deviation at any time during a partial seizure except just before the secondary tonic–clonic phase. c. Postictal signs: postictal confusion, postictal aphasia defined by interaction with an observer during and after the seizure, postictal nose wiping
Collection of data
3. If a patient exhibited exclusive unilateral signs lateralized ipsilaterally or contralaterally to the seizure focus, he was classified as unilateral. If a patient exhibited simultaneously bilateral signs, he was classified as bilateral. 4. If a patient exhibited unilateral lateralizing signs first on one side and later on the other during the same seizure, he was defined as having alternative signs and classified as bilateral, if a patient exhibited either left or right unilateral lateralizing signs during successive seizures, he was also classified as bilateral.
We retrospectively reviewed all Video-EEG and medical records for 404 patients fulfilling these inclusion criteria. Based on medical charts reviewed later and data collected at the time of EEG-video, we examined every record of every seizure and retrospectively collected for each patient the following data:
Surgery occurrence: If surgery took place, postoperative outcome was defined with a simplified Engel’s classification (Engel’s class Ia: completely seizure-free patients, Engel’s class I: seizure-free patients and those who experienced only simple partial seizures, or drug withdrawal seizures, Engel’s class II-III-IV patients: those for whom
402
Lateralizing signs in MTLE complex seizures were not abolished and surgery outcome was judged as truly unsatisfactory). Statistical analysis
Two populations of patients were analysed: 1. The complete population with MTLE-HS: lateralization of the focus was based on congruent ictal EEG and imaging (side of the hippocampal sclerosis on MRI) data. 2. The postoperative population: we examined whether the presence of reliable lateralization signs defined in the general population differed between seizure-free and not seizure-free patients. The gold standard for correct seizure focus identification was the lateralization of the focus in the global population and in the surgery group. Seizure freedom after resection was not retained as gold standard as the number of patients who were not seizure free after surgery was too low to determine whether there was any difference between seizure-free and non-seizurefree patients. Statistical analysis using either the chi-square test or Mann–Whitney test was performed with dedicated software (MedCalc). The significance level was defined as P < 0.05. Analyses were first performed for the complete population of patients with MTLE-HS and then, for statistically significant features, in the seizure-free population. Sensitivity, specificity, positive and negative predictive values were calculated for the most frequent motor lateralizing signs. As we did not determine at the time of the reviewing in how many seizures a specific sign occurred for each individual patient, variation between seizures in a patient was not accounted for. Results Global population description
Patient characteristics are detailed in Table 1. Thirteen patients were excluded because of incomplete data on seizures. A total of 183 of the remaining patients (47%) had a right hippocampal sclerosis (RTLE) and 208 (53%) exhibited a left hippocampal sclerosis (LTLE). Surgery
From these 391 patients, 323 underwent surgery (83%) and 68 (17%) did not. Among the 68
Table 1 Patients characteristics Demographic data Patients Total number Sex Female Male Pathology Hippocampal sclerosis Side of hippocampal sclerosis (HS) Right HS Left HS Manual laterality Left-handed Right handed Equal Indeterminate Epilepsy related-data Median age (years) at onset Median age (years) at video-EEG Median duration (years) of epilepsy Median number of AEDs Median number of sz/months GTC2 sz Yes Rare (
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA
Lateralizing value of semiology in medial temporal lobe epilepsy Dupont S, Samson Y, Nguyen-Michel V-H, Zavanone C, Navarro V, Baulac M, Adam C. Lateralizing value of semiology in medial temporal lobe epilepsy. Acta Neurol Scand 2015: 132: 401–409. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. Objectives – Analysing the clinical characteristics of seizures constitutes a fundamental aspect of the presurgical evaluation of patients with medial temporal lobe epilepsy and unilateral hippocampal sclerosis (MTLE-HS), the most frequent form of focal epilepsy accessible to surgery. We sought to retrospectively determine whether objective manifestations could have a reliable lateralizing value in a large population of MTLE-HS patients and if their presence could help to identify those patients who would be seizure free after surgery. Material and methods – We analysed the frequency and predictive lateralizing value of objective ictal and postictal signs in 391 patients with MTLE-HS (183 left/208 right). Data were derived from chart review and not from blinded videoEEG analysis. Correlation between the presence of reliable lateralizing signs and postoperative outcome was performed in a subgroup of 302 patients who underwent surgery. Results – Contralateral dystonic posturing was the most frequent and reliable lateralizing sign that correctly lateralized the focus in 96% of patients. Unilateral head/eye deviation was noted in 42% of the patients and predicted unilateral focus in 67%. Ipsilateral postictal nose wiping, contralateral clonus and hypokinesia correctly lateralized the focus in 75%, 81%, respectively, and 100 of patients but were less frequently depicted. Postictal aphasia was a strong lateralizing sign for left MLE-HS. The presence of reliable lateralizing signs was not a predictor of seizure freedom. Conclusion – Seizure semiology is a simple tool that may permit reliable lateralization of the seizure focus in MTLE-HS. The presence of reliable lateralizing signs is not associated with a better postoperative outcome.
Introduction
Surgery is often an effective therapy for refractory medial temporal lobe epilepsy associated with hippocampal sclerosis (MTLE-HS) (1), the most common form of refractory focal epilepsy in adults. However, only two-thirds of patients are seizure free in the 2–3 years after surgery (2). It is not clear what factors govern freedom from seizure after surgery. Obviously, the success of surgical intervention depends on a thorough preoperative evaluation in which evidence from multiple diagnostic tools is used to lateralize and localize the epileptogenic zone (3). The first
S. Dupont1,2,3,4, Y. Samson3,4,5, V.-H. Nguyen-Michel1,6, C. Zavanone2, V. Navarro1,3,4,7, M. Baulac1,3,4, C. Adam1,3 Epilepsy Unit, H^opital de la Pitie-Salp^etriere, APHP, Paris, France; 2Rehabilitation Unit, H^opital de la PitieSalp^etriere, APHP, Paris, France; 3Centre de Recherche de l’Institut du Cerveau et de la Mo€elle Epiniere (ICM), UMPC-UMR 7225 CNRS-UMRS 975 INSERM, Paris, France; 4Universite Pierre et Marie Curie, Paris 6, France; 5Stroke Unit, H^opital de la Pitie-Salp^etriere, APHP, Paris, France; 6Neurophysiology Unit of the Charles Foix Hospital, APHP, Paris, France; 7 Neurophysiology Unit, H^opital de la Pitie-Salp^etriere, APHP, Paris, France 1
Key words: epilepsy; outcome; semiology; video-EEG S. Dupont, Epilepsy Unit and Rehabilitation Unit, H^opital de la Salp^etriere 47, boulevard de l’H^opital, 75651 Paris Cedex 13, France Tel.: +33 1 42 16 41 15 Fax: +33 1 42 16 03 03 e-mail: [email protected] Accepted for publication March 20, 2015
non-invasive phase of the presurgical evaluation combines a gathering of the clinical history of the epilepsy, and the analysis of ictal semiology and ictal discharges thanks to video-EEG recordings (4, 5) and structural brain MRI. At the end of this first phase, when all the data converge towards a single focal and well-lateralized epileptogenic zone in the medial anterior temporal structures, the surgery procedure can be proposed to the patient (5). Lateralization task is mainly performed by visual inspection of interictal EEG and ictal EEG (6–8) and by the inspection of MRI (side of the hippocampal sclerosis). When lateralizing data are discordant, a 401
Dupont et al. second phase must be planed including brain hemodynamic and metabolic analysis using single emission photon tomography (SPECT) and positron emission tomography (PET) (9) and sometimes stereo-electroencephalographic (SEEG) recordings (10). With the advent of such sophisticated methods to lateralize the seizure onset zone, what is the place of ictal and postictal semiology? Objective clinical manifestations of seizures originating in mesiotemporal structures commonly consist of non-lateralized manifestations such as oroalimentary or gestural automatisms. A number of studies have shown the potential lateralizing value of some ictal and postictal manifestations, but to date no significant series of cases have been published. The purpose of this study was to examine ictal and postictal semiology in a large series of welldefined and lateralized patients with MTLE-HS to assess the lateralizing value and the frequency of clinical seizure symptoms. We also wished to determine whether the presence of reliable lateralizing clinical signs might provide a diagnostic complement to help identify patients who will be seizure free after surgery. Material and methods Patient material
The study group was identified from the Epilepsy Data Base of the Epilepsy Unit of the Pitie-Salp^etriere Hospital in Paris. It consisted of all patients suffering from refractory medial temporal lobe epilepsy (MTLE) associated with hippocampal sclerosis who underwent Video-EEG between 1991 and 2013. Criteria for inclusion were as follows: (i) welldocumented refractory MTLE, (ii) unilateral hippocampal sclerosis on structural MRI, (iii) at least one seizure recorded and documented by Video-EEG. All the patients gave their informed consent to the use and publication of data related to their epilepsy at the time of video-EEG.
1. Data related to epilepsy: side of the hippocampal sclerosis, age at seizure onset, age and number of antiepileptic drugs at the time of video- EEG recording, mean and median monthly number of seizures, mean and median duration of video-EEG recording (days), mean and median number of seizures recorded during video-EEG. 2. Seizure features: considered present if associated with at least one, not necessarily all, of the patient’s seizures: a. Objective non-motor ictal signs included gestural automatisms (including walking around), oroalimentary automatisms (chewing–swallowing–smacking–licking), mimetic automatisms (expressing a positive or negative affect such as grimacing–smiling– laughing–crying) and verbal automatisms (vocalization, ictal speech, dysphasia, dysarthria). For these signs, we only noted the presence or absence of any automatism. b. Objective motor ictal signs included dystonic posturing of the upper and lower limbs, eye blinking, tonic posturing of the upper and lower limbs, hypokinesia (= ictal paresis, partial or complete loss of muscle movement), sagging (leaning or falling to one side), clonus, head/eye deviation at any time during a partial seizure except just before the secondary tonic–clonic phase. c. Postictal signs: postictal confusion, postictal aphasia defined by interaction with an observer during and after the seizure, postictal nose wiping
Collection of data
3. If a patient exhibited exclusive unilateral signs lateralized ipsilaterally or contralaterally to the seizure focus, he was classified as unilateral. If a patient exhibited simultaneously bilateral signs, he was classified as bilateral. 4. If a patient exhibited unilateral lateralizing signs first on one side and later on the other during the same seizure, he was defined as having alternative signs and classified as bilateral, if a patient exhibited either left or right unilateral lateralizing signs during successive seizures, he was also classified as bilateral.
We retrospectively reviewed all Video-EEG and medical records for 404 patients fulfilling these inclusion criteria. Based on medical charts reviewed later and data collected at the time of EEG-video, we examined every record of every seizure and retrospectively collected for each patient the following data:
Surgery occurrence: If surgery took place, postoperative outcome was defined with a simplified Engel’s classification (Engel’s class Ia: completely seizure-free patients, Engel’s class I: seizure-free patients and those who experienced only simple partial seizures, or drug withdrawal seizures, Engel’s class II-III-IV patients: those for whom
402
Lateralizing signs in MTLE complex seizures were not abolished and surgery outcome was judged as truly unsatisfactory). Statistical analysis
Two populations of patients were analysed: 1. The complete population with MTLE-HS: lateralization of the focus was based on congruent ictal EEG and imaging (side of the hippocampal sclerosis on MRI) data. 2. The postoperative population: we examined whether the presence of reliable lateralization signs defined in the general population differed between seizure-free and not seizure-free patients. The gold standard for correct seizure focus identification was the lateralization of the focus in the global population and in the surgery group. Seizure freedom after resection was not retained as gold standard as the number of patients who were not seizure free after surgery was too low to determine whether there was any difference between seizure-free and non-seizurefree patients. Statistical analysis using either the chi-square test or Mann–Whitney test was performed with dedicated software (MedCalc). The significance level was defined as P < 0.05. Analyses were first performed for the complete population of patients with MTLE-HS and then, for statistically significant features, in the seizure-free population. Sensitivity, specificity, positive and negative predictive values were calculated for the most frequent motor lateralizing signs. As we did not determine at the time of the reviewing in how many seizures a specific sign occurred for each individual patient, variation between seizures in a patient was not accounted for. Results Global population description
Patient characteristics are detailed in Table 1. Thirteen patients were excluded because of incomplete data on seizures. A total of 183 of the remaining patients (47%) had a right hippocampal sclerosis (RTLE) and 208 (53%) exhibited a left hippocampal sclerosis (LTLE). Surgery
From these 391 patients, 323 underwent surgery (83%) and 68 (17%) did not. Among the 68
Table 1 Patients characteristics Demographic data Patients Total number Sex Female Male Pathology Hippocampal sclerosis Side of hippocampal sclerosis (HS) Right HS Left HS Manual laterality Left-handed Right handed Equal Indeterminate Epilepsy related-data Median age (years) at onset Median age (years) at video-EEG Median duration (years) of epilepsy Median number of AEDs Median number of sz/months GTC2 sz Yes Rare (
