Abstracts Z Epileptol 2010 · 23:275–278 DOI 10.1007/s10309-010-0137-7 © Springer-Verlag 2010

Epilepsiechirurgische Zweiteingriffe – Indikationsstellung & Outcome Jahrestagung 2010 der Arbeitsgemeinschaft für prächirurgische Epilepsiediagnostik und operative Epilepsiechirurgie

Automated morphometric MRI analysis for improved detection of focal cortical dysplasia in epilepsy patients Kröll-Seger J Swiss Epilepsy Centre, Zurich, Switzerland Introduction: Focal cortical dysplasia (FCD) is increasingly recognized as an underlying cause of formerly cryptogenic epilepsy and accounts for up to 25% of patients in epilepsy centres. Most of these patients are pharmacoresistent, and epilepsy surgery seems to be the best available treatment option. However, before surgery it is necessary to localize the lesion and to delineate its extent. Typical MRI features of FCD include abnormal gyral contours, thickening of the cortex, abnormal differentiation of the grey-white matter boundary, and sometimes signal hyper intensity in T2-weighted images. In subtle cases diagnosis is timeconsuming and difficult, and conventional MRI can be unrevealing in a high percentage of patients. Therefore, attempts have been made to facilitate lesion detection by modern image post-processing strategies. Methods: In the past we have developed a new voxel-based method for post-processing of MRIs based on algorithms of the freely available software for statistical parametric mapping (SPM) and on additional simple calculations and filters. From a high-resolution T1-weighted 3D MRI dataset three new feature maps are derived which characterize in comparison with a normal database three different potential features of FCD: the abnormal extension of grey matter into white matter, blurring of the grey-white matter junction, and abnormal thickness of the cortical ribbon. This morphometric MRI analysis has proven to facilitate the detection of subtle FCD and to increase the diagnostic yield of MRI. However, the recognition of FCD in the new feature maps is still based on visual examination, and a clear cut-off differentiating between truly dysplastic brain areas and false-positive findings highlighted in these maps is missing. Project aim: The research project is aimed at developing and validating a method for automated detection of FCD. Realization: In a first step, a variety of parameters are developed which characterize suspicious brain regions highlighted in the morphometric maps by quantifying their extent, volume, geometric shape, degree of deviation to the normal database, extent of overlap between the different feature maps etc. These quantitative parameters are tested on the MRI data of FCD patients and healthy controls to determine appropriate

cut-off values which are best suitable to distinguish between dysplastic tissue and normal variations of grey and white matter distribution. After establishing a set of favourable parameters and corresponding cut-offs they are combined and integrated into a procedure for automated FCD detection. This method is finally validated in a separate large population of healthy controls and epilepsy patients with histological proven FCD recruited in different epilepsy centres. The diagnostic yield will be assessed by comparing the locations of potential FCD with the resection areas and by determining sensitivity, specificity, and predictive values of this method. Conclusion: The project shall extend the method of morphometric MRI analysis by implementing and validating an additional procedure for automated FCD detection. This allows for a more time-saving, objective and observer-independent MRI analysis and may further improve the sensitivity of MRI for detecting subtle dysplastic lesions. Ultimately, these efforts shall increase the chances for early and successful epilepsy surgery in patients with pharmacoresistent focal epilepsy.

Hemispherectomy in adults Delev D, von Lehe M and Schramm J University Hospital Medical Center, Bonn University, Germany Introduction: Hemispherectomies are performed less frequently in adults, little has been published and thus, knowledge about these patients and results for functional hemispherectomies are less well known. We therefore reviewed our data base for adult hemispherectomy patients with the intent to describe the patient group and outcome. Methods and patients: This is an analysis of 27 adult patients who had functional hemispherectomy in a total series of 130 patients between 1989 and 2009. Data were collected prospectively usually by annual or bi-annual follow-up performed by a telephone call or letters or follow-up visits at the hospital. Mean age at operation was 29 years, range 19 to 54. Patients with follow up under 12 months (n=2) were excluded from most analyses. One patient died, having been seizure free four months after the surgery in a traffic accident as a pedestrian. Etiologies were: Post-infarction cyst or porencephaly n=15 (58%), post encephalitic with or without cyst n=4 (15%), Rasmussen encephalitis or other acute encephalitis n=2 (8%), hemiatrophy n=2 (8%), Zeitschrift für Epileptologie 4 · 2010 

| 275

Abstracts hemimegalencephaly n=1 (4%), dysplasia or cortical malformations n=2 (8%). Four patients were operated on using Rasmussen’s functional hemispherectomy technique, three using our own perisylvian transcortical method, and 17 using the transsylvian keyhole disconnection procedure (Schramm et al). Follow-up over one year was available in 24 cases (mean 6 yrs, range 2-12 yrs). Engel class I outcome was achieved in 88% of cases, 88% in the transsylvian technique, 100% in the transcortical perisylvian technique, and 75% in the Rasmussen technique. Of the 24 patients 17 (71%) were always Engel class I after surgery, four (16%) were initially not seizure free but became Engel class I later, and three (12%) were initially seizure free and later had seizures again. Thus, recurrent seizures occurred in 12% of all patients. Two patients already had a shunt before the hemispherotomy. Five patients developed surgical complications related to the hemispherotomy: one wound infection, subdural hematoma, and infection of a pre-existing shunt. Two late surgical complications: one intraventricular cyst needed resection and one patient needed a later shunt with several revisions. One postoperative pneumonia and one recurrence of depressions and episodes of fear, known previously. None of the complications led to permanent sequelae except for the one case with a new shunt implantation. The shunt rate is thus 4%. Conclusion: Outcome of functional hemispherectomy in adults appears to be rewarding with Engel class I outcome of 88%. There was a relatively high rate of complications, although significant deficits did not result from this. Later re-occurrence of seizures occurred in 12%. It appears worthwhile to perform functional hemispherectomies in adults. Reference: Schramm J, Kral T, Clusmann H: Transsylvian Keyhole Functional H­emispherectomy. Neurosurgery 49:891-900; 2001

Hemispherotomy in 3 adult patients: Indication and follow-up Hermsen A1, Hamer HM1, Reif PS1, Bauer S1, Haag A1, Schramm J2, Rosenow F1 1 Epilepsy Center Marburg, Department of Neurology, Philipps-University Marburg, Germany, 2Department of Neurosurgery, University of Bonn, Germany We present Indication for and and follow-up after hemispherotomy in three adult patients Case 1, female, 17-11 yrs A) right-hemispheric epilepsy since age 13 yrs & B) IGE Focal deficits: left homonymous hemianopsia Seizures (sz): I) Dialeptic sz  GTCS, II) Dyscognitive status epilepticus Etiology: Sturge-Weber-Syndrome EEG: Interictal: A) sharp wave right temporal, B) spike-wave complex, generalized Ictal: A) Seizure-pattern right parieto-occipital, B) Seizure-pattern generalized (polyspike&wave-complexes) MRI: angiomatosis temporo-parieto-occipital, suspected hippocampal sclerosis, right hemiatrophy Neuropsychology (restriction mother tongue): Right-handed, IQ still average though school for learning disabled, short term-memory below average, verbal & nonverbal memory unremarkable; No lateralising functional deficits, frontal or parietal dysfunctions.

276 | 

Zeitschrift für Epileptologie 4 · 2010

11/09: modified right hemispherotomy (parieto-occipito-temporal disconnection) 6-months-Follow-Up: Three early postoperative sz, seizure free since (ILAE1). New spastic hemiparesis left leg/foot EEG: continuous slow and sharp wave, regional right temporal (T6, rare) Neuropsychology: Improvement: short-term memory (especially verbal) Losses: nonverbal memory, new spatial perception deficit. Positive attitude about operation. Case 2, male, 20 yrs Right hemispheric epilepsy since age 14 yrs Focal deficits: Left spastic hemiparesis, homonymous hemianopsia left, hydrocephalus, ventriculo-peritoneal shunt Seizures:    I) epigastric aura  automotor sz  clonic sz left face     GTCS,   II) tonic sz left arm  bilateral tonic sz Etiology: early prenatal porencephalic cyst, DD schizencephaly EEG: Interictal: Sharp wave right parietal (max.P8,P4), continuous slow, lateralized right Ictal: EEG-seizure-pattern right parietal MRI: right-hemispheric defect (MCA territory) with polymicrogyria. Neuropsychology: Right-handed, secondary school, low average IQ, deficits in attention, verbal memory & executive functions, preserved visuospatial functions & nonverbal memory contrasting with the right hemispheric lesion. Wada-Test (right injection only): no language deficit, preserved memory, no increase of hemiparesis 08/07: Rt. Hemispherotomy 2-yrs-Follow-Up: Seizure free (ILAE 1) No new neurological deficit. EEG: Sharp wave, right temporal Neuropsychology: vocational training for clerk, improvement in attention, decline in visuopatial short-term memory and nonverbal memory Case 3, male, 28 yrs Rasmussen syndrome since age 10 yrs Etiology: Rasmussen encephalitis right, Seizure: Unspecific aura  dialeptic sz  nystagmus to left  tonic sz left arm  versive sz to left  GTCS EEG: Interictal: Spike right parietal; asymmetry, decreased background activity right Ictal: EEG-seizure pattern right frontal (F4/F8) MRI: Rt. Hemiatrophy after Rasmussen encephalitis Neuropsychology (restrictions: vigilance & hemi spastic): right-handed, normal school until 10 yrs, then Learning disabled school, now: sheltered workshop, low average IQ, better in figural material, deficits in verbal & nonverbal memory. No psychiatric comorbidity. Wada-Test (right): no speech deficit, partial right-hemispheric representation of memory possible 08/08: Hemispherotomy

1-yr-follow-up: Seizure outcome: none with loss of consciousness (Engel 1) No new neurological deficits EEG: Continuous slowing generalized & in basic activity, Sharp wave, regional right frontal (F8) Neuropsychology: no standardized FU due to new anxiety & psychotic symptoms (wants to die, because of intense fear to die), currently under remission (psychotherapy, risperidone) Discussion: Seizure onset was relatively late 10-14 yrs. All patients had a good seizure outcome (2 Engel 1a, 1 Engel 1). Quality of life has improved in two patients, one initially had severe psychiatric symptoms that are currently under remission.

„EpiSurgDat“: Datenbank der AG für präoperative Epilepsiediagnostik und operative Epilepsietherapie – Stand der Dinge Haag A, Rosenow F Interdisziplinäres Epilepsiezentrum Marburg, Klinik für Neurologie, Universitätsklinikum Gießen und Marburg GmbH, Deutschland Hintergrund: – In der prächirurgischen Epilepsiediagnostik wird eine Vielzahl von Patienten mit unterschiedlichen, z.T. sehr seltenen Epilepsiesyndromen unter Anwendung zahlreicher diagnostischer Verfahren genau phänotypisiert – Die monozentrisch zu erreichenden Fallzahlen an vergleichbaren Gruppen gewährleisten meist keine ausreichende statistische Power für die Beantwortung von Forschungsfragen – Für die Epileptologie gibt es bisher keinen deutschsprachigen/europäischen Forschungsverbund (z.B. analog zu den Kompetenznetzen) Aufgaben: – Aufbau eines multizentrischen Forschungsverbundes der AG für prächirurgische Epilepsiediagnostik und operative Epilepsietherapie gem. e.V. mit einer epilepsiechirurgischen Datenbank („EpiSurgDat“) –  Nutzung der bereits für Studien verwendeten Software SecuTrial® – Anpassung an die spezifischen Bedürfnisse der epileptologischen Forschung – Anträge bei Ethikkommissionen und Datenschutzbeauftragten – Aufbau einer formalen Struktur für multizentrische Forschungsprojekte und einer Rechtsgrundlage zur Nutzung der Daten anderer Zentren Ergebnisse: – „Minimal-Dataset“: Bereits veröffentlichtes multidisziplinär erstelltes Set von Datenfeldern als Grundlage für die Datenbank (Zeitschrift für Epileptologie 4 / 2007) – Web-basierte Datenbank: Steht in der Trainingsversion zur Verfügung, die beispielhafte Eingabe von anonymisierten Daten ist möglich – Ethikantrag & Datenschutz: Antrag bei der Ethikkommission Marburg und Datenschutzkonzept ist vorbereitet – Nutzung der Daten für Forschungsprojekte: Vorschlag zur Verfahrensweise vorbereitet

– „EpiSurgDat“ muss hierbei datenschutzrechtlichen und ethischen Maßgaben Rechnung tragen – Das Verfahren zur Nutzung der Daten anderer Zentren für Forschungsprojekte muss transparent, fair und zeitökonomisch gestaltet sein

Evaluation of health care utilisation among epilepsy patients in the district of Marburg-Biedenkopf Sektorenübergreifende Evaluation der Versorgungsstrukturen von Patienten mit Epilepsie im Landkreis Marburg-Biedenkopf Strzelczyk A, Nickolay T, Bauer S, Haag A, Oertel WH, Reese JP, Reif PS, Rosenow F, Dodel R and Hamer HM Department of Neurology and Interdisciplinary Epilepsy Center, Philipps-University, Marburg, Germany PURPOSE: Epilepsy is a common neurological disorder and imposes a substantial burden on individuals and society as a whole. In contrast there are only few studies on the health care utilisation among epilepsy patients in Germany. We performed a cost of illness study with general practitioners, neurologists and epileptologists in the district of MarburgBiedenkopf. PATIENTS AND METHODS: A “prevalence-based,” cross-sectional convenience sample of consecutive adults with epilepsy irrespective of seizure severity, duration of illness and epilepsy syndrome was evaluated in 2008. The patients were attending general practitioners, neurologists in private practice or a hospital-based epileptologist as an outpatient in the district of Marburg-Biedenkopf. Data on socioeconomic status, course of epilepsy and health care costs were recorded using questionnaires. RESULTS: During a four month period, 472 outpatients with epilepsy (age 46.4 ± 17.1 years, 254 male) were included. 24,4% were seen by a general practitioner, 14,4% by a neurologist in private practice and 61,2% by an epileptologist at a tertiary epilepsy center. Overall 3.6% of the patients were newly diagnosed with epilepsy, 31.2% were in seizure remission, 21,5% had occasional seizures not requiring AED changes, 22,4% had active non-drug resistant epilepsy and 20,7% had a drugresistant epilepsy. The minority of patients treated by general practitioners (39,5%) or neurologists in private practice (45%) had an active epilepsy with at least one seizure in the past 12 months while this applied to 74,3% of patients treated at the epilepsy center. Patients with focal epilepsy were the main group treated at the epilepsy center (75% of all patients treated at the epilepsy center) compared to neurologists (55%) and general practitioners (36%). Patients with idiopathic generalized epilepsy were the main group attending general practitioners (47%) compared to neurologists (30%) and the epilepsy center (13%). The mean three months costs for anticonvulsive treatment amounted to 148 € ± 250 (SD) per patient seen by general practitioners, 159 € ± 249 per patient seen by neurologists and 504 € ± 705 per patient attending the epilepsy center. DISCUSSION: The results confirm that patients with active or drugresistent epilepsy seek treatment at an epilepsy center while the majority of patients treated by general practitioners and neurologists in private practice is in seizure remission.

Diskussion: – Die Epilepsiedatenbank „EpiSurgDat“ soll zu einer Verbesserung der Standardisierung, Kommunikation und Kooperation zwischen Epilepsiezentren führen Zeitschrift für Epileptologie 4 · 2010 

| 277

Abstracts Anfallskontrolle nach Reoperationen Rosenow F Marburg, Deutschland Einführung: Nach einem epilepsiechirurgischen Eingriff bleiben etwa 70% der Patienten mit einer Temporallappenepilepsie (TLE) und 55% der Patienten mit einer extratemporalen Epilepsie (ETE) (unter Medikation) anfallsfrei, d.h. umgekehrt, dass 30-45% der Patienten weiterhin an einer medikamentenrefraktären Epilepsie leiden. Dieses Patientengut erfüllt zwar nach wie vor die Kriterien von Kandidaten für einen epilepsiechirurgischen Eingriff, wird aber nur zu 3-15% erneut operiert. Hierzu wurde eine Literaturübersicht erstellt. Ergebnisse: Die Literatur zu epilepsiechirurgischen Reoperationen besteht aus monozentrischen Fallserien von maximal 70 erfolglos erstoperierten oder erneut operierten Patienten und beschränkt sich i. W. auf die TLE. Nach diesen Studien wurden etwa 20-50% der untersuchten Kollektive mit einem Engel outcome der Klassen III und IV reoperiert. Durchschnittlich 60% (40-81%) erreichten nach der Zweitoperation Engel Class I oder II (Ia nur selten berichtet). In einzelnen Studien fand sich ein kontralateraler Anfallsbeginn in 25% der Patienten die eine Temporallappenresektion erhalten hatten, in anderen Serien war der Anfallsbeginn zu 100% ipsilateral und ging meist von resektionsnahem mesio- oder latero-temporalem Gewebe aus Positive Prädiktoren für dieses outcome (Engel I+II) waren: Tumor als Pathologie, pathologischer MRT-Befund, Fieberkrämpfe in der Vorgeschichte, anterior-temporale Anfallsursprungszone (AUZ), temporo-mesiale und -basale Nachresektion, Nachweis von verbliebenem mesiotemporalem Gewebe, nur ipsilaterale epilepsietypische Potentiale (ETP). Negative Prädiktoren waren: Fokale kortikale Dysplasie oder mesiotemporale Sklerose als Pathologie, traumatisch Hirnschädigung oder Enzephalitis als Ätiologie, posterior-temporale AUZ, multiregionale ETP. Schlussfolgerung: Bemerkenswert ist die geringe Rate an Reoperationen bei Fortbestehen der Selektionskriterien für eine prächirurgische Diagnostik und das Fehlen multizentrischer (prospektiver) Studien.

Epilepsiechirurgische Reoperationen bei Kindern im Epilepsiezentrum Bethel 1990 - 2009 Axer S, Kalbhenn T, Pannek HW, Tuxhorn I, Woermann FG, Ebner A, Polster T Abteilung für präoperative Intensivdiagnotik und Epilepsiechirurgie bei Kindern und Jugendlichen, Klinik Mara, Epilepsiezentrum Bethel, Deutschland Hat eine epilepsiechirurgische Intervention nicht den angestrebten Erfolg erbracht, stellt sich die Aufgabe, die Chancen und Risiken eines erneuten chirurgisch-resektiven Vorgehens gegenüber anderen Optionen abzuwägen. Um diesbezüglich eine genauere Beratung zu ermöglichen, haben wir die epilepsiechirurgischen Reoperationen bei Kindern am Epilepsiezentrum Bethel zwischen 1990 und 2009 analysiert. Unter allen epilepsiechirurgisch behandelten Patienten identifizierten wir 57 Kinder (25 Mädchen, 32 Jungen), bei denen in diesem Zeitraum Reoperationen durchgeführt wurden. Die meisten Kinder wurden 1x reoperiert, wenige Kinder 2x und 1 Kind erhielt 3 Reoperationen. Die Verläufe wurden retrospektiv anhand der klinischen Krankenakten untersucht. Reoperationen wurden in allen Regionen und bei unterschiedlichen Ätiologien durchgeführt. Der Anfallsbeginn der Epilepsien lag zwischen dem ersten Lebenstag und dem 12. Lebensjahr. Das Alter bei erstem epilepsiechirurgischem Eingriff zwischen dem 5. Lebensmonat und dem 12. Lebensjahr.

278 | 

Zeitschrift für Epileptologie 4 · 2010

Im Einzelnen fanden wir folgende lokalisatorische Verteilung der Nachresektionen und der Ätiologien: 1. 17 Frontallappen-Resektionen bei Fokaler Corticaler Dysplasie (FCD), Polymikrogyrie, Tuberöser Sklerose, Z. n. Meningitis, DNT. 2. 14 Temporallappen-Resektionen bei Hippocampussklerose, Gangliogliom, Oligoastrozytom, DNT, FCD, Z. n. SHT. 3. 12 Mehrlappen-Resektionen bei FCD, Tuberöser Sklerose, DNT, Z.n. Encephalitis. 4. 14 Hemisphärische Operationen bei Hemimegalencephalie, Sturge-Weber-Syndrom, Rasmussen-Enzephalitis, FCD, Z. n. Infarkt oder postpartaler Blutung. In allen Gruppen waren zudem Patienten, bei denen sich histologisch keine Läsion nachweisen ließ. Es kam in keinem Fall zu einer unerwarteten irreversiblen Komplikation. Diskutiert wird die Erfolgsrate von Reoperationen unter besonderer Berücksichtigung der für das Kindesalter bedeutsamen Größe des neuropsychologischen Outcomes.