SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page

Presidential Symposium

1 E. Pataraia 1University Hospital of Neurology (Vienna, A)

Goals, methodology: Epilepsy surgery is defined as any neurosurgical intervention with the primary goal to relieve intractable epilepsy. On the other hand essential brain regions like primary motor and sensory cortex as well as brain areas supporting language and memory functions have to be spared to avoid neurological deficits caused by the operation. Thus, the exact localization of the epileptogenic zone and of essential brain regions is crucial for the successful surgical treatment of seizures that can only be accomplished during a thorough presurgical work-up. A new noninvasive brain mapping procedure magnetoencephalography (MEG) was employed during the presurgical evaluation for localization of the epileptogenic zones and for the determination of hemispheric dominance and intrahemispheric localization of linguistic functions in patients with drug-resistant focal epilepsies.

Results: The role of MEG for the localization of the epileptogenic zone in the noninvasive evaluation of patients with focal drug-resistant epilepsies:

We evaluated the sensitivity and selectivity of interictal MEG versus prolonged ictal and interictal scalp video- EEG in order to identify patient groups that would benefit from preoperative MEG testing. One hundred thirteen consecutive patients with medically refractory epilepsy who underwent surgery were included. The epileptogenic region predicted by interictal and ictal Video-EEG (V-EEG) and MEG was defined in relation to the resected area as perfectly overlapping with the resected area, partially overlapping, or nonoverlapping. Using MEG, we were able to localize the resected region in a greater proportion of patients (72.3%) than with noninvasive V-EEG (40%). MEG contributed to the localization of the resected region in 58.8% of the patients with a non-localizing V-EEG study and 72.8% of the patients for whom V-EEG only partially identified the resected zone. Overall, MEG and V-EEG results were equivalent in 32.3% of the cases, and additional localization information was obtained using MEG in 40% of the patients. MEG was most useful for presurgical planning in patients who had either partially or nonlocalizing V-EEG results.

Functional organization of interictal spike complex in medial temporal lobe epilepsies:

Thirty patients with mesial temporal lobe epilepsy (MTLE) using combined MEG and EEG recordings were icluded. Spikes could be recorded in 14 patients (47%) during the 2- to 3-h MEG/EEG recording session. The MEG and EEG spikes were subjected to separate dipole analyses and the spike dipole localizations were superimposed on MRI scans. All spike dipoles could be localized to the temporal lobe with a clear preponderance in the medial region. Based on dipole orientations in MEG, patients could be classified into two groups: patients with anterior medial vertical (AMV) dipoles, suggesting epileptic activity in the mediobasal temporal lobe and patients with anterior medial horizontal (AMH) dipoles, indicating involvement of the temporal pole and the anterior parts of the lateral temporal lobe. Whereas patients with AMV dipoles had strictly unitemporal interictal and ictal EEG changes during prolonged video-EEG monitoring, 50% of patients with AMH dipoles showed evidence of bitemporal affection on interictal and ictal EEG. Nine patients underwent epilepsy surgery so far: all five patients with AMV dipoles became completely seizure-free postoperatively (Class Ia) and two out of four patients with AMH dipoles experienced persistent auras (Class Ib).

Plasticity of the brain mechanisms for receptive language in patients with mesial temporal lobe epilepsy and structural lesions:

We examined brain activation profiles for receptive language function in patients with left hemisphere space occupying lesions and patients with left temporal lobe epilepsy due to mesial temporal sclerosis (MTS) to assess whether cross- and intrahemispheric plasticity for language varied as a function of lesion type or location. We evaluated 44 patients: 21 patients with MTS and 23 lesional patients. All patients underwent preoperative language mapping while performing a word recognition task. The location of the activity sources was subsequently determined by co-registering them with MRIs. The number of clustered, contiguous activity sources located in temporal and inferior parietal regions (excluding sources in somatosensory cortices) was then assessed. Hemispheric lateralization of language-specific magnetic activity was determined as left hemispheric, right hemispheric and bilateral according to relation of the acceptable late activity sources in left and right hemispheres. Patients were classified into two groups based on the location of the cluster(s) of language-specific activity sources within the dominant hemisphere: typical localization of receptive language-specific cortex (if the cluster of activity sources fell within the cortical region that is commonly identified as Wernicke's area) and atypical localization of receptive language-specific cortex (if the cluster of activity sources did not overlap with Wernicke's area).

A higher incidence of atypical language lateralization was noted among patients with MTS compared with lesional patients (43% vs. 13%). The majority of MTS patients with early seizure onset (before 5 years of age) showed atypical language lateralization. In contrast, the precise location of receptive language-specific cortex within the dominant hemisphere was found to be outside of Wernicke's area in 30% of lesional patients and only 14% of MTS patients. There is an increased probability of a partial or total displacement of key components of the brain mechanisms responsible for receptive language function to the nondominant hemisphere in MTS patients. Early onset of seizures was strongly associated with atypical language lateralization. Lesions in the dominant hemisphere tend to result in an intrahemispheric reorganization of linguistic function.

Organization of receptive language-specific cortex before and after left temporal lobectomy: In the present study we documented the reorganization of brain areas mediating receptive language function in patients with left temporal lobe epilepsy after a standard anterior temporal lobe resection. We evaluated which patients were most likely to show a change in the lateralization and localization of the mechanisms supporting receptive language and if such changes were associated with neuropsychological function. The results of preoperative Wada-testing and pre- and post-operative neuropsychological testing and MEG language mapping were compared. Patients with atypical (bilateral) hemispheric dominance pre-operatively were significantly more likely than patients with (typical) left-hemisphere dominance to show evidence of increased right hemisphere participation in language functions after surgery. Patients with left hemispheric dominance preoperatively were more likely to show intra-hemispheric changes involving a slight inferior shift of the putative location of Wernicke's area. Patients with bilateral representation tended to perform worse on neuropsychological test measures obtained both pre- and postoperatively. Interhemispheric functional reorganization of language-specific areas may occur in patients undergoing left anterior temporal lobectomy. Intrahemispheric reorganization may take place even when the resection does not directly impinge upon Wernicke's area.

Conclusions: Combined MEG/EEG dipole modeling can identify subcompartments of the temporal lobe involved in epileptic activity and may be helpful to differentiate between subtypes of mesial temporal lobe epilepsy noninvasively.

MEG is most useful for presurgical planning in patients who have either partially or nonlocalizing V-EEG results in the noninvasive evaluation phase. We predict the replacement of the more invasive procedure with MEG in the near future for temporal lobe epilepsies, subsequent to the optimization of the conditions under which preoperative MEG is performed.

MEG can be especially helpful in the localization of language-critical cortex in sites other than those expected within the dominant hemisphere. Our findings also suggest that not only structural elements, but also functional factors have an effect on receptive language organization in the brain. Factors influencing atypical language lateralization have theoretical importance for understanding the organization and reorganization of higher cognitive functions, as well as practical implications, especially in brain surgery and neurological rehabilitation.

MEG is a useful method in clinical practice, as it has the capacity to provide reliable images of the working brain of individual subjects, and it is capable of capturing relevant aspects of brain activation by reflecting the actual participation of a particular area in the function under investigation. Finally, it is capable of capturing both the spatial as well as the temporal features of that activation.

Abstract

  1. Top of page

Session: Pharmacogenetics

1 F. Zimprich 1University Hospital of Neurology (Vienna, A)

Genetic variations in drug targets are thought to influence an individual's response to pharmacotherapy. This is considered to be particularly important in the field of epilepsy, where the response to treatment varies considerably between patients with otherwise identical clinical phenotypes. Most of the licensed antiepileptic drugs (AEDs) target either:

  • i) 
    voltage-gated ion channels (Na+, Ca2+, K+),
  • ii) 
    components of the inhibitory GABAergic synaptic machinery or they
  • iii) 
    modulate glutamate mediated excitatory neurotransmission.

Functional variations in the genes coding for these proteins are considered prime candidates in the search for factors that explain pharmacodynamic variations (i.e., variations in the interaction between AEDs and their targets at the molecular level).

In this review I will focus on some of the advances made in this field in recent years. The best evidence so far exists for genetic variations in voltage-gated Na+-channels. Voltage-gated Na+-channels occupy a central position in the pathogenesis of epilepsies. They are responsible for the rapid rising phase of the neuronal action potential. Several AEDs (such as carbamazepine) block Na+-channels in a use-dependent manner and therefore prevent the high-frequency firing of these channels. In an elegant study Remy et al. (1) could show that the electrophysiological characteristics of sodium currents (tested on hippocampal surgical specimens) differ between patients responsive to carbamazepine treatment and those who were resistant to this AED. In pharmacoresistant patients the use-dependent block of carbamazepine on Na+-channels was completely lost whereas it was clearly detectable in pharmacoresponsive patients. The molecular changes underlying this important phenomenon are as yet unclear, but are likely to involve (post-)transcriptional changes in Na+-channels.

It is known from animal models that i) seizures may induce a dynamic shift in the alternative splicing of Na+-channels and that ii) an alternative splicing of sodium channel genes can result in marked functional differences of the finally displayed sodium current (2-4). Tate et al. (5,6) have recently investigated a polymorphism in the human SCN1A gene (a sodium-channel gene) in the consensus 5′ splice donor site of a highly conserved, alternatively spliced exon. By disrupting this consensus sequence the major allele (A) was reported to reduce the expression of the “fetal” exon (5N) relative to its “adult” version (5A) in humans. (Both exons are normally expressed in the adult brain at different proportions). In rodents seizures seem to induce an up-regulation of channels containing the 5N-exon. This up-regulation of the 5N-exon was also seen in patients with refractory epilepsy, but only in those who were homozygote for the nondisrupting minor G-allele. Interestingly, patients with the AA-genotypes were found to have been in need for significantly higher doses of AEDs that target Na+-channels (carbamazepine and phenytoin) in comparison to patients with the other genotypes (5,6). The implication of these findings is that this is a functional variant in a drug target gene that determines how efficiently the respective drug works.

Our own group has further investigated this polymorphism. In contrast to Tate et al. we studied patients with temporal lobe epilepsy who were largely pharmacoresistant. (In so far our study can not be considered an exact replication study.) In our cohort we could not find the reported association between the genotype and the dose of clinically prescribed carbamazepine. However we did find a significant overrepresentation of the (disrupting) A-alleles among patients in comparison to controls. Further studies are certainly needed to clarify the relevance of the polymorphism.

We still have very little direct evidence on how genetic variations modify other drug targets. However there are interesting circumstantial data for several other molecules of the above mentioned pathways. GABAA-receptors, which mediate most of the inhibition in the CNS and are the target for several AEDs such as benzodiazepines could be cited as one example. It has been shown both in experimental models and in humans that seizures cause changes in the subunit composition of GABAA-receptors. These subunit changes can go along with a reduced efficacy of AEDs acting via GABAA receptors (7, reviewed by ref. 8). In the light of these results it will be particularly interesting to investigate genetics factors that quantitatively influence the expression of individual GABAA-receptor subunits. Polymorphisms in cis-regulatory gene regions, although difficult to identify, would be obvious candidates in the search for such genetic modulators.

  • References:

  • 1
    Remy, Ann Neurol 2003, 53:469
  • 2
    Song, J Biol Chem 2004, 279:32554
  • 3
    Copley, Trends Genet 2004, 20:171
  • 4
    Gastaldi, BRMBR 1997, 44:179
  • 5
    Tate, PNAS 2005, 102:5507
  • 6
    Tate, Pharmacogenet Genomics 2006, 16:721
  • 7
    Brooks-Kayal, Nat Med. 1998, 4:1166
  • 8
    Schmidt & Loscher 2005, 46:858

1 U. Klotz 1Dr. Margarete Fischer-Bosch-Institute of Clinical Pharmacology (Stuttgart, D)

Different factors including pharmacogenetics contribute to the interindividual variability in drug response. Several antiepileptic drugs (AEDs) are metabolized by the cytochrome P-450 (CYP) superfamily. Some of these CYPs exist in form of genetic (allelic) variants (e.g. CYP2C9 and CYP2C19) which might affect the plasma concentrations or drug exposure (AUC) of AEDs. Therefore such polymorphisms are of clinical interest.

With mephenytoin defects in its metabolism could be attributed to more than 10 mutated alleles (designated as *2, *3 and others) of the gene expressing CYP2C19. Consequently poor (PM) and extensive (EM) metabolizers could be differentiated whose frequencies vary among ethnic populations. CYP2C19 contributes to the metabolism of diazepam and phenytoin, the latter drug representing also a substrate of CYP2C9 with its predominant variants defined as *2 and *3. For both AEDs there is maximally a twofold difference in the hepatic elimination or AUC between the extremes of EM and PM which in the case of phenytoin would suggest a dosage reduction only for patients who are carriers of mutated alleles of both, CYP2C19 and CYP2C9, a subgroup which is very rare (about 1%) within Caucasians but more frequent (about 10%) in Asians. The minor contribution of CYP2C19 to the metabolism of phenobarbital can be neglected. In rare cases valproic acid (VPA) can be metabolized to the reactive (hepatotoxic) metabolite 4-ene-VPA. It is not yet clear whether genetic variants of the involved enzyme (CYP2C9) are responsible for this problem. Likewise, the active metabolite of carbamazepine (CBZ) CBZ-10,11-epoxide is transformed by the microsomal epoxide hydrolase, an enzyme which is also highly polymorphic, but the pharmacokinetic and clinical consequences still need to be evaluated (1).

As for old and especially new AEDs the pharmacogenetic influence on their metabolism is not very striking it is not surprising that there are no treatment guidelines taking pharmacogenetic data into account. Therefore the traditional and validated therapeutic drug monitoring approach representing a direct “phenotype” assessment still remains the method of choice when an individualized dosing regimen is anticipated.

Reference: 1. Klotz U. The role of pharmacogenetics in the metabolism of anticonvulsants-pharmacokinetic and therapeutic implications. Clin Pharmacokinet, in press (2007).

1 W. Löscher 1University of Veterinary Medicine Hannover (Hannover, D)

Drug transporters are increasingly recognized to be important to drug disposition and response. The efflux transporter P-glycoprotein (Pgp), the encoded product of the MDR1 (ABCB1) gene, was first extensively studied for its role in mediating the multidrug resistance (MDR) phenotype associated with certain cancers. However, this transporter is constitutively expressed by many normal tissues, including brain capillary endothelial cells that form the blood–brain barrier (BBB). At the BBB, Pgp acts as a protective barrier to restrict the penetration of lipophilic toxins into the brain. At normal, physiological expression levels of Pgp at the BBB, it does not significantly restrict the brain uptake of antiepileptic drugs (AEDs). However, enhanced expression or functionality of Pgp may restrict the brain entry of various AEDs, thus resulting in sub-therapeutic concentrations at their brain targets. Tishler et al. (1995) were the first to report that MDR1 and Pgp expression are significantly increased at the BBB of patients with AED-resistant epilepsy. Since then, this observation has been reproduced and substantiated by many studies and has led to the “multidrug transporter hypothesis” of refractory epilepsy.

Why is the expression of Pgp increased at the BBB of patients with AED-refractory epilepsy? In animal models of epilepsy, it was shown that seizures lead to a transient, regionally restricted overexpression of Pgp, i.e., an acquired (induced) upregulation of this transporter. This could explain that high seizure frequency before onset of treatment is often associated with AED resistance. It would, however, not explain that resistance may occur de novo, i.e., before the first seizure or after one or a few short seizures. De novo drug resistance suggests that constitutive (intrinsic) factors cause resistance. Ulrich Brinkmann's group in Bernried described in 2000 a silent, exonic polymorphism of MDR1 that is associated with increased expression and functionality of Pgp in the human intestinal tract. This single nucleotide polymorphism (SNP) in exon 26 (C3435T) was proposed to affect the absorption and tissue concentrations of numerous drugs in C-allele-carrying individuals. This prompted a subsequent genetic association study in which Brinkmann, Sisodiya and colleagues genotyped the C3435T polymorphism in 315 patients with epilepsy, demonstrating that patients with drug-resistant epilepsy were more likely to have the CC genotype than the TT genotype (Siddiqui et al., 2003). Although the C3435T polymorphism is silent, i.e., it does not change the encoded amino acid, there is a strong linkage disequilibrium (i.e., a nonrandom association) with SNPs at exon 21 (G2677T/A) and exon 12 (C1236T), which may explain the alterations in Pgp expression and functionality associated with the C2435T polymorphism. In fact, subsequent studies found a three SNP haplotide including the C3435T variant along with G2677T and C1236T to be more significantly associated with multidrug resistance than the C3435T polymorphism alone.

The finding of Siddiqui et al. (2003) led to the suggestion that drug resistance in epilepsy might be genetically predicted and suggested new therapeutic avenues. However, several further case-control studies and a more recent prospective cohort study failed to replicate any association with the C3435T polymorphism or the three-SNP haplotype. There are various potential explanations for these discordant results, including inconsistent phenotype definition, inclusion of various AEDs that might not be Pgp substrates, overlap in substrate specificity between Pgp and other drug efflux transporters, population substructure, ethnic differences, genotyping error, inadequate power or potential confounding by comorbidity and comedication. However, taken overall, the available genetic studies do not support that the MDR1 C3435T variant or other MDR1 polymorphisms are an important determinant of multidrug resistance in epilepsy. Similarly, in animal experiments, in which confounding factors can be much more minimized than in population-based studies, we did not find any association between Mdr1 polymorphisms and AED resistance (Baars et al., 2006). This does not mean that overexpression of Pgp is not an underlying mechanism of AED resistance, because acquired (induced) Pgp overexpression, e.g., in response to seizures, may occur independently of any polymorphism in the MDR1 gene. Clinical trials with Pgp inhibitors are needed for evaluating the potential role of Pgp in drug-resistant epilepsy. Furthermore, the role of other efflux transporters (such as multidrug resistance-associated proteins; MRPs) in drug resistance needs to be explored in more detail.

References:

Baars C, Löscher W, Leeb T, Becker A, Potschka H. (2006) Eur J Pharmacol 550:54-61. Baars C, Löscher W, Leeb T, Becker A, Potschka H. (2006) Eur J Pharmacol 550:54-61. Hoffmeyer S et al. (2000) Proc Natl Acad Sci U S A 97:3473-3478. Siddiqui A et al. (2003) N Engl J Med 348:1442-1448. Tishler DM et al. (1995) Epilepsia 36:1-6.

1 D. Schmidt 11Epilepsy Research Group (Berlin, D)

This review discusses the influence of genetic factors on dose-related and idiosyncratic adverse reactions to AEDs. Genetic influences on AED pharmacokinetics may lead to adverse reactions through AED accumulation and other mechanisms. As an example, a decreased enzyme capacity to metabolize phenytoin or mephenytoin resulting in diplopia, ataxia and other intoxication symptoms has been described in relation to the CYP2C9/CYP2C19 genotype in some patients (Herrlin et al., 1998). Genetic influences may also play a pathogenic role in idiosyncratic adverse reactions of AEDs. Although pharmacogenetics are suspected to play a role in many idiosyncratic side effects such as neural tube defects in women using valproate (Duncan et al., 2001), the delayed AED hypersensitivity syndrome reactions (HSS) has received the most attention recently and will be briefly discussed here. HHS are one of the most feared idiosyncratic drug reactions and are most common with exposure to AEDs. HSS is associated with chemotoxic and T-cell mediated inflammatory injuries in barrier tissue systems that contain cytochrome oxidases (e.g., skin, mucosa, liver and lungs). HSS can be viewed as a disturbance in the defense system against bioactive foreign molecules, also called xenobiotics. The putative mechanism of AED HSS involves genetic susceptibility to T-cell mediated responses to bioactive molecules with accumulation of AEDs and oxidized metabolites causing major histocompatibility complex (MHC) and non-MHC-dependent clonal activation of T-cells and subsequent cytokine/chemokine production in T-cells, keratinocytes and other target cells as recently reviewed by Krauss (2006). Since MHC proteins expressed by human chromosome 6 help determine whether molecules such as AEDs are presented as foreign antigens, it may be possible to develop a MHC library to determine individual susceptibility to HSS. An example is the study of Han Chinese with antigen HLA-B*1502, placing patients at risk for HSS when exposed to carbamazepine (Chung et al., 2004). However, this finding was not confirmed in caucasian patients with HSS (Alfirevic et al., 2006b). Serious skin hypersensitivity to carbamazepine has also been associated with a polymorphism at position 308 of the TNFalpha promotor region gene and with variants of the HLA gene allels DR3 and Q2 (Pirmohamed et al., 2001). The same group showed that serious carbamazepine-induced hypersensitivity reactions are associated with the heatshock protein (HSP70) gene cluster in the MHC class III region (Alfirevic et al., 2006a). Although association studies have inherent weaknesses and have not provided unifying conclusions, it is reasonable to expect that once libraries of MHC and enzyme-subtypes associated with serious drug reactions have been established, it may be possible to develop individual panels indicating susceptibility to HSS and other serious drug reactions of AEDs. However, there is a lack of good correspondence between results from different laboratories, and more recent findings are awaiting attempts at confirmation. Thus, there are currently no AED treatment guidelines that are informed by pharmacogenetic data (Ferraro et al., 2006). From a pragmatic clinical perspective, the major current treatment of serious idiosyncratic reactions is immediate removal of the offending AED. In the future, genetically determined dose-related and idiosyncratic adverse reactions may be avoided by single drug treatment with modern AEDs such as gabapentin, pregabalin or levetiracetam which are neither metabolised nor involved in idiosyncratic reactions.

References:

Alfirevic A. et al. (2006) Serious carbamazepine-induced hypersensitivity reactions associated with the HSP70 gene cluster. Pharmacogenetics and pharmacogenomics. 16:287-296. Alfirevic A, Jorgensen AL, Williamson PR, Chadwick DW, Park BK, Pirmohamed M. (2006b) HLA-B locus in Caucasian patients with carbamazepine hypersensitivity. Pharmacogenomics. 7:813-818. Chung WH et al. (2004) Medical genetics: a marker for Stevens-Johnson syndrome. Nature. 428: 486. Duncan S. et al. (2001) Repeated neural tube defects and valproate monotherpay suggest a pharmacogenetic abnormality. Epilepsia 42:750-751. Ferraro TN, Dlugos DJ, Buono RJ. (2006) Challenges and opportunities in the application of pharmacogenetics to antiepileptic drug therapy. Pharmacogenomics. 7(1):89-103. Herrlin K et al. (1998) Bantu Tanzanians have a decreased capacity to metabolize omeprazole and mephenytoin in relation to their CYP2C19 genotype. Clin Pharmacol Ther 64:391-401. Krauss G. (2006) Current understanding of delayed anticonvulsant hypersensitivity reactions. Epilepsy Currents 6:33-37. Pirmohamed M. et al. (2001) TNFalpha promotor region gene polymorphisms in carbmazepine hypersensitive patients. Neurology 56:890-896.

Abstract

  1. Top of page

Session: Epilepsy – Emerging models, mechanisms and concepts

1 A. Lüthi , 1 L. Cueni , 1 M. Canepari , 2 R. Lujan , 3 M. Watanabe , 4 C.T. Bond , 4 J.P. Adelman ( 1University Basel (Basel, CH) ; 2 University Castilla-La Mancha (Albacete, E);3 Hokkaido University (Sapporo, JP);4 Vollum Institute (Portland, USA)

Oscillatory burst firing in thalamic neurons, generated by the low-threshold T-type Ca2+ current, underlies electroencephalographic waves characterizing sleep states. A common feature in a number of generalized absence epilepsy models is an augmented T-current, accompanied by an enhanced expression of the corresponding Cav3 Ca2+ channel subunits. In spite of the well-established significance of T-currents in the normal and epileptic thalamocortical system, we lack elementary quantitative information about T-current-dependent Ca2+ signaling in the dynamics and exacerbation of oscillatory burst discharges. Using electrophysiological recordings and high-speed Ca2+ imaging in brain slices, as well as immunohistochemical and computational techniques, we studied oscillatory burst discharges in neurons of a sleep rhythm pacemaker, the nucleus reticularis thalami, in normal, 3-weeks-old C57Bl/6 mice. Neuronal oscillations in the nRt exhibit complex burst patterns, which are accompanied by a gradual dampening of bursting or abrupt transitions between bursting and silent states. T-type Ca2+ currents rapidly activated Ca2+-gated SK2-type K+ channels, which were localized exclusively on dendrites, effectively terminated low-threshold Ca2+ bursts, and generated an after-hyperpolarization that supported activation of the next low-threshold burst. T-currents dominated the burst-induced [Ca2+]i increases, and reached ∼0.7 μM throughout proximal dendrites. In successive oscillatory cycles, [Ca2+]i elevations strongly decreased, in a manner consistent with cumulative T-channel inactivation. Moreover, sarco/endoplasmic Ca2+ ATPases (SERCAs) selectively sequestered Ca2+ ions entering through T-channels. Both processes limited SK channel gating and disfavored the generation of an additional low-threshold burst. In a computational model of nRt cells, incorporating cumulative inactivation and intracellular Ca2+ sequestration was sufficient to reproduce oscillatory dampening. In conclusion, we identified a biophysical machinery by which T-currents, through dual Ca2+ signaling, shape the dynamics of oscillatory activity, and a role for endoplasmic Ca2+ sequestration in controlling the timing of low-threshold burst discharges. Altogether, this study provides the first biophysical analysis of the [Ca2+]i transients and Ca2+ signaling during a natural, sleep-related thalamic oscillation. Additionally, it presents novel candidate mechanisms that, when dysfunctional, may promote the perversion of thalamic oscillations into hypersynchronous activity typical for generalized epilepsies.

1 Y. Yaari , 1 H. Su , 1 C. Yue , 2 J. Pitsch , 2 D. Sochivko , 2 S. Schoch , 2 H. Beck , 1 A. Becker ( 1Hebrew University–Hadassah School of Medicine (Jerusalem, IL) ; 2 University of Bonn (Bonn, D)

A single episode of status epilepticus (SE) induced in rodents by the convulsant pilocarpine, produces, after a latent period of two or more weeks, a chronic epileptic condition. In its neuropathological, electroencephalographic and behavioral features, this condition is reminiscent of human temporal lobe epilepsy (TLE). During the latent period of epileptogenesis (within 3 days after SE), most rat CA1 pyramidal cells that normally fire in a regular pattern acquire low-threshold bursting behavior, generating high-frequency clusters of 3-5 spikes as their minimal response to depolarizing stimuli. A Ni2+-sensitive T-type Ca2+ current (ICaT), shown to be upregulated after SE, plays a critical role in intrinsic burst generation in most cases. The ICaT driving low-threshold bursting is located in the apical dendrites and is recruited by backpropagating somatic spikes. Of the three identified alpha1 subunits generating ICaT (CaV3.1, CaV3.2 and CaV3.3), transcription of only CaV3.2 (alpha1H) is upregulated in SE-experienced CA1 pyramidal cells, implicating this channel subunit in the de novo emergence of intrinsic bursting. Intriguingly, knockout mice lacking CaV3.2 develop acute SE after pilocarpine injection just like their wild type counterparts, but are significantly less likely to undergo epileptogenesis and acquire TLE. Together, these findings identify CaV3.2 as a potential target for pharmacological and molecular treatments aimed at halting epileptogenesis during its latent phase.

1 M. De Curtis 1National Neurological Institute “Carlo Besta” (Milan, I)

Hippocampal-entorhinal-hippocampal loop is considered important in the development of epileptiform ictal activity in the temportal lobe. Peculiar pattern of fast activity at epileptiform discharges onset in the hippocampus and medial EC was observed in isolated guinea pig brain preparation (Uva et al., Epilepsia 2005; 46:1914.) and is observed in human intracerebral recordings from the seizure-onset area. We demonstrated that hippocampus-mediated activation of medial EC superficial neurons is characterized by direct feedforward inhibition (Gnatkovsky and de Curtis J. Neurosci 2006; 26(3):873). Focal epileptiform discharges in the EC-H region of the isolated guinea big brain were induced by 3-min arterial perfusion of the GABAa receptor antagonist, bicuculline methiodide (50 μM). Superficial neurons in the medial EC were recorded with biocityne-filled sharp electrode simultaneously with field potentials from both medial EC and hippocampus during interictal and ictal epileptiform events. At ictal discharge onset a sequence of brief IPSPs correlated with fast (20–30 Hz) extracellular activity was observed in medial EC neurons. These IPSPs were coupled to a slow depolarizing shift of the membrane potential, followed by action potential firing with further transformation into burst pattern. Slow membrane depolarization correlated with negative shift of field potential. We demonstrate that superficial and deep layer cells of the medial EC do not fire during the generation of fast activity at ictal onset, whereas putative interneurons generate high frequency firing. IPSPs into superficial cells may be generated by fast inputs mediated by the hippocampus area CA1. These findings suggest that local mechanisms of fast oscillations at seizure onset are maintained by interneruronal activity.

1 H. Beck 1University Clinic of Epileptology (Bonn, D)

Virtually all CNS synapses display the potential for activity-dependent long-term potentiation (LTP) and/or depression (LTD). Intriguingly, the potential to exhibit LTP or LTD at many central synapses itself is powerfully modulated by prior synaptic activity. This higher-order form of plasticity has been termed metaplasticity. Metaplasticity appears to be invoked both following abnormal activity such as seizures, as well as following more physiological activity patterns. This raises the question whether physiological and pathological metaplasticity share the same cellular mechanisms or not.

We first examined hippocampal metaplasticity following priming stimulation at low (10 Hz) frequencies. We find that inhibitory autophosphorylation of Ca2+-calmodulin-dependent kinase II (CaMKII) is required for hippocampal metaplasticity at the lateral perforant path-dentate granule cell synapse. Brief 10 Hz priming, which does not affect basal synaptic transmission, caused a dramatic, pathway-specific, and long-lasting (up to 18 hours) reduction in subsequently evoked LTP at lateral perforant path synapses. In contrast, LTD was unaffected by priming. The induction of lateral perforant path metaplasticity required the activation of NMDA receptors during priming. In addition, metaplasticity was absent in knockin mice expressing alpha CaMKII that cannot undergo inhibitory phosphorylation, indicating that inhibitory autophosphorylation of alpha CaMKII at threonines 305/306 is required for metaplasticity. Metaplasticity was not observed in the medial perforant pathway, consistent with the observation that CaMKII activity was not required for the induction of LTP at this synapse. Thus, modulation of alpha CaMKII activity via autophosphorylation at Thr305/Thr306 is a key mechanism for physiological metaplasticity that may be of importance in the integration of temporally separated episodes of activity.

We went on to examine changes in perforant path synaptic plasticity in the perforant path in chronic experimental epilepsy. We find a similar loss of LTP in the lateral but not the medial perforant path, however, the cellular mechanisms for this loss seem to be distinct. Available evidence indicates that the loss of LTP in chronic epilepsy is caused by a deficit in the glioneuronal cotransmitter D-serine. These results suggest that physiological and pathological forms of metaplasticity may not rely on identical mechanisms. These findings are of potential clinical relevance because they suggest pharmacological treatment avenues that selectively target pathological metaplasticity.

Abstract

  1. Top of page

Session: Status epilepticus

1 R.A. Sälke-Kellermann 1Swiss Epilepsy Centre (Zurich, CH)

Introduction: Status epilepticus (SE) in young children has several special characteristics. The etiology, clinical symptoms, and prognosis of SE present distinctive features depending on the stage of brain development. SE is divided into two categories: convulsive SE and nonconvulsive SE (NCSE). This presentation will give an overview about NCSE in childhood, and it will emphasize its devastating effects on cognitive and neurological development in children leading to “epileptic encephalopathies” (EEs), a new category in the Proposed Diagnostic Scheme of the ILAE Task Force on Classification and Terminology 2001. EEs are severe conditions, in which very frequent or long-lasting seizures as well as subcontinuous interictal epileptiform discharges are, partly or entirely, causative of cognitive and neurologic deterioration.

Classification of NCSE: Up till now, there is no valid classification of NCSE in children. On the basis of phenomenology we can distinguish between generalized NCSE (GNCSE), focal NCSE (FNCSE), and SE, undetermined whether focal or generalized. This classification may be correct in epilepsies in adulthood, but it does not sufficiently reflect the heterogeneous character of NCSE and its role in epileptic syndromes of childhood. The GNCSE includes the following categories which still are controversial: 1. Absence status (AS) with three separate types: typical absence status, atypical absence status, and minor epileptic status; 2. Myoclonic status; 3. Atonic SE. GNCSE occurs in the following syndromes: early myoclonic encephalopathy, Ohtahara syndrome, West syndrome, Dravet syndrome, and Lennox–Gastaut syndrome. Additionally, most authors include the myoclonic astatic epilepsy in this group. FNCSE includes SE with simple or complex focal seizures, for example Rasmussen's encephalitis. The last group, SE, undetermined whether focal or generalised, encompasses the following syndromes: the syndrome of electrical status epilepticus in slow sleep (ESES), Landau–Kleffner syndrome, atypical benign partial epilepsy (“Pseudo-Lennox syndrome”), and obtundation status in severe myoclonic epilepsy in infancy (Dravet syndrome).

Etiology: The etiology of malignant encephalopathies remains unknown. There probably are multiple factors leading to intractable epilepsy as well as to severe global retardation. Structural, genetic, and metabolic causes can play an important role. Other factors are frequent seizures and continuous or nearly continuous epileptiform discharges as shown in the electroencephalographic (EEG) recordings. However, mechanisms or processes leading to encephalopathy only can be assumed. There is some experimental evidence that GABA A receptors may be involved in the maintenance and termination of absence seizures and, probably, of ASE, too. After an FNCSE, higher concentration of the enzyme Enolase was estimated in blood and cerebrospinal fluid. Some reports provided the idea that the plasticity of the immature brain may remain unutilised, or gets unfavourably misdirected. Elaboration of abnormal synaptic connections may be detrimental for development.

Clinical manifestations of NCSE: GNCSE: Clinical symptoms are subtle such as clouding of consciousness, reduced reactivity, and ataxia. EEG recordings show continuous or nearly continuous slow spike-and-wave (SW) activity, or rhythmic recurrent epileptiform discharges bilaterally or generalized. GNCSE has a higher incidence than FNCSE. Several seizure types can occur during a GNCSE in children with EEs: atypical absences, atonic seizures, astatic seizures, epileptic spasms, myoclonic-atonic seizures, bilateral myoclonic jerks, and negative myoclonus. The duration may last hours or even days. Two syndromes, the Dravet syndrome and the myoclonic astatic epilepsy will be presented in detail.

FNCSE: In Rasmussen's encephalitis an infectious aetiology with perivascular lymphocyte infiltrations is proposed. The children aged from 1 to 15 years suffer from focal seizures. The disease is progressive with intractable seizures which evolve into FNCSE. A mental regression will gradually become obvious.

SE, undetermined whether focal or generalised: Children suffering from ESES or Landau–Kleffner syndrome have motor deficits, negative myoclonus, and orofacial dyspraxia. Their previous normal cognitive development deteriorates during the acute phase of the illness. Seizures occur only seldom. The cause of the regressive development is the electrical status of continuos SW discharges during sleep. The epilepsy will recede during puberty, but cognitive deficits may recover only partly.

Prognosis: The prognosis with NCSE and EEs is mostly poor concerning both epilepsy and cognitive development. Some reports concerning West syndrome or myoclonic astatic epilepsy suggest a prognosis positively modified by successful therapy. Seizure control seems to be essential for a favourable developmental outcome.

1 F. Rosenow , 1 H. Hamer , 1 S. Knake 1University Hospital Giessen and Marburg GmbH (Marburg, D)

The dilemma: Status epilepticus (SE) is a life threatening neurological emergency, which occurs about 14.000 times/year in Germany and it is associated with 1.300 deaths (1). Therefore, SE is a relevant medical problem. On the other hand there are several reasons why there are only 2 randomized studies on the treatment of SE so far:

  • - 
    Considering a case fatality rate of 10% studies in SE will lead to unwanted entries in the drug information brochure even of the p.o. – preparations of a drug.
  • - 
    The lack of the possibility to obtain informed consent from the patient in SE presents an ethical problem which requires certain measures.
  • - 
    14.000 indications/year for a one time emergency therapy is a very small market.
  • - 
    Most medications are approved for i.v. therapy only after the patent protection has expired, therefore the financial interest of a company in approval for status therapy is low.

For the reasons mentioned there are only 2 studies of evidence class 1. The first of the studies considers the out of hospital treatment by medical assistants. The authors found that a benzodiazepine (diazepam or lorazepam) was significantly superior to placebo in obtaining control of SE at admission to the hospital (2). The second study regarding the in house treatment of SE found that control of SE within 20 minutes occurs significantly less frequently after phenytoin (18 mg/kg) monotherapy (3) as compared to 3 others standard therapies:

  • 1) 
    Lorazepam (0,1 mg/kg)
  • 2) 
    Phenobarbitol (15 mg/kg) and
  • 3) 
    Diazepam (0,15 mg/kg) followed by phenytoin (18 mg/kg).

Considering this evidence basis all further studies will have to use one of these standard therapies (4).

Ways out of the dilemma:

  • 1) 
    A recommendable study design for the future would be to apply one of the 3 standard therapies (e.g., lorazepam 0,1 mg/kg) combined with either placebo or a novel drug (e.g., valproic acid 30 mg/kg or levetiracetam 30 mg/kg). Primary endpoint would be the control of Status epilepticus within 30 minutes after initiation of therapy.
  • 2) 
    The Ethics committee of the Philipps-University Marburg for instance would allow inclusion of patients in Status epilepticus considering that these patients themselves can profit from a evidence based improvement of treatment because recurrence of SE does occur in about 10 – 20% of the cases. The condition for the approval of the Ethic committee would, however, be the inclusion of the sufficient number of cases to allow a significant result. An adaptive study design would be feasible, e.g., including some 80 patients into a pilot study. If this pilot study showed a clinically relevant effect (e.g., 10% higher control rate) one would continue to include the patients based on case number estimated from the data of the initial 80 patients (5).
  • 3) 
    With Lacosamide which will be approved in early 2008 we have for the first time a substance that will obtain i.v. approval very shortly after the general approval for use as an antiepileptic drug. This means that there is a higher financial interest in developing this drug for the Status indication.
  • 4) 
    All medical doctors concerned with the care of epilepsy patients should demand the conduction of controlled randomized trials of i.v. antiepileptic medications. The use of i.v. drugs not approved for SE should be minimized. One should always consider that Status carries a 10% case letality rate and that it may be a legal risk to treat a patient with a drug not approved for this indication.
  • 5) 
    Should the pharmaceutical industry still not be willing to conduct a regulatory trial it would be possible to ask the German Research Foundation (DFG) to sponsor such a study.

References:

1. Knake S et al. for the Status Epilepticus Study Group Hessen (SESGH). (2001) Incidence of Status Epilepticus in adults in Germany – a prospective, population based study Epilepsia 42:7148. 2. Alldredge BK et al. (2001) A comparison of lorazepam, diazepam, and placebo for the treatment of out-of-hospital status epilepticus. N Engl J Med 345:6317. 3. Treiman DM et al. (1998) A comparison of four treatments for generalized convulsive status epilepticus. Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J Med 339:7928. 4. Rosenow F et al. (2002) Recent developments in treatment of status epilepticus – a review. Epileptic Disorders 4(Suppl.2):4151. 5. Schafer H et al. (2006) An overview of statistical approaches for adaptive designs and design modifications. Biom J. 48:50720. 6. Krämer G et al. (2005) Stellenwert intravenöser Valproinsäure in der Therapie des generalisierten tonisch-klonischen Status epilepticus. Aktuelle Neurologie 32:26374.

Abstract

  1. Top of page

Session: Health economic aspects of epilepsy

1 H. Hamer 1Phlipps-University Marburg (Marburg, D)

Purpose: There are only few studies on the costs of epilepsy in Germany. Therefore, we performed a pilot study to estimate the direct and indirect costs of refractory epilepsy in a German epilepsy center. In addition, a systematic review of published cost of illness studies on epilepsy was done.

Methods: A ‘prevalence-based’, cross-sectional convenience sample of adults attending the outpatient clinic of our tertiary epilepsy center was evaluated. Seizure-free patients and patients presenting with their first seizure were excluded. Direct and indirect costs were prospectively recorded over a three months period using questionnaires and a patient diary. Cost driving factors were identified.

Results: 101 patients were included (40.7 ± 15.2 years; disease duration: 18.1 ± 15.3 years; 6 patients had focal epilepsy with simple partial seizures only, 28 with complex partial seizures, 43 with secondarily generalized tonic–clonic seizures; 20 had idiopathic generalized epilepsy with generalized tonic–clonic seizures). The total costs of epilepsy per patient were in average € 2610 ± 4200 over the three months period. Direct cost contributed 39% to the total costs. Costs of anticonvulsant medication were the main contributor to the direct costs while indirect costs were caused mainly by losses due to early retirement. Cost driving factors included higher seizure frequency, longer disease duration, ictal falls and situationally inappropriate complex behavior during or after the seizure.

Conclusions: Indirect costs were higher than direct costs in adult patients with active epilepsy attending a German epilepsy center. Medication contributed the most to the direct costs and early retirement was the main factor for the indirect costs. The costs of poorly controlled epilepsy in this German study were above average of the European costs of epilepsy. A comparison of the different cost of illness studies in epilepsy is hampered by subbstantial limitations of the studies in respect to patient selection criteria, diagnosis, and methods. Therefore there is an urgent need for studies which evaluate direct and indirect costs in a comparable fashion.

Abstract

  1. Top of page

Session: Epileptic seizures as symptom of rare genetically determined syndromes

1 G. Uyanik , 1 U. Hehr , 1 J. Winkler 1University of Regensburg (Regensburg, D)

Cortical malformations are observed with an increased frequency related to symptomatic epilepsies not only in children but in particular also in adult patients. The development of the human cortical architecture is a complex process requiring fine tuning of many intervening elements like i) proliferation of neural precursor cells, ii) migration of the neuroblasts to their final destination, and iii) differentiation and organization of the neuroblasts to mature neurons and integration into neuronal circuits. Numerous genes are important for the neuronal migration resulting in different forms of cortical malformations. The migratory process consists of the initiation of the migration, cell soma translocation towards the cortex, and halting of the migration after reaching the final destiny. Our understanding of this complex process with its different, spatial and temporal restricted modes is rapidly increasing. The combined approach using imaging technologies and molecular genetics allows the elucidation of different forms of cortical malformations. Patients with neuronal migration disorders present with mental retardation and frequently refractory epilepsy.

The disturbed initiation of the migratory process leads to an accumulation of heterotopic neurons along the walls of the lateral ventricles, their site of origin, and causes periventricular nodular heterotopia (PVNH). Mutations of the FLNA gene lead to the most common X-linked form of PVNH affecting mostly females. Further, in rare cases of autosomal recessive PVNH associated with microcephaly mutations of the ARFGEF2 gene have been detected.

Disturbances of the migratory process at the step of cell soma and nucleus translocation lead to the agyria-pachygyria-band spectrum. Autosomal dominant mutations of the LIS1 gene cause isolated lissencephaly type I, a brain malformation associated with absent or reduced gyration and thickened cortex. Another form of lissencephaly type I is caused by mutations of the DCX gene. Males harboring mutations of the DCX gene develop a lissencephalic cortex, whereas heterozygous DCX mutations in females result in subcortical band heterotopia (also called Double Cortex syndrome). In patients with lissencephaly type I the genetic testing strategy can be based and efficiently directed from the obtained MR images. LIS1 mutations cause more severe malformation in the posterior regions, whereas DCX mutations result in anterior agyria.

Another form of lissencephaly type I, the X-linked lissencephaly with abnormal genitalia (XLAG), where in difference the corpus callosum is absent, is associated with mutations of the ARX gene. Lissencephaly with cerebellar hypoplasia type b (LCHb) is the result of the failure of the migrating neuroblasts to enter the cortical plate. Autosomal recessive mutations of either the RELN gene or the VLDLR gene lead to LCHb.

Disturbances to stop the migratory process lead to “overmigration” of the neuroblasts resulting in lissencephaly type II (cobblestone lissencephaly) with agyric, pachygyric or polymicroygyric appearance of the cortex depending on the severity of the malformation. Autosomal recessive lissencephalies type II is mostly associated with congenital muscular dystrophy and varying degrees of eye abnormalities (like in Walker-Warburg syndrome, muscle-eye-brain disease or Fukuyama congenital muscular dystrophy). The underlying defect is the hypogylcosylation of alpha-dystroglycan by disturbances of the O-mannosylation of alpha-dystroglycan. Up to date six genes are identified (POMT1, POMT2, POMGnT1, FKTN, FKRP, LARGE) accounting for 30% of these alpha-dystroglycanopathies.

Recently, mutations of the GPR56 gene have been associated with bilateral frontoparietal polymicrogyria (BFPP1). Autosomal recessive mutations of the GPR56 gene should also be considered in patients with so called “cobblestone only” lissencephalies, where muscular dystrophy and eye abnormalities are absent.

The identification of the causative mutation in patients with malformation of the cortical development not only allows classifying the individual symptoms to a specific disorder but also allows a specific genetic counselling of the parents and the families according to recurrence rates. Further, the identification new genes and involved pathways increase our understanding about mental retardation and epileptogenesis.

1 G. Kurlemann , 1 B. Fiedler 1University Pediatric Clinic Munster (Munster, D)

EEG plays a decisive role in the diagnosis and monitoring of epilepsy: over and above this, EEG patterns can provide evidence of other, often rarer neuropediatric diseases in which epilepsy is not always a symptom from the onset of the illness. Knowledge of such trailblazing EEG patterns can avoid further diagnostic investigations which can be particularly stressful in infancy.

Pattern 1: slow frequency photic stimulation provoking occipital spikes provides proof of the late infantile form of neuronal ceroid lipofuscinosis (Jansky – Bielschowsky, NCL 2) in children with seizures and loss of motor function. This always lethal disease is confirmed molecularly or by curvilinear inclusion bodies in the skin.

Pattern 2: high-amplitude beta activity without medication in retarded children should be linked to diagnosis of type 1 lissencephaly. Whenever EEG activity is too fast for the age of the child type 1 lissencephaly should be a tentative diagnosis to be confirmed by ultrasound or MRI. Other possibilities could be infantile axonal dystrophy, disorders of organic acid metabolism and amino acid metabolism, giant axonopathy and Lowe syndrome, a genetic variant or a nonspecific finding.

Pattern 3: high-amplitude rhythmic theta – delta activity and occipital running spikes after eye closure in children with mental and motor retardation, happy puppet and missing speech are suspects for Angelman syndrome. Angelman syndrome is caused by defects in the maternally inherited imprinted domain located on chromosome 15q11-q13; the most common defect being a de novo deletion of maternal origin. In dependence of the size of deletion there are severe treatable epilepsies in these children. Interestingly this EEG pattern is resolved in many cases after puberty. A similar EEG pattern is found in children with Wolf–Hirschhorn syndrome.

Pattern 4: burst suppression EEG in awake mature newborns after exclusion of severe asphyxia can indicate the following: molybdenum cofactor deficiency and sulfite oxidase deficiency, nonketotic hyperglycinaemia, pyridoxine dependency, pyridoxal phosphate dependency and Ohthahara syndrome. Diagnosis of metabolic diseases is conducted via the metabolic pattern, diagnosis of Ohthahara syndrome is always an exclusion diagnosis.

Pattern 5: rebuild-up phenomenon in Moyamoya disease. High-amplitude rhythmic slow waves appear and intensify during hyperventilation called physiological build-up phenomenon; nonsynchronous polymorphous slow waves that appear a few minutes after hyperventilation are characteristic of childhood Moyamoya syndrome. This is called rebuild-up phenomenon. Therefore hyperventilation attempt should be accomplished in mid-EEG, in order not to miss this age-specific EEG pattern occurring nearly only in childhood. Rebuild-up is associated with reduction of PaCO2 and constriction of cerebral arteries.

Pattern 6: periodic bi- or tri-phasic complexes of high amplitude represent highly specific findings for subacute sclerosing panencephalitis (SSPE) following measles infection in early childhood in children without vaccination. Theses complexes are called Rademecker complexes. SSPE is not treatable, can however be avoided by consistent measles vaccination.

1 B. Neubauer , 2 S. Waldegger , 1 A. Hahn , 1 F. Eberhardt , 3 G. Kurlemann , 1 S. Garkisch , 4 U. Müller , 5 U. Stephani , 6 T. Sander 11 Justus-Liebig University (Giessen, D) ; 2 Philipps University Marburg (Marburg, D) ; 3 University Hospital Munster (Munster, D) ; 4 Institute of Human Genetics (Giessen, D) ; 5 Children's Hospital Kiel (Kiel, D) ; 6 Max-Delbrück-Centrum (Berlin, D)

Neuronal M-type channels (which are subject to inhibition by muscarine, hence the name), are composed of KCNQ2, KCNQ3 and KCNQ5 heterotetramers. These channels contribute to the neuronal resting potential and influence cortical excitability. Mutations in KCNQ2 and 3 have been found to cause benign familial neonatal convulsions, a rare autosomal dominant inherited epilepsy syndrome that remits spontaneously during infancy. We investigated KCNQ2, 3, and 5 genes in 60 nuclear pedigrees with Rolandic epilepsy and the associated EEG trait because benign neonatal seizures precede this epilepsy syndrome in a significant number of cases (5%). We identified 6 coding sequence variants in KCNQ3, 2 in KCNQ2, and 2 in KCNQ5. Functional significance was investigated by a Xenopus oocyte expression system. A formerly unknown sequence variant of KCNQ3 (Ser574Pro) encompassing a conserved carboxyterminal “A” domain was found to cosegregate with the phenotype in two small families with rolandic epilepsy without neonatal seizures. The same sequence variant was significantly associated with a spectrum of phenotypes in a large sample (n = 578) of individuals with common idiopathic generalized epilepsies (p-values for: IGE = 0.008, JME = 0.016, IAE = 0.003, CAE = 0.023) whereas it was absent in 461 controls. Several available drugs increase M-type current in man and may deserve systematic evaluation in idiopathic generalized epilepsies.

1 G. Wohlrab 1University Children's Hospital Zurich (Zurich, CH)

Summary: The Aristaless-related homeobox gene, ARX, is an important transcription factor with crucial roles in development. At least fifty-nine mutations have been described in the ARX gene in seven X chromosome linked disorders involving mental retardation. Recent studies with ARX-screening suggest that the gene is mutated in 9.5% of X-linked families with these disorders (1). The ARX gene is emerging as one of the more important disease-causing genes on the X chromosome and ought to be considered for routine screening.

Introduction: The Aristaless-related homeobox gene, ARX, is an important transcription factor with a crucial role in forebrain, pancreas and testes development (2). ARX gene mutations were identified in seven X chromosome linked disorders showing the common symptom of mental retardation:

  • - 
    X-linked West syndrome ISSX (X-linked infantile spasms syndrome; 3,4)
  • - 
    Partington syndrome PRTS (3)
  • - 
    X-linked lissencephaly with ambiguous genitalia XLAG (5,6)
  • - 
    X-linked myoclonic epilepsy with generalized spasticity and intellectual disability XMESID (3)
  • - 
    Nonsyndromic X-linked mental retardation NS-XLMR (3,7)
  • - 
    Agenesis of corpus callosum with abnormal genitalia ACC (Proud syndrome, 8)
  • - 
    Hydranencephaly with abnormal genitalia HYD-AG (8)

Although a general picture of genotype-phenotype correlation is emerging, the striking pleiotropy of ARX mutations is still poorly understood.

ARX mutations: The ARX gene is located at Xp22. It is a relatively small gene of ∼12. kb and has five coding exons. Fifty-nine ARX gene mutations were reported since 2002. These 59 cases (9) represent 32 different mutations, including 2 splice site mutations, 3 nonsense mutations, 12 missense mutations, 14 indels and one duplication. The majority of mutations (76%) are located in exon 2 of the ARX gene. Only 4 mutations were found more than once.

Genotype–phenotype: There are about seven different conditions, with or without malformations, in which ARX gene mutations have been identified. In general truncation mutations and missense mutations in crucial, evolutionarily conserved residues lead to more severe outcomes, whereas the polyalanine expansion mutations and some missense mutations have less severe consequences. The “nonmalformation group” (ISSX, XMESID, PRTS, NS-XLMR) is predominantly caused by polyalanine expansion mutations. Mutations in the “malformation group” (HYD-AG, XLAG, ACC) are diverse with 32 mutations in 32 families (8). All patients of the 59 cases reported have been mentally retarded. The severity of mental retardation and other clinical manifestation, for instance dystonic hand movements vary, both between and within families (9).

ARX-screening: The ARX gene is one of the estimated ∼100 genes involved in X-linked mental retardation. Molecular screening of exon 2 already was performed in males with mental retardation with positive results in 4 of 143 men (10). It is difficult to make firm recommendations for screening for mutations in ARX: If screening tests were cheap and easy available, then they could be incorporated into a molecular scan of all mentally retarded patients without a cytogenetic, biochemical or firm dysmorphic syndrome diagnosis. This might be possible in future, but presently it is not the case. At present, the most practical and productive screening is fort he the most frequent mutation, the c.428-451 (24 bp) duplication.

Conclusions: Mental retardation is a highly heterogeneous condition with a variety of contributing factors, both genetic and environmental. The ARX gene is emerging as an important factor. ARX mutations show striking pleiotropy and often puzzling intra- and interfamilial variation. The group of syndromes, caused by ARX-mutations, comprises developmental disorders with a spectrum from hydranencephaly to lissencephaly, agenesis of the corpus callosum without other malformations and overlapping syndromes with apparently normal brain structure. Thus, mutations of ARX contribute to the pathogenesis of each of these disorders, and will need to be considered in the evaluation of children with mental retardation with or without brain and genital malformations.

References:

1. Poirier K, Lacombe D, Gilbert-Dussardier B et al. (2006) Screening of ARX in mental retardation families: consequences fort he strategy of molecular diagnosis. Neurogenetics 7:3946. 2. Banerjee-Basu S, Baxevanis AD. (2001) Molecular evolution of the homeodomain family of transcription factors. Nucleic Acids Res 29:32583269. 3. Stromme P, Mangelsdorf ME, Shaw MA, et al. (2002) Mutations in the human ortholog of Aristaless cause X-linked mental retardation and epilepsy. Nat Genet 30:441445. 4. Wohlrab G, Uyanik G, Gross C, et al. (2005) Familial West syndrome and dystonia caused by an Aristaless related homeobox gene mutation. Eur J Pediatr 2005 164:326328. 5. Kitamura K, Yanazawa M, Sugiyama N, et al. (2002) Mutation of ARX causes abnormal development of forebrain and testes in mice and X-linked lissencephaly with abnormal genitalia in humans. Nat Genet 32:359369. 6. Uyanik G, Aigner L, Martin P, et al. (2003) ARX mutations in X-linked lissencephaly with abnormal genitalia. Neurology 61:232235. 7. Bienvenu T, Poirier K, Friocourt G, et al. (2002) ARX, a novel Prd-class-homeobox gene highly expressed in the telencephalon, is mutated in X-linked mental retardation. Hum Mol Genet 11:981991. 8. Kato M, Das S, Petras K, et al. (2004) Mutations of ARX are associated with striking pleiotrop< and consistent genotype-phenotype correlation. Hum Mutat 23:147159. 9. Gécz J, Cloosterman D, Partington M. (2006) ARX: a gene for all seasons. Curr Opin Genet Dev 16:308316. 10. De Souza Gestinari-Duarte R, Barros Santos-Reboucas C, Mattos oncalves Pimentel M. (2006) Mutational screening of ARX gene in Brazil males with mental retardation of unknown etiology. J Hum Genet 51:737740.

1 T. Dorn 1Swiss Epilepsy Centre (Zurich, CH)

Background/Methods: Epileptic seizures can be a symptom of rare genetic syndromes, especially when they associated with mental retardation and dysmorphic signs. In some of these syndromes special epileptic features point to the underlying entity. The exact identification of the underlying entity can have implications for therapy and care. This is exemplified by the case of a woman with colpocephaly, a dilation of the occipital horns and atria of the lateral ventricles, at first diagnosed as hydrocephalus, seizures and mild neuropsychological impairment The knowledge about epileptology and genetics of colpocephaly is summarised.

Case Report: The family history of the 19 years old woman was unremarkable. She was born at date with normal weight, but she was cyanotic and required oxygene because the umbilical was wrapped around her neck, but no artificial ventilation was necessary. At week 8 she was restless and vomited. Hydrocephalus was diagnosed and a ventriculo-peritoneal shunt system was implanted, which had to be revised for the first time at the age of 6 years due to too low CSF pressure. The findings of an MRI at the age of 8 years were summarised as colpocephaly. A second shunt revision became necessary 3 years later for unknown reasons. Psychomotor development was delayed. She had to go to a special school. Neuropsychological examination at the age of 8 years revealed normal intelligence, but relatively low non-verbal skills compared to the normal verbal IQ and bifrontal and left-temporal impairments. Affective instability, anxiety and aggressive behaviour additionally have disturbed her psychosocial development until now. Strabismus divergens, pendular nystagm and reduced visus of the right eye were early mentioned in the reports. A first probably generalised seizure occurred at the age of 2 years and 2 month. Further seizures with oro-alimentary automatisms, bulbus deviation and myoclonic jerks in the right hand and mouth acompanied by hypersalivation and vomiting were again mentioned about two years later. Later the patient reported a sick feeling in her neck, a sour taste in her mouth and retching as intial symptoms of her seizures. In EEG multifocal spikes and bilateral synchronous spike wave complexes lasting up to several seconds were visible. The changes were more pronounced on the reight side. At the age of 11 years continuous epileptic activity during sleep was described which was not seen in a sleep EEG at the age of 17 years. Several antiepileptic drugs were tried (CBZ, VPA, VGB, PB, STH). Their effect is not clear due to difficulties in compliance which also resulted in three status epileptici. However, pharmacoresistance could be demonstrated against OXC, which also caused hyponatriemia. With adding LEV to OXC the patient became seizure free and stayed seizure free with reduction of OXC dose.

Discussion: Several epileptological features of our patient especially vomitus during complex partial seizures and sleep EEG findings resemble the findings in four patients with colpocephaly out of five ones with 6q terminal deletion described by Elia et al. (Epilepsia 2006;47:830–838) recently. However, those cases had lower IQ values than our patient. Furthermore, strabism but not pendular nystagmus as in our patient is described in the paper. Despite of these differences we believe that our patient also has a 6q terminal deletion. Unfortunately we could not prove our hypothesis so far since the parents have not given consent to FISH analysis. Of course other (genetic) etiologies could be possible, because colpocephaly has shown to be a symptom in other syndromes with cerebral and other malformations. Colpocephaly could be considered as persistance of the fetal configuration of the lateral ventricles and related to agenesis of corpus callosum and therefore be related to different insults occurring between one and four months of gestation. Since the development of colpocephaly is not related to a disturbance of CSF circulation and since the patients of the above-mentioned paper have no shunt the necessity of the shunt implantation in our patient can be questioned.

Conclusion: Our findings underline the existence of a syndrome with colpocephaly, mild mental retardation, occulomotor disturbances and complex partial seizures with vomitus as outstanding semiological feature. Wether this syndrome is solely due to terminal deletions of chromosome 6q or can have other etiologies has still to be clarified. Colpocephaly should not be diagnosed as hydrocephalus. Patients with colpocephaly possibly need no shunt implantation.

Abstract

  1. Top of page

Session: Update epilepsy 2007

1 J.-M. Fritschy 1University Zurich (Zurich, CH)

The development of nine novel antiepileptic drugs since 1993 did not provide effective treatment for many patients with refractory epilepsy. Improvements in the therapy of seizure disorders will require a better understanding of the pathophysiology underlying acquired epilepsy, therapy resistance, and epileptogenesis. Studies of mice carrying targeted gene deletions have shown that unrelated mutations can lead to epilepsy, suggesting that seizures are a common outcome of molecular and cellular dysfunctions of the brain. Familial epilepsies mainly arise from mutations in ion channels and are characterized by typical clinical features. However, the link between a mutation and occurrence of seizures is often not established. Furthermore, similar clinical features can occur in the absence of known mutations. For instance, although mutations in the GABAA receptor gamma2 subunit gene are associated with GEFS+, they are not found in most patients suffering from focal epilepsies subsequent to febrile seizures. Animal models carrying mutations found in acquired epilepsies have therefore been established to analyze the molecular mechanisms underlying seizures.

Here, we review four animal models with mutations in synaptic proteins (voltage-gated sodium channels, nicotinic acetylcholine receptors, GABAA receptors, and a secreted synaptic protein, LGI1). These models illustrate how molecular defects affecting specific aspects of synaptic transmission result in diverse seizure disorders by sometimes unexpected mechanisms. In a mouse model of severe myoclonic epilepsy in infancy, the reduction of Nav1.1 sodium channels paradoxically resulted in hyperexcitability of hippocampal circuits, leading to spontaneous seizures. This effect was due to a selective reduction of sodium channels in interneurons, which caused a decrease in synaptic inhibition (Yu et al., Nature Neuroscience 2006; 9:1142–9). Similarly, the key role of interneurons in regulating neuronal activity was demonstrated in mice carrying point-mutations in the alpha4 subunit of nicotinic acetylcholine receptors associated with autosomal dominant nocturnal frontal lobe epilepsy. In this case, however, the mutation increased the function of nicotinic cholinergic receptors, which are highly expressed in GABAergic interneurons, leading to enhanced inhibitory function. As a result, acetylcholine increased neuronal synchronization in neocortical circuits causing interictal spiking and recurrent seizures (Klaassen et al., Proc Natl Acad Sci USA 2006; 103:19152–7.). The GABAA receptor gamma2 subunit mutation R43Q associated with GEFS+ has been reported to impair assembly and cell surface expression of GABAA receptors (Berkovic and Petrou, Trends Mol. Med. 12, 343). The precise effects on GABAergic transmission remain controversial, although recent data suggest that tonic inhibition, produced by extrasynaptic receptors, might me reduced. Intriguingly, the effect of the mutation was shown to be temperature-dependent, with cell surface expression being reduced in vitro at temperatures higher than 37°C (Kang et al., J Neurosci 2006; 26:2590–7.). However, since most GEFS+ patients do not carry this mutation, such a mechanism alone is not sufficient for explaining the onset of seizures. In the last example, reduced expression of LGI1, a secreted molecule regulating glutamatergic transmission, was sufficient to cause seizures in mice. LGI1 mutations have been associated with rare familial idiopathic temporal lobe epilepsies, suggesting that disturbances of glutamatergic synapses might underlie seizure activity without causing morphological alterations or neuronal loss (Fukata et al., Science 313:1792). Altogether, these studies highlight the relevance of experimental studies for understanding the pathophysiology of acquired epilepsy and seizure disorders in general. The possibility that regulatory molecules such as LGI1, contribute to seizures provides novel targets for the development of therapeutic strategies.

1 B.J. Steinhoff 1Epilepsy Centre Kork (Kehl-Kork, D)

This lecture presents the most innovative and interesting clinical papers published during the last 12 months. The selection is entirely subjective without any objective evidence but reflects scientific highlights every epileptologist should be aware of. The review comprises leading papers on the fields of genetics, imaging, clinical neurophysiology, comorbidity research, anticonvulsant drug therapy and epilepsy surgery. The most interesting papers concerning these fields will be summarized.

Abstract

  1. Top of page

Oral session 1

1 Y.G. Weber , 2 A. Storch , 1 T.V. Wuttke , 3 K. Brockmann , 1 J. Kempfle , 1 D. Blazevic , 3 A. Pekrun , 1 M. Fauler , 1 F. Lehmann-Horn , 1 H. Lerche 11 University Ulm (Ulm, D) ; 2 University Dresden (Dresden, D) ; 3 University Gottingen (Gottingen, D)

Paroxysmal dyskinesias are characterized by attacks of involuntary movements classified as kinesiogenic, nonkinesiogenic or exertion-induced. The pathophysiology of these syndromes is still unclear. We describe a novel autosomal dominant syndrome in a three-generation family with four affected individuals presenting with a combination of a paroxysmal exertion-induced dyskinesia, hemolytic anemia and epilepsy. The syndrome was characterized by extensive clinical, genetic and experimental studies. Three individuals suffer from attacks of involuntary movements with dystonic and choreoathetotic components after heavy workload since early childhood. The two sons of the index case show mild neuropsychological deficits, mild permanent motor problems and a focal epilepsy with myoclonic and atonic seizures with status epilepticus in fasting state. A good response of the symptoms to glucose infusion and long term ketogenic diet was observed. Affected individuals presented with decreased CSF glucose, hemolytic anemia, and increased intracellular sodium and decreased potassium concentrations in red blood cells, which showed an increased percentage of echinocytes. Genetic investigation of SLC2A1 encoding the glucose transporter type 1 (Glut1) revealed a mutation in the pore region. Functional analysis in Xenopus laevis oocytes injected with wildtype or mutated cRNA revealed a decreased glucose uptake and a cation leak of the mutant transporter compared to the wild type. Erythrocytes analysis from patients and healthy controls confirmed the electrolyte rearrangement. The epileptic activity can be well explained by the energy deficit. In analogy to genetic defects of the anion exchanger of the erythrocyte membrane which can also induce deformed erythrocytes with intracellular ionic changes, the cation leak caused by the mutation may also explain the abnormal erythrocyte morphology and secondary hemolytic anemia. We propose that the dyskinesias are caused by a combination of a metabolic defect and an electrolyte shift in glial cells expressing the transporter. The cation leak may decompensate during prolonged exercise when not enough glucose as an energy source can be delivered across the blood–brain barrier.

1 T. Schmitt-Mechelke , 2 D. Bartholdi , 3 R. Touraine 11 Pediatric Clinic (Lucerne, CH) ; 2 Institute of Medical Genetics (Zurich, CH);3 Laboratory of Molecular Genetics (Saint-Etienne, F)

Background: Mutations of the ARX-Gene (aristaless-related homeobox gene, Xp22.11) have been observed in a variety of X-linked neurogenetic diseases: lissencephaly with ambious genitalia (XLAG), mental retardation with agenesis of the corpus callosum and ambiguous genitalia (Proud-Syndrome), mental retardation with progressive dystonia (Partington-Syndrome) and X-linked infantile spasms with mental retardation. We observed a male neonate with a myoclonic epileptic encephalopathy and a severe multisystem disorder due to a novel mutation in the ARX-gene.

Case Report: After an uneventful pregnancy and birth, a male term infant presented with myoclonic neonatal seizures on the first day of life. Physical examination showed an unilateral cryptorchism and was unremarkable otherwise; his head circumference was normal. Ultrasound and MRI showed an agenesis of the corpus callosum and a lissencephalic cortical migration disorder. The EEG showed an asynchronous burst-suppression pattern with frequent myoclonic and tonic seizures unresponsive to treatment with phenobarbital, vigabatrin or benzodiazepines equivalent to an early myoclonic epileptic encephalopathy (Ohtahara-Syndrome). The neonate developed hypothalamic dysfunction with temperature instability and an enteropathy with potassium-loosing secretory diarrhea. Considering the severely impaired prognosis, comfort care was offered and the patient died after 3 weeks.

Analysis of the ARX gene disclosed a 1414C>T mutation in exon 4 resonsible for a nonsense mutation R472X not described before. The mutation was not detected in DNA of the mothers leukocytes.

Conclusion: ARX-mutations can present as a neonatal epileptic encephalopathy with lissencephaly and agenesis of the corpus callosum without obvious genital malformation. The condition can affect the gastrointestinal tract and carries a poor prognosis. Prenatal diagnosis is feasible.

1 K. Jann , 1 M. Hauf , 1 J. Mathis , 2 K. Meyer , 1 T. Dierks , 1 T. König , 1 R. Wiest 11 University Berne (Berne, CH) ; 2 Bethesda Clinic (Tschugg, CH)

Purpose: Interictal combined EEG and fMRI is of gaining importance for the focus localization and understanding of the pathophysiology in epilepsies. We report on a new approach for the continuous acquisition of functional MRI data and EEG using an Independent Component Analysis (ICA) based technique.

Method: 18 patients, 14 with focal epilepsies (FE) and 4 with idiopathic generalized epilepsies (IGE) were examined with simultaneous 96-channel EEG-fMRI recordings in a 3 Tesla MR unit. ICA derived factors coding for time varying epileptiform activity were convolved with a hemodynamic response function (HRF) to predict the BOLD signal. Voxelwise correlations between the ICA-predictors and the BOLD signal were computed. Regions with significant correlations were interpreted as having been active during the time course of the interictal discharges.

Results: In 12 subjects with FE, the interictal focus was detected in accordance with previously and simultaneously recorded EEG sources and fitted the clinical semiology and structural lesions on high resolution MRI, if present. In three patients with IGE, symmetrical cortical (de-)activation patterns and bilateral thalamic changes were recorded, supporting the corticoreticular theory of epileptogenesis in IGE. Three patients had to be excluded due to uneventful EEG recordings or motion artifacts.

Conclusion: Our data suggest that simultaneous EEG/fMRI recordings are a safe, noninvasive and promising technique to provide additional information about the irritative zone in FE and different propagation patterns in FE and IGE.

With the support of the Swiss National Foundation (Grant no. 320000-108321/1).

1 M.T. Lutz , 2 C. Helmstaedter 11 Epilepsy Center Kleinwachau (Radeberg, D) ; 2 University Clinic of Epileptology (Bonn, D)

Purpose: One of the newest models of intelligence is the Cattell-Horn-Carroll model (CHC model). This model is a hierarchical framework of cognitive abilities and consists of different strata describing varying levels of generality. Performing a comprehensive evaluation of intelligence in addition to a neuropsychological testbattery for answering questions of diagnosis, prognosis, or outcome in patients with epilepsy, often surpasses available resources on time and manpower. Therefore we raised the question whether it would be efficient to deduce information about intelligence from a neuropsychological test battery. For that purpose, we analysed if data of a neuropsychological test battery can be described in the framework of the Cattell-Horn-Carroll model.

Methods: Testing of attention, visual and verbal memory, language, higher verbal and visual reasoning, and executive functions was performed in 190 healthy subjects recruited for a normative study and in 190 patients with refractory epilepsy. Both samples were derived from the University Clinic of Epileptology, Bonn, Germany. First, we contrasted alternative models of intelligence in the sample of healthy subjects using confirmatory factor analyses (CFA) and then we tested the factorial invariance of the best fitting model across the groups using multigroup CFA.

Results: The best fitting model in the sample of healthy volunteers was a reduced CHC-model with five broad cognitive abilities, namely crystallized intelligence, visual processing, long-term retention and retrieval, processing speed, and short term memory (goodness of fit characteristics: Chi-square = 206.48 (df = 140); CFI = 0.923). A g-factor was not included. Invariance of the measurement and structural model across groups could be demonstrated after erasing two test parameters. The fit characteristics of this multigroup CFA were good (chi-square = 344.98 (df = 228); CFI = 0.925).

Conclusion: This cross-battery study showed that the Cattell-Horn-Carroll model of cognitive abilities is suited for describing the data of a neuropsychological test battery. This holds true for healthy volunteers as well as for patients. The results add information about the factorial validity of the test battery and overcome out of date concepts of intelligence. Furthermore, the results show an economic way for combining the assessment of intelligence with neuropsychological diagnostics in presurgical patients with pharmacoresistant epilepsy.

1 K. Krakow , 2 F. Rosenow , 1 M. Sitzer , 1 C. Foerch 11 J.W.Goethe-University (Frankfurt, D) ; 2 University Hospital Giessen and Marburg GmbH (Marburg, D)

Background: Estimates of the rates of early epileptic seizures after stroke range from 2% to 33%. Aside from cortical location, no risk factors for poststroke seizures have been consistently demonstrated. In this study, we tried to identify predictors of acute poststroke seizures in different stroke subtypes.

Methods: We analyzed a large prospective hospital-based stroke registry with more than 100 hospitals enrolling stroke patients into this computerized database. Within a 7-year period (1999-2005), 58 954 patients with the diagnosis transient ischemic attack (TIA, n = 16 718), cerebral infarction (CI, n = 37 360), and intracerebral hemorrhage (IH, n = 4876) were documented, who were admitted within 24 hours after stroke onset. In multivariate analyses we adjusted for age, gender, neurological deficit at admission (Modified Rankin Scale, impairment of consciousness), type of stroke, arterial hypertension and diabetes mellitus.

Results: The mean age of the population was 72 ± 13 years. 50.1% were female. Acute poststroke seizures (defined as the occurrence of epileptic seizures during hospitalization) occurred in 0.8% of patients with TIA (mean duration of hospitalization 8 days), in 2.4% of patients with CI (11 days), and 5.1% of patients with IH (12 days). Multivariate analysis revealed that in patients with CI the risk of an acute symptomatic seizure decreases by 1.7% for each year of life (odds ratio (OR) 0.983 [0.977-0.989], p<0.001). Further predictors of acute seizures for all stroke subtypes were a severe neurological deficit and in particular diabetes mellitus (CI: OR 2.2 [95% CI 1.9-2.6], IH: OR 1.6 [1.2-2.2], TIA: 5.6 [3.8-8.3].

Conclusions: Beside young age and severity of the neurological deficit, diabetes mellitus is an independent risk factor for the occurrence of acute epileptic seizures after stroke. This effect is particularly strong in ischemic subtypes of stroke. Further studies are necessary to elucidate the mechanism of this association.

1 E. Rodin 1University of Utah (Salt Lake City, USA)

Purpose: Precise determination of seizure onset is essential for “non-lesional” epilepsy surgery. Yet this is at times difficult when conventional EEG frequencies of 0.5-75 Hz are used. Since it is known that slow baseline shifts accompany seizures, and can have lateralizing significance, infraslow activity (ISA) between 0.01-0.1 Hz was investigated.

Methods: The data from fifteen adult patients with partial complex seizures (N35) were compared with those from six children with classical 3 Hz Spike-Wave (SW) discharges accompanied by Absence seizures (N30). The adults were recorded on a Grass-Telefactor system and the children on a Nihon-Kodhen instrument. Data analysis was performed with the BESA program. When the low frequency filter was left open an offset was observed in a number of channels regardless which system was used. This disappeared when the low frequency filter was set at 0.01 Hz (6db forward). For the demonstration of ictal ISA a window of 30 seconds was used. It contained ten seconds prior to the conventionally determined seizure onset and 20 seconds of the beginning seizure. The data were then displayed in a topographic manner on a common average reference.

Results: In the adult patients two types of infraslow distributions were observed: 1) a clear propagation of activity within a given hemisphere with subsequent spread to the opposite side; 2) the ISA showed no appreciable latency differences between the two hemispheres. In these instances the frontal areas showed negative potentials while the posterior head regions were positive. At times both types of ictal ISA could be seen in different seizures from the same patient.

The 3 Hz SW pattern in the children was always accompanied by an ISA transient. It started several seconds before the seizure and subsequently outlasted it. The topographic distribution was identical in all seizures. There was synchrony throughout all channels with frontal negativity and posterior head regions positivity.

Conclusions: Ictal ISA can be clinically meaningful. Its major use may reside in the presurgical evaluation of medically refractory patients because it is likely that those who show the ictal ISA generalized topographic distribution may have a more guarded long-term prognosis when non lesional removal of cerebral tissue is performed. The data also suggest that when epilepsy monitoring units upgrade existing equipment that systems with the widest possible frequency range should be considered.

Abstract

  1. Top of page

Session: Update –“Investigator-initiierte klinische Studien (IIT)” by epilepsy

1 J. Rémi , 1 P. Axtner , 1 A. Hüttenbrenner , 1 K. Schüler , 1 K. Won , 1 S. Noachtar 1Neurological Clinic and Polyclinic (Munich, D)

In prospective studies we compared the effects of the new antiepileptic drugs (AED) oxcarbazepine (OXC), levetiracetam (LEV), gabapentin (GBP), pregabalin (PRE) and the established AED carbamazepine (CBZ) on eye movements and posture control in twelve healthy volunteers who received single doses of 600 mg OXC, 1000mg LEV, 900 mg GBP, 75mg PRE and 400mg CBZ in a double-blind, cross-over, randomized trial.

The drug effects were evaluated for averaged plasma levels, for individual highest to lowest plasma levels and for individual highest to lowest effect.

All AED affected eye movements significantly, with highest effects on saccadic eye movements. For GBP and OXC, the strongest impact was seen on vertical eye movements, with GBP slowing the peak vertical saccade by a maximum of 20.4% and OXC by 29.7%. Both CBZ and PRE also slowed the peak vertical saccade (CBZ by 17.6% and PRE by 27.3%), but showed even stronger impact on the peak horizontal saccade (CBZ by 35.6%, PRE by 28.5%). LEV was different from the other AED as it did not slow saccadic eye movements significantly, but had a significant impact on slow pursuit eye movements with a maximum slowing of 32% on vertical slow pursuit.

GBP affected body sway significantly in most conditions. The maximum impairment was seen in the difficult conditions (reclined head, eyes closed on foam plated showed 25.0% increase in body sway, tandem stance with closed eyes 67.0%). CBZ and PRE both also affected body sway, but significant impairment was mostly seen in difficult conditions (reclined head, eyes closed on foam plate showed a 24% increase in body sway for CBZ, 24.7% increase for PRE). OXC increased body sway in only very few conditions (reclined head, eyes closed on foam plate showed a 6.0% increase, tandem stance with closed eyes 9.4%). LEV increased the RMS in only one condition (open eyes on foam plate) by 36.7%.

Although OXC, LEV, GBP, PRE and CBZ may all cause the subjective feeling of giddiness, the effects on eye movements and body sway are different. All AED effected eye movement, with fast eye movements being similarily affected by the AED with similar mechanism, the sodium channel blockers OXC and CBZ, and the AED with a proposed effect on the GABAergic system GBP and PRE. Only LEV with its different mechanism of action – impairing synpatic vesicle action – had a different impact, impairing slow eye movements. For postural control, results differed, GBP and PRE differed in their results as did CBZ and OXC, and LEV as a special case. AED seem to have a less homogenous effect on postural control as they have on eye movements.

Our results enable new insights on physiological and pathophysiological mechanisms of dizziness, and can be applied in the development and testing of new AEDs.

1 H. Karakizlis , 1 K.-M. Klein , 1 S. Bauer , 1 P. Müller , 2 S. Reinecker , 2 N. Burchardi , 2 C. Schade-Brittinger , 3 S. Evers , 4 Y.G. Weber , 5 S. Arnold , 6 J. Springub , 7 I. Eisensehr , 8 M. Schröder , 9 S. Hallmeyer-Elgner , 1 F. Rosenow 11 University Hospital Giessen and Marburg GmbH (Marburg, D) ; 2 Coordination Centre for Clinical Trials (Marburg, D) ; 3 University Hospital Munster (Munster, D) ; 4 University Hospital Ulm (Ulm, D) ; 5 Neurological practice 1 (Munich, D) ; 6 Neurological practice (Westerstede, D) ; 7 Neurological practice 2 (Munich, D) ; 8 University Hospital Regensburg (Regensburg, D) ; 9 University Hospital Dresden (Dresden, D)

Objective: The purpose of this study is to compare the efficacy, safety and tolerance of the drugs lamotrigine and levetiracetam in the initial monotherapy of patients with newly diagnosed epilepsy or a first seizure with a high risk of recurrence.

Main end points are: rate of seizure-free patients in the first 6 weeks of the trial (main outcome criterion) as well as rate of seizure-free patients during the last 16 weeks and the total 26 weeks of the observation period; time to first seizure; time on study medication; safety and quality of life during treatment are compared.

Inclusion Criteria:

  • - 
    Age ≥ 12 years
  • - 
    Body weight ≥ 30kg (patients 12-15 years of age) and ≥ 40kg (patients above 16 years of age), respectively
  • - 
    Either one epileptic seizure with high risk of recurrence (focal seizure semiology, MRI lesion or pathological EEG findings) or a newly diagnosed epilepsy (≥ 2 unprovoked seizures) with at least 1 seizure within the past 3 months before start of study therapy
  • - 
    Treatment with a maximum of one AED for at most four weeks before start of study therapy
  • - 
    Women of ≥ 16 years must use ≥ one of the following contraceptives: Oral contraceptive (for at least 1 one month before start of study therapy), barrier with spermicide (condom or diaphragm), intrauterine contraceptive device, tube ligation. For girls between 12 and 15 years of age a written confirmation of sexual abstinence is sufficient.
  • - 
    Informed consent.

Exclusion Criteria:

  • - 
    Nonepileptic seizures or acute symptomatic seizures caused by an acute treatable condition
  • - 
    Absence seizures or simple partial seizures without motor signs (aura) only
  • - 
    History of chronic focal epilepsy or status epileptic
  • - 
    Progressive neurological, degenerative or malignant diseases which are clinically relevant from the investigator's point of view (e.g., neoplastic or cardiovascular disease)
  • - 
    Prior treatment with levetiracetam or lamotrigine
  • - 
    Renal insufficiency (creatinine clearance < 50 mL/min [adults] or 80 mL/min [children])
  • - 
    known hypersensitivity to levetiracetam, lamotrigine or another component of the trial drugs
  • - 
    Patients who are attended by a legal guardian
  • - 
    Psychiatric disease or affective disorders (within the past 6 months), which had to be treated with electric convulsive therapy, tranquilizing agents, monoamine oxidase inhibitors or CNS-active sympathomimetics (e.g., methylphenidate)
  • - 
    Alcohol- or drug-addiction within the past 12 months
  • - 
    Pregnant or breast-feeding women
  • - 
    Patients who participated in another clinical trial within the past 30 days

Study Design: The study includes for all patients in levetiracetam-group and lamotrigine-group three visits and two phone contacts. The first visit is a baseline visit (Visit 1). A second visit will follow after six weeks (primary outcome criterion) and the third visit after 26 weeks. For regulatory reasons a fourth visit is planned for the levetiracetam-group after 36 weeks. The first telephone contact is scheduled after 4 weeks (titration of levetiracetam) and the second telephone contact after 11 weeks (titration of lamotrigine).

The main analysis of the primary outcome criterion is based on the intention-to-treat (ITT) population. For the confirmatory statistics a two-tailed Fisher's exact test with a level of significance of a = 0.05 is used. The calculation of these secondary results has to be regarded as descriptive and exploratory. Further analyses, in which stratification criteria were taken into account or in which per-protocol comparisons were aimed, served to explore the robustness of the obtained results.

Recruitment: A total number of 406 patients will be included in the LaLiMo trial. At present, 54 centers are participating, which together recruited 306 patients as of 2/20/07. So far eight centers (five hospitals and three medical practices) have recruited more than 10 patients per center. 35 of all patients (11,5%) are < = 18 years old and 44 (14,5%) are ≥ 65 years.

Until the last patient is randomized the recruitment will go on. Last patient is expected in December 2007.

Abstract

  1. Top of page

Session: Historical aspects of epileptology

1 H. Schneble 1German Epilepsy Museum Kork (Offenburg, D)

To reconstruct the history of epilepsy and epileptology throughout the centuries and millenniums, one can follow the traces of physicians and scientists that have been investigating intensively epilepsy and its impacts during their lifetime.

Such chronological review based on the historical personalities could begin with Hippocrates and then include Galen, Avicenna, Paracelsus, Tissot, Hughlings-Jackson in the past and, in our modern times, e.g., Henri Gastaut and Dieter Janz.

The same procedure might be chosen for a historical abstract of epileptology in Germany during the last 2 centuries. (This temporary limitation seems to be justified by the historical and political conditions and the scientific evolution during this time.)

1855: H. A. Reimer opens in Görlitz the first “Heilanstalt für Epileptische” (‘Nursing Home for Epileptics’). It is the first medical institution that is exclusively dedicated to epileptics. (Before, specialized hospitals for patients suffering from psychiatric and epileptic disorders had been founded in France, e.g., Salpêtrière and Bicêtre.)

1857: A. Kussmaul and his assistant A.D. Tenner publish the results of the provocation of an epileptic reaction in a whole animal model: they achieved to trigger epileptic fits by interruption of the animals cortical blood flow.

1867: The “Pflegeanstalt für epileptische Knaben,” a nursing home for epileptic boys, is inaugurated near Bielefeld. In 1972, Friedrich Bodelschwingh becomes the manager of the institution that will later become famous all over the world under the name “Bethel”.

1870: For the first time in history, G. Th. Fritsch and E. Hitzig achieve to provoke epileptic seizures by electric stimulation of a canine brain.

This, together with the results of Kussmaul's and Tenner's work, provided the proof that the brain is the origin of every epileptic activity.

1899: O. Binswanger writes the first extensive German text book about epilepsy. Before, mainly french physicians had published similar works.

1912: A. Hauptmann who is at this time junior officer on the university hospital in Freiburg, discovers the antiepileptic properties of phenobarbital. Phenobarbital is the second efficient antiepileptic drug after the discovery of the antiepileptic virtues of bromine half a century before, and it has considerably less side effects.

1928: In Breslau, O. Foerster teaches his method of intraoperative cortical stimulation and epilepsy-surgical techniques (performed on only locally anesthetized patients) to the American neurosurgeon W. Penfield. Foerster is therefore considered as one of the pioneers who prepared the amazing evolution of epilepsy surgery that will soon begin in Montreal under Penfield's direction.

1929: H. Berger published in the “Archiv für Psychiatrie und Nervenkrankheiten” the first article about the “Elektrenkephalogramm” in the human. Until 1938, 13 other articles will follow.

1931/32: F. Krause gives a detailed description of the surgical treatment of epileptic disorders in “Spezielle Chirurgie der Gehirnerkrankungen” that relies on his own yearlong experience in epilepsy surgery.

1969: D. Janz publishes his extensive text book “Die Epilepsien”

1975: Foundation of the “Königssteiner Arbeitskreis.”

1978: H. Penin becomes the director of the first Epilepsy Hospital of a German University in Bonn. At the same time, he becomes the full professor of the first chair for epileptology in the German-language area.

This summary cannot be exhaustive and many milestones of the History of Epileptology in Germany, especially after World War II, have not been considered sufficiently. However, one can see that two conditions are always necessary for the progress of medicine and epileptology: first a specific spirit of the time that does not only allow, but even encourages independent thinking, and second an outstanding personality that is persistently seeking new knowledge.

1 K. Karbowski 1University Hospital (Berne, CH)

Epilepsy in Switzerland was already a concern of the famous travelling doctor Paracelsus (1493-1541), whose real name was Theophrastus von Hohenheim, and who was born in Einsiedeln in the canton of Schwyz. His ideas about epilepsy were, however, mostly influenced by a mystical mediaeval way of thinking. Nearly 250 years went by before a scientific epileptology came into existence. Its founder was Samuel-Auguste Tissot (1728-1795), born in Grancy, canton Waadt, physician with a practice in Lausanne. In his “Traité de l'épilepsie,” published in 1770, he not only described the symptoms of nearly all the forms of epileptic attacks known today, but expresses also his views on the pathogenesis, course and prognosis of epilepsy. Nearly 100 years later, Théodore Herpin (1799-1865) from Geneva made an important contribution to epileptology. In his first book, published in 1852, he set up a series of general treatment guidelines including monotherapy, and a gradually increasing dosage. In his second teaching book, which was published posthumously in 1867, Herpin formulated the “double lois d'identité,” the double law of identity. Its first especially significant part declares that the less severe or partial seizure manifestations – which occur in epileptics in the intervals between Grand mal seizures – are in fact different degrees of aborted or incomplete symptoms of a Grand mal seizure. Herpin also described the “commotions épileptiques,” a syndrome which is now known as benign juvenile myoclonic epilepsy. The Bernese Nobel prize winner Theodor Kocher (1841-1917) was one of the pioneers of the surgical treatment of epileptic patients. In 1899 he demonstrated the importance of relieving the increased intracranial pressure by the excision of the Dura mater in posttraumatic epilepsy. At the end of the 19th century the long term care of epileptic patients also got more attention. In 1886 both the welfare association “Bethesda” in Berne, with its institution in Tschugg, and the Swiss Institution for Epileptics in Zürich were founded. “L'Institution de Lavigny” was founded 20 years later in the French part of Switzerland. It should be noted that Dr. Hans v. Heinrich Landolt, who was head of the Swiss Institution for Epileptics (now known as the Centre for Epilepsy, “Epilepsiezentrum”) in Zurich from 1955-1971, published fundamental papers about temporal lobe epilepsy and about epileptic psychoses. He discovered that often with the occurrence of psychotic states, the EEG becomes more normal or entirely normal as compared with previous and subsequent EEG findings. The term “forced normalisation” is linked worldwide to the name Landolt. The Institute for Epilepsy in Zürich, as well as the University neurosurgical clinic have had EEG units since 1948. This in the neurosurgical clinic was run by Dr (later Professor) Rudolf Hess. It became independent in 1972 under the name of the Institute for electroencephalography, and was the teaching centre for a whole series of Swiss and foreigner electroencephalographists and epileptologues. Professor Hess was a leading worldwide expert in this field up to his retirement in 1981. His successor Professor Heinz Gregor Wieser, with his expert knowledge of stereo EEG, has made an important contribution to the operative treatment of epilepsy. Together with the neurosurgeon M.G.Yasargil he developed the method of selective amygdala-hippocampectomy. In the 50s and 60s of the 20th century, EEG units were also set up in other Swiss university and cantonal hospitals. They were the regional diagnostic and treatment centres, as well as the home of research and teaching in the field of epilepsy. Currently, the operative treatment of epilepsy is practised not only in Zürich, but also in Berne and Geneva/Lausanne. At the social level, in 1931 the Swiss Help Association for Epileptics was founded, which was replaced in 1963 by the Swiss League against Epilepsy. Dr. Günter Krämer, the current president of this League, and since 1994 the head of the Zürich Epilepsy Centre, has developed a busy teaching and research activity. He has also been, since 2003, the chief editor of the renowned Swiss journal “Epileptologie” which currently appears 4 times a year.

Abstract

  1. Top of page

Working group for presurgical epilepsy diagnostics and epilepsy surgery – part 1

1 A. Schulze-Bonhage , 2 R. Wohlfarth , 3 J. Saar , 4 C. Helmstaedter 11 University Hospital Freiburg (Freiburg, D) ; 2 Clinic for Neurological Rehabilitation (Krozingen, D) ; 3 Epilepsy Centre Kork (Kork, D) ; 4 University Clinic of Epileptology (Bonn, D)

Background: Epilepsy surgery is an efficient treatment of focal epilepsy but seizure freedom is often achieved at the cost of cognitive impairments due to surgery. This study analyses short-term effects of cognitive rehabilitation after temporal lobe epilepsy surgery.

Methods: Two groups of temporal lobe resected patients matched with regard to age, sex, type of surgery, and seizure outcome, one with (n = 55) and one without postoperative rehabilitation (n = 57) were evaluated with respect to memory and attention before and three months after temporal lobe surgery.

Results: After surgery, 78% of the patients were seizure free. MANOVA revealed a significant "side X surgery effect on verbal recognition and a “rehabilitation X surgery” effect on verbal learning and recognition. There were no effects for loss in verbal delayed recall or figural memory. Detailed analyses indicated gains due to rehabilitation particularly after right surgery. Attention generally improved. The risk to have loss in verbal memory was 4 times higher without than with rehabilitation.

Conclusions: Rehabilitation can reduce the risk of verbal memory decline after temporal lobe surgery. Rehabilitation affected the cortically associated aspects of verbal learning rather than mesially associated aspects of long term consolidation/retrieval, figural memory was not affected at all, and attention improved independent on rehabilitation. Furthermore, left resected patients, who would need an efficacious rehabilitation most, showed less profit than right resected patients, indicating that left sided surgery may impair capacities needed for efficient training of verbal memory. Alternative rehabilitative strategies may be necessary for this patient group.

1 R. Thorbecke 1Epilepsy Center Bethel (Bielefeld, D)

  • 1
    It is well established that QL after ATL is significantly improved 1 – 2 years after surgery. This holds however only for PWE completely seizure free (Engel class I) (Vickrey et al., 1992; Markand et al., 2000). Another finding is that factors closely related to depression are strong predictors for QL outcome (Loring et al., 2004). There are however until now no studies which evaluate the effects of pre- and postoperative psychosocial interventions on QL outcome e.g., preoperative shaping of expectations or close postoperative monitoring and support of patients at risk for psychiatric complications.
  • 2
    The findings on the effects of surgical treatment on employment are contradictory. There are some studies mostly from the early times of epilepsy surgery (e.g., Green, 1966; Taylor, 1968) showing strong effects which may however be explained by patient selection or inadequate control groups (e.g., Jones et al., 2002). The studies from the last 10 years show only modest or no effects. There seem also to be methodological difficulties e.g., when unemployment is reported to decrease, however at the same time the proportion of patients on early disability increases. In the U.S. collaborative study in which 396 patients were included only a net improvement of 7% in employment 24 month after surgery was observed (Chin et al., 2006). The authors therefore urge to make early rehabilitation efforts to reintegrate epilepsy surgery patients into employment.
  • 3
    In the epilepsy centre Bethel in 1997 a short term rehabilitation unit was opened and a special program for patients undergoing surgical treatment developed. A comparison of 103 patients operated between 7/1991 – 3/1996 and 115 patients operated 2/1998 – 5/2002 showed that 2 years after surgery significantly more patients of the latter group were in gainful employment (p = 0.03), the net difference between the two groups being 15%.
  • 4
    Analysis of the effects showed that the main factors are: a) common preoperative definition of employment aims in relation to the surgical intervention together with the patient; b) medical, psychiatric, psychological and social work support aiming at successful work integration immediately after surgery; c) second intake of patients who are not yet integrated into work or enrolled in a vocational rehabilitation program six months after surgery; and d) continuous monitoring and support of patients at risk for psychiatric complications.

References:

R. Thorbecke, Barbara Loer, Postsurgical Rehabilitation, in: H.Lüders (ed.) Epilepsy Surgery, 3d edition, forthcoming.

Abstract

  1. Top of page

Working group for presurgical epilepsy diagnostics and epilepsy surgery – part 2

1 A. Ebner 1Epilepsy Center Bethel (Bielefeld, D)

Objective: The goal of this study was to evaluate the long-term outcome of extratemporal epilepsy surgery and preoperative prognostic factors associated with seizure outcome.

Methods: This retrospective study included 154 consecutive adult patients who underwent epilepsy surgery at Bethel Epilepsy Centre, Bielefeld, Germany between 1991 and 2001.

Seizure outcome was categorized on the modified Engel classification. Survival statistics were performed using Kaplan–Meier curves, life table and Cox regression models to evaluate the risk factors associated with outcomes. Sixty-one patients (39.6%) underwent frontal resections, 68 (44.1%) posterior cortex resections (PCEs), 15 (9.7%) multilobar resections, 6 (3.9%) parietal resections and 4 (2.60%) occipital resections.

Results: The probability of an Engel Class I outcome for the overall patient group was 55.8% (95% confidence interval (CI 52-58%) at 6 months, 54.5% (CI 50-58%) at 1 year, and 51.1% (CI 48-54%) at 14 years.

If a patient was in Class I at 2 years postoperatively, the probability of remaining in class 1 for 14 years postoperatively was 88% (CI 78-98%). Factors predictive of poor long-term outcome after surgery were previous history of surgery (p = 0.04), tonic-clonic seizures (p = 0.02) and the presence of an auditory aura (p = 0.03). Factors predictive of good long-term outcome were surgery within 5 years after onset (p = 0.015) and preoperative invasive monitoring (p = 0.002).

Conclusions: Extratemporal epilepsy surgery is effective in long-term follow-up. The outcome at the first 2-year follow-up is a reliable predictor of long-term Engel Class I postoperative outcome.

1 C. Helmstaedter , 2 J. Schramm , 1 C.E. Elger 11 University Clinic of Epileptology (Bonn, D) ; 2 Neuroscience Center – University of Bonn (Bonn, D)

Seizure freedom and preservation if not improvement of the patients cognitive capabilities are the declared aims of the presurgical evaluation and surgical treatment of pharmacoresistant epilepsy. With a series of 926, 6-68 years old patients, operated in Bonn between 1988 and 2003, the question was raised as to whether an improvement of cognitive and seizure outcome can be observed over time. Patient selection was determined by availability of pre- and postoperative memory data. Three cohorts with comparable group sizes were differentiated, once according to the year of surgery, and again with consideration of the type of epilepsy, i.e., frontal (N = 117), mesiotemporal (N = 398), temporal (N = 334), posterior (N = 49), multifocal/hemispheric (N = 28). In each cohort temporal lobe epilepsies TLE represented about 80% of the patients. However, the diagnosis of a mesiotemporal (mTLE) increased significantly from the first to the second cohort (from 10% to 60%). At the same time selective surgeries increased from 65% to 90% in mTLE, but also in nonmesial TLE (from 12% to 39%). Independent on the cohort 10–16% of the epilepsies were frontal, 4–6% posterior, and 3–4% multifocal or hemispheric. The latter were not followed any longer. WADA tests decreased from 87% to 18%. The indication changed due to better predictors, changing surgical procedures and introduction of fMRI. Different from all other groups, patients with mTLE were steadily older over time, they had longer lasting epilepsies and they showed poorer baseline performance (in 4 of 7 domains). Cognition was evaluated via domainrelated categorial data. This reduces the number of variables and but bears the risk of a bottom effect in retesting. Preoperatively memory deficits were the prevailing impairment (60%–70%), 90% of the patients showed an impairment in at least one of 7 domains (m < 1SD).

One year after surgery 63% of all patients were completely seizure free (no single seizure, no aura, Engel/Ia). Seizure free patients increased significantly from initially 56% to 68% in the third cohort. Selective temporal lobe surgeries were more sucessful (69%) than standard 2/3 resections (58%). However also 2/3 resections became more successful over time (from 49% to 66%).

Significant cognitive change after surgery was preferentially observed in memory (33%), 50% of all patients deteriorated in at least in one memory domain. Veral memory losses were greater after left than right temporal resections, and within the left resected group moresignificant after 2/3 anterior temporal lobectomy. Baseline performance, IQ, seizure freedom, side of surgery and age but not cohort affiliation were predictive for cognitive change in the different domains.

Thus while cognitive outome of epilepsy surgery appears quite stable over time, for temporal lobe surgery a positive trend regarding seizure control becomes evident with the introduction of more selective and individual surgeries. This trend appears significantly determined by improved diagnostic facilities. Within the individual surgical groups, the results are very stable over time.

1 G. Widman , 1 C. Helmstaedter , 1 C.E. Elger , 1 C.G. Bien 1University Clinic of Epileptology (Bonn, D)

Objectives: Acquisition of structured clinical data during and after presurgical evaluation of epilepsy patients is helpful for quality assurance, investigation of outcome, and can provide a starting point for clinical research. However, after more than 15 years experience with acquisition of clinical data (and 6 attempts to establish a stable database for presurgical evaluation), we discovered many shortcomings that impede completeness, integrity, and accuracy of data. The presented database tries to overcome these shortcomings.

Methods: To assure completeness, integrity, and accuracy of data, the following steps were taken:

  • 1) 
    Data acquisition is included into the normal clinical workflow on the ward.
  • 2) 
    Clinical departments creating data for presurgical evaluation (e.g., EEG, neuropsychology) feed corresponding data into the database on their own.
  • 3) 
    All datatypists have a direct personal benefit from promptly entering correct data.
  • 4) 
    Using Personal Digital Assistants (PDAs) with a bidirectional synchronisation, patient related datasets can be taken along in the “white coat” of the physician and are therefore available, controlled, and updated during the ward round, shift transfers, and conferences.
  • 5) 
    To disburden datatypists (e.g., physicians) from typewriting, digital dictations (mobile and stationary) as well as speech recognition (stationary) are provided for entering data into the database
  • 6) 
    The database uses an “XML-structure” (extended makeup language) and can easily be integrated into, or synchronized with the local clinical information system.
  • 7) 
    Integration in our clinical information system allows to automatically include the acquired data into the patient report on patient discharge.
  • 8) 
    To simplify data structure and increase long term stability of the database, only a single table is used to store data of an outpatient visit or a stay on the ward.

Results and conclusions: We present a database for presurgical evaluation that stores a minimal data set for each visit. Since data is created during normal clinical workflow with benefits for all data typists and without severe additional expenses, a high level of completeness, integrity, and accuracy of data can be achieved. A pseudonymisation of datasets for a multicenter database is possible. Within 6 month up to now, about 500 data sets have been created.

Abstract

  1. Top of page

Session: MR imaging and post-processing in epileptology

1 W. Serles , 2 D. Prayer 11 University Hospital of Neurology (Vienna, A) ; 2 University Hospital of Radiology (Vienna, A)

Magnetic resonance imaging (MRI) is the method of choice to detect structural lesions in patients with seizures. In general, MRI should be performed in any patient with a first unprovoked seizure, in patients with epilepsy of any age and in children with prolonged febrile seizures. Relative indications for MRI are acute symptomatic seizures (i.e., provoked seizures) if the patient has no neurological deficit and a normal x-ray computed tomography (CT). This has to be decided on an individual basis. In children with clear cut benign focal or idiopathic generalized epilepsy syndromes MRI should only be done if anesthesia is not necessary, except for patients with neurological deterioration or medical intractability. MRI is not required in children with simple febrile seizures.

We propose an age-dependent MRI protocol which is based on brain maturation, incidence of structural lesions within age groups with emphasis on the most prevalent temporal lobe epilepsy syndrome, and time constraints in clinical practice. The basic protocol takes about 30 minutes and can be extended if indicated (e.g., with diffusion weighted imaging, MR angiography or gadolinium enhanced imaging).

Age group 0–2 years (performed under anesthetic): To differentiate grey and white matter adequately in this age group (contrast is different to mature brain because of higher water content) transversal T2- weighted fast spin echo sequences with long repetition (TR-) and echo (TE-) times, a slice thickness of 3 mm and inclined along the anterior-posterior commissure (AC-PC) – plane should be done (e.g., Fast Spin- or Turbo Spin-Echo with TR 3000/TE 120 ms). A transversal T*2- weighted gradient recalled echo sequence (slice thickness 3-4 mm, interslice gap 0–50%) should be added to detect blood or calcifications. Finally, a three dimensional (3D)-T1- weighted volume sequence (with isotropic voxels and a slice thickness of 1-1.5 mm) in the coronal orientation and perpendicular to the hippocampal plane concludes the basic protocol.

Age group 2–50 years: A transversal fluid attenuated inversion recovery (FLAIR) sequence (slice thickness 4 mm, interslice gap 0–50%, inclination along the AC-PC plane) is performed first, followed by a transversal T2* gradient recalled echo sequence. A 3D -T1- weighted volume sequence in the coronal orientation (perpendicular to the hippocampal plane, with isotropic voxels and a slice thickness of maximum 2 mm) covering the whole brain is added. Given the high prevalence of temporal lobe epilepsy, imaging is continued with coronal FLAIR fast spin-echo (slice thickness 3 mm), coronal T2- weighted fast spin-echo (slice thickness 2 mm) and coronal T1 weighted inversion recovery fast spin-echo sequences (perpendicular to the hippocampal plane), covering the brain from the temporal lobe pole to the splenium of the corpus callosum) (‘temporal lobe protocol’).

Age group over 50 years: A transversal FLAIR sequence, followed by a transversal T2* gradient recalled echo sequence and the temporal lobe protocol should be performed.

For more information you can look up the following web pages http://www.kup.at/journals/ilae/leitlinien.html or http://www.kup.at/journals/summary/1510.html

1 J. Spreer 1Institute of Neuroradiology (Freiburg, D)

Hippocampal sclerosis: Hippocampal sclerosis (HS) is the most common cause of focal temporal lobe epilepsy. Histopathologically, it is characterized by the loss of neurons and consecutive gliosis. The MRI findings are the volume loss of the ammonshorn and an increased signal in T2-weighted and FLAIR-sequences. There are several additional diagnostic findings, as the widening of the temporal horn, the atrophy of the fornix and corpus mamillare, and the volume loss of the temporal lobe, which may point to a HS.

Dual pathology: In about 20% HS is associated to another potentially epileptogenic lesion, e.g., a malformation of cortical development. It is essential to diagnose these socalled “dual pathologies”especially in view of epilepsy surgery.

Focal cortical dysplasia: Focal cortical dysplasias (FCD) are disturbances of the physiologic cortical layering caused by an impairment of neuronal proliferation, migration and differentiation during the embryologic development. Until the implement of MRI these malformations could be diagnosed only by autopsy. With the increasing quality of MRI FCD are diagnosed more and more; in patients with focal epilepsies FCD are found in 5 to 25%. Presumably many epilepsies which formerly have been classifed as “cryptogenic” in fact are caused by FCD.

MRI signs of FCD are the broadening of the cortex and the blurring of the transition zone between gray and white matter. In many cases the findings are subtle and require extensive diagnostic investigations, including high-resolution MR-sequences, if possible with high magnetic fields (> 1,5 Tesla). Automated procedures are helpful in ambiguous cases. A special form is the "transmantle-type” FCD. In this type of malformation a bandlike gliotic zone extends from the involved cortex through the deep white matter to the ventricle.

Vascular malformations–cavernomas: Cavernomas are by far the most common cerebral vascular malformations in patients with epilepsy. Cavernomas are venous malformations with a comparatively low risk of bleeding (about 0.5% per year, if there has been no previous bleeding). In MRI, cavernomas show a characteristic central structure with mixed signal intensities, caused by irregularly thrombosed compartments, and a low-signal rim. The signal loss in the periphery is caused by degradation products of hemoglobin (hemosiderin) and is best to be seen in gradient-echo sequences. Diagnostic problems may arise in the early phase after a bleeding, if the cavernoma is masked by the hematoma. In these cases control examinations after the resorption of the hematoma are mandatory.

Epilepsy tumors: This rubric refers to a small group of brain tumors which are extremely rare in the general population but quite frequent in patients with focal epiepsies. Common to these tumors is the preferential localisation in the temporal lobes.

Gangliogliomas/Gangliocytomas (about 40% of the tumors in epilepsy surgery) and dysembryoplastic neuroectodermal tumors (DNET) are tumors of neuronal origin. Histopathologically, DNETs show similarities to focal cortical dysplasias. A characteristic finding is the superficial localisation with arrosion of the adjacent tabula interna, indicative of an extremly slow growth. Calcifications, cysts, and focal contrast enhancement may be present in both entities. Pleomorphic xanthoastrocytoma may mimick a glioblastoma multiforme in imaging studies. However, prognosis is much more favourable in the former.

1 H. Urbach 1University of Bonn (Bonn, D)

Hippocampal sclerosis, long-term epilepsy associated tumours (LEATs), focal cortical dysplasias, vascular malformations and gyral scars contribute to the vast majority of epiletogenic lesions in large epilepsy surgery series. However, even within these subgroups new entities have been described recently (e.g., angiocentric neuroepithelial tumours), the spectrum of other lesions is likely to be extended (e.g., focal cortical dysplasia IIb), some lesions are still incompletely understood. Furthermore, high-resolution (3T) MRI may demonstrate overlappings between formerly seprarated lesions (e.g., holoprosencephalies, polymicrogyrias, and heterotopias) and subtle imaging findings that are more likely the consequence of seizures (e.g., thalamic seizure spread). Finally, recent moleculargenetic findings point to epilepsy associated diseases with a broad spectrum of imaging findings (e.g., Alpers disease).

References:

1. Lellouch-Tubiana A, Boddaert N, Bourgeois M, et al. (2005) Angiocentric Neuroepithelial Tumor (ANET): A new Epilepsy-related clinicopathological entity with distinctive MRI. Brain Pathol 15:2816. 2. Barkovich AJ, Kuzniecky RI, Jackson GD, Guerrini R, Dobyns WB. (2005) A developmental and genetic classification for malformations of cortical development. Neurology 65:187387. 3. Mochida GH, Walsh CA. (2004) Genetic basis of developmental malformations of the cerebral cortex. Arch Neurol 61:63740. 4. Tzoulis C, Engelsen BA, Telstad W, et al. (2006) The spectrum of clinical disease caused by the A467T and W748S POLG mutations: a study of 26 cases. Brain 129:168592.

1 T. Hammen , 1 F. Kerling , 1 M. Dölken , 1 A. Doerfler , 1 I. Bluemcke , 1 H. Stefan 1University Hospital Erlangen (Erlangen, D)

1H-MR Spectroscopy (1H-MRS) is a noninvasive method for detecting various brain metabolites containing N-acetylaspartate (NAA), Cholin (Cho), Creatin (Cr) GABA and Glutamat (Glx). It becomes a clinical diagnostic tool for evaluating epilepsies (1-4).

The following topics are discussed:

  • 1) 
    1H-MRS provides additive information for lateralizing the affected hemisphere especially in patients with temporal lobe epilepsy (TLE) that remain without pathological findings in high resolution MRI (5-7). Up to 30% of patients with TLE remain without remarkable changes in MRI. In this patient group metabolite alterations in 1H-MRS relate to focal epileptic activity recorded by intensive video EEG monitoring. In up to 66% of TLE patients that are graded as unilateral by Video-EEG monitoring NAA in hippocampal structures is reduced ipsilateral to the affected hemisphere.
  • 2) 
    Results in 1H-MRS help to differentiate between mesial and lateral, neocortical TLE (8-10). Patients who were classified as mesial TLE by preoperative diagnostics including neurological and neuropsychological examinations, video-EEG monitoring and high resolution MRI demonstrated pronounced metabolite alterations in temporomesial, hippocampal structures whereas patients classified as lateral, neocortical TLE demonstrated no pathologic metabolite alterations in hippocampal structures. The absence of spectroscopic differences in NE may help to distinguish neocortical from mesial TLE.
  • 3) 
    Metabolite alterations in preoperative 1H-MRS are able to predict postoperative outcome in patients with TLE (11-14). In these studies TLE patients mainly diagnosed with hippocampal sclerosis were scheduled for epilepsy surgery. Metabolite spectra were generated preoperatively from hippocampal and parahippocampal regions ipsi- and contralateral to the hemisphere scheduled for lesionectomy. In our study the NAA/Cho ratios of the affected hemisphere were reduced compared with results of contralateral hemisphere or normal controls. The results indicate that changes in 1H-MRS classified as bilateral or contralateral to the operated side were predictors for poor postoperative seizure outcome whereby 92.3% of patients who became seizure free had no severe bilateral or contralateral metabolic deviations. Comparable results have been reported by Kuzniecky et al. and Li et al. (12,13). In the study of Kuzniecky et al.. the postoperative outcome was classified as seizure free or not seizure free after 24 months. As in our study worse postoperative outcome was associated with contralateral or bitemporal abnormalities of 1H-MRS spectra. The results show that ipsilateral metabolite abnormalities which are in unison to the EEG focus stand for a good prognostic outcome whereas contralateral or widespread, bilateral abnormal metabolite spectra are associated with a bad postoperative outcome. The same results were investigated in MRI negative patients by Suhy et al. (15). The authors performed 1H-MRS in 15 TLE patients without pathological findings in MR imaging who underwent seizure surgery. The nonseizure-free patients (NSF) demonstrated lower ipsilateral hippocampal N-acetylaspartate (NAA)/(Cr+Cho) z-scores compared to contralateral scores. The NSF ipsilateral z-scores were lower than the seizure-free patients' (SF) ipsilateral z-scores. Similarly, NSF contralateral scores were lower than contralateral SF. These findings suggest that metabolite alterations predict surgical outcome in patients with TLE without evidence of mesial temporal sclerosis on MRI.
  • 4) 
    In addition to predicting postoperative outcome, Proton MR Spectroscopy is used to evaluate conservative treatment schedules in epilepsies. The effects of vigabatrin on brain gamma-Aminobutyric acid (GABA) in patients with epilepsies and the monitoring of responder characteristics by 1H-MR-spectroscopy is a topic of recent studies (16-19). Mueller et al.. investigated responder profiles in patients who were under the treatment of vigabatrin (VBG) by analysing GABA to a Cr ratio (GABA+/Cr) by short term 1H-MRS. Measurements were performed before (baseline) and after a titration period of 1 month. A third measurement followed a maintenance period of 3 months. The authors were able to correlate responder groups depending on the therapeutic efficacy of VGB to changes in the GABA+/Cr signal. The nonresponders which showed no increase in seizure reduction under the treatment of vigabatrin showed no significant change in the GABA+/Cr signal during the treatment compared with baseline. The full responders had a significant increase of the GABA+/Cr signal during the whole treatment phase and a lower ipsilateral level at baseline. The partial responders had also a lowered ipsilateral GABA+/Cr signal before treatment which increased at the beginning of the treatment and decreased when seizures started again under medication.

References:

1. Hammen T, Stefan H, Eberhardt KE, BH WH, Tomandl BF. (2003) Clinical applications of 1H-MR spectroscopy in the evaluation of epilepsies–what do pathological spectra stand for with regard to current results and what answers do they give to common clinical questions concerning the treatment of epilepsies? Acta Neurol Scand 108(4):22338. 2. Kuzniecky R. (1997) Magnetic resonance and functional magnetic resonance imaging: tools for the study of human epilepsy. Curr Opin Neurol 10(2):8891. 3. Achten E, Boon P, Van De Kerckhove T, Caemaert J, De Reuck J, Kunnen M. (1997) Value of single-voxel proton MR spectroscopy in temporal lobe epilepsy. AJNR Am J Neuroradiol 18(6):11319. 4. Cendes F, Andermann F, Dubeau F, Arnold DL. (1995) Proton magnetic resonance spectroscopic images and MRI volumetric studies for lateralization of temporal lobe epilepsy. Magn Reson Imaging 13(8):118791. 5. Hammen T, Kerling F, Schwarz M, Stadlbauer A, Ganslandt O, Keck B, et al. (2006) Identifying the affected hemisphere by (1)H-MR spectroscopy in patients with temporal lobe epilepsy and no pathological findings in high resolution MRI. Eur J Neurol 13(5):48290. 6. Hajek M, Dezortova M, Komarek V. (1998) 1H MR spectroscopy in patients with mesial temporal epilepsy. Magma 7(2):95114. 7. Woermann FG, McLean MA, Bartlett PA, Parker GJ, Barker GJ, Duncan JS. (1999) 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 Neurol 45(3):36976. 8. Hammen T, Stefan H, Pauli E, Schafer I, Huk W, Tomandl B. (2003) 1H-MR spectroscopy: a promising method in distinguishing subgroups in temporal lobe epilepsy? J Neurol Sci 215(1-2):215. 9. Vermathen P, Ende G, Laxer KD, Knowlton RC, Matson GB, Weiner MW. (1997) Hippocampal N-acetylaspartate in neocortical epilepsy and mesial temporal lobe epilepsy. Ann Neurol 42(2):1949. 10. Li LM, Caramanos Z, Cendes F, Andermann F, Antel SB, Dubeau F, et al. (2000) Lateralization of temporal lobe epilepsy (TLE) and discrimination of TLE from extra-TLE using pattern analysis of magnetic resonance spectroscopic and volumetric data. Epilepsia 41(7):83242. 11. Stefan H, Pauli E, Eberhardt KE, Schafer I, Hopp P, Huk WJ. (2000) [MRI spectroscopy, T2 relaxometry, and postoperative prognosis in cryptogenic temporal lobe epilepsy]. Nervenarzt71(4):2827. 12. Li LM, Cendes F, Antel SB, Andermann F, Serles W, Dubeau F, et al. (2000) Prognostic value of proton magnetic resonance spectroscopic imaging for surgical outcome in patients with intractable temporal lobe epilepsy and bilateral hippocampal atrophy. Ann Neurol 47(2):195200. 13. Kuzniecky R, Hugg J, Hetherington H, Martin R, Faught E, Morawetz R, et al. (1999) Predictive value of 1H MRSI for outcome in temporal lobectomy. Neurology 53(4):6948. 14. Eberhardt KE, Stefan H, Buchfelder M, Pauli E, Hopp P, Huk W, et al. (2000) The significance of bilateral CSI changes for the postoperative outcome in temporal lobe epilepsy. J Comput Assist Tomogr 24(6):91926. 15. Suhy J, Laxer KD, Capizzano AA, Vermathen P, Matson GB, Barbaro NM, et al. (2000) 1H MRSI predicts surgical outcome in MRI-negative temporal lobe epilepsy. Neurology 58(5):8213. 16. Mueller SG, Weber OM, Duc CO, Meier D, Russ W, Boesiger P, et al. (2003) Effects of vigabatrin on brain GABA+/Cr signals in focus-distant and focus-near brain regions monitored by 1H-NMR spectroscopy. Eur J Neurol 10(1):4552. 17. Petroff OA, Behar KL, Rothman DL. (1999) New NMR measurements in epilepsy. Measuring brain GABA in patients with complex partial seizures. Adv Neurol 79:93945. 18. Petroff OA, Hyder F, Collins T, Mattson RH, Rothman DL. (1999) Acute effects of vigabatrin on brain GABA and homocarnosine in patients with complex partial seizures. Epilepsia 40(7):95864. 19. Petroff OA, Hyder F, Mattson RH, Rothman DL. (1999) Topiramate increases brain GABA, homocarnosine, and pyrrolidinone in patients with epilepsy. Neurology 52(3):4738.

1 S. Knake 1Philipps-Universiy Marburg (Marburg, D)

MRI is an important diagnostic tool in the evaluation of epilepsy patients. In patients with medically refractory focal epilepsies, MRI has an impact on predicting the postsurgical outcome [Knake 2005a]. Many efforts are made to improve structural imaging, either by improving image quality (signal-to-noise ratio) or by using innovative quantitative image analysis techniques. To improve routine clinical scanning, high-field, high-resolution surface-coil MRI is frequently used [Knake 2005b].

The use of novel semi-automated quantitative image analysis techniques has the potential to improve lesion detection to more accurately assess lesion burden, to characterize cortical abnormalities and to determine the location and extent of associated cortical and deep grey nuclei involvement.

Semi-automated techniques such as voxel based morphometry (VBM), cortical thickness analysis and automated methods for identification and volumetry of subcortical structures have been developed and have the potential becoming part of a routine clinical assessment in future (Huppertz, 2005; Bernasconi, 2004; Dale, 1999; Fischl, 2000).

Diffusion imaging has been used interictally to study patients with focal epilepsies and to postictally identify areas of transient postictal changes (Szabo, 2005; Wieshmann, 2000; Wieshmann, 1999). Novel techniques such as Arterial Spin Labelling (ASL) carry the potential to delineate areas of ictal hyperperfusion / interictal hypoperfusion without using contrast agents (Wolf, 2001). ASL magnetically labels blood in the main arteries and uses the magnetic properties of blood as an endogenous contrast agent. ASL might as well be used to delineate the eloquent cortex in future.

With increasing interest in epileptogenesis and methods of seizure propagation, diffusion tensor imaging may yield useful information about changes in white matter organization (Grant, 2005; Grant, 2004). Diffusion tensor imaging (DTI) is a new technique that might be used to evaluate white matter integrity. Like DWI, DTI uses the diffusion properties of water in different in the brain. DTI provides information about the rate, magnitude, and directionality of water diffusion in the brain and is influenced by microstructural factors such as myelin and other fiber components. DTI may be used to delineate areas of microstructural white matter changes that extend beyond the visible cortical abnormality in patients with cortical dysplasia and with temporal lobe epilepsy (Eriksson et al., 2001; Rugg-Gunn et al., 2001).

The discussed postprocessing techniques are promising but still investigational and require validation and pathological correlation before their clinical potential can be realized in future.

References:

Bernasconi N. Neuroimage. 2004;23:717723. Dale AM. Neuroimage. 1999;9:179194. Eriksson SH. Brain 2001;124:617626. Fischl B. PNAS 2000;97:1105011055. Grant PE. Epilepsia 2004;45 Suppl 4:416. Grant PE. Epilepsia 2005;46 Suppl 7:714. Huppertz HJ. Epilepsy Res 2005;67:3550. Knake S. Neurology 2005a;65:10261031. Knake S. J Clin Neurophysiol 2005b;494502. Rugg-Gunn FJ. Brain 2001;124:627636. Szabo K. Brain 2005;128:13691376. Wieshmann U. MRI 1999;17:12691274. Wieshmann U. JNNP 2000;68:501503. Wolf RL. AJNR 2001;22:13341341.

Abstract

  1. Top of page

Session: Cognition and epilepsy: from experimental studies to the patient

1 T. Kirschstein 1University Rostock (Rostock, D)

Patients with chronic temporal lobe epilepsy often suffer from persisting cognitive dysfunction in terms of declarative memory deficits. The fundamental molecular mechanisms of learning and memory in the central nervous system (CNS) have been recognized as different forms of synaptic plasticity. Depending on the intensity of the neuronal activity, synaptic strength can be potentiated or depressed, termed long-term potentiation (LTP) and long-term depression (LTD), respectively. Moreover, synaptic plasticity itself is modulated by the prior history of neuronal activity resulting in a higher-order plasticity termed metaplasticity. Thus, we asked how chronic epileptic seizures can influence synaptic plasticity in the chronically epileptic hippocampus. Following a prolonged status epilepticus induced by pilocarpine the rats developed spontaneous epileptic seizures, and long-term extracellular recordings were performed in the hippocampal brain slice. First, we studied NMDA receptor-dependent synaptic plasticity in the dentate gyrus. Whereas LTP in the medial perforant path was unaltered, we found a significant impairment of LTP in the lateral perforant path of epileptic rats. This lack of LTP could be rescued by D-serine bath application indicating NMDA receptor dysfunction in the epileptic lateral perforant path. Then we focused on NMDA receptor-dependent synaptic plasticity at the Schaffer collateral-CA1 synapses. Whereas LTD was unaltered in chronic epilepsy, we found an unexpected increase of LTP. As this discrepancy might be due to a differential expression of the most abundant NMDA receptor subtypes NR2A and NR2B, we went on to block these subtypes pharmacologically. We found that application of low micromolar zinc (1-10 μM), which primarily inhibits NR2A, virtually abolished LTP in controls, whereas epileptic rats still showed significant LTP. With a higher zinc concentration (100 μM) which blocks NR2A and NR2B, LTP was inhibited in both groups. In contrast, when a specific NR2B blocker was used, LTP remained inducible in controls, but was completely lost in epileptic rats. These results suggested a dominant role of NR2B in LTP induction in rats that have experienced status epilepticus. In conclusion, chronic epileptic seizures lead to a persistent alteration of synaptic plasticity in the hippocampus that can be seen as a kind of seizure-induced metaplasticity. These metaplastic changes could be in part involved in the cognitive impairment seen in many patients. A more precise understanding of the underlying mechanisms will be essential to develop new therapeutic strategies.

1 M.P. Richardson 1King's College London (London, UK)

Background: Temporal lobe epilepsy (TLE), usually caused by hippocampal sclerosis (HS), is a common disease which frequently fails to respond to antiepileptic drugs (AEDs). Hence, surgical removal of the epileptogenic zone, usually including the mesial temporal lobe (MTL), is an important treatment option and is usually curative. As a direct result of removal of MTL, however, memory is often significantly worse following surgery. Estimation of the risk for memory decline following surgery, and its severity, is therefore an important part of planning surgery in TLE. At the present time, there is no accurate tool established for this purpose. Functional magnetic resonance imaging (fMRI) of memory has recently shown great promise as a tool to predict memory decline following surgery in TLE, but considerable work is required to implement memory fMRI as a reliable clinical tool. Furthermore, although fMRI is widely used in neuroscience research, there is no existing methodology to facilitate a rigorous translation of “research-based” fMRI tools to the clinical domain.

Method: We studied 30 patients with mesial temporal lobe epilepsy (mTLE) and left HS, 12 of whom subsequently underwent surgery, and 13 normal control subjects. The patients who had surgery underwent neuropsychometric evaluation pre- and postoperatively. All subjects underwent a verbal memory encoding event-related fMRI study. First, we compared the pattern of verbal memory encoding activity in the group of patients with the group of normal controls, to establish typical patterns of activity for this task in normals and left HS patients. Secondly, we examined the group of patients who underwent surgery, to find the brain region(s) where preoperative memory encoding fMRI activity correlated with postoperative memory decline. This was assessed at the voxel level, using established methodology for examining group fMRI data. We also compared the predictive value of fMRI with established tools (structural MRI, neuropsychometry). Thirdly, we examined the data for each individual patient who underwent surgery, to establish a methodology to analyse data on a case-by-case basis, as would be required in a clinical setting. This approach used a new methodology to examine each case, to establish robust quantitative parameters of memory encoding activity which could be used prospectively to predict postoperative memory decline. Finally, in a qualitative manner only, we compared our approach with those of other groups who have also successfully implemented memory fMRI as a clinical tool in TLE.

Results: Verbal memory encoding involved activation of left hippocampus in normals, but was associated with reorganisation to right hippocampus and parahippocampal gyrus in the patients. In the patients who underwent surgery, a voxel-level analysis of the group showed that hippocampal memory encoding activity, measured from preoperative fMRI, provided strong prediction of postoperative memory change. Multiple regression analyses showed that fMRI provided the strongest independent predictor of memory outcome after surgery, compared with structural MRI and preoperative neuropsychometry. Examining individual patient data, visual inspection of individual patient activation statistic maps revealed noisy data that did not afford visual interpretation. Stepwise multiple regression analysis of predetermined regions of interest revealed left hippocampal activity was the strongest predictor of postoperative verbal memory outcome; greater left hippocampal activity predicted a greater postoperative decline in memory.

Conclusions: Preoperative memory fMRI provides a promising tool to evaluate hippocampal functional integrity prior to resective temporal lobe surgery in TLE. It is now essential to undertake studies to determine the clinical role of memory fMRI versus the intracarotid amobarbital test, and also to undertake studies comparing the different approaches to memory fMRI that have shown a clinical role in TLE.

1 J. Fell , 1 C.E. Elger 1University of Bonn (Bonn, D)

Background: Human declarative memory, i.e., the consciously accessible long-term memory for events and facts, crucially depends on two structures within the medial temporal lobe (MTL), the rhinal cortex and the hippocampus. But there had been no direct evidence for an interaction of both structures during memory formation. Transient coupling of neural assemblies may be accomplished by phase synchronisation of oscillatory activity, in particular gamma activity, i.e., EEG activity in the frequency range above 30 Hz.

Methods: In a group of 9 presurgical patients with unilateral MTL-epilepsies the event-related EEG was recorded with depth-electrodes during a word-memory task using common nouns as stimuli (1,2). In another group of 10 patients a similar memory task was performed contrasting common nouns with rare nouns (high versus low frequency of occurrence) (3). We compared the EEG-responses of the nonpathological MTL during word encoding of later remembered and forgotten words (subsequent memory effect). To address the question, whether the reduced facility to memorize dreams is associated with altered electrophyiological characteristics during sleep, all-night EEG was recorded in a group of 8 patients with depth- and scalp-electrodes (4). Moreover, 12 patients were awakened during REM phases and dream reports were inquired (5). Phase synchronization and spectral coherence within different frequency bands were quantified as measures of ongoing and stimulus-related neural connectivity between MTL substructures.

Results: We found that successful memory formation is accompanied by an initial stimulus-related increase of phase synchronization of gamma activity between rhinal cortex and hippocampus and a later decrease (1). These memory-related synchronisation changes are interindividually correlated with increases of rhinal-hippocampal theta (4-7 Hz) coherence (2). However, a pronounced broad-band rhinal-hippocampal coupling, including the gamma range, is only observed for words with a high frequency of occurrence (common words), but not for rare words (3). Compared to the waking state ongoing rhinal-hippocampal coherence decreases during sleep, most pronounced within the gamma-band (4). Finally, we detected that ongoing rhinal-hippocampal coherence is more than twice as large for patients with good dream memory compared to patients with poor dream memory (5).

Conclusion: Synchronisation of gamma oscillations may accomplish fast coupling and decoupling processes, which initiate and later terminate the information transfer between rhinal cortex and hippocampus (6). Moreover, early changes in synaptic plasticity may be triggered by synchronized gamma activity (7). Coherent EEG oscillations in the lower frequency range probably reflect a slowly modulated connectivity between rhinal cortex and hippocampus, which supports processes in the gamma range. However, a pronounced broad-band coupling, including the gamma range, may only occur, when significant semantic information is being processed within rhinal cortex, as is the case for common words. The reduced ongoing rhinal-hippocampal coupling during sleep may represent an indirect electrophysiological correlate of the diminished ability to encode memories during sleep. In accordance with this interpretation, ongoing rhinal-hippocampal connectivity was found to predict the memorization of dreams. Alltogether, these findings support the idea that ongoing, as well as stimulus-related rhinal-hippocampal synchronisation processes are crucial for successful declarative memory formation.

References: 1. Fell J, Klaver P, Lehnertz K, Grunwald T, Schaller C, Elger CE, Fernández G. (2001) Human memory formation is accompanied by rhinal-hippocampal coupling and decoupling. Nat Neurosci 4:1259–1264.

2. Fell J, Klaver P, Elfadil H, Schaller C, Elger CE, Fernández G. (2003) Rhinal-hippocampal theta coherence during declarative memory formation: interaction with gamma synchronisation?Eur J Neurosci 17:1082–1088.

3. Fell J, Fernández G, Klaver P, Axmacher P, Mormann F, Haupt S, Elger CE. (2006) Rhinal-hippocampal coupling during declarative memory formation: dependence on item characteristics. Neurosci Lett 407:37–41.

4. Fell J, Staedtgen M, Burr W, Kockelmann E, Helmstaedter C, Schaller C, Elger CE, Fernández G. (2003) Rhinal-hippocampal EEG coherence is reduced during human sleep. Eur J Neurosci 18:1711–1716.

5. Fell J, Fernández G, Lutz MT, Kockelmann E, Burr W, Schaller C, Elger CE, Helmstaedter C. (2006) Rhinal-hippocampal connectivity determines memory formation during sleep. Brain 129:108–114.

6. Fell J, Klaver P, Elger CE, Fernàndez G. The interaction of rhinal cortex and hippocampus in human declarative memory formation. Rev Neurosci 13:299–312.

7. Axmacher N, Mormann F, Fernández G, Elger CE, Fell J. Memory formation by neuronal synchronisation. Brain Res Rev 52:170–182.

1 M. Seeck 1University Hospital of Geneva (Geneva, CH)

Decrease of cognitive functions is a well known, unfavorable side effect of chronic epilepsy, due to the active epileptic focus and/or concomitant antiepileptic drug treatment. In the context of presurgical epilepsy evaluation, procedures using neuropsychological testing are applied to accomplish several tasks: to determine overall functioning, to localize the focus, or to determine vital cortex which needs to be spared when resecting cerebral tissue. More recent approaches which were developed to achieve these tasks are: fMRI based methods and postictal neuropsychological testing. They help to avoid invasive procedures, such as the intracarotid amytal test (IAT), in many patients. In a recent own study, postictal memory assessment was found superior in lateralizing memory performance compared to the interictal testing in patients with temporal lobe epilepsy. While there is a large body of fMRI studies on language lateralization, fMRI memory studies are few and lateralization of verbal/nonverbal memory with fMRI is less well established. Hopefully, lateralization and localization of cognitive functions will be carried out solely with non-invasive techniques, in particular since there is evidence that the corticography and IAT may not be the gold standard in every patient.

Abstract

  1. Top of page

Oral session 2

1 S. Ruf , 1 I. Krägeloh-Mann , 1 M. Wolff 1University Children's Hospital (Tubingen, D)

Purpose: For many years ACTH and steroids have been used in the treatment of drug resistant childhood epilepsies, especially in children with West syndrome. However, severe side effects have been reported using ACTH and continuous treatment with steroids. We present efficacy and adverse effects of an alternative scheme using pulsatile steroids per os in outpatients.

Patients and methods: 20 patients with drug resistant epilepsies aged five months to 16 years were selected: West syndrome N = 11, Continuous spike and waves during slow sleep (CSWS) N = 5 (Landau- Kleffner syndrome N = 2, atypical benign focal epilepsy N = 3), focal epilepsy with secondary bisynchrony N = 2, and childhood absence epilepsy N = 2. The children were treated using a pulsatile steroid scheme per os (25mg/kg/d for three days once a week for four weeks followed by a variable tapering phase). Epilepsy and EEG data as well as adverse side effects were documented before treatment, after four weeks and at the end of the therapy.

Results: From the 11 children with West syndrome, six became seizure-free, and hypsarrhythmia disappeared in four of them. In three children seizure frequency was reduced > 50%. In the children with CSWS, EEG improved markedly in one child and slightly in two. In the two children with focal epilepsy and secondary bisynchrony no effect occurred. The two children with absence epilepsy became seizure free during follow up. There were no serious side effects in all of the children.

Conclusion: In this cohort of children with extremely pharmacoresistant epilepsies the pulsatile steroid therapy was partly effective. Severe side effects did not occur. Therefore, we suggest a broader and earlier use of steroids in pharmacoresistant childhood epilepsies.

1 G. Ramantani , 1 O. Wunderlich , 1 D. Friebel , 1 H. Ikonomidou 1Technical University Dresden (Dresden, D)

Encephalocraniocutaneous lipomatosis (ECCL), also known as Fishman syndrome, is a rare congenital neurocutaneous disorder comprising unilateral cranial lipomas, lipodermoids of the eye and brain abnormalities. Epileptic seizures and a consequent significant developmental delay are usually present. We report the clinical, electroencephalographic and neuroimaging findings over the years in a presently 6-year-old girl diagnosed with ECCL in the first months of life. The patient showed infantile spasms at the age of two months and remained seizure free over 2 years under vigabatrin treatment. At the age of 18 months, a mild spastic hemiparesis was diagnosed, which improved markedly with physical therapy. The cognitive development was estimated in the lower normal range. The cutaneous and ocular lesions remained unchanged over time. The EEG initially showed a burst-suppression pattern in the neonatal period, which evolved into an asymmetric activity with a continuous spike wave focus corresponding to the cerebral malformation. The MRIs showed a hemimegalencephaly with an enlargement of the lateral ventricle on the affected side, widening of the subarachnoid spaces, a lack of normal insular opercularisation, dysplastic cortex in the temporoparietooccipital region, corticopial calcifications, arachnoid cysts of the middle cranial fossa and thinning of the corpus callosum. The progressive enlargement of the lateral ventricle in the first years of life showed spontaneous regression, so that no ventriculoperitoneal shunt was necessary. Functional magnetic resonance imaging revealed transfer of motor functions to the unaffected hemisphere, in accordance to the hypothesis of functional reorganization. Furthermore, the results of functional imaging, including PET, SPECT and MR-Spectroscopy pointed towards an increased cerebral blood flow in the affected hemisphere, possibly due to intermittent ictal events. These features were consistent with ECCL. Most children with ECCL are reported to have a significant developmental delay; however in our case seizure control is associated with a considerable improvement in developmental outcome.

1 T. Gerstner , 1 D. Büsing , 1 E. Longin , 1 S.A. König 1University Children's Hospital (Mannheim, D)

Introduction: Valproic acid (VPA) is a commonly prescribed medication for generalized and focal epilepsy, migraine, neuropathic pain and bipolar disorder. The common side effects associated with VPA are typically benign, less common but more serious adverse effects may occur. These include hepatotoxicity, hyperammonemic encephalopathy, coagulation disorders and pancreatitis.

Method: We mailed a questionnaire to all 1200 members of the “German Section of the International League against Epilepsy,” containing questions about VPA-induced side effects from the years 1994 to 2003.

Results: We received 450 answers (37.5%) to our questionnaire. To our surprise 254 (56.4%) of the physicians never noticed any severe side effects of VPA.

31 cases of VPA associated-hepatotoxicity were seen by German physicians from 1994 to 2003, including nine cases of fatal liver failures and 22 reports of reversible hepatopathies, 44 cases of VPA-associated pancreatitis were reported, 27 cases of encephalopathy and 42 patients suffered of a VPA-associated coagulopathy. Additionally, one case of change of the hair colour and 10 events of massive weight gain (>15%) were reported.

Conclusion: In Germany 208,320 persons are under VPA-medication (date: November 2005). Due to this amount of patients, VPA-associated side effects should be recognized observant. Our query showed more than 50% of all questioned physicians never recognized any side effect, suggesting a very good tolerance of this drug. On the other hand, sever side effects (hepatotoxicity, pancreatitis and encephalopathy) are more likely as the literature supposed. Concomitant the outcome seemed to be better, if the side effect was recognized in time.

1 J. Dobesberger , 1 G. Brössner , 1 K. Seppi , 1 G. Walser , 1 R. Ehling , 1 I. Unterberger , 1 E. Schmutzhard , 1 E. Trinka 1Medical University Innsbruck (Innsbruck, A)

Purpose: Status epilepticus (SE) as well as seizure clusters are emergencies which require immediate and efficacious treatment. To date, intravenous (IV) benzodiazepines (BZD) as well as IV phenytoin (PHE) and IV valproic acid (VPA) are widely used antiepileptic drug (AED) therapies in these situations. This study aimed to investigate the effectiveness and tolerability of IV levetiracetam (LEV) in treatment of seizure clusters and SE.

Methods: All patients treated with IV LEV between August and December 2006 were retrospectively enrolled in our study. Indications for treatment, dosage, responsiveness and side effects were evaluated.

Results: 13 patients (six women and seven men, median age 60 ± SD 25 years) were consecutively treated with IV LEV. Indications for IV LEV were (1) nonconvulsive SE refractory to standard first-line therapy (i.e., BZD, PHE and VPA) in four patients, (2) IV LEV administration after successful treatment of nonconvulsive SE (i.e., BZD) in three patients and (3) seizure clustering triggered by AED reduction during video-EEG monitoring in six patients (rapid titration group). 10/13 patients suffered from focal epilepsies before study enrolment; 3/13 had no previous history of seizures. Dosage evaluation revealed that in 11 patients a bolus of 1000 mg IV LEV was initially administered and the two other patients were initiated with 500 mg IV LEV. All patients received oral LEV after treatment with IV LEV.

Outcome: Nonconvulsive SE was terminated with IV LEV in 3/4 patients after the initial dosage, one patient was resistant to all therapeutic advances. In the other 9 patients, IV LEV was effective in terminating SE/seizure clusters and no relapses occurred following treatment. No side-effects were observed.

Conclusions: These data suggest that IV LEV can be used as a safe alternative to standard AED therapies for acute treatment of seizure emergency situations, although larger studies are required to confirm these findings.

1 A. Reinshagen , 1 M. Holler , 1 H. Fritz , 1 F. Hoffmann 1Martha-Maria Hospital Halle-Dölau (Halle/Saale, D)

Purpose: Levetiracetam (LEV) is recently available as an intravenous (IV) formulation in Germany for use in patients unable to receive oral therapy. The use in status epilepticus (SE) is not yet reported; are there implications to improve status epilepticus therapy.

Method: We report on the use of IV LEV in six patients with different types of refractory status epilepticus, well documented why not to receive other anticonvulsive medication for status epilepticus.

Results: We present 1) the case of a of focal myoclonic SE due to a cortical ischaemic infarction in a 52 years old female with severe hepatopathy with coagulation disturbances, 2) the case of a complex-focal nonconvulsive status epilepticus in a male patient aged 41 years with symptomatic epilepsy and multiple sclerosis, who was on treatment with lamotrigine (LTG), 3) the case of a subtle status due to thrombopathia in a 80-year-old female, 4) the case of a 66-year-old male patient in persistent vegetative state with EEG-proved generalized SE on treatment with LTG, 5) a case of an ICU female patient, 81-year-old, with persistent myoclonic seizures and 6) a multimorbid ICU male patient aged 36 years with presumed vegetative state, but burst-suppression pattern EEG.

All of them were insufficiently treated with benzodiazepins and other anticonvulsive drugs before the decision for IV LEV (last case high dose LEV via gastrostoma) was made.

In 5 of 6 cases IV LEV stopped or helped stopping the refractory status with more or less well documented time courses, patient 1 after 30 min, patient 6 after at least 2 hrs. Patient 3 died without achieving consciousness, the patient (6) with high-dose LEV (6 g) experienced optical hallucinations after a long lasting coma.

Conclusion: There are poor data for SE therapy, for therapy of refractory status an aggressive treatment with sedation seems to be necessary. IV LEV appears to be a promising treatment for refractory status epilepticus without sedation and with no need for ICU as shown in the reported cases. There is need for more data and especially the time course of anticonvulsive effect of IV LEV in SE.

1 E. Gelpi , 1 L. Urak , 1 T. Czech , 1 H. Mayer , 1 G. Patzl , 1 M. Freilinger , 1 A. Dressler , 1 D. Prayer , 1 G. Kasprian , 1 J. Hainfellner , 1 M. Feucht 1Medical University of Vienna (Vienna, A)

Objective: To correlate clinical and radiolgical parameters with neuropathological findings of children with intractable seizures associated with abnormalities of cortical development.

Material and methods: Children <18 years who underwent surgery for refractory epilepsy associated with abnormalities of cortical development and other nontumorous lesions were selected for this study. Structural and functional brain imaging was performed in all patients. Clinical follow-up data were available for all patients. Neuropathological workup of surgical specimens was blinded to clinical and outcome data. Malformations caused by abnormalities of cortical development (MCD) were classified according to Palmini et al.

Results: Of a total of 93 children operated at the Medical University of Vienna, Austria, between 1996 and 2006, 21 (13 males, 8 females) met inclusion criteria. Mean age at onset of epilepsy was 2.3 years (range 1 day-9 years). Mean age at surgery was 7 years and 2 months (median 5y6m; range 3 months-23.4 years). Mean interval between onset of seizures and epilepsy surgery was 57.5 months (median 33 months; range 2-259 months). Lobar distribution was as follows: 8 temporal, 4 frontal, 1 postcentral, 2 occipital, 1 parietooccipital. 3 children showed hemimegalencephaly, 1 schizencephaly and polymicrogyria and 1 patient hemiatrophy. Surgery types comprised cortical resection, extended lesionectomy, and functional hemispherotomy. Neuropathological findings confirmed radiological interpretation in 18/21 cases. Histological diagnoses comprised: 9 focal cortical dysplasia (FCD) type II B according to Palmini's classification, 3 FCD type II A, 1 FCD IA, 1 FCD 1B, and 2 cases with mild MCD type II. Additionally, 2 cases showed diffuse astrogliosis and 3 cases normal grey matter. One patient had additionally polymicrogyria. Dual pathology (FCD type IIB and hippocampal sclerosis) was observed in only 1 case. Seizure free status after operation was achieved in 81% of patients. Postoperative outcome was class 1 in 17, class 3 in 2 and class 4 in 2/21 children according to Wieser's classification. Postoperative complications appeared only in 1 patient who developed an hygroma.

Conclusion: Epilepsy surgery in children with abnormalities of cortical development achieved excellent outcomes in term of seizure control and neurodevelopmental performance in the Austrian patient cohort. Early operation from onset of seizures and extended cortical resections yielded better outcome.

1 H. Clusmann , 1 O. Schijns , 1 C.G. Bien , 1 M. von Lehe , 1 H. Urbach , 1 M. Majores , 1 T. Kral , 1 J. Schramm , 1 C.E. Elger 1University of Bonn (Bonn, D)

Purpose: With the evolution of high-resolution MR-techniques signal alterations mainly in the area of the temporal pole are detected more frequently in patients with temporal lobe epilepsy (TLE). The impact of these findings, often accompanied by hippocampal sclerosis, is unclear.

Methods: We identified 370 patients undergoing different subtypes of temporal resection for TLE between 2000 and 2006 in a prospectively collected epilepsy data base. Radiological and histological reports were screened for the presence of temporo-polar gray-white matter abnormalities (GWMA).

Results: Abnormalities of “gray- and white-matter differentiation” were identified in 75 patients (rate 20%). Mean seizure onset was 7 (0,5-28) yrs, compared to 12 years (0.5-42) yrs in a matched control group without this feature, however preoperative epilepsy duration was similar. All patients underwent non-invasive video-EEG-monitoring and 28 patients (37%) had invasive EEG-monitoring with chronically implanted electrodes. Presurgical evaluation resulted in one of the following strategies: Of 58 patients with hippocampal sclerosis and presence of GWMA 29 patients underwent selective amygdalohippocampectomy (SAH), without resection of the area of GWMA. Another 29 underwent tailored temporal resections including the GWMA and the hippocampus. Seizure control was virtually equal (72–79% Engel Class I outcomes) with the two different approaches and compared to the matched controls without temporo-polar GWMA. In 17 patients GWMA was found to be the only pathology: only 58% of these patients had an Engel Class I outcome. If present, histological examination revealed gradual disturbances of gray- and white-matter differentiation in the majority of cases. There was no mortality and no permanent morbidity, except for expected visual field defects. Initial precipitating injuries (IPI) were predominantly associated with the presence of isolated hippocampal sclerosis.

Conclusions: Detailed presurgical evaluation enabled either limited or tailored resections with similar success rates, in patients with or without the presence of GWMA. Hippocampal sclerosis and concomittant temporo-polar abnormalities should be recognized and considered upon presurgical evaluation, however, the presence of GWMA does not exclude the use and potential benefit of limited resections.

1 M.A. Nitsche , 2 S. Thome-Souza , 2 S. Freedman , 2 K.D. Valente , 2 A. Pascual-Leone , 1 W. Paulus , 2 F. Fregni 11 Georg-August-University (Gottingen, D) ; 2 Harvard Medical School (Boston, USA)

Purpose: Transcranial direct current stimulation (tDCS) has been shown to be an effective noninvasive tool for inducing prolonged cortical excitability diminutions in healthy humans. Since cortical hyperexcitability is a critical condition for epileptic seizures in patients with cortical malformations, tDCS might diminish cortical epileptiform activity in these patients and reduce seizure frequency and severity. In this pilot study, we aimed to study the effects of cathodal DC polarization in patients with refractory epilepsy and malformations of cortical development (MCDs) as indexed by seizure frequency and epileptiform EEG discharges.

Methods: Nineteen patients with MCDs and refractory epilepsy underwent one session of DC polarization (20 min, 1 mA) targeting the epileptogenic focus. The number of epileptiform discharges (EDs) in the EEG and seizures were measured before (baseline), immediately after, and 15 and 30 days after either sham or active DC polarization. Seizure frequency after the treatment was compared with baseline.

Results: Active compared with sham DC polarization was associated with a significant reduction in the number of epileptiform discharges [mean ED reduction of −64.3% (95% CI, −122.5% to −6.0%) for the active treatment group and -5.8% (95% CI, −26.8% to 15.2%) for the sham treatment group]. A trend (p = 0.06) was noted for decrease in seizure frequency after active compared with sham treatment [mean seizure frequency decrease of −44.0% (95% CI, −95.0% to 7.1%) for the active treatment group and −11.1% (95% CI, −22.2% to 44.4%) for the sham treatment group].

Conclusions: This randomized, controlled study shows that cathodal DC polarization does not induce seizures and is well tolerated in patients with refractory epilepsy and MCDs. Furthermore, the results suggest that this technique might have an antiepileptic effect based on clinical and electrophysiological criteria.

1 J. Sperner , 1 J. Mertin , 1 H. Fiebelkorn , 1 H.J. Friedrich 1University Hospital Schleswig-Holstein (Lubeck, D)

Purpose: To investigate the immediate EEG changes during VNS on-time compared to VNS off-time in patients with pharmacoresistant epilepsy in correlation to VNS outcome.

Methods: 64 EEGs from 15 patients treated with VNS were investigated using quantitative automated, Fast-Fourier-Transformation-analysis (FFT). Routine scalp-EEGs (10-20-System) were analyzed to compare VNS on- (30 s) and off-time (3 or 5 min) in relation to output current and clinical outcome. FFT was applied in each EEG with at least 20 epochs (3 s duration) during on- and off-time. Changes of the power-spectra (amplitude2) and the gamma frequency band (20-32 Hz) during VNS on time were calculated for each of the 19 electrodes of the EEG.

Results: Eight patients (receiving 28 EEGs) were termed as VNS nonresponders because of seizure reduction less than 50%. Seven patients (receiving 36 EEGs) were termed as VNS responders. The overall significance of gamma-frequency increase was p = 0,001 (Mann-Whitney U test) in VNS responders. The differentiation into three classes of VNS output current revealed a significant increase of gamma frequency in the low (0,5-1,5 mA) and the medium (1,75-2,0 mA) current group, but no change in the high group (2,25-3,25 mA). This was taken as a striking argument, that the increase of fast EEG frequency band is not due to the stimulation artifact, but is a VNS induced, brain generated oszillation.

Discussion: These results clearly show quantitative EEG differences in VNS responders compared to nonresponders. The gamma frequencies are well known to originate from the locus coeruleus, which plays an important role in transmitting the anticonvulsive VNS effect in animal models of epilepsy. When the locus coeruleus is removed in experimental epilepsy models, the VNS therapy does not work any longer. Further studies are indicated to clarify the role of fast EEG frequencies during VNS therapy.

1 B. Huber , 1 M. Knoop 1Epilepsy Center Bethel (Bielefeld, D)

Purpose: Epilepsy patients who continue having seizures despite therapeutic efforts are at an increased risk of suffering injuries. Our aim was to collect data on the number, severity and circumstances of seizure-related injuries occurring in residential patients of an epilepsy centre. A further aim is to examine if part of the injuries can be avoided by protective measures.

Methods: Questionnaires were distributed and collected monthly over a period of 16 months in a defined part of the residential department for staff to report all injuries (requiring any kind of treatment) happening in the context of seizures.

Results: 273 injuries were reported; in 65 cases the relation to a seizure was uncertain – these were excluded from further evaluation. 122 injuries were mild (like abrasions, contusions, sprains), 80 were moderate (lacerations, cuts, one luxated shoulder), and 20 were severe (all fractures). No very severe events (like intracranial bleeding, other life-threatening conditions or permanent damage) were reported but one patient died from SUDEP. Head lacerations were the most frequent type of injury. More injuries occurred within the home (either in the patient's private room or in the common rooms) than at the workplace or out of doors. There appeared to be a core group of approx. 40 patients (out of 500 covered by this investigation) with repeated and/or severe injuries.

Discussion: Injuries remain a major burden for the person with therapy-resistant epilepsy. A majority of them happens at home – an environment that can be shaped by the person affected and his/her carers. In a second step of this project we are going to examine individually all of the patients of the above mentioned “core group” with respect to possible protective measures in their personal environment as well as for remaining medical therapeutic options.

1 M. Bergmann , 1 J. Ndayisaba , 1 W. Oberaigner , 1 G. Kuchukhidze , 1 E. Trinka 1University of Innsbruck (Innsbruck, A)

Introduction: Sudden unexpected death in epilepsy patientes (SUDEP) accounts for up to two third of fatalities in selected populations with drug resistant epilepsy. Incidence rates vary substantially between 0.3 and 10/1000 person years (PY) according to the design of the studies. The incidence of sudden unexpected death in epilepsy patientes (SUDEP) in Tyrol is unknown therefore we aimed to analyze SUDEP cases using a hospital based record linkage system.

Methods: All patients with definite epilepsy (n = 3334) treated at the outpatient epilepsy clinic of the Universitätsklinik für Neurologie, Innsbruck, Tyrol, between 1.1.1970 and 31.12.2000 were included in the study. Epilepsy diagnosis was based on the ILAE classification. Living or dead status was obtained from the Institut für Klinische Epidemiologie, TILAK, Innsbruck and the Landesstatistik Tirol using a specialized probabilistic record linkage. Patients were followed until death or 31.12.2003. A total of 48.595 person years were analyzed. All patients with ICD-9 codes G40.0-G41.9 (i.e., “epilepsy”) were further analyzed in detail (n = 43). Based on expert chart review we classified SUDEP cases as: definite, probable, possible and no-SUDEP cases. Incidence rates and proportional mortality of definite and probable SUDEP were calculated in relation to the total mortality of the cohort in the same time period.

Results: Five pts. had definite, 7 probable and 22 possible SUDEP. Eight pts had no SUDEP; (one of them was a near SUDEP case). In our cohort incidence rates of SUDEP (definite and probable) was 0.2/ 1000 person years. SUDEP (definite and probable) accounted 1.8% of all deaths caused by epilepsy in the Tyrol cohort.

Conclusion: SUDEP is a rare cause of death in the general epilepsy population. The results of Tyrol cohort reflect population based incidence rates and are compareable to population based studies about SUDEP.

1 A.C. Lang-Dullenkopf , 1 P. Weber 1University Children's Hospital Basel (Basel, CH)

Introduction: The hippocampal formation is an essential region in temporal lobe epilepsy. The development of hippocampal sclerosis is in discussion. The relevance of this gliosis for temporal seizures seems clear, if a hippocampal sclerosis could be developed due to seizures is unclear. Reversible posterior leukoencephalopathy syndrome (RPLS) is a brain disorder most commonly associated with malignant hypertension, toxemia of pregnancy, or the use of immunosuppressive agents. It is associated with an inflammation of the brain.

Case report: We describe a male patient with hippocampal sclerosis developing after RPLS. Normal development until the age of 5 years. Diagnosis of a leukemia at the age of 5 years. Initiation of chemotherapy according to study protocols after documentation of a normal EEG and brain MRI. Some days after the end of the first chemotherapy he developed visual hallucinations with small manikins walking at the wall. Some days later the boy suffered from 2 complex partial seizures on the right side with disorientation, yawning and speech difficulties. A therapy with phenobarbital and aciclovir intravenously was initiated. The MRI findings were consistent with RPLS. In the initial MRI the hippocampal formation was normal. Further on the patient developed seizures with swallowing and smacking. Therefore we started a medication with oxcarbazepin and later on with phenytoin. Two years later worsening of seizure frequency and new seizure types (epigastric aura, visual and psychomotoric phenomenons) developed. At this time chemotherapy was stopped according to study protocols. Now the MRI findings were consistent with a left hippocampal sclerosis. Now the patient is seizure free treated with lamotrigine 300mg/d.

Discussion: We assume that in this case a healthy boy without any preexisting neurological signs developed hippocampal sclerosis after inflammation and seizure activity. In animal studies it is described that hippocampal sclerosis can evolve after high seizure activity or after inflammation. In this case we cannot differentiate what was the origin of the new pathologic entity but we can properly show the development of hippocampal sclerosis.

Abstract

  1. Top of page

Session: Mitochondrial diseases in epileptology

1 W. Sperl 1Paracelsus Private Medical University (Salzburg, A)

Mitochondria are organelles of the cell with a key function in energy supply. Furthermore mitochondria play a role in apoptosis and can generate reactive oxygen species (ROS). These three components, energy deprivation, induction of apoptosis and radical damage have major input for pathophysiological processes. Mitochondrial dysfunction is usually gradual and may occur as metabolic disease but is also part of civilisation diseases and important in tumour formation and aging. Proper function of the mitochondrial energy metabolism requires more than 100 enzymes, including membrane transport, metal ion-, vitamin- and lipid metabolism, which highly complicates the diagnosis. Secondary disorders can occur by inhibition of various enzymes due to toxic metabolites or substances. Oxidative phosphorylation, the central part of the metabolism, consists of multisubunit enzyme complexes; parts of them are encoded on the nuclear, parts on the mitochondrial genome. Defects of the mitochondrial DNA only affect the energy metabolism and mitochondrial genetics differs in several aspects from nuclear genetics.

Mitochondrial dysfunction can affect various organs, especially those with high energy demand. Therefore diseases with (cardio-)muscular and CNS involvement have been found first (mitochondrial encephalomyopathies). Mitochondrial diseases usually affect more than one organ, however tissue-specific forms are also known. Some disorders do not present primarily with an affection of the neuromuscular system, e.g., mitochondrial hepatopathy, cardiomyopathy. A number of mitochondrial syndromes like MELAS, MERRF, NARP, KSS, Pearson, CPEO, LHON, MNGIE, Barth, Leigh, Alpers-Huttenlocher etc. have been defined. Some of them have a unique genetic cause but in some syndromes also defects in different enzymes can result in the same phenotype. Many patients do not fit into syndrome criteria and therefore diagnostics has to be performed stepwise. Lactate elevation in body fluids can be a hallmark for diagnosis, but can be unspecific and may be absent in several patients. Magnetic resonance imaging, spectroscopy and other organ investigations can point to a mitochondrial disease. In most cases the diagnosis can be made by functional biochemical analysis (functional investigations in isolated mitochondria from fresh muscle tissue and respiratory chain enzyme investigations including ATPase and PDHC) and histological and histochemical investigation of a tissue (muscle) biopsy. Genetic analysis of target genes proofs such a defect and is the basis for genetic counselling.

Therapy of disorders of the mitochondrial energy metabolism is limited in most cases. It is mainly based on the enhancement of remaining enzyme activities by increased cofactor supplementation. Furthermore, toxic radicals are tried to be reduced by antioxidant support. Pyruvate oxidation defects can be treated remarkably by ketogenic diet, which provides an alternative energy fuel for these patients. Gene therapy has only been tried in vitro till now.

In summary diagnostics of mitochondrial disorders is complex and requires an individually adapted diagnostic cascade performed in cooperation with highly experienced diagnostic centers with quality criteria and international cooperation. Randomized prospective multicenter studies are required to improve therapy.

1 S. Gallati , 1 A. Schaller 1University Berne (Berne, CH)

Hereditary mitochondrial disorders (MCDs) are caused by mutations in the mitochondrial DNA (mtDNA) or nuclear DNA, resulting in extremely variable single- or multisystem diseases with onset between birth and senescence. Even the same molecular defect can associate with diverse phenotypes. One of the most common mtDNA mutations, the A3243G transition in the tRNALeu(UUR) gene, is mainly associated with mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS), but also with various other types of multisystemic diseases such as progressive external ophthalmoplegia (PEO), diabetes and deafness.

Clinical and molecular genetic findings of three patients and family members carrying the A3243G mutation will be presented:

  • 1
    M.A. with epilepsy, admitted to the hospital because of sudden severe exacerbation, died two weeks later at the age of 11 years.
  • 2
    B.B. with diabetes and deafness, 3 spontaneous abortions, one son developing first symptoms (bilateral hearing loss, muscle weakness, seizures) at the age of 10 years.
  • 3
    G.W. with epilepsy, diabetes, hearing loss, progressive dementia, ataxia, brain atrophy, death at the age of 62 years. Although the noninvasive genetic testing of blood samples is common and successful in many cases, A3243G mutation levels are known to decrease in this tissue during ageing and to exist at consistently higher mutation levels in postmitotic tissues.

Patients M.A., B.B. and her son showed 64%, 20%, and 47% respectively of mutated mtDNA in blood, whereas the A3243G amount in muscle tissue from the son was 67%. For patient G.W. the mutation was only detectable in mtDNA isolated from brain parenchyma but not in muscle tissue and blood. This phenomenon may be explained by the fact that embryologically, muscle derives from mesodermal whereas CNS derives from ectodermal germ layers. Although, molecular genetic analysis of skeletal muscle is widely accepted as to be the ‘gold standard’ in routine diagnostics of patients with suspected MCD, our findings demonstrate, that testing of muscle mtDNA still includes the risk to yield false negative results.

In conclusion, our findings underscore the fact that phenotypic expression of MCDs depend on a threshold effect based on mitotic segregation and polyploidy on the one hand and on the nature and localization of a mutation, its tissue distribution as well as the energy needs of an organ on the other hand. Family members were provided with this complex information in the context of genetic counselling and the extreme difficulty to interpret testing results of at-risk individuals and to predict phenotypes and clinical courses was discussed. Moreover, the phenotypic presentation of the A3243G mutation, originally described as MELAS mutation, is extremely heterogeneous, meaning that, in fact, A3243G is associated with but not diagnostic for MELAS per se.

1 T. Dorn 1Swiss Epilepsy Centre (Zurich, CH)

Background: Mitochondrial cytopathies (MC) are a huge group of genetic diseases having respiratory chain dysfunction in common which causes dysfunction of tissues using oxydative phosporylation such as nervous system, muscles and heart. Epileptic seizures are a frequent symptom when cerebral cortex is involved. One subgroup of MC is caused or associated with mutations, deletions or rearrangements of mitochondrial DNA (mtDNA). These alterations of mtDNA, however, are usually not present in each mitochondrium, cell or tissue, i.e., they are heteroplasmic. Furthermore, the number of altered mtDNA copies usually increases during life time explaining the progressive nature of these entities with symptoms occurring after the number of altered mtDNA copies has exceeded a tissue specific threshold value. Thus, patients with the same mutations can differ considerably with respect to type, age at onset and severity of symptoms as well as with respect to velocity of disease progression. Despite these differences the involvement of different parts of the nervous sytem (especially occipital cortex, retina, n. opticus, inner ear), muscles and heart in course of disease progression often enables the suspicion of mitochondrial disease which should cause further diagnostic procedures including mtDNA analysis of cells of different tissues (usually blood and muscle). But one can assume that due to heteroplasmy patients with, for example, symptoms generated by only one part of the brain – at least at the beginning of their diesease – should exist. Therefore as epileptologists we have to ask whether there are MC patients with epileptic seizures as the outstanding symptom for a long time. If this is the case we have to clarify how MC can be diagnosed in these patients because in MC valproic acid being often used as initial monotherapy in generalized and focal epilepsies should not be used because of ist potential to cause fatal liver failure. In order to answer these questions two MC patients are presented with a predominat epileptological phenotype showing course of and difficulties in the diagnostic process.

Case Reports: Patient 1, female, born 1974 had a mother with a left branch bundle block and a father suffering from migraine. After a syncope 1986 a slowly progressive syndrome emerged with multifocal myoclonias, rare generalized tonic–clonic and frequent occipital seizures, headache, ataxia and slight neuropsychological impairment. Finally hypacusis was diagnosed in 1999. Most important paraclinical findings were giant visual and somatosensory evoked potentials, mild cerebellar atrophy in MRI, slightly elevated cerebrospinal fluid lactate levels and absence of ragged red fibres in muscle biopsy. 1995 analysis of lymphocyte mt-DNA showed no point mutation between positions 8167 and 9049, thus, MERRF was excluded. In 1999 a mutation in the EPM1 (CSB) gene had been excluded. However, a sensitive quantitative allele-specific PCR which was performed because of hypacusis revealed a heteroplasmic point mutation T3271C in lymphocyte mt-DNA. After the diagnosis of MC had been established the patient suffered from stroke like episodes for several times underpinning the diagnosis of MC. Patient 2, male, born1950 with partial seizures and sometimes myoclonic jerks since 1960, seizure freedom since 2000 under an antiepileptic polytherapy with oxcarbazepine, primidone and clobazam, normal MRI, mild left frontotemporal neuropsychological signs. Repeatedly observed slight increases of creatine kinase without clinical signs of myopathy led to an EMG which showed only mild neurogenic changes. An ophthalmological examination including VEP and perimetry because of increasing visual disturbances since 2001 disclosed optic nerve atrophy without ocular movement disorders. After exclusion of an autoimmune disease an extended analysis of mtDNA in blood cells demonstrated a homoplasmic polymorphism (T9582C) which was also found in his two asymptomatic brothers. Finally in mtDNA of a muscle biopsy which otherwise showed only mild neurogenic changes, no ragged red fibers and normal oxphos activity a heteroplasmic 7.4 kb deletion going beyond the Kearns-Sayr deletion was found.

Discussion/Conclusion: These two cases demonstrate that epileptological features can be predominant in the phenoptype of MC and be almost the sole symptom for a long time in the course or at least at the beginning of the disease. Thus, in each patient with first seizures we closely have to look for mild mitochondrial signs and symptoms, especially when treatment with valproic acid is planned.

1 W.S. Kunz 1University of Bonn (Bonn, D)

The specific pathophysiological significance of mitochondria in epilepsy is documented by the genetic association of mutations in mitochondrial tRNA genes with certain forms of myoclonic epilepsy. Frequent mitochondrial tRNA mutations which are relevant for routine genetic testing are the A8344G MERRF mutation in the tRNA lysin gene and the A3243G MELAS mutation in the tRNA leucin(UAA) gene. Recent investigations confirm also in Alpers syndrome, a myoclonic epilepsy in children, a direct mitochondrial cause. Here, mutations in the mitochondrial DNA polymerase gamma (POLG1) have been associated with the disease. Especially frequent are the pathogenic mutations A467T or W748S, which are met homozygeous or in the compound heterozygeous state with other rare POLG1 mutations in nearly all Alpers patients. These mutations lead by inhibition of replication of mitochondrial DNA in postmitotic tissues to multiple deletion of mitochondrial DNA or to the depletion of mitochondrial DNA. The functional consequences of depletion of mitochondrial DNA on mitochondrial oxidative phosphorylation are similar to the effects of pathogenic mitochondrial tRNA point mutations. It has to be noted, however, that in the Alpers patients the effects of depletion of mitochondrial DNA are especially severe in liver, which causes the extremely high toxicity of valproic acid.

  • - 
    Even in hippocampal tissue of patients with temporal lobe epilepsy and Ammons horn sclerosis local pathologies of mitochondrial respiratory chain were detected. In the hippocampal CA3 region a deficiency of mitochondrial respiratory chain complex I was shown which molecular cause is – similar to the Alpers syndrome – a depletion of mitochondrial DNA. The direct cause for this depletion of mitochondrial DNA is however not a nuclear mutation but a status epilepticus-associated increased oxidative stress leading to increased DNA turnover.
  • - 
    Thus it has to be concluded that pathologies of mitochondrial function appear to have a direct causal relationship to certain forms of epilepsy.

Abstract

  1. Top of page

Session: Immune mediated epilepsies

1 J. Bauer , 2 C.E. Elger , 1 H. Lassmann , 2 C.G. Bien 11 Medical University Vienna (Vienna, A) ; 2 University of Bonn (Bonn, D)

The current histopathological criteria of Rasmussen encephalitis (RE) include the presence of T cell dominated inflammation, microglial activation, neuronal loss and astrocyte activation. Several pathophysiological mechanisms have been suggested. Early studies described viral infections although none of the studies could conclusively link a specific virus to RE. A second mechanism propagated to induce neuronal death in RE was an antibody-mediated cytotoxicity by antiglutamate receptor-3 antibodies. Recent studies, however, revealed that anti-GluR3 antibodies are not specific for RE since they are found also in noninflammatory focal epilepsy cases. Furthermore, it has been shown that numerous RE cases are GluR3 antibody negative. Recently, we provided evidence for another mechanism of cell death in RE by showing that cytotoxic T cells may destroy neurons by the release of granzyme-B (GrB). Here, we show evidence that not only neurons but also astrocytes are attacked and killed by these cytotoxic T cells. In RE we found apoptotic astrocytes and the presence of lesions deficient of astrocytes. On the border of these astrocytic lesions we could show the presence of GrB positive T lymphocytes in close apposition to astrocytes. Such apoptotic astrocytes and astrocyte deficient lesions were not present in other forms of epilepsy such as Ammon's horn sclerosis or focal cortical dysplasia suggesting that T-cell mediated death of astrocytes is a specific feature of RE.

1 C.G. Bien , 1 C.E. Elger 1University of Bonn (Bonn, D)

Limbic encephalitis (LE) was described in the 1960s as a clinical-pathological syndrome in adults. Initially, the paraneoplastic form was the centre of interest. An increasing number of diagnostically valuable autoantibodies in patients' sera (and CSF) have been identified. Lately, the impact of nonparaneoplastic LE cases has been acknowledged. In the serum of some of these patients antibodies against voltage-dependent potassium channels (VGKC-antibodies) have been detected. The characteristic MRI course of LE patients has recently been described in detail: Hippocampal swelling and T2-/FLAIR-signal increase are early findings. Usually, the swelling regresses after a few months. This is followed by hippocampal atrophy with continuous signal increase, i.e., hippocampal sclerosis.

The following formal diagnostic criteria for LE are suggested: clinical “limbic” signs, which manifested themselves in adulthood no longer than five years previously, must be regarded as an indispensable diagnostic prerequisite. The patient concerned must have at least one of the following symptoms: disturbance of episodic memory, epileptic seizures of temporal semiology or affective disturbances with a prominent lability of mood and lack of inhibition. Additional systems of the central or peripheral nervous system may be involved. In addition to the clinical characteristics, one of the following additional features must be observed: verification of a neoplasma or of an LE associated auto-ab or a temporomedial FLAIR-/T2-signal increase shown in an MRI scan which is not otherwise explainable, or a chronic lymphocyte-microglial limbic encephalitis shown on histopathological examination of a brain specimen.

Treatment should start as soon as the tentative diagnosis of LE is made – that means, even before the antibody test results are available. As a first line treatment, methylprednisolone pulses (3-5 times 500-1000 mg on consecutive days), repeated once every month are recommended. If a tumour is detected (paraneoplastic LE), its treatment is mandatory. Decisions on long-term treatment depend on the response to the initial therapy and the final classification of the LE subsyndrome and its prognosis.

1 S. Rüegg 1University Hospital Basel (Basel, CH)

The interplay between immune responses and epilepsy is complex; increasing data tries to link immune responses (humoral or cellular) with epileptic syndromes. Various clinico-immunological situations might be differentiated within this context. In some of these conditions, patients with known focal or generalized epilepsy harbour different well characterized autoantibodies (Abs) (like antinuclear or antithyroid Abs), but the proof of their causative role in the development of epilepsy is absent. These Abs may be either coincidental or an epiphenomenon. In other patients, the presence of Abs and seizures is a core feature of several specific syndromes, like juvenile neuronal ceroid lipofuscinosis (Batten's disease), and steroid-responsive encephalopathy with autoimmune thyroiditis (SREAT; formerly Hashimoto's encephalopathy). However, it is difficult to directly link the detected Abs with the pathogenesis of these disorders and epilepsy in particular.

With regard to epilepsy, Abs against glutamic acid decarboxylase (GAD) are of special interest. Glutamic acid decarboxylase catalyzes the formation of the main cerebral inhibitory transmitter gamma-amino butyric acid (GABA) from the main cerebral excitatory transmitter glutamic acid by one single step, i.e., decarboxylation, dependent on pyridoxal (vitamin B6) as a cofactor. Autoantibodies against GAD (anti-GAD-Abs) may inhibit or slow this process leading to increased levels of glutamic acid and lowered levels of GABA. It is tempting to speculate that this combined hyperexcitatory and disinhibitory effect may contribute to the epileptogenesis in syndromes with anti-GAD-Abs, like juvenile neuronal ceroid lipofuscinosis (Batten's disease).

Intriguingly, other neurological disorders, like stiff-person syndrome or non-neurological disorders, like insulin-dependent diabetes mellitus (IDDM), were associated with anti-GAD-Abs. However, the risk for seizures does not seem to be markedly increased in these diseases. Further analysis of subtypes of anti-GAD-Abs recognizing distinct epitopes, and of differences in anti-GAD-Abs titers may help to elucidate the exact pathophysiological mechanisms. Conversely, data is limited about the prevalence of anti-GAD-Abs in primary epileptic syndromes. Case control and cohort studies reported the presence of anti-GAD-Abs in a small part of patients with pharmacoresistant focal epilepsy, and an almost complete lack of these Abs in patients with idiopathic generalized epilepsy (IGE). On the other hand, the recent observation of a 4.4-fold increased risk of IDDM in patients with IGE raise the suspicion that anti-GAD-Abs might still play a role in the epileptogenesis of IGE syndromes.

Autoimmune thyroiditis with anti-tyhroid Abs is mandatory for making the diagnosis of SREAT. Seizures are present in about 80 to 90% of patients with SREAT and they figure among the essential features of this rare syndrome. Pathogenesis of SREAT remains enigmatic. While autoimmune thyroiditis is one of the most prevalent autoimmune disorders, only a very small proportion develops SREAT and seizures. Additionally, the course of the disease is neither closely related to the titers of the anti-thyroid Abs nor to the level of the thyroid hormones. Nevertheless, the striking and immediate efficacy of steroids in the majority of patients with SREAT suggests an inflammatory, and probably autoantibody-mediated, humoral pathogenesis.

Thyroid hormones are essential for the differentiation of oligodendroglial cells; thus, disturbances of myelination and myelin maintenance might alter the propagation of neuronal currents and facilitate the generation of seizures. Hyperthyroidism or substitution of L-thyroxin in hypothyroidism experimentally lowers the seizure threshold in animals and clinically induces focal or absence and myoclonic seizures in some patients with IGE syndromes, but the exact pathway and specific contribution of anti-thyroid Abs to the epileptogenesis of SREAT still remains largely unclear.

To conclude, autoantibody-induced, humoral immune-mediated mechanisms may play an important role in the pathogenesis of epilepsy in several disorders. Antibody-mediated inflammation is associated with the production and release of cytokines with proconvulsive properties, like interleukin-1-beta. Anti-GAD-Abs specifically target a pivotal enzyme essential for the formation of principal neurotransmitters and they subsequently shift the balance of excitability to the excitatory side. The anti-thyroid Abs may affect the propensity of the brain to seize by the induction of proconvulsive cytokines too. They also cause alterations of neuroglial structures important for the generation and propagation of electrochemical neuronal activity, like membranes and myelin sheets. However, the precise mechanisms why and how anti-GAD-Abs and anti-thyroid Abs influence epileptogenesis in several syndromes remain unknown and warrant further experimental and clinical investigation.

1 A. Vezzani 1Mario Negri Institute for Pharmacological Research (Milan, I)

Inflammatory reactions occur in brain in various CNS diseases, including autoimmune and neurodegenerative disorders. Proinflammatory cytokines and related molecules have been described in CNS and plasma in experimental models of seizures and in clinical cases of epilepsy. These inflammatory reactions may involve either the innate or the adaptive immune systems, or both, and share molecules and pathways also activated by systemic infection. The activation of the innate immune system, and the associated inflammatory reactions, appear to mediate some of the molecular and structural changes occurring in brain during epileptogenesis. Human studies have shown that intrinsic brain inflammatory reactions occur in patients with neuropathologies associated with the late development of epilepsy. Experimental studies in mice with genetically impaired cytokines functions and application of cytokines in normal rodent brain have shown alteartions in neuronal excitability, cell survival, and blood–brain barrier function. Moreover, specific anti-inflammatory treatments can affect seizures in experimental models, and in some instances, in clinical cases of epilepsy.

This presentation will overview the current knowledge in this field to critically discuss the possibility that inflammation in the brain may be a common factor contributing, or predisposing, to the occurrence of seizures and cell death, in various forms of epilepsy of different etiologies. The elucidation of this aspect may open new perspectives for the pharmacological treatment of drug-resistant epilepsies.

Abstract

  1. Top of page

Session: Evidence-based medicine in epileptology and off-label use of antiepileptic drugs

1 F. Rosenow 1University Hospital Giessen and Marburg GmbH (Marburg, D)

Evidence-based medicine (EBM) – why bother? (inspired by Gordon Guyatt (1))

There is a tendency among epilepsy experts to question the need of high level evidence in the treatment of status epilepticus (SE). Discussing the off-label use of valproic acid for instance I have heard remarks such as: We never had high level evidence for the use of our standard – phenytoin, why should we need a call I evidence study for valproate. The question arises: What were our decisions based on before we tried to base them on evidence? The answer: on expert opinion. Expert opinion again is based on small samples judged before a background of personal belief and interest and thus quite biased. There are may examples showing that the majority of expert opinions failed to recognize or chose to ignore the evidence from clinical trial recommending a therapy with no benefit but real risks for many years if nor decades (e.g., the lidocaine story). From this we learned that treatment decisions need to be based on the current best evidence (implying a hierarchy) from clinical trials and are further influenced by health care decisions which are influenced by (patient) values.

What is the evidence for off-label therapy in status epilepticus?

We currently have no meta-analysis of randomized prospective trials (RCTs) and 2 RCTs sufficiently powered to come to significant conclusions. The first of the studies considers the out of hospital treatments by medical assistants. The authors found that a benzodiazepine (diazepam or lorazepam) was significantly superior to placebo in obtaining control of SE at admission to the hospital (2). The second study regarding the in house treatment of SE found that control of SE within 20 minutes occurs significantly less frequently after phenytoin (18 mg/kg) monotherapy (3) as compared to 3 others standard therapies: a) lorazepam (0,1 mg/kg), b) phenobarbitol (15 mg/kg), and c) diazepam (0,15 mg/kg) followed by phenytoin (18 mg/kg). Most other evidence we have in the treatment of SE is obtained by observational studies/case series, underpowered “pilot studies” and expert opinion (4).

The risks and advantages of applying therapy with poor evidence basis in status epilepticus

SE carries a case fatality risk of 5-30%. It is a common belief, that case fatality is mainly dependent on etiology, age and status type. However, recent evidence suggests that the treatment strategy can be one of the determinant of case fatality too. Observational studies and case series frequently report results on illdefined or non-comparable groups and expert opinion has gone wrong before we need to consider the 2 RCTs we have and we certainly need more Class one evidence. Applying off-label therapy for the first line treatment of SE, where we do have class I evidence is not only a legal but also a ethical problem. This may be different in refractory SE where we have much less evidence available.

Conclusion: The treatment of SE needs to be EB and we clearly need more high level E.

References:

1. Guyatt G, Rennie D. Users' Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice. http://www.userguides.org2. Alldredge BK et al. (2001) A comparison of lorazepam, diazepam, and placebo for the treatment of out-of-hospital status epilepticus. N Engl J Med 345:6317. 3. Treiman DM et al. (1998) A comparison of four treatments for generalized convulsive status epilepticus. Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J Med 339:7928. 4. Krämer G et al. (2005) Stellenwert intravenöser Valproinsäure in der Therapie des generalisierten tonisch-klonischen Status epilepticus. Aktuelle Neurologie 32:26374.

1 B.J. Steinhoff 1Epilepsy Centre Kork (Kehl-Kork, D)

Recent developments of new antiepileptic drugs addressed additional indications beyond epilepsy even at the stage of clinical trials prior to labelling. Examples are pregabalin or lacosamide. However, meanwhile we have examples of other and even more traditional antiepileptic drugs whose efficacy was shown in other indications based on controlled trials, so that a labelling was possible, i.e., carbamazepine, valproate or lamotrigine in psychiatric indications.

This lecture informs about controlled trials with antiepileptic drugs beyond epilepsy addressing indications the antiepileptic drugs are not yet labelled for. The objective of the lecture is to inform about those studies so that attending physicians may be more aware of study data which might help to argue about the off-label use of drugs in individual cases if this off-label use is questioned by cost-givers.

Abstract

  1. Top of page

Session: Emotion and social cognition in epilepsies

1 P. Vuilleumier 1Geneva University Medical Center (Geneva, CH)

The amygdala is a medial temporal lobe structure critically implicated in emotional and memory processes. Behavioral and functional neuroimaging studies in healthy subjects and brain-damaged patients show that amygdala activity is not only involved in the recognition of and physiological response to emotional stimuli, but may also act to influence a wide range of perceptual and cognitive processes. Feedback projections from the amygdala to sensory cortical areas may serve to enhance perceptual analysis, attention, as well as memory for emotionally or socially relevant stimuli. In turn, damage to the amygdala may alter the normal pattern of emotional responses in distant, intact brain regions that receive such feedback projections. A better knowledge of the dynamic interactions between these brain areas may help better understand the cognitive and affective consequences of temporal lobe diseases.

1 R. Winkler 1Swiss Epilepsy Centre (Zurich, CH)

The term ‘theory of mind’ (ToM) refers to the ability to attribute mental states to others in order to understand their behavior, desires, and intentions. This capacity is an important prerequisite for the highly complex, flexible social behavior of humans. As demonstrated by disorders such as autism or Asperger syndrome, deficits in reasoning about the mental states of others may disrupt successful social interactions. Imaging and lesion studies have identified a widely distributed neural system involving prefrontal, orbitofrontal, mesolimbic, and anterior and posterior temporolateral brain structures in ToM processing. These regions may also be affected in patients with refractory mesial temporal lobe epilepsy (MTLE), as has been shown in several brain imaging studies. FDG-PET studies have revealed an interictal depression of the regional cerebral glucose metabolism in the affected mesiotemporal lobe, as well as remote metabolic changes of lateral temporal, prefrontal, and orbitofrontal structures. Functional magnetic resonance imaging data point towards a diminished or absent activation of the amygdala on the pathologic side. Consistent with functional data, recent morphometric MRI studies have shown widespread volume reductions within both temporo-limbic and various prefrontal structures. Therefore, refractory MTLE may well interfere with certain aspects of social cognition that rely on the functional integrity of temporo-limbic and frontal networks. While cognitive domains such as memory, language and executive functions have been studied extensively, relatively few investigations into social cognitive abilities in general, and ToM specifically, have been carried out in this patient group. Nevertheless, the following findings regarding social cognitive abilities in MTLE provide first evidence for possible MTLE-specific deficits in this domain: several studies have found disturbances in the perception of social cues, in particular facial emotions (Meletti et al., 2003; Reynders et al., 2005). This suggests that MTLE may affect basic social perceptual abilities which are assumed to be the precursors of ToM knowledge. However, this issue remains controversial in the literature. Furthermore, there are some studies explicitly investigating ToM abilities in MTLE. In one recent study, we found that patients with unilateral MTLE were impaired in their ability to recognize a faux pas as compared to patients with extra-mesiotemporal lobe epilepsy and healthy controls (Schacher et al., 2006). The detection of a faux pas in a brief prose passage is a typical advanced ToM task and is considered to be a sensitive indicator for subtle, higher-order deficits in social cognition. Shaw and colleagues (2004) demonstrated deficits in this test and other advanced ToM tests among subjects with lesions of the amygdala associated with childhood onset of seizures. Furthermore, data from Kirsch and colleagues (2006) has provided evidence that MTLE patients show reduced emotional empathy as measured by the Interpersonal Reactivity Index Questionnaire. Empathy refers to an affective facet of ToM, describing the ability to understand the emotional states of others. One key structure underpinning this ability is the anterior insular cortex, an area crucially involved in the propagation of mesial temporal lobe seizures. Taken together, these findings suggest that MTLE may affect higher-order aspects of social cognition, namely the ability to reason about the mental states of others. There are several possible mechanisms which could account for such an impairment. The most salient explanation is that MTLE affects brain regions which play a critical role during theory of mind processing. As highlighted, MTLE is accompanied by brain abnormalities in several parts of the complex neuronal network underlying ToM processes. Alternatively, the reduced ability of MTLE patients to reason about the mental states of others might be due to degraded developmental acquisition of theory of mind abilities. Future research needs to confirm MTLE-specific impairments in ToM abilities and should help to untangle the underlying mechanisms. The previous paucity of research in this field could be due to the lack of readily apparent social deficits in patients with MTLE. Nevertheless, psychosocial maladjustment and psychiatric comorbidity are serious problems in many patients with chronic epilepsy. The fact that these conditions are more frequent in MTLE compared to other epilepsy syndromes may reflect a specific pathologic association. If MTLE as such is a specific etiology of deficits in higher-order social cognition, then ToM might provide a conceptual framework for a better understanding, improved diagnostics, and effective treatment of the elusive psychosocial problems in patients with MTLE.

1 R. Thorbecke 1Epilepsy Center Bethel (Bielefeld, D)

  • 1
    Irving Goffman in the sixties of the last century has described in a classical manner situations in which people feel embarrassed, ashamed or humiliated because implicit rules of deference and demeanour are broken. The causes for this may be very different – on the one hand social inconsistencies, on the other hand the inability of the actors to perceive and to administer these implicit rules, e.g., in people with neurological or psychiatric diseases. (I. Goffman, Interactional rituals, 1967)
  • 2
    In the reports on the social integration of persons with epilepsy (PWE) there are findings concerning the long-term social outcome of children with epilepsy showing a worse outcome in employment, living with a partner, socioeconomic status and social isolation which only partly can be explained by learning disability (Silanpää, 1998, 2004; Jalva, 1997; Camfield, 2006). Apart from negative attitudes in the environment one explanation could be deficits in social skills, related to deficits in social perception.
  • 3
    Studies on felt stigma in PWE show no good correlation to enacted stigma (Jacoby, 1992), and there are hints, that public attitudes become gradually more positive (Jacoby, 2004). Felt stigma is however highly correlated to social isolation i.e. rare contacts to friends (r = .40, p < .001) (Thorbecke, 2005). Difficulties in initiating and maintaining social contacts could be one underlying factor.
  • 4
    Studies on vocational outcome of PWE supported by general versus specialized agencies to find employment show significantly higher placements in the specialized program (Fraser et al., 1984). The core of the program, which later was implemented in several American cities as training and placement program (TAPS) (Thorbecke, Fraser, 1998) was a job-seeking skills training, in which one major component was to learn taking the role of the employer to understand better the rules of the application procedure i.e. a social skills training to improve social perception.
  • 5
    To observe the effects of social perception on integration it is necessary to have sensitive indicators, e.g., to differentiate between job finding and job keeping or between finding new friends and loosing friends. We will show data from patients of our epilepsy surgery program showing no effects of the improved seizure situation on friendship networks and employment when there are no of targeted interventions.
  • 6
    In our opinion a better understanding of deficits in the perception of social rules and expressions could be the basis for the development of social skills training programs, and help to make already existing social skills training programs more specific (see for an example section 8.14 of the Patient Education Program MOSES: “How to explain my epilepsy to other people”)

Abstract

  1. Top of page

Session: Spectrum of the absence epilepsies

1 G. van Luijtelaar 1Radboud University Nijmegen (Nijmegen, NL)

The concepts on the origin of absence epilepsy are evolving since the middle of the last century. The centrencephalic theory (Penfield and Jasper), the cortical theory (Bancaud) and the cortico-reticular theory (Gloor) have dominated the international literature for decades. We formulated recently the cortical focus theory (Meeren et al., 2002; van Luijtelaar and Sitnikova, 2006), we belief that the spike-wave discharges (SWDs) accompanying absence seizures are not likely to appear due to a general increase in cortical excitability as stated by the cortico-reticular theory, while also a thalamic origin of SWDs is less likely (Gloor, Avoli, Kostopoulos). Based on advanced nonlinear signal analyses of multiside cortical and thalamic local field potentials in WAG/Rij rats (one of the commonly used genetic models of absence epilepsy, quite similar to Genetic Absence Epileptic Rats from Strasbourg; Depaulis and van Luijtelaar, 2006) we found that SWDs orginate consistently in a small region of the cortex and reach the thalamus only after a delay. Functional studies revealed that the focal cortical area is the perioral region of the somatosensory cortex. The Reticular Thalamic Nucleus and the Thalamic Relay nuclei are getting quickly involved in generating SWDs through descending cortico-thalamic pathways. Outcomes of recent pharmacological studies in GAERS and in WAG/Rij rats from us and other groups with cortical application of drugs such as ethosuccimide and phenytoin, neurophysiological stimulation and recording data from the neocortex, immunocytochemical and in situ hybridization data from neocortical slices from genetic models are all in line with the existence of a neocortical focal zone. Some recent MEG data from children will also be presented. Also the outcomes of these data are consistent with a focal role for the cortex in initiating absence seizures.

1 R.A. Sälke-Kellermann 1Swiss Epilepsy Centre (Zurich, CH)

Purpose: Absence seizures are found in several epilepsy syndromes. Two types of absences, typical absences and atypical absences, have to be differentiated. Both of them implicate quite different therapy and prognosis. Typical absences only occur in idiopathic generalised epilepsies, and these are easy to treat. Their prognosis is very favorable.

Atypical absences are common in symptomatic or cryptogenic epilepsies, e.g., in Lennox-Gastaut syndrome and in myoclonic astatic epilepsy. Usually they are difficult to treat and resistant to medication.

The aim of this report: To study the semiology of absence seizures and to define the epilepsy syndromes they belong to.

Methods: Children and adolescents with absence seizures undergo an intense epileptological workup with long-term electroencephalographic (EEG) monitoring, neurological, and neuropsychological assessments. A symptomatic aetiology has to be excluded, among others by MRI investigation. The seizures as well as the ictal EEG recordings with videos are analysed in detail. Such thorough investigations are necessary to the identification of the epilepsy syndromes according to the Proposed Diagnostic Scheme of the ILAE Task Force on Classification and Terminology (2001). This report presents absence seizures with corresponding epilepsy syndromes including clinical symptoms with videos as well as the course of the disease concerned.

Results: The best defined syndrome with absence seizures is the absence epilepsy of childhood (AEC). The clinical features and electroclinical findings are well documented including typical absences corresponding to regular generalized bilateral 3 Hz spike-and-wave (SW) discharges in the EEG lasting about 10 sec. The patients have severe impairment of consciousness and remain unresponsive during the seizures. Ongoing activities are interrupted. Some other features, as upward deviation of the eyes and mild myoclonia of the eyelids, are possible. Seizure frequency is very high. Absences can be provoked by overbreathing, which is a reliable diagnostic method to discover these seizures. Other seizure types do not occur before adolescence. Later on, some patients have generalized tonic–clonic seizures, additionally.

According to stringent criteria, other syndromes with absences should be recognised and differentiated from AEC. The following features are critical: age at seizure onset; detailed seizure semiology, seizure frequency, duration of seizures, morphology of ictal and interictal EEG discharges, and accompanying other seizure types. In differentiation of AEC, juvenile absence epilepsy (JAE) and juvenile myoclonic epilepsy (JME) are to be identified. Besides absences, patients with JAE mostly suffer from generalised tonic–clonic seizures (GTCS) during the course of the illness as well. In JME, bilateral myoclonic seizures are the leading symptom. GTCS often occur additionally. Seizures are provoked by sleep deprivation. Some of these patients may produce absences, too. The seizures of JAE and JME mostly can become well controlled by antiepileptic drugs, but in some patients seizures continue in spite of all therapeutic effort. The diagnosis may be missed, especially in JME. Not seldom the epileptic seizures in this syndrome are confounded with focal seizures. In these cases, the choice of medical treatment may catastrophically fail. Additionally, there are some other syndromes with absences not yet included in the International Classification. This group comprises absence epilepsy of infancy, eyelid myoclonia with absences, and absences with perioral myoclonia. Seizure control in these patients is poor, and there is a great risk for developmental retardation. The diagnosis may be missed at the onset of seizures when typical features of seizures and EEG discharges remain unrecognised.

Conclusions: Concerning absence seizures, identification of the corresponding epilepsy syndrome is crucial for therapy and prognosis. Therapy is promising and prognosis favorable in AEC (complete seizure freedom) whereas in other syndromes, seizures may be difficult to treat and prognosis dubious. The exact correlation of absences to the correct epilepsy syndrome has to be made at the very onset of the disease. Atypical features of absences or abnormal patterns of EEG discharges have to be recognised in time to avoid misdiagnosis and unavailing, if not nocuous therapy.

1 H. Mayer 1Epilepsy Centre Kork (Kehl-Kork, D)

Introduction: The psychological and neuropsychological aspects of the IGE, particularly of Absence Epilepsy has attracted only little attention in the last decade in contrast to the idiopathic and symptomatic focal epilepsies. That is surprising, because still at the beginning of scientific efforts into this syndrome, in the first third of the last century, there has been scientific data, which demonstrate, that children with this syndrome are at risk for developing psychological and neuropsychological dysfunctions. Nevertheless the scientific doctrine has become more and more accepted, that absence epilepsy is a special syndrome which tends to be more easily treated than symptomatic focal or other generalised syndromes and which can be diagnosed only in patients who have made normal cognitive and neuropsychological progress. The critical discussion of that doctrine has been forgotten in the following years and decades. For that reason the doctrine can be found in later papers and textbooks up to now.

State of Art: The psychological and neuropsychological investigations in the following years, that used for the first time psychometric instruments, like intelligence test, could strengthen this doctrine for a long time. On the other side relevant clinical reports and investigations into the social and emotional development could indicate, that a significant rate of children and adolescents with absence syndrome could not achieve academic performances, based on their intelligence scores. That means, in children with absence epilepsy academic underachievement is not uncommon. It could also be found in patients, who became seizure free by antiepileptic treatment. Obviously there are risk groups, which comprise patients with an early seizure onset (> 4) or a later onset (> 8), with multiple seizure types (GTCS) or for example patients with a high relapse rate. This rate is similarly high as in patients with other generalised syndromes.

In the meantime a lot of neuropsychological deficits and special learning disabilities particularly in patients with these risk factors were proved. For example psychometric investigations could demonstrate more specific cognitive deficiencies in visoconstructive functions, memory and attentional functions or in mental speed. Also specific learning disabilities in reading, writing and arithmethics could be documented. These deficits can partly explain the high rates of academic underachievers in children and adolescents with absence epilepsy. The profile or the pattern of the neuropsychological deficits and learning disabilities is not specific. Similar profiles can be found in patients with other focal or generalized syndromes, too. More typical are the pattern of neuro-physiological disorders, particularily, SW-Burst (or variants), which can regularly be found. These patterns, that are to a great extent genetically determined could interrupt or at least trouble the general and specific information processing chain (auditory and visual), which provides the basis for functions like reading, writing or memory. Suppression or non-occurrence of SW-discharges or its variants by effective antiepileptic treatment does not guarantee freedom from cognitive impairments. Particularly thalamic brain areas play an important role in the pathogenesis of neurophysiological discharges, which do correlate with different neuropsychological disorders.

Summary: Also patients with absence epilepsy are at a greater risk for developing cognitive-neuropsychological disorders and behavioral and emotional problems. It seems evident, that the cause of the disorders is a neuropathological process, whose behavioural results can be exacerbated by the epileptic activity per se, side effects of the antiepileptic treatment and social issues. Clinically and scientifically it is important to note that this syndrome is not a unique entity. There are important differences between patients, who on the clinical level are more likely to appear typical.

Abstract

  1. Top of page

Session: Outpatient epileptology in private practices specialized on epilepsy

1 R. Thorbecke 1Epilepsy Center Bethel (Bielefeld, D)

In 1984 a workgroup for the assessment of occupational suitability of persons with Epilepsy (WG) composed of representatives of the German branch of ILAE, rehabilitation centers, occupational physicians of big companies and the state accident insurances (GAI) published recommendations for the assessment of occupational suitability of persons with epilepsy (Arbeitskreis, 1984). Epilepsy was characterized in a perspective of industrial medicine with the severity of seizures, frequency, treatment status and prognosis as the main characteristics. It was stated that suitability can not be assessed for whole professions, as on the one hand the accident risks in the different domains of the same profession are highly variable, on the other hand the occupational risks following from epilepsy vary dependent from the combination of characteristics of epilepsy. So suitability in relation to a specific profession was defined for selected combinations of seizure characteristics with the recommendations – only possible at special work places/possible at the majority of workplaces/no limitations at all.

A first revision of the WG's recommendations was published by the GAI in 1994 as official recommendation ZH 1/191. Now five categories (0 – D) of increasing seizure severity in relation to occupational risks were defined, which easily can be transformed to codes of the International Classification of Functioning, Disability and Health (ICF). As in the official regulations for driving and epilepsy, seizures occurring only during sleep and seizures only with subjective phenomena (aura) were included in the “seizure-free” 0 category. To assess the suitability for specific professions each risk category was weighted by seizure frequency and then related to the activities in the different domains of that profession. So again recommendations individualized to the type of epilepsy and the occupational risks in the different domains of the evaluated professions were made.

A second revision by the WG was published by GAI in 1999 as official recommendation BGI 585. Now the already evaluated professions from the domains of metal industry and electrical engineering were extended to professions in health care and pedagogic. In sum more than 50 professions were assessed in respect to their suitability for different risk categories.

In 2006 the WG accomplished a third revision, which also has been accepted by the GSAI and will be published as revised BGI 585. In this extension not only selected professions but also specific work activities are assessed in relation to the five risk categories. Included are activities with the risk of falling from heights, the risk when driving heavy machinery, e.g., in house construction and the risk of severe accidents by seizures when managing complex industrial control systems, e.g., in the chemical industry.

Already in 2004 the principles for medical examinations in relation to health risks at work published by the GAI (Berufsgenossenschaftliche Grundsätze für arbeitsmedizinische Vorsorgeuntersuchungen, 3. Auflage, Stuttgart 2004) in cooperation with the WG were related to the WG's recommendations. This holds for the principles G25 (driving, controlling industrial systems), G26 (breathing apparatus), G 30 (working in heat), G 41 (work with the risk of falling from heights).

1 U. Specht 1Epilepsy Center Bethel (Bielefeld, D)

A condition involving unpredictable impairment of consciousness or a fall intuitively seems to be at high risk for accidents and injuries. However, recent studies showed that the overall risk of seizure-related injury is small in people with epilepsy and most injuries are minor. As expected, epilepsy related factors including history of generalized convulsive seizures or drop attacks and higher seizure frequency were associated with higher risk for injuries. Wounds, abrasions and concussions were the predominant types of injuries. Similar to the distribution of accidents in the general population, most accidents in people with epilepsy are domestic, followed by street and work accidents.

Recommendations for preventing injuries, e.g., concerning drowning, burns, traffic accidents and fractures, are important especially in patients with pharmacoresistent epilepsy and frequent seizures; however these recommendations must be tailored individually, since unnecessary restrictions and overprotection may increase social isolation, physical inactivity, and psychological dependency.

1 R. Berkenfeld 1Neurology Practice (Neukirchen-Vluyn, D)

Following cardiopulmonary resuscitation (CPR) the consultant neurologist would comment on the patient's concurrent neurological state and provide further prognosis.

The posthypoxic myoclonus, in this context, indicate a clinical phenomenon enabling specific prognostic and therapeutic conclusions to be drawn.

Two types of posthypoxic myoclonus can be identified, acute and chronic.

Early-stage posthypoxic myoclonus is caused most likely by brainsem structures, resulting from too severe damage of the neocortex.This generalized myoclonus, characterized by its transient and self-limiting tendencies, responds poorly to medication. Occurence within 24 hours after CRP would indicate poor outcome, i.e., death or persistent vegetative state.

Chronic posthypoxic myoclonus would develop within days/weeks of patints having regained consciusness. This condition is an action and intention myoclonus, which occur when a limb is moved voluntarily and it is assumed, that it is generated by the cerebral cortex. This myoclonus, alone, can now be identified as the Lance-Adams syndrome. Successfull outcomes have been reported, in individual cases, following treatment with valproate and levetiracetam.

1 S. Arnold 1MVZ Starnberger See GmbH (Berg, D)

Epilepsy surgery is a treatment option for a small proportion of patients with difficult-to-treat focal epilepsy (approx. 5–10%). In general, the time delay between assumption of a medically refractory case of focal epilepsy and the opportunity of presurgical assessments is too long, lasting 10–20 years.

The selection of suitable candidates for epilepsy surgery is an important task, depending on good collaboration between epilepsy centres and neurologists in private practice. Improvement in this coordination leading to reduction of the time span between diagnosis and epilepsy surgery could contribute to normal live in medically refractory patients. Attempts to establish proper syndrome diagnosis as the basis for referal to epilepsy centres can also be done in private practice.

Many factors have to be taken into consideration in a given case: e.g., the underlying etiology of epilepsy, the seizure semiology, and results of EEG and MRI examinations. Based on this information a presumptive epilepsy syndrome can be diagnosed before continous EEG video examinations will be performed. Several questions can be addressed: e.g., are the results pointing to a single and surgically accessible region of seizure onset, or are conflicting results pointing to different brain regions or both hemispheres, which would reduce the chance of successful surgical treatment? Does the patient suffer from temporal lobe epilepsy or is a extra temporal epilepsy more likely? The process of presurgical evaluation in temporal lobe epilepsy is more standardized, invasive EEG investigations would be needed less frequently and even the surgical procedure tends to be less complicated than in extra temporal lobe cases. Several cases will be discussed demonstrating an example of good cooperation between an epilepsy centre and a specially trained neurologist in private practice.

Abstract

  1. Top of page

Session: Hippocampus and language

1 T. Benke 1Medical University Innsbruck (Innsbruck, A)

Semantic memory (SM, Tulving 1972) is a declarative knowledge system dealing with facts, attributes and concepts. Most theories hold that SM is neurally separated from episodic memory (EM) and has its own cognitive architecture. The assessment of SM commonly includes tests of naming, category word generation, attribute verification or semantic relatedness. In contrast, EM functions are tested by memory tasks tapping retrieval and recognition of newly learned information. Also, SM has its neural correlates in a network mainly subsuming lateral and ventral temporal, and inferior frontal regions, whereas EM is primarily related to the medial temporal lobe (MTL). However, several experimental and clinical findings point out that the MTL may also be involved in the processing of semantic materials. If this hypothesis was supported, new aspects would emerge regarding the assessment and treatment of diseases caused by lesions of the MTL, particularly temporal lobe epilepsy (TLE).

One way to elucidate the possible functional relationship between SM and the MTL is via functional magnetic resonance imaging (fMRI). In an attempt to confirm previous, coincidental observations (e.g., Binder et al., 1997), we studied fMRI activation maps when subjects performed a semantic language task, namely to verify two attributes of auditorily presented object names. The paradigm was tested in healthy normal subjects (Bartha et al., 2003), and also in a sample of patients with chronic, intractable TLE (Benke et al., 2006, Bartha et al., 2005, Köylü et al., 2006). Significant MTL activations were found in 94% of neurologically normal right and left handers in the hippocampus, parahippocampal, lingual and fusiform gyri. Activations were mostly bilateral and there was no correlation with handedness. In TLE patients, activations were also found with a high prevalence (82%), but group specific patterns of distribution emerged. Activations in right TLE patients showed an enhancement in the left MTL, whereas this pattern of increased contralateral activation was missing in the left TLE sample (see Fig. 1 & 2).

image

Figure 1. Contrast semantic vs. tone decision for left TLE patients (n = 12, second level one-sample t-test; p<0.0001; extent of cluster size (k)>10). Main activations bilaterally distributed in MTL, ventral and lateral temporal areas.

Download figure to PowerPoint

image

Figure 2. Contrast semantic vs. tone decision for right TLE patients (n = 13, second level one-sample t-test; p<0.0001; extent of cluster size (k)>10). Main activations in left > right MTL areas.

Download figure to PowerPoint

These findings indicate that MTL activations are frequent findings during semantic language processing in both, normals and TLE patients. Based on these results, SM appears to be significantly related to the MTL and adjacent ventral temporal regions, in addition to other brain areas. Furthermore, it appears that the hippocampal formation not only hosts EM but also SM functions. Lesions of the MTL therefore not only impair episodic, but also semantic memory processing, a finding which has previously been demonstrated, e.g., in TLE patients. Functional imaging shows that due to chronic MTL lesions, verbal SM in TLE patients is abnormally distributed in and outside the MTL. As a diagnostic consequence, semantic language processing may also help to test the integrity of the MTL, and possibly to assess plasticity in the MTL system.

1 B. Weber 1University Hospital Bonn (Bonn, D)

It is well recognized, that the incidence of atypical language dominance is increased in patients with focal epilepsy. But the role of the medial temporal lobes (MTL) in the processing and acquisition of language function is debated and still not fully understood. Several studies have challenged the classical view that lesions in vicinity of language areas rather lead to a shift in language dominance than do lesions more spatially apart, as e.g., in the MTL. A study comparing matched groups of patients with focal epilepsy originating either in the left inferior frontal region, the temporoparietal cortex or the MTL showed a strong influence on language lateralization only in patients with MTL-lesions. All lesions were present during early years of development, i.e., during the acquisition of language.

This study—in line with other studies showing the influence of MTL-epileptic activity on the representation of language functions—is strong evidence for the involvement of the MTL in the processing, or acquisition of language. Another, different explanation, on the other hand may be a driving force of the epileptic activity originating from the MTL, which leads to a functional disturbance of the dominant hemisphere and by this to a recruitment of the less-influenced contra lateral cortex. Further studies investigating patients with late-onset-MTL epilepsy of the relationship of MTL material specificity with language dominance could shed light on different effects of early onset vs. late onset MTL epilepsy on the representation of language dominance and might elucidate the question of the involvement of the hippocampus in language acquisition and processing.

1 A. Mecklinger , 1 P. Meyer , 2 A.D. Friederici 11 Saarland University Hospital (Saarbrucken, D) ; 2 Max Planck Institute for Human Cognitive and Brain Sciences (Leipzig, D)

A variety of amnesic patients with medial temporal lobe lesions including the hippocampus show language impairments in particular when processing syntactic structures. However, these findings are more anecdotal and have not received much attention in the neuroscience community. Here we set out to examine in more detail, how the processing of syntactic and semantic violations affects subsequent remembering in an episodic memory test using behavioral and ERP measures as depended variables. In a study phase participants listened to sentences that were either correct (Die Tür wurde geschlossen / The door was being closed) or carried a semantic (Der Ozean wurde geschlossen / The ocean was being closed) or a syntactic (Das Geschäft wurde am geschlossen / The shop was being on closed) violation. In a subsequent incidental memory test, they were visually presented with words they had heard before in one of the three conditions or which were new in the experimental context. Participants classified the words as “old” (previously heard) or “new.” Consistent with prior studies, words carrying a syntactic violation in the study phase elicited a left anterior negativity (ELAN) followed by a P600. Conversely, words carrying a semantic violation and words from correct sentences elicited an N400 component, albeit slightly smaller for correct than for semantically incorrect sentences. Performance in the episodic memory test was not affected by violation condition at study. However, words from the semantic violation condition generated a pronounced mid-frontal old/new effect (the putative ERP correlate of familiarity-based remembering). This effect was absent for words presented in one of the other study conditions. An additional analysis revealed a correlation between N400 amplitude elicited in the study phase and the magnitude of the mid-frontal old/new effect. On the basis of additional findings from a recently conducted fMRI study that show an overlap of those medial temporal lobe (MTL) regions that contribute to the N400 in the study phase and to the mid-frontal old/new effect in the test phase, we assume that MTL regions are an important interface between semantic and episodic memory in that they mediate semantic integration processes that render words accessible on the basis of familiarity in episodic memory.

1 T. Grunwald , 1 I. Oppenheim 1Swiss Epilepsy Centre (Zurich, CH)

It has often been questioned why a phylogenetical old brain structure like the hippocampus should specialize in the processing of verbal stimuli. However, pre- and postoperative memory deficits of patients with temporal lobe epilepsies (TLE) show beyond doubt that hippocampal functions are material specific, at least in part: while left-sided temporal lobe seizures (in patients with left-sided language dominance) are usually associated with verbal memory deficits, seizures originating in the right temporal lobe may – but need not – impair visual memory processes. Likewise, a resection of the left hippocampus may bear a risk of inducing additional memory deficits, while resections of the right hippocampus rarely influence memory performance.

Intracranial recordings of event-related potentials (ERPs) within the human hippocampal system have identified N400 potentials in the anterior mesial temporal lobe (AMTL-N400) that only correlate with verbal memory performance when they are recorded in the language-dominant hemisphere. We found that these potentials are associated with verbal novelty detection: Their amplitudes to “new” but not “old” words in a verbal recognition task correlate with the neuronal density of the hippocampal CA1-region (1) and can be reduced selectively by the NMDA-receptor blocker ketamine (2). Moreover, it could be shown that NMDA-receptor dependent long-term potentiation (LTP), a form of synaptic plasticity with Hebbian characteristics can be readily induced in human hippocampal slices but not in patients with hippocampal sclerosis (3). In these latter patients, we also found a reduction of AMTL-N400 amplitudes similar to the one induced by ketamine. In addition, hippocampal novelty detection is associated with successful encoding for declarative memory (4). Together, these findings suggest that successful encoding for declarative memory is at least in part mediated by NMDA-receptor dependent verbal novelty detection within the human hippocampal system within the language-dominant temporal lobe. However, AMTL-N400s to new words recorded in the non-dominant (right) temporal lobe can predict postoperative verbal memory performance after resection of the left hippocampus (5). This indicates, that the nondominant hippocampus can contribute to verbal memory processes at least to some extent.

On the other hand, while limbic ERPs related to verbal memory processes can clearly be attenuated near the epileptic hippocampus, those related to visual recognition seem not to be affected by the epileptogenic process (6). To identify hippocampal contributions to visual processing and memory we recorded intracranial ERPs directly from within the epileptic and the nonepileptic hippocampus in 12 patients with unilateral TLE during a visual object decision and naming task. While the nonepileptic hippocampus differentiated reliably between real and nonsense objects, this effect was completely eliminated within the epileptic mesial temporal lobe. This finding suggests that the hippocampus proper of both temporal lobes contributes to the semantic processing of visual objects. In fact, significant visual memory deficits seem to be present only in those patients in whom neither the left nor the right hippocampus responds differentially to real and nonsense objects.

Recent studies have shown that the human hippocampus is not only involved in declarative and episodic but also in working memory processes, especially when working memory contents are successfully encoded for long-term storage (7). In addition, findings by the Bonn group (8) demonstrate that the hippocampus plays an important role in establishing language dominance. Since working language acquisition relies on an efficient working memory and both hippocampi can participate in semantic processing, we have now begun to look for lateralized medial temporal contributions to phonological working memory.

  • 1
    Grunwald T, Lehnertz K, Heinze HJ, et al. (1998) PNAS 95: 3193-3197. Sciences USA 95: 3193-3197.
  • 2
    Grunwald T, Beck H, Lehnertz K, et al. (1999) PNAS 96: 12085-12089.
  • 3
    Beck H, Goussakov IV, Lie A, et al. (2000) J Neurosci 15: 7080-7086.
  • 4
    Fernández G, Effern A, Grunwald T, et al. (1999) Science 285: 1582-1585.
  • 5
    Grunwald T, Lehnertz K, Helmstaedter C, et al. (1998) Neuroreport 9: 3375-3378.
  • 6
    Vannucci M, Dietl Th, Pezer N, et al. (2003) Neuroreport 14: 1489-1492.
  • 7
    Ranganath C. (2006) Neuroscience 139: 277-289.
  • 8
    Weber B, Wellmer J, Reuber M, et al. (2006) Brain 129: 346-351.

Abstract

  1. Top of page

Session: Epilepsy and multiple handicaps

1 I.W. Mothersill 1Swiss Epilepsy Centre (Zurich, CH)

Patients with mental retardation and/or physical disabilities (cerebral palsy) can present with both epileptic and nonepileptic seizures, the differentiation and correct diagnosis of unclear attack disorders in these patients is often fraught with problems.

Our experience over the last 30 years of ictal recordings of both epileptic and nonepileptic seizures utilizing polygraphic Video/EEG monitoring in freely moving patients in this patient group have shown that the diagnostic conundrum are in the most cases related to the following points.

  • 1
    Our knowledge of seizure semiology as a basis for the interpretation, diagnosis and classification of epileptic seizures is based on an intact brain (neuronal systems and pathways). In patients with brain damage where these systems are functionally disturbed we often see seizures with an atypical course of events which do not fit to our notions of what an epileptic seizure should look like.
  • 2
    Compared to other patients with nonepileptic attacks those with brain damage show a higher prevalence of abnormal inter-ictal EEG findings, both unspecific and epileptiform that do not directly relate to the attacks presented.
  • 3
    The dominant symptoms presented, such as sudden changes in behavior can be an over, non communicable, reaction to a subtle epileptic seizures.
  • 4
    Falls due either to locomotor deficits abnormal startle response or epileptic seizures can not be correctly diagnosed on clinical representation alone.

We therefore have a two way problem, epileptic seizures interpreted as nonepileptic and nonepileptic attacks misinterpreted as epileptic.

It is therefore essential for the diagnosis of unclear attack disorders in patients with multiple disabilities that ictal polygraphic Video/EEG recordings be performed.

Illustrative examples of problematic cases will be presented.

1 S. Springer , 2 G. Kluger , 1 M. Noterdaeme 11 Heckscher Clinic (Munich, D) ; 2 Treatment Centre (Vogtareuth, D)

Question: Pervasive developmental disorders have an incidence of 0,4%. There is a frequent association with mental retardation, epilepsy as well as externalising behavioral problems. Patients with these comorbid conditions often show a major impairment in their psychosocial adjustment and treatment requires an interdisciplinary approach. Key issues are differential diagnosis and the coordination between treatment of the epilepsy and the psychiatric syndrome.

We investigate the significance of neuropediatric and psychiatric comorbidity as well as the degree of mental retardation in patients with autism und epilepsy with respect to their psychosocial integration and treatment.

Methods: We selected a group of 60 patients with autism and epilepsy out of a sample of 400 children with autism. For all the patients, diagnoses were made according to the multiaxial classification scheme. We recorded the psychiatric syndrome, neuropsychological profiles, begin and type of the epilepsy, actual and planned treatment options (psychotherapy, medication and special educational interventions).

We compared the two groups of patients (group 1: autism, epilepsy and mental retardation, group 2 autism, epilepsy, without mental retardation) with respect to psychiatric symptoms, treatment and psychosocial adjustment.

Results: Epilepsy was diagnosed in 15% of the main sample. This is in accordance with the current literature. In 80% of these cases, autism and epilepsy were associated with mental retardation. 50% of the patients did not have active speech and language skills, and 60% had a severe degree of mental retardation. There was a clear association between an early begin of the epilepsy (before age three) and the degree of mental retardation. In this subgroup, we found significantly more syndromes such as tuberous sclerosis and ESES.

All patients with comorbid conditions (autism, epilepsy and mental retardation) were severely impaired in their psychosocial adjustment. Most of them had additional psychiatric problems such as self-injurious or aggressive behaviors, as well as attention deficits. Therefore, these patients required a complex pharmacological treatment with a combination of antiepileptic drugs, neuroleptic drugs and stimulants. The modified pharmacological sensitivity of children with mental retardation had to be considered.

These patients also needed a comprehensive, multiprofessional psychiatric and educational treatment. 20% of the sample was in psychiatric inpatient treatment or lived in special residential settings.

Discussion: Patients with the multiple comorbid conditions such as epilepsy, autism and mental retardation show more neurological and psychiatric problems than patients with autism and epilepsy without mental retardation. These patients show major problems in their psychosocial adjustment. They need a comprehensive treatment approach with an extensive cooperation between child neurology and child psychiatry as a key element.

Abstract

  1. Top of page

Session: Dissociative disorders

1 M. Schmutz 1Swiss Epilepsy Centre (Zurich, CH)

Background: Patients with psychogenic nonepileptic seizures (PNES) form a significant and still challenging subgroup amongst all seizure patients treated in tertiary epilepsy centers. At the diagnostic level simultaneous EEG-Video-Monitoring of a typical seizure is generally accepted as the gold standard, however the findings relating to the psychological and psychiatric basis of PNES remain unclear and heterogeneous. Old concepts of hysteria—after their slow and drawn out agony during the 20th century—can no longer serve as an etiological background. A pure neurological approach in which PNES are considered as a nosological entity in its own right also seems to fail in delivering better perspectives for understanding and treating PNES. Outcome data are also very heterogeneous. There are only scant empirically based hints at relevant factors predicting outcome. Therefore indication and prognosis for psychotherapeutic treatment remain clinical evidence based and lack stronger empirical foundation.

Method: In the framework of a prospective outcome study started 2004 at the Swiss Epilepsy Center in Zurich, the predictive power of relevant features related to psychodynamics, relationship and personality is investigated. All patients with EEG-Video confirmed diagnosis of PNES are examined with a semistandardized psychodynamic oriented interview (OPD, Operationalized Psychodynamic Diagnostics) to determine their psychodynamic conflicts and the level of structural personality integration. In addition two self-rating questionnaires (IIP-D and Neo-FFI) are administrated in order to evaluate quality and degree of interpersonal conflicts and basic personality traits. Outcome at one, six and twelve months after diagnosis is measured by seizure frequency, self-rated seizure intensity and seizure related interruption of activity. Furthermore an adaptation of the QOLIE-31 is administrated to evaluate self-rated quality of life as well as tendencies towards depression and anxiety. The development of psychosocial and vocational status is followed too.

Goal: The study is to determine psychological factors predicitve for outcome of PNES patients. Furthermore a classification of PNES patients based on psychological and psychiatric features and with a useful relation to therapy and prognosis is intended.

Preliminary results and conclusions of the still ongoing study will be presented and discussed at the May 2007 Epilepsy Congress in Basel/Switzerland.

Abstract

  1. Top of page

Oral session 3

1 M. Majores , 1 S. Schoch , 1 J. Schramm , 1 G. Engels , 1 C.E. Elger , 1 A. Becker 1University Clinic Bonn (Bonn, D)

Rationale: Focal cortical dysplasias (FCD) are characterized by a localized malformation of the neocortex and the underlying white matter. Balloon cells, similar to those observed in tuberous sclerosis (TSC) are present in a significant number of cases (FCDIIb). Histopathological similarities indicate that FCDIIb may be pathogenetically related to the tuberous sclerosis complex (TSC), caused by mutations in either TSC1 or TSC2, that normally constitute a tumor suppressor mechanism. Accumulation of coding allelic variants affecting exons 5 and 17 of TSC1 has been observed in many FCDIIb patients (Becker et al., Ann Neurol 2002). Here, we have studied the potential functional relevance of coding allelic variants of TSC1 found in FCDIIb with respect to the interaction of the TSC1 (hamartin)/TSC2 (tuberin) complex.

Methods: Sequence variants of the TSC1 allele found in FCDIIb patients were introduced in the full-length TSC1 cDNA (kindly provided by Dr. M. Nellist, Amsterdam) by site-directed mutagenesis. We have further generated TSC1 cDNA variants with mutations found in TSC patients, i.e., by introducing premature stop codons. Protein interaction assays were performed to analyze a potentially compromised interaction between the hamartin and tuberin.

Results: TSC1 sequence variants were successfully introduced by site-directed mutagenesis. Compromised interaction of the hamartin/tuberin complex was most pronounced in TSC1 mutants bearing the premature stop codon at position R692X (p<0,01). Sequence alterations of TSC1 present in FCDIIb (H732Y) also resulted in significantly compromised TSC1/2-interaction compared to wild-type TSC1 (p<0,05). The latter was comparable to findings obtained by the expression of the TSC-determining allele variant R786X.

Conclusion: Our data suggest, that allelic variants with amino acid exchange of the TSC1 gene accumulated in FCDIIb may be associated with compromised hamartin/tuberin interaction. The functional relevance is currently studied in cell culture and animal systems.

Supported by DFG (SFB TR3, Emmy-Noether-Program), BONFOR, the BMBF German

Israeli program and the Deutsche Krebshilfe.

1 R. Meier , 1 U. Häussler , 2 A. Aertsen , 3 A. Depaulis , 1 U. Egert 11 Bernstein Center for Computational Neuroscience (Freiburg, D) ; 2 Albert-Ludwigs-University (Freiburg, D);3 Inserm U836 University Joseph Fourier-CEA (Grenoble, F)

The Mesial Temporal Lobe Epilepsy (MTLE) syndrome is among the most prevalent forms of focal epilepsies, however, network structures and dynamics involved in the generation of seizures are still fairly unknown. MTLE is associated with severe changes in the hippocampal network histology, termed hippocampal sclerosis and recurrent epileptic seizures occurring in temporal brain areas. MTLE is usually pharmacoresistant and surgical resection of involved brain areas is often needed to suppress seizures.

To advance less invasive therapy options it is necessary to investigate on the time scale of processes initiating epileptic seizures.

We addressed these questions using the in vivo intrahippocampal kainate model for MTLE in mice. These mice show histological changes comparable to human hippocampal sclerosis in the injected hippocampus. We recorded recurrent epileptiform activity (EA) in the injected and in the contralateral, intact hippocampus. We measured interhippocampal coherence to investigate short term seizure generation processes on a timescale that would allow acute intervention.

Interhippocampal coherence decreased significantly in high frequency bands (> 80 Hz) up to 12 seconds before the onset of EA. This indicates an early decoupling of the ipsilateral hippocampus from the contralateral, intact hippocampus during the seizure initiation phase. Additionally, this time scale limits the possible range for cellular and mechanisms leading to increased synchronicity in the network, ulitmately initiating the seizure.

Acknowledgements: This work was supported by the German Federal Ministry of Education and Research (BMBF grant 01GQ0420 to BCCN Freiburg), INSERM, Fondation pour la Recherche Medicale, Fondation de l'Avenir and DAAD (U.H.).

1 R. Hopfengärtner , 1 V. Bauer , 1 F. Kerling , 1 H. Stefan 1University Erlangen-Nuremberg (Erlangen, D)

Objective: An efficient method for the offline-detection of epileptic seizures for long-term scalp EEG is described. The goal was to develop an algorithm for seizure detection applicable to different types of seizures, which provides high sensitivity and acceptable values for the selectivity (number of false detections) without requiring any a priori information on the individual seizures of the patient.

Methods: For the detection of characteristic electrographic changes during the evolution of seizures we developed an efficient and robust method based on power spectral analysis techniques. No additional pre-processing of EEG data is required. The integrated power for three different standardized multichannel seizure detection montages for two characteristic frequency bands has been investigated and compared to each other. The seizure performance measures sensitivity and selectivity have been determined. For the adequate evaluation of the dynamics of seizures we introduced the seizure decision value (SDV). For the calculations we used the same parameters for all patients taking an appropriate artifact rejection into account. The parameters have been derived from “test data” not included in the study.

Results: Fourteen patients who underwent preoperative evaluation for medically intractable epilepsies were included in the study. A total of 2358 h scalp EEG recordings containing 123 seizures without any preselection of data sets were evaluated. The highest averaged sensitivity was 88.7% and selectivity (false-positive errors/hr) was 0.38/h of the referenced Fz-Cz-Pz montage. The other montages (bipolar longitudinal, common average) yielded better selectivity values but lower sensitivities.

Conclusions: The proposed method provides already for identical parameters used for all EEG data for the referenced Fz-Cz-Pz montage high values for the sensitivity while the rate of false-positve errors is quite low. In addition, the method is very fast and provides an easy user tuneability of parameters for the detection of individual patient specific seizures. It is assumed, that the proposed method is capable to make a significant contribution to the diagnostic gain of long-term scalp EEG monitoring during presurgical evaluation.

1 M. Hauf , 1 C. Ozdoba , 1 J. Mathis , 1 R. Wiest 1University Berne (Berne, CH)

Purpose: Altered mental status due to nonconvulsive status epilepticus (NCSE) is clinically difficult to discriminate from a postictal state, hence it is of crucial importance for therapy decision. To test the hypothesis whether dynamic brain perfusion CT (P-CT) identifies increased regional cerebral blood flow in patients during NCSE, we compared P-CT findings in patients with NCSE to patients in different postictal conditions.

Method: Twenty patients (mean age 49 years, range 9-80) underwent P-CT and EEG immediately after an epileptic seizure. Two methods of evaluation were performed. As quantitative assessment of brain perfusion, asymmetry indices (AI) between the affected cortical region and the corresponding region of the unaffected hemisphere were calculated for regional cerebral blood volume (rCBV), regional cerebral blood flow (rCBF), and mean transit time (MTT). Regional perfusion changes were compared to EEG findings. For qualitative assessment, images were individually rated by 3 board-certified neuroradiologists.

Results: Eight patients in NCSE had increased regional cortical perfusion. Six patients showed postictal slowing on EEG corresponding to an area of regional hypoperfusion. CT and EEG were normal in six patients with a first epileptic seizure. K-means Cluster analysis separated NCSE from other groups in rCBF, MTT and rCBV parameters statistically. Consensus for the visual evaluation of the perfusion maps was reached for every patient with NSCE.

Conclusion: P-CT may help to identify patients with NCSE during emergency workup. This technique provides important information to neurologists and emergency physicians in the difficult clinical differential diagnosis of altered mental status after a convulsive epileptic seizure.

1 M.T. Lutz , 1 T. Mayer 1Epilepsy Center Kleinwachau (Radeberg, D)

Purpose: EpiTrack is a short neuropsychological measure for the assessment of attention and executive functions in patients with epilepsy (Lutz MT et al. Epilepsy Behav 2005;7(4):708-714.). It has been proved to be sensitive to adverse cognitive side effects of antiepileptic drugs (AEDs). After its introduction, the question arose if it is a one-dimensional or a two-dimensional measure.

Method: We performed a maximum likelihood confirmatory factor analysis of the EpiTrack in a sample of 220 patients with epilepsy. This sample was independent to the standardization sample. Two hypothesized latent variable models underlying the covariation of the six subtests were evaluated. The first model included a verbal and a nonverbal factor, the second model was a general factor model. For both models, the clinical validity was analyzed in a subgroup of 154 patients.

Results: The goodness of fit of the two-dimensional model was slightly superior to the general factor model (GFI = 0.956 versus 0.945). With respect to the clinical validity, the age-corrected correlation between the visual factor and the number of AEDS was −.258 (p = .001); the respective correlation of the verbal factor was -.210 (p = .008). The correlation between the one-factorial composite score and the number of AEDs was in the same range (r =−.255, p = .001).

Conclusion: Even if the fit measures of the two-factor model were slightly better in explaining the factor structure of the EpiTrack, the additional factor did not provide significant gain in clinical validity. As the more parsimonious general factor solution provides comprehensive clinical sensitivity, the use of a composite score seems to be justified.

1 T. Falkenstetter , 1 E. Trinka , 1 J. Dobesberger , 1 A.R. López-Rolón , 1 G. Walser , 1 M. Delazer , 1 T. Benke 1University Hospital Innsbruck (Innsbruck, A)

Background: Intractable temporal lobe epilepsy (TLE) is associated with cognitive impairment. Preoperative neuropsychological evaluation helps to identify spared and impaired cognitive functions (i.e., the functional deficit zone) and to lateralize and localize the seizure focus. Reading skills are essential for education, training and career and for using diverse media to gather information—thus, they have an influence on the social position a person attains. Disorders affecting the temporal lobe may compromise reading abilities—we hypothesize that this may also be the case in pharmacoresistent TLE. Up to date, few studies have investigated interictal reading skills in adult patients with TLE. The lateralizing value of assessing reading abilities is still largely unknown.

Specific Aims: The present pilot project prospectively collects and analyzes data on reading abilities in a sample of patients with TLE. Particularly practical skills relevant to effective reading such as reading speed and text comprehension are evaluated to test the impact of a reading impairment and its possible effect on the patients' everyday lives.

Methods: Patients with left sided and right sided TLE are compared with each other, as well as with healthy controls, regarding their reading skills (reading of nonwords, regular and irregular words, reading fluency, text repetition, apprehension of grammatical structures and text comprehension using standardized and newly designed tests; reading habits and reading socialization using questionnaires). This data is correlated with demographic (age at testing, sex, education), medical (aetiology, age at epilepsy onset, duration of epilepsy, seizure frequency, seizure severity, therapy) and neuropsychological (profile of other higher cognitive functions) data gained from the Innsbruck Epilepsy Surgery Program (INES).

Preliminary results: At present, the group of healthy controls (n = 10) generally shows a better performance in all tested reading skills in comparison with the group of TLE patients (n = 12). A significant difference is found in spelling, in oral repetition of a read text and in the apprehension of grammatical structures. Although our results are preliminary, they support our hypothesis that reading deficits can be found among TLE patients. Increasing patient data will allow us to reveal supposed differences between left sided and right sided TLE patients.

1 E. Pauli , 1 C. Däubler , 1 M. Schwarz , 1 H. Stefan 1University Erlangen (Erlangen, D)

Purpose: We focussed on emotional-affective features of patients with temporal lobe epilepsy (TLE) as well as on aspects of personality in contrast to the general population and postoperative changes taking into account the focus' hemispheric location.

Methods: Pre- and postoperative examination of 120 TLE patients (N = 52 focus in the speech-dominant, N = 68 focus in the nondominant hemisphere) using standardized psychometric inventories (BDI, STAI, FPI, EPI) and subjective rating scales.

Results: TLE patients showed significantly higher depression (p = .01) and anxiety scores (p = .00) than the general population. Clinical relevant symptoms of depression were seen in 9% of the patients, 53% showed substantial elevated anxiety levels. Concering personality aspects TLE patients scored lower in satisfaction with life, effort orientation and extraversion (each p<.01). Scores on inhibition (p<.01), physical discomfort, emotional lability (each p<.05), excitability and health concern (each p<.10) were higher compared with the general population.

Patients improved after surgery and showed significantly lower depression and anxiety rates (each p<.01) and amelioration in neuroticism, emotional lability (each p<.01), satisfaction with life, level of stress perception (each p<.05) and extraversion (p<.10). Furthermore patients quoted better on subjective rating scales.

In presurgical evaluation, TLE-patients with epileptogenic focus in the speech-dominant hemisphere scored significantly worse than the non-speech-dominant group in extraversion, effort orientation (each p<.05) and inhibition (p<.10). No significant hemisphere-specific influence was detected in postsurgical evaluation.

Interestingly there have been noticeable sex differences: On distinct personality dimensions (inhibition, extraversion, health concern (each p<.01), physical discomfort, emotional lability and neuroticism (each p<.05)) men scored significantly worse than women. After surgery womens anxiety levels were in average significantly lower, 10% of women and 39% of men showed postsurgically clinical relevant anxiety scores.

Conclusion: TLE patients benefit from epilepsy surgery in regard to depression and anxiety comorbidity. In comparison to women, men gain less and seem to constitute a risk group concerning emotional-affective disorders. Contrary to the literature on hemispheric specification, our study showed a weaker influence of focus lateralization on emotional-affective variables than expected.

1 R. Meier , 2 A. Schulze-Bonhage , 2 A. Aertsen 11 Bernstein Center for Computational Neuroscience (Freiburg, D) ; 2 Albert-Ludwigs-University (Freiburg, D)

Epileptic seizures can cause a variety of temporary changes in perception and behavior. In the human EEG they are reflected by multiple ictal patterns, where epileptic seizures typically become apparent as characteristic, usually rhythmic signals, often coinciding with or even preceding the earliest observable changes in behavior. Their detection at the earliest observable onset of ictal patterns in the EEG can, thus, be used to start more detailed diagnostic procedures during seizures and to differentiate epileptic seizures from other conditions with seizure-like symptoms. Recently, warning and intervention systems based on the detection of ictal EEG patterns have attracted increasing interest.

Since the workload involved in the detection of seizures by human experts is quite formidable, several attempts have been made to develop automatic seizure detection systems. So far, however, none of these found widespread application.

Here we present a novel procedure for generic, online and real-time automatic detection of multimorphological ictal-patterns in the human long-term EEG and its validation in continuous, routine clinical EEG recordings from 57 Patients with a duration of approximately 43h. We analysed 91 seizures (37 focal, 54 secondarly generalized) representing the 6 most common ictal morpgologies (Alpha, Beta, Theta and Delta- rhythmic activity, Amplitude depression and Polyspikes).

We found that taking the seizure morphology into account plays a crucial role in increasing the detection performance of the system. Moreover, besides enabling a reliable (mean false alarm rate <0.5/h, for specific ictal morphologies <0.25/h), early and accurate detection (average correct detection rate >96%) within the first few seconds of ictal patterns in the EEG, this procedure facilitates the automatic categorization of the prevalent seizure morphologies without the neccessity to adapt the proposed system to specific patients.

Acknowledgements: We thank Armin Brandt and Carolin Gierschner for supplying the EEG long term data from the Freiburg Epilepsy Center. Partial funding for this research was supplied by the Committee for Research, University Clinics, Freiburg and the German Federal Ministry for Education and Research (BMBF, grant 01GQ0420 to BCCN Freiburg).

1 B. Feddersen , 1 M. Kilian , 2 C. Deransart , 1 S. Noachtar 11 University of Munich (Munich, D) ; 2 University of Grenoble (Grenoble, F)

Introduction: The aim of the present study was to evaluate whether in temporal lobe epilepsy (TLE) patients seizures with unilateral hand dystonias are less likely to generalize secondarily than seizures with head versions.

Methods: We searched the data base of the University of Munich Epilepsy Monitoring Unit for the terms dystonia, version and temporal lobe epilepsy. Forty-nine TLE patients with seizure evolutions characterized by either dystonia (n = 24) or head version (n = 25) were included in the study. All patients were off antiepileptic medication. Seven patients had seizures with both, dystonia and version in the same seizure evolution and other seizures during which only dystonias or versions occurred. Only patients with high quality MRI and EEG-video recordings were included. Sixty-four seizure evolutions with dystonia were intraindividually compared with 140 seizures of these patients without dystonia. In addition, 54 seizures with version were compared with 147 seizures of the same patients without version. All seizure evolutions have been classified prospectively independent of the other clinical results based on a semiological seizure classification system (Lüders et al., 1998).

Results: The rate of secondary generalization was significantly higher in seizures with version (96%, 52 of 54) than seizures with dystonia (6%, 4 of 64) (p<0.05). The comparison of the rate of secondary generalization in seizures of the same patients with and without dystonia (and no version) revealed that the rate of secondary generalization was lower in seizures with dystonia (6% vs. 12%; not significant). However, the difference in rate of generalization was much higher (p<0.05) in the seizures with version: 96% of the seizures with version generalized secondarily as opposed to 17% if no version was present p<0,05. All 17 seizures, during which both dystonia and version occurred had dystonias prior to the version. They generalized more often than compared when dystonia and version were absent (71% versus 29%) p<0,05.

Conclusions: In TLE, epileptic spread to the frontal eye field, which is involved in generating head version is more likely to eventually lead to secondary generalization than spread to the basal ganglia, which is considered the generator of unilateral hand dystonia. These findings support a differential fazilitating and inhibiting role of different spread patterns of epileptic activity.

References:

Lüders HO, et al. Semiological seizure classification. Epilepsia 1998;39:100613.

Abstract

  1. Top of page

Session: Mortality of the epilepsies and SUDEP

1 E. Trinka 1University Hospital of Neurology (Innsbruck, A)

Patients with epilepsy carry a risk of premature death which is on average two to three times higher than in the general population. Deaths can be attributed to the underlying cause of epilepsy, to epilepsy itself, or to the treatment of epilepsy. The mortality rate, case fatality, standardized mortality ratio (SMR) are calculated from cohort studies, which may be either prospective or retrospective, population based or based on selected populations. The proportional mortality consideres only the proportion of deaths in patients with epilepsy and may be calculated, when denominators are not available, like in developing countries. Studies on newly diagnosed patients (incident cohorts) come to different estimates than those analysing prevalent cases. The problem of standatisation, cohort effects and competing risks have to be adressed adaequately when analysing mortality in patients with epilepsy. Finally the mortality measures have to be translated into life expectancy and lost years of life which are a more meaningful measure of increased mortality to both the patients and the health system.

Despite different study designs and populations patients with refractory seizures carry the highest risk compared to those who are seizure free whose risk is compareble to the general population. Sudden unexpexted death in epilepsy patients (SUDEP) is the most important cause of epilepsy-related deaths especially in the young and middle aged groups with drug resistent epilepsy. Population based studies with long-term follow up demonstrated that the first years after diagnosis carry the highest risk of death, while in the later year the mortality decrases. Recent studies suggest that improved seizure control may help to reduce the risk of deth in patients with epilepsy. The epidemiological data on mortality, their determinants and risk factors as well as methodological issues are critically discussed.

Abstract

  1. Top of page

Poster session 1.1. Clinical epileptology: new antiepileptic drug

1 G. Kuchukhidze , 1 M. Rauchenzauner , 1 T. Gotwald , 1 E. Haberlandt , 1 S. Felber , 1 E. Trinka 1Innsbruck Medical University (Innsbruck, A)

Purpose: Joubert syndrome (JS) is an autosomal recessive neurodevelopmental disorder characterized by molar tooth malformation on MRI and clinical presentation with neonatal hypotonia, episodic hyperpnea, ataxia, developmental delay, mental retardation and oculo-motor apraxia. Abnormalities of dentate nucleus are rare in JS and their clinical significance is poorly understood (Maria, 1999). Epilepsy found in concert with infratentorial dysplasias or in association with JS is also rarely reported.

We present three patients with JS, agenesis of dentate nucleus and epilepsy in an attempt to shed more light on this rare association.

Patients and methods: Patient 1, a 7-year-old, mentally retarded boy with an unremarkable family history of neurological diseases had muscular hypotonia, infantile spasms, mild ataxia and global developmental delay.

Patient 2, a 23-year-old woman suffering from symptomatic epilepsy with generalized tonic–clonic seizures, learning disability, convergent squint, mild gait ataxia and dysmorphic facial features. Her seizures began at the age of 12 and were poorly controlled by different anticonvulsants. Her family history was negative with regard to neurological diseases.

Patient 3, a 12-year-old boy with delayed developmental milestones, mental retardation, generalized epilepsy with tonic seizures and absences, hypoacusis, convergent squint and gait ataxia. His family history of epilepsy or other neurological diseases was unremarkable.

All patients underwent thorough clinical examination, EEG recordings, neurophychological testing and high resolution MRI (1.5 T). MRI sequences included T1-weighted gradient echo 3D MPRAGE with and without IV contrast application, axial and coronal T2-weighted turbo spin echo, T1 inversion recovery, T2-weighted fast FLAIR and diffusion weighted sequences.

Results: Three patients presented with typical clinical and MRI features of JS. Additionally all patients had agenesis of dentate nucleus on MRI and all suffered from symptomatic generalized epilepsies. Parents were not related in any case and patients were born after uneventful pregnancy and labor. Cerebellar foliation, basal ganglia, cerebral cortex and pituitary gland were normal. No white matter abnormalities were observed and MRI changes were not progressive over time.

Conclusion: Abnormalities of deep brain structures (brainstem and cerebellum) might contribute to the pathophysiology of symptomatic generalized epilepsies in patients with JS.

1 S. Dempewolf , 1 S. Bunten , 1 S. Happe 1Hospital Bremen East (Bremen, D)

Objective: Sleep deprivation (SD) is a commonly used method to increase the diagnostic yield of the electroencephalogram (EEG), especially in the evaluation of patients with suspected epilepsy. In different studies, the observed rate of epileptic activity achieved by SD is varying from 13% to 52%.

Patients and methods: We retrospectively studied all EEGs of patients aged 16 years or older examined in our department between January 2000 and June 2006, who, after having undergone an initial routine EEG recording, had been examined with a second recording after full-night SD (n = 831). Most of the patients were admitted to our hospital because of a loss of consciousness of unknown origin (n = 634). Forty-three patients were suffering from known epilepsy of unknown etiology. One hundred eight patients were referred to our department because of a focal neurological deficit and 46 patients because of other complaints.

Results: Altogether, routine EEG showed epileptic activity in 4.9% and EEG following SD in 14.6%. The percentage of occurrence of epileptic activity did not differ in routine EEG (1.9%) and EEG after SD (2.6%) in patients without epileptic symptoms (n = 313). In patients diagnosed a first epileptic seizure (n = 270), however, the proportion was 2.6% vs. 6.3%, in focal epilepsy (n = 153) 9.8% vs. 40.5%, in generalized epilepsy (n = 38) 31.6% vs. 73.7%, and in unclassified epilepsy (n = 31) 6.5% vs. 12.9%. Epileptic discharges were recorded in a routine setting in 3.8% and after SD in 7.7% in patients (n = 26) whose symptoms remained unexplained.

Conclusion: Our data confirm the diagnostic value of EEGs following sleep deprivation. In many cases, it is helpful to distinguish between epileptic and non epileptic seizures and to specify epileptic syndromes.

1 I. Unterberger , 1 R. Unterpertinger , 1 G. Bauer , 1 E. Trinka 1Medical University Innsbruck (Innsbruck, A)

Background: Familial myoclonic epilepsies are characterized by an adult onset of epileptic seizures [myoclonias, generalized tonic–clonic seizures (gtcs)], a family history of seizures and a benign course of the disease. Genetic heterogeniety could be demonstrated.

Aim: The aim of our study was to further characterize a large pedigree comprising 237 family members with a history of familial myoclonic seizures.

Results: Thirty-four family members could be studied. All of them underwent a detailed clinical workup including neurological exam, natural history, history of seizures, history of cognitive dysfunction or any movement disorder, family history and EEG. Probands were rated as “clinically affected,” if they exhibited a seizure history of myoclonias only, myoclonias plus gtcs, myoclonias plus febrile seizures (fs) or fs only. Fifteen alive patients (12 m, 3 w, mean age 38 yrs, mean age at myoclonia onset 39 yrs) fulfilled the criteria “clinically affected.” Five out of fifteen showed stimulus sensitive myoclonias, six had a history of myclonias plus gtcs, one exhibited myclonias plus fs, two suffered from fs only and two further described “ a sudden thrill or shaking of the body” in adolescence, which could not be further classified. Three patients showed mild dysmetria. Patients were members of basically three subfamilies within the large pedigree. All patients showed a benign, nonprogressive course of the disease with a good response to antiepileptic drug treatment.

Discussion: We present a family with a benign familiar myoclonic epilepsy with adult onset and possible autosomal dominant inheritance. Worldwide more than 50 families are described but confusingly different terms (e.g., beningn adult familial myoclonic epilepsy 1, autosomal dominant cortical myoclonus and epilepsy 2, familial adult myoclonic epilepsy 3, familial cortical myoclonic tremor with epilepsy 4) are used. A more standardized description would be helpful.

References: 1 Mikami et al. Am J Hum Genet 1998; 2 Guerrini et al. Brain 2001; 3 Labauge et al. Neurology 2002; 4 van Rootselaar et al. J Neurol 2002.

1 D.-M. Altenmüller , 1 J. Vesper , 1 V. Van Velthoven , 1 J. Zentner , 2 H.-J. Huppertz 11 Neurological Clinic of Freiburg (Freiburg, D) ; 2 Swiss Epilepsy Centre (Zurich, CH)

Pharmacoresistant focal epilepsies which are regarded as ‘cryptogenic’ because of normal findings of high resolution magnetic resonance imaging (MRI) often pose a particular diagnostic and therapeutic challenge.

In order to identify reliably the epileptogenic zone and its extension for epilepsy surgery, an invasive video EEG monitoring with implanted intracranial electrodes is frequently required.

A morphometric MRI analysis, i.e., postprocessing of the MRI data and comparison with a normal database regarding cortical gyration, cortical thickness and differentiation of the grey-white matter junction, may help to detect and delineate potential focal cortical dysplasias. This can decisively influence the planning of electrode implantation with respect to the precise location of subdural grid and strip electrodes and intracerebral depth electrodes.

Exemplarily, we report on a patient with cryptogenic focal epilepsy in whom the combined use of a morphometric MRI analysis and—for the first time targeted through this—invasive EEG including a depth electrode implanted stereotactically at the base of a left frontal sulcus in addition to a subdural grid on the overlying cortex permitted a stereo-EEG evalution of the primary epileptogenic area and allowed for a very circumscribed resection of a focal cortical dysplasia, which led to seizure freedom. Spatiotemporal EEG analysis and histopathologic findings confirmed that the depth electrode guided by morphometric MRI analysis was placed in the epileptogenic core of the focal cortical dysplasia.

1 H. Korall 1Center of Metabolic Diagnostics (Reutlingen, D)

Introduction: Clinical symptoms of many inborn errors of metabolism (IEOM) are partially unspecific like intractable grand mal and absence seizures, ataxia, muscular hypotonia, mental retardation and progressive extrapyrimidal movement disorders. For the clinician it may be difficult to decide which biochemical test may be useful to detect epilepsies based on a metabolic disorder. Using tandem mass spectrometry (TMS) we developed multiple disease categories screening analysis (MULTISCAN)—an advanced selective screening method in urine, dried blood spots and plasma for detecting inborn errors of metabolism (IEOM). This broadband screening method facilitates the diagnosis of epilepsies based on metabolic disorders.

Method: Urine metabolites were separated by a column and quantified by TMS. We show new applications with TMS screening for common and rare metabolic diseases by partially using keymetabolites for specific categories of IEOM. Plasma metabolites like homocysteine, phytanic acid and very long-chain-fatty acids (VLCFA) were detected, too. In most cases quantification of pathognomic metabolites is based on stable isotopes dilution technique as internal standards.

Results: In just a few minutes run all relevant aminoacids and additional pathognomic metabolites are detected. Key metabolites e.g., pipecolic acid are used for a urine preselection in peroxisomal disorders, guanidinoacetate and creatine for the diagnosis of guanidinoacetate methyl transferase deficiency (GAMT). GAMT-deficiency shows epileptic seizures, is automatically screened in each urine sample by TMS and may be treated with oral creatine substitution and additional dietary arginine restriction and ornithine supplementation. We made the diagnosis of 5 GAMT patients of 28 patients described worldwide. Fabry disease by quantification of globotriaosylceramides (GL-3) and disorders of purine and pyrimidine metabolism are screened, too. Peroxisomal disordes may be confirmed by quantification of phytanic acid and VLCFA in plasma. These disorders are not detected by extended newborn screening in dried blood spots. Furthermore by this applications we found new approaches in quick, reliable and adequate therapy control of IOEM.

Summary: Especially tandem mass spectrometry with MULTISCAN in dried blood spots, plasma and urine and gas chromatography are rapid, sensitive and reliable diagnostic tools for an effective, selective screening and therapy monitoring of IEOM associated with epilepsy.

1 R. Kraus , 1 B. Budig , 1 S. Klarmann , 1 M. Naumann 1Hospital Augsburg (Augsburg, D)

Objective: Myoclonic status epilepticus is rarely observed in patients with primary generalized idiopathic epilepsy. It is also uncommon in adult-onset idiopathic epilepsy or in senile myoclonic epilepsy described in patients with Alzheimer-type dementia. We present the case of an 87-year-old woman with typical myoclonic jerks and EEG features resemble juvenile myoclonic epilepsy to discuss whether a de novo impulsive petit mal status epilepticus may appear in an 87-year-old woman.

Methods and results: An 87-year-old woman was admitted to our hospital because of an acute disturbance of behavior and myoclonic jerks of her upper limbs occurring every ten seconds since several hours. She had the history of severe Alzheimer and vascular dementia. Epileptic seizures were never seen before. The myoclonic jerks were associated with brief bursts of generalized polyspike discharges of 9-10 Hz. Background EEG activity shows a diffuse slowing about 5 Hz. Random interictal spikes frontal were seen. The jerks were reduced after injection of 2.5 mg midazolam and completely abolished after intravenous administration of 1000 mg levetiracetam.

Discussion: Although the clinical and electroencephalographic findings at onset resemble classical impulsive petit mal status, we wonder whether de novo primary idiopathic epilepsy may appear in late age. Symptomatic myoclonic status associated with metabolic, toxic and post anoxic encephalopathies are much more common in this age. But these encephalopathic myoclonias usually are associated with absence of epileptiform discharges and are much more difficult to treat. Because of vascular cerebral lesions, we discuss the possibility of primary frontal lobe epilepsy with secondary bilateral synchrony. Furthermore the cortical and subcortical dysfunctions in Alzheimer disease itself could cause epileptic myoclonias. Myoclonic epilepsy is being increasingly recognized as a late-onset complication in patients with Alzheimer-type dementia. In series of patients with Alzheimer's disease 10% had seizures or myoclonus but they never occurred together. Interestingly in our case the bursts of polyspike waves on the first EEG were replaced by continuously periodic triphasic sharp waves occasionally occurring in patients with rapidly progressive Alzheimer dementia.

Conclusion: Some findings suggest that primary myoclonic status epilepticus can occur even in late age, but it needs further observations to distinguish them from symptomatic encephalopathic myoclonias.

1 H. Muhle , 2 C. Dreiwes , 2 R. Boor , 1 A. Van Baalen , 1 U. Stephani 11 University of Kiel (Kiel, D) ; 2 Northern German Epilepsy Centre Raisdorf (Kiel, D)

Introduction: In 1977 Jeavons described a photomyoclonic reflex epilepsy that manifests with 1. eyelid myoclonia (rapid blinking) immediate after eye closure (not occurring in the dark), 2. upwards deviation of the eyes, 3. accompanied by generalized spike-wave (sw) discharges in the electroencephalogram, and 4. independent brief absences (also accompanied by sw). All patients exhibited photosensitivity during intermittent photic stimulation (IPS). The “Jeavons syndrome” (eyelid myoclonia with or without absences, EMA) represents an idiopathic generalized epilepsy (IGE).

Eyelid myoclonia can be separated from myoclonic absences, where myoclonia occur more regular (3-4/s) and often synchronously with the spikes and waves but not associated with eye closure or photostimulation. Loss of consciousness is necessary for diagnosis.

Patients: We describe two children who both developed pure eyelid myoclonia (without loss of consciousness) after a period of time with myoclonic absences (with loss of consciousness) some years ago.

In patient 1 epilepsy started with myoclonic absences at age seven years. A tonic–clonic seizure appeared at age 13 years. At the age of 16 years, she showed eyelid myoclonia after spontaneous eye closure diagnosed as EMA. The family history includes several IGE and photosensitivity. Her brother exhibited juvenile myoclonic epilepsy with tonic–clonic seizures. The index patient, her mother, her epileptic brother, and two healthy siblings are photosensitive. The pedigree is shown below (dark quadrant = photosensitive).

Patient 2, a boy, exhibited idiopathic generalized epilepsy with myoclonic astatic seizures (Doose syndrome), tonic–clonic seizures and myoclonic absences. Later he developed epileptic eyelid myoclonia (without absences) and occasional pure absences. He was also photosensitive. A syndrome shift from Doose syndrome to EMA was diagnosed.

inline image

Conclusion: The accurate discrimination between epileptic eyelid myoclonia and myoclonic absences differentiates the Jeavons-syndrome from pyknoleptic and spanioleptic absence epilepsies. Therefore consciousness should be tested during the occurrence of sw. This is also important for distinguishing these epilepsies for genetic analyses.

1 M. Fischera , 1 K. Anneken , 1 C. Kellinghaus , 1 I.W. Husstedt , 1 S. Evers 1University Hospital Munster (Munster, D)

Objective: To determine the etiological and clinical aspects of seizures in patients with human immunodeficiency virus (HIV) infection.

Methods: The patient charts of all HIV infected patients with epileptic seizures who presented to the neurological clinic of the University Hospital of Muenster between 2004 and 2006 were enrolled. Seizure type was classified and seizure frequency as well as seizure cause was determined.

Results: 12 Patients infected with HIV (9 male, 3 female) suffered from epileptic seizures, mean age was 45 years (32–57 years). Ten patients had new onset of epileptic seizures. Two Patients suffered from epileptic seizures before diagnosis of HIV infection. Mean latency (±SD) between diagnosis of HIV infection and the first seizure was 4.5 (±4.7) years. Seizure etiology was progressive multifocal leukencephalopathy (PML) in four patients (33%), AIDS dementia complex in one patient, CNS toxoplasmosis in one patient, HSV encephalitis in one patient, concurrent etiology (PML and subarachnoidal haemorrhage / AIDS dementia complex and CNS toxoplasmosis / AIDS dementia complex and HIV associated encephalitis / AIDS dementia complex and benzodiazepine withdrawal) in four patients (33%) and unknown etiology in one patient. Seizures were generalized tonic-clonic in two patients (16%), partial motor in five patients (42%) and partial motor with secondary generalization in five patients (42%). One of them suffered from a focal status epilepticus with secondary generalisation. Recurrence of seizures was observed in 50% of patients in spite of anticonvulsive medication.

Conclusions: The majority of patients with HIV infection and seizures have secondary brain lesions as seizure etiology. Interestingly, 80% of patients with partial motor seizures suffered from PML. Special care should be given to the choice of the anticonvulsive medication considering the seizure type and the manifold interactions of anticonvulsants and antiretrovirals.

1 T. Polster , 2 S. Athanassopoulos 11 Epilepsy Center Bethel (Bielefeld, D) ; 2 Children's Hospital Bethel (Bielefeld, D)

Background: Munchausen-Syndrome-by-Proxy is an established differential diagnosis in difficult-to-explain medical situations. Although seizures are reported to be among the most common presenting symptoms, little attention is paid towards this topic in the literature on pediatric neurology or epilepsy.

Method: The case of a toddler is presented, who was repeatedly admitted in a pediatric hospital with a history of seizures followed by transient ataxia.

Results: 1. The use of a prescribed medication (diazepame) to induce the symptoms complicated the diagnostic workup. 2. The suspicion of a factitious disorder was proven by toxicological screening of a urine specimen with gas chromatography-mass spectrometry, that detected N-Methyl-3,4-methylenedioxyamphetamine (commonly referred to as MDMA or ecstasy). 3. This case of factitious disorder by proxy does not fulfill the criteria of Munchausen-Syndrome-by-Proxy.

Discussion: 1. Meticulous screening for toxic substances is mandatory in all cases of an unclear presentation of a seizure disorder in childhood. 2. The use of the terminology “factitious disorder by proxy” is helpful to describe a number of conditions with induced medical symptoms, Munchausen-Syndrome-by-Proxy being only the best known of them. The differentiated look at these conditions is a prerequisite for adequate diagnostic workup and therapy.

1 M. Sparmann , 1 P. Hopp , 1 T. Mayer 1Epilepsy Center Kleinwachau (Radeberg, D)

Introduction: Gaucher's disease is an autosomal-recessive inherited lipid storage disorder based on a deficiency of the lysosomal enzyme glucocerebrosidase and resulting storage of glucocerebrosides. Type 3a of Gaucher's disease is a subacute neuropathic form with myoclonic seizures.

Methods: We recently characterized a patient suffering from type 3a Gaucher's disease with specific interictal electroencephalographic pattern.

Results: This 36 year old woman developed at age of 33 years anemia, throbocytopenia and splenetic enlargement. The diagnosis of Gaucher's disease was established by bone marrow biopsy. Since youth the patient has very slow horizontal sakkades. She developped supranuclear gaze palsy, dysarthria, trismus, mild spastic and cerebellar symptoms. Initially the patient had brief myoclonic jerks triggered by emotional agitation. At the age of 33 years she had one seizure with fall and loss of consciousness. She also developped myoclonic seizures of both arms lasting up to one minute (1/month).

EEG examination revealed occipital predominant rapid polyspike activity which appeared after eye closure and increased by photic stimulation. In addition she had intermittent slow backround activity.

Conclusions: We suggest that clinical and the previously rarely discribed electrophysiological findings of this case are suggestive of type 3a of Gaucher's diasease.

1 C. Kurth , 1 B.J. Steinhoff 1Epilepsy Centre Kork (Kork, D)

Subject: Sleep deprivation and hyperventilation are well known provocation methods in routine EEG. In this study we examined the benefit of these techniques for video EEG monitoring of drug resistant epilepsy patients.

Method: Monitoring protocols of 107 patients (51 men, median age 34.5 years; minimum 18 and maximum 62 years) were evaluated. Occurrence of seizures within 2 hours after hyperventilation was determined. For those patients who had a partial sleep deprivation time between beginning of sleep and first seizure after sleep deprivation was examined. Seizure occurrence was compared to the distribution of the other seizures of the patients, too.

Results: 79 patients performed a hyperventilation. In one patient an epigastric aura occurred within a few minutes afterwards. In a second patient a series of 3 short tonic seizures was induced. 20 patients had a partial sleep deprivation. In 11 of them (no seizures before sleep deprivation) no seizure developed after a median latency of 88.5 hours (26-304 hours). 1 patient without previous seizures got seizures 75 and 87 hours after the provocation. 7 patients had a median seizure frequency of 2 (1–11) before sleep deprivation. They exhibited 2 seizures (1–5) in median after provocation with a median latency of 17 hours (8–29). In one patient with psychogenic seizure 3 of 6 seizures occurred before sleep deprivation, each of them around noon without clear relation to the provocation.

Discussion: Hyperventilation has almost no impact on occurrence of seizures in video EEG monitoring. In patients having no seizures before sleep deprivation, no seizures occurred after provocation without one exception. Median seizure frequency before and after sleep deprivation was comparable for those patients having seizures before provocation. Median latency was long. Therefore relationship between provocation and seizure occurrence is very questionable.

Conclusion: Benefit of hyperventilation and sleep deprivation seems to be little in presurgical video EEG monitoring of epileptic patients. But further investigations are necessary define subpopulations of patients in which provocation of seizures may be reasonable.

1 B. Feddersen , 1 C. Bernhard , 1 C. Stoyke , 1 J. Remi , 1 C. Vollmar , 1 S. Noachtar 1University Munich (Munich, D)

Introduction: Evaluation for temporal lobe epilepsy surgery aims to identify candidates for partial temporal lobe resection and needs to distinguish unilateral from bilateral temporal lobe epilepsy (TLE). We wished to investigate whether the analysis of seizure semiology and lateralising seizure phenomena helps to identify bilateral TLE.

Methods: We searched the data base of the University of Munich Epilepsy Monitoring Unit for the terms bitemporal and TLE. Twenty-one patients with seizures arising independently from either temporal lobe and the first consecutive 21 patients of our database with left and right temporal lobe epilepsy, who were seizure free after temporal lobe resection, were included in the study. Only patients with high quality MRI and EEG-video recordings were included. The analysis of seizure semiology was based on video recorded objective seizures. Thus, patients who only had auras were excluded.

Results: Patients with bilateral TLE had more often seizures with a different seizure semiology and seizure evolution than the patients with unilateral TLE (90% vs. 43%; p<0,05). We recorded a total of 450 lateralising signs. Patients with bilateral TLE had more often lateralising signs arising from both hemispheres than the patients with unilateral TLE (80% vs. 50%; p<0,05).

Conclusions: Seizure semiology provides clues as to whether a patient has unilateral or bilateral TLE.

1 B.J. Steinhoff 1Epilepsy Centre Kork (Kehl-Kork, D)

Purpose: Vigabatrin (VGB) may lead to permanent visual field defects in about 1/3 of patients. In patients treated with VGB regular perimetric controls are therefore recommended. However, we had the impression that in patient being treated with VGB for a longer period without visual field problems this devastating adverse event occurs almost rarely.

We therefore assessed the incidence of visual field defects in patients treated with VGB for more than three years.

Patients and methods: We addressed the percentage of patients who developed visual field defects after at least 3 years of VGB treatment. Patients were identified from our patient database. Only patients treated with VGB for at least three years were considered.

Results: We identified 27 adult patients from our database who were treated with VGB longer than three years. Six of them were severely mentally handicapped so that regular visual field controls could not get obtained. Thus, the data of the remaining 21 (mean 39.29 ± 10.7 years, range 21–59 years) were used. All of them still underwent regular ophthalmological and especially perimetric investigations beyond the treatment period of three years. The mean VGB treatment time was 12.1 ± 2.1 years, range 8–15 years). Since these patients were out-patients from all over the country they were advised to see the same ophthalmologist in their home area to ascertain that Goldmann perimetry was always performed according to the same standard.

Mean dosage was 2012 ± 943 mg, range 1000–4750 mg). Regular ophthalmological investigations revealed that no patient developed visual field defects.

Conclusions: We identified no patient who developed visual field defects after an uneventful VGB treatment time of three years. Some of our patients continued VGB for up to 12 years. Apparently in these subjects the necessity of regular ophthalmologic controls appears to be questionable. In patients being treated with VGB for more than three years without ophthalmologic problems the risk to develop visual field defects appears to be minor. VGB responders therefore should not be automatically withdrawn only due to the fear that this side effect could still occur. For safety reasons our data should not be misinterpreted. Regular ophthalmological controls should still be performed until bigger series may confrim our observations.

1 C. Brandt , 1 T.W. May , 1 E. Nieder , 1 B. Huber , 1 V. Böhme , 1 K. Witte-Bölt , 1 B. Pohlmann-Eden 1Epilepsy Center Bethel (Bielefeld, D)

Zonisamide (ZNS) was licensed as add-on treatment for partial epilepsy in adults in Germany in March 2005. Controlled clinical trials demonstrated that ZNS is an efficient and well tolerated antiepileptic drug (AED). The aim of this open label study was to evaluate efficacy and tolerability of ZNS in patients with refractory epilepsy under clinical conditions.

We included all patients of the Bethel Epilepsy Centre who started with ZNS between June 2005 and October 2006. Most of these patients suffered from severe refractory epilepsy. They were interviewed after 3, 6 and 12 months. However, due to the currently small number of patients with a follow-up of twelve months, data for a follow-up of six months will be presented. We assessed the (dis)continuation of ZNS, type of epilepsy, seizure type and frequency, comedication etc. The primary parameter for efficacy and tolerability was the retention rate of ZNS.

75 adult patients (aged 32.9 ± 15.1 years, range: 4.5 – 69.6) were treated with ZNS as add-on therapy, on average in combination with 2.1 AED. The mean observation period was 167 days. After six months, 60 patients were available for follow-up: 2% (n = 1) became seizure free, 19% were responders (≥ 50% reduction of seizure frequency), in 46% the seizure frequency remained unchanged or increased. In 34%, ZNS had been withdrawn, 66% (n = 40) remained on ZNS. Obviously, in addition to seizure frequency other aspects of outcome (e.g., tolerability, changes of severity of seizure) have had also an impact on continuation of ZNS therapy. Changes of the comedication have to be taken into account.

Reasons for withdrawal were: lack of efficacy (13.3%, n = 8), side effects (11.7%, n = 7) or both (8.3%, n = 5). Average weight loss was 1.6 and 2.9 kg after 3 and 6 months of treatment (p<0.01), respectively. Adverse effects most frequently mentioned by the patients were weight loss (18.3%, n = 11), tiredness (13.3%, n = 8), psychiatric (13.3%, n = 8), cognitive side effects (10%, n = 6), aggressive behavior (8.3%, n = 5) and agitation (6.7%, n = 4).

The retention rate of 66% after 6 months indicates that add-on treatment with ZNS is effective and well tolerated even in patients with refractory epilepsies, although the effect on total seizure frequency was less marked.

1 B. Huber , 1 E. Robertson , 1 M. Bocchicchio , 1 W. Wagner , 1 E. Feuerbaum , 1 E. Wilking , 1 T. Meinert , 1 H. Schorlemmer , 1 M. Seidel 1Epilepsy Center Bethel (Bielefeld, D)

Purpose: To gain information on the efficacy and tolerability of the novel antiepileptic drug pregabalin (PGB) in therapy-resistant patients with intellectual disability.

Methods: Retrospective evaluation of all inpatients started on PGB for epilepsy between December 2004 and July 2006. After 6 months of PGB treatment, seizure reduction was calculated comparing 3 months on PGB (4th–6th month on PGB) with 3 months baseline before PGB. In addition, efficacy was clinically judged using the Clinical Global Impression (CGI) scale. Unwanted side effects were evaluated using a list of the ten most frequently observed side effects in the regulatory studies. The CGI scale was also administered to judge tolerability.

Results: After 6 months, 24 of 32 patients included were still on PGB (75% retention rate). 6 patients (18,75%) were responders (50% seizure reduction and/or “moderate” or “very good” effect on CGI). 16 patients or their carers reported no side effects at all and 7 had only side effects that “do not impair essentially.” 8 patients/carers reported “side effects which impair essentially,” mainly sleepiness, lack of drive, mental slowing, loss of cognitive or daily life capacities, and one case of increased auto-aggression. Weight increase occurred in approx. one third of the patients. Other potential side effects out of the “list of ten” were rare.

Discussion: After 6 months of treatment, the efficacy of PGB proved to be moderate (18,75% responder rate). However, three out of four patients were still on PGB, so we assume that dose escalation or the adaptation of comedication was not finished yet. At a second time point of evaluation after one year the results may become more unequivocal.

Tolerability was good in the majority of patients. However, one quarter experienced impairing (mainly central-nervous) side effects.

1 E. Nieder , 2 T.W. May , 1 C. Brandt , 2 K. Witte-Bölt , 1 B. Pohlmann-Eden 11 Evangelisches Krankenhaus Bielefeld gGmbH (Bielefeld, D) ; 2 GfE – Gesellschaft für Epilepsieforschung (Bielefeld, D)

Pregabalin (PGB) was licensed as an add-on treatment for partial epilepsy in adults in Germany in September 2004. Controlled clinical trials demonstrated that PGB is an efficient and well tolerated antiepileptic drug (AED). The aim of this study (open label, mostly phone interviews) was to clinically evaluate the efficacy and the tolerability of PGB in patients with refractory epilepsy.

We included all patients of the Bethel Epilepsy Centre who started with PGB between September 2004 and December 2005. The patients' interviews were carried out after 3, 6 and 12 months. We assessed e.g., the rate of (dis)continuation of PGB, the type of epilepsy, seizure type and frequency, comedication. The primary parameter for both efficacy and tolerability was the overall retention rate of PGB.

In total, 111 adult patients (aged 37.5 ± 13.2 years) were treated with PGB in combination with 1 to 4 AED's (average 2.1). The patients had already been treated with a mean of 6 to 8 AED (range 1–14) in different mono- or combination therapies, before they received PGB as additional therapy. Twelve patients had a vagal nerve stimulator. The mean observation period was 224 days. 105 patients were available for 12 months follow-up, of which 5% became seizure free; 20% were responders (> 50% reduction of seizure frequency), in 14% the seizure frequency remained unchanged or increased and in 60% (n = 63) PGB was withdrawn. Accordingly, 40% (n = 42) of the patients remained on PGB.

Reasons for withdrawal were lack of efficacy (30%, n = 32), side effects (8%, n = 8) or both (20%, n = 21). Weight gain was the most frequently mentioned adverse effect (25%, n = 26). Average weight gain was 2.2, 3.7 and 4.1 kg after 3, 6 and 12 months of treatment, respectively. Other adverse effects mentioned by the patients included tiredness (17%, n = 18), dizziness/unsteady gait (12%, n = 13), cognitive (10%, n = 11) and psychiatric side-effects (10%, n = 10), and skin reactions/irritations (7%, n = 7).

The retention rate of 40% after 12 months indicates that add-on treatment with PGB is effective and well tolerated even in patients with refractory epilepsies.

1 V. Böhme , 2 B. Rambeck , 2 T.W. May , 2 R. Neb , 1 B. Pohlmann-Eden , 1 C. Brandt 11 Epilepsy Center Bethel (Bielefeld, D) ; 2 GfE – Gesellschaft für Epilepsieforschung (Bielefeld, D)

Introduction: Pregabalin (PGB), a recently licensed antiepileptic drug (AED), is rapidly absorbed and renally eliminated. In patients with normal renal function, its elimination half-life is about 6 hours. Therefore, considerable changes in serum concentration during the day are to be expected. However, only few pharmacokinetic data have been published for PGB so far. The purpose of this study was to measure the fluctuations of PGB serum concentrations during the day and to identify possible contributing factors.

Methods: We retrospectively examined the data of 16 patients with daily doses of PGB between 500 and 1000 mg (mean 702 mg) as an add-on AED. PGB serum levels were determined at hourly intervals for up to 5–9 hours after drug intake. All patients had normal renal function.

Results: During the day, PGB serum concentrations increased to 324 ± 109% of the baseline (morning value); the mean trough PGB concentration was 3.5 ± 1.3 mg/l (range: 1.3–6.3 mg/l), the mean peak PGB concentration was 10.5 ± 2.0 mg/l (range: 8.0–13.9 mg/l). The highest concentrations were found 2.3 hours after drug intake with a span from 1 to 5 hours. In accordance with the literature, the mean estimated elimination half-life of PGB was 5.7 ± 1.3 hours; however, the half-lives varied between 3.2 and 7.8 hours. In patients on enzyme-inducing comedication the half-life of PGB was shorter and the fluctuations of PGB concentrations were greater. Neither age nor PGB dose had an influence on half-lives or (percentage) fluctuations.

Discussion: As expected, marked fluctuations of PGB serum concentrations during the day were observed. However, fluctuations and half-lives of PGB were found to vary considerably among patients. Enzyme-inducing comedication might be a contributing factor, surprisingly enough in an AED with renal elimination. It has to be considered that the extent of fluctuations could actually affect clinical response. Due to limitations of our retrospective study (e.g, open-label study, relatively small number of patients) our results should be verified in a larger prospective study under standardized conditions.

1 T.W. May , 2 R. Schulz , 2 M. Hoppe , 1 B. Rambeck , 2 A. Ebner 11 GfE – Gesellschaft für Epilepsieforschung (Bielefeld, D) ; 2 Epilepsy Center Bethel (Bielefeld, D)

Levetiracetam (LEV), a new antiepileptic drug, is usually applied in dosages up to 3000 mg/day. Some studies indicate that the efficacy and tolerability of LEV is dose-dependent.

We present a case report of an adult woman (36 yr.) suffering from symptomatic, focal epilepsy treated with extremely high LEV dosages. The patient underwent epilepsy surgeries (1994 / 2004) due to therapy resistant epilepsy, but auras (in clusters) still remained. As the patient reported positive effects of LEV on frequencies of auras, LEV was stepwise increased up to 15000 mg/day (according to the patient's request and under closely clinical examination for side-effects). LEV concentrations up to 240 mg/l were measured without severe adverse effects.

A daily profile was performed in this patient on an extremely high LEV dosage in order to investigate pharmacokinetics of LEV and to examine whether side effects during the day (dependent on fluctuations of LEV) did occur.

A daily profile was performed in this patient for clinically reasons when she received a LEV dosage of 11000 mg/day t.i.d. (4000/3000/4000 mg). Blood samples were taken at 8:00, 9:30, 11:00, 14:00, 17:00, 20:00, 23:00 h and 8:00 h at the next day. Serum concentrations of LEV were determined by HPLC. The patient was asked to assess side effects during the day using a standardized questionnaire. Furthermore, body sway was investigated by posturography.

The LEV concentrations fluctuated markedly during the day (45.7-166.2 mg/l) in this patient on a LEV dosage of 11000 mg/day. However, even on the highest LEV concentrations no acute, adverse effects occurred and no or only slight complaints were reported by the patient. Body sway (posturography) was in the normal range and did not correlate with LEV concentrations during the day. The elimination half-life estimated from the daily profile was 5.7 h and clearance was 3.8 l/h in this patient. Thus, the pharmacokinetic parameters were comparable to published data.

In some extraordinary cases, LEV seems to be tolerated even in extremely high dosages (up to 15000 mg/day) and high LEV concentrations (up to 240 mg/l) without serious adverse effects over a long period (> 2 years). Pharmacokinetics of LEV does not seem to change at high LEV dosages.

This case demonstrates that LEV dosage and serum concentrations show considerable individual variability with regard to tolerability.

1 E. Korn-Merker , 1 A. Hofmann-Peters 1Epilepsy Center Bethel (Bielefeld, D)

Rational: In Japan zonisamide (ZNS) is admitted for the treatment of focal and generalized seizures in adults and children as mono- and add on-therapy since 1989. Since 2005 ZNS is available in Germany as add on-therapy for the treatment of focal epilepsies in patients older than 18 years. Because we treat children and adolescents with pharmacoresistant epilepsies mostly, we used ZNS off-label.

Method: 22 patients (6 girls, 16 boys) aged 2–19 years with mostly focal epilepsies (symptomatic 18, cryptogenic 2) but also 2 with IGE received ZNS add on-therapy. Mean dosage was 4,5 mg/kg BW/d (max. 7,9 mg/kg BW/d). Comedications have been VPA, OXC, CBZ, TPM, LEV, LTG, PB, BR, FBM, CLN and CLB. 2 patients had a Vagal-Nerve-Stimulator too. In mean all patients had had 9 antiepileptic drugs before without persisting effect. They have had different seizures types. First ZNS was titrated up to 5 mg/kg BW/d.

Results: 7 patients have had a seizure reduction of 75–90%. 2 children showed a reduction of 50-75%. 13 patients didn't benefit from ZNS, 2 of them have had an transient effect only. One female adolescent suffered from rheumatic pain, which disappeared after discontiunation of ZNS. One boy had repeating vomiting, which stopped some time after discontitnuation of ZNS according to the half life time. One boy who already suffered from attention deficit under several other anticonvulsant drugs before, showed the same symptoms with ZNS. We did not see any hyperthermia or kidney stones.

Conclusion: Nearly 41% of our patients had a benefit from ZNS-treatment. Regarding the pharmacoresistance, proofed before, this is an amazing effect. One girl with myoclonic absences benefit from ZNS since more than one year and now receives a ZNS-monotherapy. In 2 patients only we saw a transient effect.

1 M. Bacher , 1 B.J. Steinhoff 1Epilepsy Centre Kork (Kehl-Kork, D)

Objective: Assessment of long-term efficacy and tolerability based on retention rates of zonisamide (ZNS).

Patients and methods: Year-by-year ongoing assessment of clinical data files of all adult epilepsy patients treated with add-on ZNS since June to December 2005.

Results: We observed 57 adult patients most of them suffering from intractable focal epilepsy syndromes. The actual retention rate is 53% as compared to 75% one year ago. Long-term ZNS add-on treatment still appears to be not complicated. In most cases the tolerability was good with a satisfying efficacy profile.

Conclusion: Based on our experience ZNS is a favorable addition to our antiepileptic drug treatment options.

1 U.A. Kopp , 1 V. Gaus , 1 B. Wandschneider , 1 B. Schmitz 1University Hospital Charité (Berlin, D)

Objective: Adverse cognitive side effects have been reported for all major antiepileptic drugs (AED; Aldenkamp 2001). However, so far little is known about the impact of Levetiracetam (LEV) on cognitive performance. The aim of this study was to assess the effect of LEV, Carbamazepine (CBZ) and Valproate (VPA) on cognition after beginning an antiepileptic monotherapy in de novo patients.

Methods: We investigated cognitive performance in 15 patients of our outpatients department. Patients with newly diagnosed epilepsy were randomly assigned to monotherapy with LEV, CBZ or VPA respectively. Two of the patients with idiopathic generalized epilepsy (n = 4) were randomised to VPA, and 2 to LEV. Twelve patients had localisation-related epilepsies (CBZ n = 6; LEV n = 4; VPA = 1).

Neuropsychological assessment was performed at three points: prior to introduction of AED therapy, 6 weeks and 12 weeks after beginning of AED medication. Psychometric tests comprised attention (simple reaction time task, divided attention), short term and working memory (digit span forward and backward), learning and retention of verbal and non-verbal information (Rey-Auditory-Verbal-Learning Test and Rey-Osterrieth Complex Figure Test), planning abilities (Tower of London) and inhibition (Stroop Colour-Word-Interference Task).

Results: Multivariate analyses of variance were conducted for each cognitive variable separately. No significant effect of type of medication or time of assessment could be found.

Conclusions: There were no significant side effects on attention, memory, executive functions and overall intellectual abilities of monotherapy with LEV as compared to CBZ or VPA over a 12 week period. Thus, LEV appears to be a safe antiepileptic drug with respect to cognition. However, the sample examined so far is small so that continuing work is needed to consolidate these preliminary data.

1 C. Reinsberger , 2 T. Dorn , 2 G. Krämer 11 Julius-Maximilian-University (Wurzburg, D) ; 2 Swiss Epilepsy Centre (Zurich, CH)

Smoking has been shown to induce glucuronidation and therefore influence serumlevels of drugs metabolized by glucuronidation such as lorazepam and oxazepam. In this study a possible influence of smoking on lamotrigine (LTG), a widely used drug to treat epilepsies with simple- and complex-partial as well as generalized seizures that is metabolized by glucuronidation is examined for the first time.

LTG serumlevels of 43 patients with epilepsy treated with LTG in monotherapy were retrospectively examined. Fifteen of the patients (ten females & five males) were smokers (three cigarettes/month to three packages/day) and 29 (15 females, 14 males) nonsmokers. Patients who smoked more than three cigarettes per month were considered as smokers. Allowing repeated measurements in single patients, 204 samples were analyzed. All patients were recruited as in- or outpatients in our epilepsy center within two years. The statistical comparison between LTG serum levels was performed by one-factor ANOVA.

The comparison between LTG serumlevels revealed a highly significant (p = 0,00138) difference of the quotient of LTG serumlevel and dosage between smokers and nonsmokers. Analyzing only male patients, the same relationship with an almost equally high level of significance could be demonstrated (p = 0.008). Although there was no statistically significant correlation between the number of cigarettes per day and LTG serum level-to-dose ratio, there was a trend pointing to a dose effect of smoking on LTG serumlevels in the Pearson Correlation (r = -0,127 μmol/l*mg, p = 0,069).

Smokers showed a significantly lower LTG serumlevel-to-dose ratio than nonsmokers. Because our patients were examined on various occasions and daytimes and because a strict compliance, especially in the outpatients of our sample, cannot be generally assumed, this effect may even be underestimated. The inclusion of patients with a very mild consumption of cigarettes (three per month) into the group of smokers may also contribute to an underestimation of this effect. In addition, due to the relatively low number of smokers, possible dose effects of smoking could not have been detected with statistical significance. Whether this effect is of clinical significance and directly related to glucuronidation has to be confirmed by further prospective studies with a higher number of patients and a more detailed assessment of smoking habits.

1 J.-P. Ernst , 1 D. Neumann , 1 J. Fahrbach , 1 A. Wiemer-Kruel 1Epilepsy Centre Kork (Kehl-Kork, D)

Introduction: The rate of intractable epilepsies is high in specialized epilepsy centers. Besides the symptomatic focal epilepsies, West and Lennox-Gastaut syndromes, the severe myoclonic epilepsy (SMEI) and the myoclonic-astatic epilepsy (MAE) play an important role. The need for new antiepileptic drugs (AED) is high. Since December 2001 Stiripentol is available as orphan drug for the treatment of SMEI in combination with sodium valproate and clobazam. It is unrelated to other AED and belongs to the group of aromatic allylic alcohols. It has been shown to prevent the uptake of GABA and inhibits its metabolism. More important it improves the effectiveness of other AED by inhibition of several cytochrome P 450 isoenzymes.

Methods: In the year 2006 patients with refractory epilepsies, in whom all other available AED had failed, were treated with stiripentol add-on. The data were analyzed retrospectively.

Patients: 10 patients (2 female, 8 male) with a mean age of 10;3 years (5,5–16) received stiropentol add-on to a pre-existing therapy with 2 to 3 AED, mostly valproic acid, bromides, phenobarbitone and topiramate. Classification: Severe myoclonic epilepsy 5, myoclonic-astatic epilepsy 1, CSWS 1, symptomatic focal epilepsy 3.

Results: Seizure-free 1 (MAE), seizure-reduction > 90% 1 (SMEI), 50-75% 1 (SMEI). In all other cases we saw only transient, slight improvement. No one worsened. No serious events occurred.

Discussion: Stiripentol seems to be an effective drug in the therapy of severe generalized epilepsy of infancy and childhood. It should be introduced early, if valproic acid, bromides, phenobarbitone and topiramate have failed.

1 R. Besser , 1 M. Eckert 1Hospital Krefeld (Krefeld, D)

Purpose: The injectable Levetiracetam (LEV) formulation has been developed for intravenous (IV) use in epileptic patients, if oral administration is temporarily not possible. Fast infusion rates even in elderly epileptic patients are required for its use in the emergency situation.

Methods: 13 patients with and without previous antiepileptic medication other than LEV were treated with LEV IV 2000 mg for focal convulsive or nonconvulsive status epilepticus (SE) and generalized nonconvulsive SE. The infusion rate was 15 minutes in six patients and 5 minutes in seven. Consciousness, cardiovascular function and respiration were monitored. Blood samples were collected 30 to 60 minutes, 6, 12 and 24 hours later. The plasma was separated within 12 hours and LEV concentrations were determined by HPLC.

Results: The median age of the patients was 79 years ranging from 50 to 93 years. Despite higher age, all patients showed creatinin values within normal range. Following LEV IV, regardless the infusion rate, none of the patients developed additional clouding of consciousness. No alterations in cardiovascular function or respiration occurred. LEV plasma concentrations were available from 12 patients. Thirty to 45 minutes after LEV IV the mean plasma concentration was 48 mg/l ranging form 21-81 mg/l. Thereafter the concentration dropped in an exponential manner. The respective values for 6, 12 and 24 hours were 31 (14.5–54), 20 (8.5–46) and 17 (6.7–44) mg/l. Three patients had received an additional dose of LEV IV 500 mg before the last blood sample. Efficacy was inconsistent. LEV tended to be more effective in patients with repetitive seizures, than in those with generalized nonconvulsive SE.

Conclusions: LEV IV 2000 mg within 5 to 15 minutes was well tolerated in elderly epileptic patients and achieved high plasma concentrations for at least 12 hours.

Abstract

  1. Top of page

Poster session 1.2. -Clinical Epileptology (Therapy)

1 C. Tilz , 2 Y. Wang-Tilz , 2 A. Backof , 2 H. Stefan 11 Convent Hospital "Barmherzige Brüder” (Linz, A) ; 2 University of Erlangen-Nuremberg (Erlangen, D)

Objectives: According to the Chinese medical literature acupuncture is most effective in the treatment of generalized epilepsy. For years there is still a gap between the standardization of the treatment according to seizure type and an objective evaluation of the effects. In this study the effect of acupuncture as add-on therapy in the treatment of absence seizures was quantified during the monitoring of long-term EEG.

Methods: 6 outpatients suffering from primary generalized absence seizures based on the ILAE criteria (1981), who were referred to the epilepsy centre, University of Erlangen-Nuremberg, Germany, were enrolled. The whole study lasted 11 days and the concomitant antiepileptic drugs were maintained stable. After 48 hours long-term EEG monitoring the patients were randomised to either the real-points group or sham-points group to receive seven sessions of acupuncture treatment, daily 30 minutes at the same time. Immediately after the last session followed another 48 hours long-term EEG monitoring. The selection of the acupuncture points (acupoints), the treatment duration and the depth of the needle-insertion were based on Chinese literature and experts' experience from P.R China. A total of 11 acupoints were used in every session, including five scalp points and six extremity points. The parameter of effect was the changes of the number of spike-wave-complex before and after the seven-day acupuncture treatment.

Results: A total of six patients finished the study: 4 in the real-points group and 2 in the sham-points group. The age ranged from 4 yrs. to 63 yrs with a mean of 29 ± 19.73. Before acupuncture there were totally 14066 spike-wave-complexes in the real-points group (median 2125, ranged from 20 to 9795) and 10558 in the sham-points group (median 5279, ranged from 94 to10464) during the 48-hour long-term EEG recording. After the acupuncture the total number reduced to 6596 (53% reduction) in the real-points group with the median of 397 (range from 0 to 5802) and 9566 (8.4% reduction) in the sham-points group with the median of 4783 (range: 97 to 9488). Two patients in the real-points group were completely free from spike-wave-complexes during the post-acupuncture EEG monitoring.

Conclusion: This pilot study showed a preliminary effect of acupuncture in reducing the number of spike-wave-complexes in the EEG monitoring. But because of the small sample seize, further study with more patients are definitely necessary.

1 M. Rauchenzauner , 1 A. Griesmacher , 1 T. Tatarczyk , 1 G. Falkensammer , 1 L. Zimmerhackl , 1 G. Luef , 1 E. Haberlandt , 1 W. Högler 1University Hospital Innsbruck (Innsbruck, A)

Introduction: Valproic Acid (VPA) therapy causes weight gain and hyperleptinemia. Both, VPA and newer antiepileptic drugs (AED), have been associated with vitamin D insufficiency and increased bone turnover or even reduced bone mass.

Objective: To determine the influence of VPA and new AEDs on body composition, insulin resistance, leptin, vitamin D and novel markers of bone metabolism in children and adolescents.

Methods: This study included 142 non-institutionalized, ambulatory children >6 years with idiopathic epilepsy on chronic AED monotherapy. Body-Impedance analysis, HOMA-index, leptin, 25OH vitamin D, osteoprotegerin (OPG), soluble receptor activator of NF-kappaB ligand (sRANKL) and tartrate-resistant acid phosphatase 5b (TRAP5b) were determined.

Results: 87 children (mean age 12.3 ± 3.6) were on treatment with VPA and 55 (mean age 12.3 ± 3.1) on other AEDs, comprising the control group. VPA-treated children, vs. controls, had greater weight standard deviation score (SDS; 0.31 ± 1.3 vs. -0.65 ± 1.1, p<0.001), body-mass-index SDS (0.43 ± 1.3 vs. -0.44 ± 0.9, p<0.001), percentage body fat (22.6 ± 9.3% vs. 12.2 ± 6.6%, p<0.001), leptin concentrations (25.9 ± 19.7 vs. 7.2 ± 7.6ng/mL, p<0.001) and HOMA index (3.2 ± 2.0 vs. 2.2 ± 1.6, p = 0.008). No difference was found between groups in 25OH vitamin D (107.5 ± 63.8 nmol/l vs. 116.3 ± 77.2 nmol/l), OPG (2.8 ± 0.7 pmol/l vs. 3.0 ± 0.8 pmol/l), Trap5b SDS (0.48 ± 1.9 vs. 0.54 ± 1.9) with a tendency for higher RANKL concentrations in VPA treated children (0.47 ± 0.7 vs. 0.39 ± 0.9, p = 0.071). None of the children had Vitamin D insufficiency (<50nmol/L). No correlations between markers of bone turnover and anthropometric variables, leptin or HOMA index were found.

Conclusion: VPA therapy was associated with higher body weight, body fat and serum leptin concentrations as well as impaired glucose homeostasis. Neither the VPA nor the AED group showed evidence of vitamin D insufficiency. There was no significant effect of AED treatment, hyperleptinemia and high body fat on novel markers of bone turnover.

P. Borusiak , G. Kluger , G. Wohlrab , T. Bast for the “Valentinskreis pädiatrischer Epileptologen”

Objective: Idiopathic focal epilepsies generally are a benign condition. However in some children mental deterioration and difficult to treat seizures occur. Some authors suspect an aggravation by certain anticonvulsive drugs.

Methods: We performed an extensive literature review on the question of drug-induced electroencephalographical and/or seizure aggravation in idiopathic focal epilepsies with respect to carbamazepine, gabapentin, lamotrigine, levetiracetam, oxcarbazepine and sulthiame. These data were compared with the results of studies on spontaneous evolution of children with idiopathic focal epilepsies.

Results: One systematic retrospective study finds a probable association of seizure aggravation in 2/98 patients treated with carbamazepine. Otherwise only case reports exist for carbamazepine, lamotrigine and oxcarbazepine. Atypical course in the beginning of epilepsy, Landau-Kleffner-syndrome and bilateral sharp waves might be risk factors.

Conclusion: Given the variable spontaneous course we did not find any data for a systematic aggravation due to certain drugs in idiopathic focal epilepsy.

B. Schäuble , S. Wang , F. Wiegand for the EPMN-106/INT-28 Investigators

Objectives: Analyses of data on adolescent subjects from dose-controlled studies evaluating the efficacy and tolerability of topiramate as initial monotherapy.

Methods: Subset analyses of a previously published trial (Arroyo et al., 2005) including patients between 10-15yrs with newly diagnosed epilepsy or untreated patients with epilepsy relapse and ≥2 lifetime seizures and 1–2 partial or generalized tonic–clonic seizure during a 3 months retrospective baseline. The primary efficacy end point was time to first seizure; a secondary efficacy measure was the seizure-free rate at 6 months and 1 year. The safety population was pooled with patients receiving 50mg/day in a very similar pivotal trial (Gilliam et al., 2003).

Results: A total of 470 patients were included in the initial efficacy analysis, 114 patients between 10–15 years of age (50% female) are presented in this analysis. Kaplan-Meier survival analysis for time to first seizure in the adolescent subset of the IIT population favored 400mg/day (n = 57) over 50mg/day (n = 57) (P = 0.001). The probability of being seizure free at 6 months was 79% in those randomized to 50mg/day and 94% in patients randomized to 400mg/day (P = 0.001). Average daily dose of topiramate during the double blind phase was 43mg/day (0.92mg/kg/day) in the 50mg group and 335 mg/day (7.47mg/kg/day) in the 400mg/day group. Adverse events leading to study discontinuation were anxiety, chest pain and migraine in one patient in the 50mg arm. In the 400mg arm, 7 patients overall discontinued due to cognitive problems, fatigue and mood complaints. No treatment related serious adverse events occurred. Mean weight change from baseline to final visit in the double blind phase (6 months) were a 2.5% weight gain in the 50mg arm and a 0.7% weight loss in the 400mg arm.

Conclusion: The results of this analyses support the efficacy and safety of topiramate for initial monotherapy of partial-onset or generalized tonic–clonic seizures in adolescents with epilepsy.

B. Schäuble , P. Levisohn , K. Holland , F. Wiegand for the JME Capss-107 Investigators

Objective: Open label study (CAPSS-107) to evaluate the effectiveness of topiramate (TPM) in patients with juvenile myoclonic epilepsy (JME).

Method: Open-label study enrolling patients age 12-65 years with a confirmed diagnosis of JME (myoclonic jerks and PGTCS or an EEG with generalized epileptiform abnormalities consistent with a clinical diagnosis of JME), age of disease onset between 8–26 years. Patients had to have myoclonus and/or >1 PGTCS within the 3-month retrospective baseline. Eligible patients were randomized (2:1) to 26 weeks treatment with TPM (target 3–4 mg/kg/day; maximum, 9 mg/kg/day) or VPA (target 10 mg/kg/day; maximum 60 mg/kg/day or 750 mg/day if >16 years of age). Seizure frequency, tolerability, and patient and physician evaluations of response were recorded during each visit (Baseline and Weeks 4, 8, 14, and 26).

Results: 28 patients were randomized to TPM (N = 19; median age 15 years; range 9-42) or VPA (N = 9; median age 16 years; range 12-34). 12 patients had one baseline AED. 12/19 patients receiving TPM and 7/9 patients receiving VPA completed the study; mean dose among completers was 189 ± 78 mg/day and 897 ± 375 mg/day, respectively. At the end of the study, all patients but one patient achieved TPM (OXC) or VPA monotherapy. Reasons for discontinuation in the TPM group included adverse events (N = 2), patient choice (N = 1), or loss to follow-up (N = 1); two patients discontinued VPA due to adverse events or other reasons. During the 3-month maintenance phase, seizure-free rates were 47% with TPM and 33% with VPA. Seizure-free rates for myoclonic, PGTCS, and absence seizures in the TPM and VPA groups were 7/14 (50%) vs 6/9 (67%), 8/12 (67%) vs 3/4 (75%), and 2/2 (100%) vs 1/2 (50%), respectively. Physician and patient global evaluation of improvement, alertness, and improvement in seizure severity were similar in both treatment groups (TPM, 75%; VPA, 71%). Neurotoxicity scores over time were similar, but systemic toxicity scores were higher with VPA. Most TPM-treated patients lost (mean, -4.1 kg), while patients receiving VPA gained weight (mean, 5 kg); the between group difference was statistically significant (p<0.001).

Conclusion: TPM was effective in the treatment of patients with JME with less systemic toxicity compared to VPA. The results suggest similar effectiveness in seizure control, but because of the small sample size, clinically meaningful differences cannot be ruled out. These results are supported by recently published data from a randomized controlled trial [1, 2].

1 B. Schäuble , 1 A. Schreiner 1Janssen-Cilag GmbH (Neuss, D)

Objective: To evaluate quality of life (QUOLIE-10) and weight change of epilepsy patients transitioned from valproate (VPA) to topiramate (Topamax, TPM) monotherapy.

Methods: Multicenter, prospective, open-label, noninterventional trial (TOPMAT-EPY-403). Patients with epilepsy independent of seizure type ≥ 12 years of age and unsuccessfully treated with VPA due to lack of efficacy and/or lack of tolerability were eligible for this study. Patients were prospectively followed for 20 weeks after initiation of, and transition to a TPM monotherapy.

Results: Overall, 147 patients (59% women, mean age 41 ± 19 years) were enrolled. Reasons for transition from VPA onto TPM were side effects in 81% (mostly tremor, somnolence or cognitive symptoms) and/or lack of efficacy (61%). 23% of patients were seizure free at baseline. 84% of patients completed the study. Mean weight at baseline was 75.9 ± 17.9 kg and decreased on average by 3.7kg ± 4.5kg; mean BMI improved from 26.5 to 25.4 kg/m2. Seizure frequency significantly decreased after transition to TPM (p<0.0001 vs. baseline). QUOLIE-10 subscores mental health, role functioning and severity of epilepsy all improved significantly (p<0.0001 vs. baseline for all items). Tolerability of TPM was rated as ‘good’ or ‘very good’ in 83%. TPM was well tolerated with 14% of patients experiencing treatment-emergent adverse events (AEs). AEs >/= 3% were paraesthesia (4%) and weight decrease (5%). The only cognitive AE reported was speech disorder in 3% of patients.

Conclusion: Transition from valproate to topiramate monotherapy was associated with a substantial improvement in quality of life, a considerable weight decrease and a beneficial tolerability profile including few cognitive side effects.

1 B. Schäuble , 1 A. Schreiner 1Janssen-Cilag GmbH (Neuss, D)

Objective: To evaluate quality of life (QUOLIE-10), tolerability and weight change in patients with epilepsy after transition from carbamazepine (CBZ) or oxcarbazepine (OXC) to topiramate monotherapy (Topamax, TPM).

Methods: Multicenter open-label noninterventional trial (TOPMAT-EPY-405). Patients >/= 12 years of age with epilepsy unsuccessfully treated (due to lack of efficacy and/or lack of tolerability) with CBZ (72%) or OXC (28%) were eligible for this study. Patients were prospectively followed for 26 weeks after initiation of, and transition to, TPM.

Results: 140 patients (54% women, mean age 47 ± 18 years) were enrolled. Main reasons for CBZ/OXC discontinuation were side effects (80% of patients) and/or lack of efficacy (75%). At end point, median TPM dose was 100 mg/day. Mean seizure frequency decreased from 6/month at baseline to 1.4/month at endpoint (p < 0.0001). 81% of patients completed the study. All QUOLIE-10 items including subscores mental health, role functioning, and severity of epilepsy improved significantly (p < 0.0001 vs baseline for all items and subscores). Mean baseline BMI slightly improved from 26 to 25.5 kg/m2 at endpoint, mean weight decrease was –1.9 ± 3.7 kg. TPM was well tolerated: 26% of patients reported treatment-related adverse events (AEs). AEs >/= 5% were paraesthesia (9%), memory difficulties (5%) and weight decrease (8%). 12% of patients discontinued treatment due to an AE. The only additional cognitive side effect reported was speech disorder in 4% of patients. Tolerability of TPM was rated as ‘good’ or ‘very good’ in 80% by physicians.

Conclusion: Transition from CBZ or OXC to topiramate monotherapy was associated with a significant improvement in quality of life and was well tolerated.

1 A. Carius , 2 A. Schreiner , 2 B. Schäuble 11 University Hospital Freiburg (Freiburg, D) ; 2 Janssen-Cilag GmbH (Neuss, D)

Objective: To evaluate seizure frequency and tolerability in patients with epilepsy treated with topiramate (TPM, Topamax) transitioning from valproic acid (VPA)

Methods: Multicenter open-label noninterventional trial (TOPMAT-EPY-403) following patients ≥ 12 years of age with epilepsy previously unsuccessfully treated with VPA. Patients were prospectively followed for 20 weeks after transitioning to TPM. A 12-week retrospective seizure frequency was used as baseline.

Results: 147 patients (59% female; mean age 42 yrs (± SD 19)) were followed. Median duration of epilepsy was 9 yrs (range, 0-60 yrs). 77% had seizures during the 12-week retrospective baseline. 70% of patients were on TPM monotherapy at end point and 77% continued TPM therapy. Most frequent seizure types at baseline were generalized tonic–clonic (52%), complex partial (23%), and simple partial (12%). Main reasons for transition from VPA to TPM were insufficient efficacy (61%) and/or side effects (81%). Mean dose of VPA at time of first administration of TPM was 1286 ± 629 mg. Median TPM dose was 125 mg/day at endpoint. Mean (± SD) seizure frequency decreased significantly from 32 ± 248 seizures per month during the retrospective baseline to 3 ± 16 seizures/month during the maintenance period (p < 0.001). The responder rate (≥ 50% seizure reduction in patients with seizures during their retrospective baseline) during the last three months of observation was 75%, and 51% patients remained seizure-free during this period. 94% of the patients who were seizure-free at baseline continued to do so during the study. 16% of patients discontinued TPM, 8% due to an adverse event (AE), and 3% due to insufficient efficacy. 14% of patients reported treatment-emergent adverse events (TEAE). TEAE occurring in > 3% were paraesthesia (4%) and weight decrease (5%). The only cognitive adverse events was speech disorder occurring in less than 3% of patients.

Conclusion: In patients previously unsuccessfully treated with VPA, mono – or add-on therapy with topiramate was well tolerated according to patient's reports and was associated with a substantial reduction in seizure frequency and a high seizure-free rate.

1 B. Schäuble , 1 A. Schreiner , 2 H. Adam 11 Janssen-Cilag GmbH (Neuss, D) ; 2 Practice (Rostock, D)

Objective: To explore seizure frequency and tolerability in patients with epilepsy treated with topiramate (TPM; Topamax) transitioning from carbamazepine (CBZ) or oxcarbazepine (OXC).

Methods: Multicenter open-label noninterventional trial (TOPMAT-EPY-405) prospectively following patients with epilepsy ≥12 years of age who were previously unsuccessfully treated with CBZ (72.1%) or OXC (27.9%) for 26 weeks after transitioning to TPM. A 12-week retrospective seizure frequency was used as baseline.

Results: 140 patients (54% female; mean (± SD) age 47 ± 18 yrs) were enrolled. 72.1% of patients had unsuccessfully been treated with CBZ and 27.9% with OXC. Mean (± SD) duration of epilepsy was 14.3 ± 13.7 yrs and patients had had an average of 2.2 AEDs (range, 1–10) prior to observation. 84% had seizures during the 12-week retrospective baseline. Most frequent seizure types at baseline were generalized tonic–clonic (52%), complex partial (25%), and simple partial (16%). Main reasons for transition from CBZ/OXC to topiramate were insufficient efficacy (75%) and/or side effects (80%). Mean dose of CBZ at treatment initiation of TPM was 825mg ± 397mg/day and 1254 ± 555mg/day with OXC. At endpoint, the median TPM dose was 100mg/day. 73% of patients received TPM monotherapy by the end of observation. Mean (± SD) seizure frequency decreased significantly from 6 ± 21/month at baseline to 1.4 ± 5/month during the observation period. The responder rate (>/= 50% seizure reduction) during the last 3 months of observation was 91%, and 62% of patients remained seizure-free during this period. 19% of patients discontinued TPM, in 12% due to an adverse event (AE), and in 3% due to insufficient efficacy. The only treatment-emergent adverse events reported in ≥5% were paraesthesia (9%) and weight decrease (8%).

Conclusions: In patients previously unsuccessfully treated with CBZ or OXC, topiramate was well tolerated and associated with a substantial reduction in seizure frequency and a high seizure-free rate.

1 H. Stefan , 2 A. Schreiner , 1 F. Kerling , 1 B. Kasper , 2 B. Schäuble 11 Epilepsy Center – University Erlangen-Nuremberg (Erlangen, D) ; 2 Janssen-Cilag GmbH (Neuss, D)

Objective: To assess effectiveness of topiramate (Topamax, TPM) in different age groups of elderly patients (≥ 60) with epilepsy.

Methods: In this prospective, open label, multicenter phase IV flexible dose clinical trial (TOP-GER-13), elderly patients (60-64, 65-74 and ≥ 75 years (yrs)) with epilepsy irrespective of seizure type were included and followed for a median of 12 months. Doses of TPM and concomitant AEDs could be adjusted individually. Seizure frequency and adverse events were assessed at each visit.

Results: 107 patients (53% male, mean age 69 ± 7 years) were enrolled encompassing 32 patients 60–64yrs, 54 patients 65–74yrs and 21 patients ≥ 75yrs. 102 patients had at least one seizure during the retrospective 12-week baseline. Mean duration of epilepsy was 11.8, 9.7 and 5.5 years, respectively. Most frequent seizure types at baseline were generalized tonic–clonic (58%) and complex partial (25%) with the highest seizure frequency in the oldest group (9.3 ± 34.1 at baseline). Mean seizure frequency for all was 3.5 ± 14.6 at baseline and decreased to 1.6 ± 7.7 at endpoint (p < 0.0001). At endpoint, the mean TPM dose given as monotherapy was 98 mg/day and 153 mg/day if given adjunctively. Mean monotherapy dose was lowest in the oldest elderly group (86mg/day) and the percentage of monotherapy was highest in the oldest group. The proportion of responders (seizure reduction ≥ 50%) was highest in the youngest group (87% responders), and overall, 78% of patients were responders and 44% remained seizure-free throughout the study.

46 patients (43%) had at least one treatment-emergent adverse event (TEAE). The number of TEAEs was highest amongst the oldest elderly. TEAEs ≥ 5% were somnolence (9.4%), dizziness (7.5%), paraesthesia (5.6%), and memory difficulties 5.6%. Main reasons for study discontinuation (40% overall) were an TEAE (15.9%) or loss to follow-up (12.2%).

Conclusion: In elderly patients with epilepsy, TPM was well tolerated amongst all age groups, was associated with a significant decrease in seizure frequency, and was well tolerated. Doses used were slightly lower than the recommended target doses for adults.

1 A. Owega , 2 A. Schreiner , 3 K. Rettig , 2 B. Schäuble 11 Practice (Cologne, D) ; 2 Janssen-Cilag GmbH (Neuss, D) ; 3 G.E.M. (Meerbusch, D)

Objective: To describe differences in effectiveness and safety profile in patients treated with the most commonly prescribed AEDs in Germany transitioning to topiramate monotherapy (Topamax, TPM).

Methods: Multicenter, open label, observational study (TOPMAT-EPY-0001) examining patients ≥ 6 yrs diagnosed with epilepsy and prior insufficient treatment (lack of effectiveness and/or tolerability) with PHT, OXC, CBZ or VPA monotherapy and planned transition to TPM monotherapy. Patients were followed for 16 weeks after initiation of TPM.

Results: The ITT analysis included 407 patients (53% female, no significant difference between groups) including 148 patients on VPA, 168 on CBZ, 42 on OXC and 49 on PHT. Mean age (± SD) over all groups was 45.8 ± 16.8 yrs. Between-group comparisons were performed by the Chi2- or the Kruskal-Wallis-H-test. Patients on VPA or PHT were younger at diagnosis than patients on CBZ or OCX (p < 0.005), but epilepsy duration was longer in the PHT treated group (p < 0.001). BMI differences among patients were non-significant. The number of previous AEDs ranged from 1–10 (mean 2.4 ± 1.6), being highest among PHT patients (p = 0.018). Overall, 75% of patients transitioned to TPM due to lack of efficacy (no differences between groups) and 61% due to insufficient tolerability (Chi square- test: p = 0.067). Reason for switch were side effects listed most frequently for CBZ (64% somnolence) or VPA (55% weight gain). TPM median dose at endpoint was 100mg/day. Overall, seizure frequency decreased from 2.35 ± 5.4 per 4 weeks during the 12-weeks retrospective baseline to 1.14 ± 4.05 during the prospective observational period (Wilcoxon- test: p < 0.001). Regardless of the prior AED (OXC, CBZ, VPA or PHT), all patients showed a significant seizure reduction while transitioning. 64.2% of all patients had an at least 50% seizure reduction and 41.8% were seizure free during the entire documentation period. A total of 58 adverse events occurred in 38 patients and 51 of these AEs had at least a possible relationship to TPM treatment. Treatment-related AEs (≥ 3% out of 58 AEs) were: tiredness, nausea, weight loss, dizziness, lack of concentration, restlessness, ataxia, exanthem, and development of seizures.

Most common reasons (≥ 3%) for discontinuation of TPM (15.8% overall) were AE (3.1%) or “unknown” (8.9%). 87% of physicians rated the effectiveness of TPM “very good” or “good” regardless of previous AED and 91% considered the patient's situation as very much or much improved. The number of seizure related physician visits or loss of workdays decreased (p < 0.001). 82% of patients continued on TPM treatment.

Conclusion: Transitioning from PHT, VPA, CBZ or OXC to TPM was associated with a substantial seizure reduction and good tolerability regardless of the AED previously used. In addition, there was a significant reduction in seizure related physician visits and loss of workdays.

1 B. Huber 1Epilepsy Center Bethel (Bielefeld, D)

Introduction: VPA-associated liver failure most often affects severely handicapped young children (some with metabolic defects) on antiepileptic polytherapy. Most cases happen during the first half year of treatment.

We report on a case of fatal liver disease after the extraordinary long period of 21 years of VPA therapy.

History: The 54 year old patient suffered from epilepsy, moderate to severe intellectual disability, and physical handicaps due to pre- or perinatal brain hypoxia. He was seizure free since VPA had been added to his medication in 1985. Phenytoin and later barbexaclon (a phenobarbital-derivative) were gradually discontinued. Since 1991 he remained on VPA monotherapy (1200 mg/d continuously since 1996) with his seizures still completely controlled.

Clinical course: In autumn 2006 the patient developed jaundice, pruritus and loss of appetite. Liver enzymes (transaminases) were elevated (ASAT and ALAT in the range of several hundred, GGT > 1000; alcalic phosphatase approx. 3 fold of normal) while parameters of liver synthesis were normal. Thorough laboratory examinations showed no plausible cause of liver dysfunction. VPA serum concentration was low (35,4 μg/ml). Pancreatic enzymes were normal. Abdominal ultrasound was little informative due to obesity but CT did not show signs of extrahepatic cholestasis. There was no evidence of fever or infection preceding jaundice.

Histology obtained by liver puncture, although unspecific, was compatible with drug toxicity. VPA was discontinued and replaced with clobazam. Thereupon transaminases improved but bilirubin elevated to extremely high values. Only in this stage the patient became somnolent and later soporous. He developed high fever. Only then parameters of liver synthesis were reduced; ammonia was elevated twofold. A generalized tonic–clonic seizure had to be aborted with midazolam. He died from bilateral pneumonia and cardiac failure.

Discussion: Although our patient finally died from pneumonia, we believe that VPA hepatotoxicity was decisive in this fatal course (decrease of liver enzymes after stopping VPA; failure to determine any other cause). Concomitant medication was not known to be hepatotoxic. However, postmortem autopsy was not performed and thus the final cause of death could not definitely be established.

Our patient is unusual in several aspects: adult age, monotherapy and relatively long period with isolated jaundice. The most extraordinary feature, however, is the very long period of VPA therapy.

Conclusion: There is jeopardy of VPA hepatotoxicity to be overlooked or underestimated in patients who do not belong to the typical risk group, especially in long lasting treatment. VPA should be considered as a possible aetiology in every VPA-treated patient with unexplained hepatic disorder, including adult patients with long lasting VPA therapy.

1 G. Möddel , 1 P. Zwanzger , 1 R. Dziewas , 1 W.-R. Schäbitz , 1 C. Kellinghaus , 1 K. Anneken , 1 M. Fischera , 1 S. Kovac , 1 M. Dogan , 1 S. Evers 1University Hospital Munster (Munster, D)

Refractory status epilepticus (SE) comprises a therapeutic challenge and poses the question whether nonmedical treatment options are feasable. To date, eight cases of SE have been reported which were supposedly successfully terminated by electroconvulsive therapy (ECT). Here, we present three EEG-documented cases in which ECT was not effective in terminating SE.

Methods: Retrospective chart review and EEG analysis of all patients who received ECT for the treatment of SE at our institution between 2003 and 2006.

Results: Patient 1 was a 59-year old female with nonconvulsive SE due to bitemporal encephalitis. Despite repeated CSF analyses and lesion biopsy, no infectious agent was detected. The patient was treated with midazolam, propofol, phenytoin, valproate, levetiracetam, topiramate, oxcarbazepine, and lidocaine. SE was terminated three times but recurred after reduction of the anaesthetic dose. On days 62 through 69 after SE onset, three ECT were done, without effect on EEG status. Two weeks after ECT, SE ceased and did not recur. This was correlated with extensive bilateral cortical necrosis on MRI. Patient 2 was a 65-year old male with SE after resection of multiple left convexity meningeomas. SE was not terminated by lorazepam, phenytoin, propofol, valproate, levetiracetam, lamotrigine, topiramate, lidocaine, and thiopental. On day 40 after SE onset, the patient received one ECT without cessation of SE. Patient 3 was a 80-year-old female with SE due to right frontoparietal subdural hematoma. The patient was unsuccessfully treated with lorazepam, phenytoin, propofol, thiopental, levetiracetam, and valproate, and subsequently received three ECT, 28 days after SE onset. SE was not terminated. All three patients were finally in a persistent vegetative state and transferred to palliative care.

Discussion: ECT was not effective in terminating refractory SE in three out of three patients. In all three patients, functional outcome was bad. To date, there are eight cases in the literature that report termination of SE by ECT. Only one of these case reports [1] documents termination of SE by showing respective EEG recordings. Despite cessation of SE, outcome was bad in this patient as well. The previous positive case reports suggesting effectiveness of ECT in terminating SE are not exceeding anecdotic relevance. Further, scientifically based data is lacking.

Reference:[1] Lisanby et al. ECT in the treatment of status epilepticus. J ECT. 2001 Sep;17(3):210-5.

1 G. Möddel , 1 W.-R. Schäbitz , 1 R. Dziewas , 1 F. Bösebeck , 1 C. Kellinghaus , 1 K. Anneken , 1 M. Fischera , 1 S. Kovac , 1 M. Dogan , 1 S. Evers 1University Hospital Munster (Munster, D)

Only few substances employed for the treatment of status epilepticus (SE) are used on the basis of controlled studies, i.e., lorazepam, diazepam, midazolam, phenytoin, and phenobarbital. Several substances are used on the basis of uncontrolled studies and case-series, such as valproate and propofol. The local anesthetic drug lidocaine, which acts as a sodium channel blocker like phenytoin, was first used in the treatment of status epilepticus in 1955. Until the late 1980s, approximately 150 cases were published, with the largest series comprising some 50 subjects. Since then, lidocaine has been used only sporadically, despite promising early data.

Methods: Retrospective chart review and EEG analysis of all patients with SE refractory to standard treatment who were treated with lidocain i.v. at our institution between December 2004 and August 2005.

Results: Six patients with refractory SE received lidocaine i.v. Lidocaine was applied as a continuous infusion (2-4 mg/kg/h). All patients had previously been administered standard doses of benzodiazepines and phenytoin. Five subjects had been intubated and received anesthetic doses of propofol and midazolam, two patients were additionally treated with thiopental narcosis. SE was terminated in three patients following lidocaine loading (etiologies: 1. ischemic MCA infarct, 2. hypoxic brain damage after delayed CPR, 3. inflammatory lesion secondary to AIDS CDC C3). Intubation was avoided in one patient (AIDS). In one patient, SE was terminated following lidocain infusion but recurred after dose reduction. In two patients (etiologies: hypoxic brain damage, and multiple left-hemispheric meningeomas, respectively), SE could be terminated by neither adequate doses of midazolam, propofol, phenytoin, valproate, and levetiracetam, nor by lidocaine and thiopental anesthesia. One of these patients died of multiple organ failure during hospitalisation, the other was transferred to palliative care in persistent vegetative state. In none of our patients treated with lidocaine at a maximum dose of 4 mg/kg/h, high-grade cardiac arrhythmias were noted. In none of the patients, significant lowering of blood pressure or increase of the necessary catecholamine dose occurred.

Conclusion: Lidocaine, a sodium channel blocker like phenytoin, may constitute an effective and safe alternative option for the treatment of refractory SE. Case series as this one are promising, however controlled trials are lacking so far.

1 G. Möddel , 1 C. Kellinghaus , 1 R. Dziewas , 1 K. Anneken , 1 M. Fischera , 1 S. Kovac , 1 M. Dogan , 1 S. Evers 1University Hospital Munster (Munster, D)

Status epilepticus (SE) is a neurological emergency with high mortality and morbidity. We present here our first experiences with the recently available intravenous (IV) formulation of levetiracetam (LEV).

Methods: We screened all patients admitted to our hospital between April and November 2006 with a diagnosis of either epilepsy or SE and retrospectively reviewed the records and EEGs of the patients found to have been treated with IV LEV.

Results: 7 patients had received IV LEV. 4 patients were treated for refractory SE after failure with standard doses of benzodiazepines and phenytoin. 3 were previously treated with valproic acid, 2 topiramate and 2 lidocaine; 3 patients were subsequently intubated and received propofol to induce EEG burst suppression. IV LEV was administered at 1500–9000 mg daily. SE was terminated in 2 (50%) patients; 1 with convulsive SE due to pre-existent focal epilepsy consequent to early hypoxic brain injury, one with nonconvulsive SE secondary to remote left MCA infarction. In 2 patients (50%) SE was not terminated; one patient had traumatic subarachnoid hemorrhage and left PCA infarction, the other had encephalitis of unknown etiology. IV LEV was generally well tolerated with no associated cardiovascular side effects. 1 patient experienced nausea and vomiting after receiving a 1500 mg IV LEV loading dose, resulting in mild aspiration pneumonia, successfully treated with antibiotics. 1 patient with remote right parietal infarction was admitted with generalised tonic–clonic seizures without regaining consciousness between seizures. After acute lorazepam administration this patient received 1000 mg IV LEV over one hour, followed by a 2000 mg/24 hr continuous infusion. Intubation was unnecessary and no complications occurred. 2 days after admission, the patient was discharged home in good condition on oral LEV. 2 patients with focal epilepsy were administered IV LEV in preference to oral therapy because of swallowing difficulties. There was no change in seizure frequency and no specific side effects.

Conclusions: IV LEV terminated refractory SE in 2 out of 4 patients with severe brain lesions, without cardiovascular side effects. IV LEV also appears well tolerated and effective when used to initiate antiepileptic therapy in clinical settings and as a replacement for oral antiepileptic drugs in patients with transient swallowing difficulties. Although further data are required, IV LEV may be a promising new treatment option for SE.

1 M. Steinert , 1 B. Henkel , 1 E. Korn-Merker 1Epilepsy Center Bethel (Bielefeld, D)

Rationale: Ketogenic diet (KD) is a therapy which is well established in the United States as a treatment option for children with pharmacoresistant epilepsies. In Europe, KD becomes more and more accepted, too.

In our tertial epilepsy center we treat children with difficult to control epilepsies. Since 2003 we established KD. We wanted to know whether different epilepsy syndromes respond different to KD and how many AEDs the patients took before the introduction of KD.

Method: Retrospective analysis of 14 patients who were treated with KD from 11/2003 until 12/ 2006. Responders were defined as patients with seizure reduction of at least 50%.

KD is always established as an inpatient setting. Usually, at least one member of the family is trained thoroughly during a 14 day hospital stay by our dietician. During the hospital stay blood gas analysis, glucose and ketones are monitored closely. Patients were discharged after two weeks only if ketone bodies were at minimum of 2 + in the urine and/ or > 2 mmol/l in the blood. Evaluation of KD took place with inpatient setting after 2-3 months.

Results: 14 Patients were treated with KD. 7 patients are responders, 7 are nonresponders. 3 of them quit KD because of noncompensated acidosis and /or recurrent severe hypoglycaemia during the first three days of KD.

Responders had the diagnosis of the following epilepsies: symptomatic focal (2), cryptogenic focal (1), idiopathic generalized (2) idiopathic epilepsy with generalized and focal signs (1), CSWS and myoclonic seizures (1). Mean age of this group was 8. 4 years (range: 3.4–13.2). The median of AEDs before the introduction of KD was 8 (range: 5–11).

Non-responder had the following epilepsy syndromes: symptomatic focal (4), cryptogenic focal (1), epilepsy with focal and generalized seizures (1) and Lennox-Gastaut syndrome (1).

Mean age in this group was 6.3 years (range:1.5–9.8). The median of AEDs before KD was 7 (range: 5–13).

Conclusions: According to our small series, we could not find a clear correlation between a specific epilepsy syndrome and positive effect of KD. Both groups are overlapping.

Responders as well as nonresponders took a lot of AEDs before KD was started. According to the current literature, KD should be offered as soon as 2 or 3 appropriate AEDs fail to control the epilepsy. But even if KD is started after multiple AED treatment attempts we can report of a responder rate of 50%.

1 S. Rupprecht , 1 K. Franke , 1 S. Fitzek , 1 O.W. Witte , 1 G. Hagemann 1Friedrich-Schiller-University (Jena, D)

Purpose: Nonconvulsive status epilepticus (NCSE) is not rare but certainly underdiagnosed. There is some debate about the sequelae of NCSE and how aggressively it should be treated, as high dose antiepileptic drugs (AEDs) may be an influential factor on prognosis especially in the elderly. We evaluate whether levetiracetam (LEV), a well tolerated highly effective AED may be a safe treatment option in complex partial status epilepticus (CPSE), one type of NCSE.

Methods: Retrospective analysis and follow-up of all patients with CPSE in a University Hospital treated with oral doses of LEV (8 patients) compared with conventional intravenous (IV) AEDs including valproate, phenytoin and benzodiazepines (11 patients) during a 5 year period. Both groups were statistically compared for age, hospitalisation time, time in intensive care unit and outcome. Follow-up was collected in both groups to assess seizure frequency and subjective quality of life.

Results: All 8 patients showed a marked clinical improvement with final cessation of ictal EEG-activity and clinical symptoms of NCSE after LEV initiation within 3 (mean 1.5) days. LEV was titrated from a starting dose of 500–1000 mg per day to a maximum dose of 2000 mg per day within 2 days. Conventional treatment in the control group was similarly effective but there were severe side effects (phenytoinintoxication (n = 1), toxic hepatitis after valproate (n = 1), valproate-induced hyperammonaemic encephalopathy with elevated ammonia levels (n = 3), whereas no relevant side effects in the LEV-treated group were reported. Long-term follow-up (6-36 months postdischarge) revealed 6 patients being treated with LEV achieved persisting seizure frequency reduction. One patient changed LEV medication because of inefficacy and the other died from causes other than epilepsy 2 months postdischarge. We found no significant differences in hospitalisation time, time in intensive care unit and outcome between the LEV group and the control group.

Conclusions: These data suggest that LEV may be a well tolerated, effective treatment option in NCSE and highlights the need for a prospective controlled study, particularly as an IV LEV formulation is now available.

1 K. Anneken , 1 M. Fischera , 1 C. Kellinghaus , 1 G. Möddel , 1 I.W. Husstedt , 1 S. Evers 1University Hospital Munster (Munster, D)

Introduction: About 5% of patients with HIV infection develop epileptic seizures. Anecdotal reports on antiepileptic treatment in HIV infected patients are available (1), however, systematic data is rare. We examined efficacy and safety of antiepileptic drug (AED) treatment in HIV infected inpatients with epileptic seizures.

Methods: The patient charts of all HIV infected patients admitted to the neurological inpatient clinic of the University of Münster, Germany, between 2004 and 2006 were analysed concerning the following items: demographic data, HIV related data, epileptic seizures, AED drug regime, efficacy and side effects of antiepileptic treatment.

Results: In this population, 12 out of 40 HIV infected patients (30%) had at least one epileptic seizure. All patients suffering from epileptic seizures received HAART and AED treatment, nine were put on AED monotherapy and three on AED polytherapy. Gabapentin was used in 11 cases, carbamazepine in 2 cases, and each of the following drugs in one case: lidocaine, levetiracetam, pregabalin, valproic acid, clonazepam, oxcarbazepine, and phenytoin. Nine patients had significant reduction (>50%) of seizure frequency, all of them were treated with gabapentin. Four patients suffered from side effects (SE), three of them received gabapentin. The most common SE was sleepiness. All SE were mild in nature.

Conclusion: Epileptic seizures are a relevant complication in HIV infected patients. Our data suggest that AED treatment with gabapentin is effective and safe in this population.

Literature: 1. Bhigjee AI, Rosemberg S. (2006) Optimizing therapy of seizures in patients with HIV and cysticercosis. Neurology 67:19-22.

1 S. Bunten , 1 B. Kroeger , 1 S. Dempewolf , 1 M. Lanz , 1 H. Rawert , 1 S. Happe 1Hospital Bremen East (Bremen, D)

Introduction: Multimorbid patients with status epilepticus are particularly challenging regarding treatment as they can present several problems when the common treatment measures are applied. Most recommendations start with intravenous lorazepam or another benzodiazepine followed by phenytoin or valproate. Administration of phenobarbital or primidone should come last as it can lead to respiratory problems and intubation.

Methods: Evaluation of multimorbid patients treated with levetiracetam due to EEG confirmed persisting status epilepticus and failure of initial treatment.

Results: Eight patients in status epilepticus were treated with levetiracetam after initial therapy failed. In six cases levetiracetam ended the status epilepticus after a treatment with lorazepam had failed. One patient received levetiracetam after an ineffective treatment with clonazepam. One patient was treated with a combination of valproate and phenytoin, another with valproate and lorazepam before the decision was made to alter the medication.

Seven times treatment took place due to nonconvulsive status epilepticus, once due to a series of generalized seizures. Twice, the medication was administered orally, seven times intravenously. All patients were multimorbid, two suffered from liver cirrhosis, two of the patients were treated with levetiracetam twice, each time with status epilepticus.

All patients profited by the decision to switch to levetiracetam as it ended the status epilepticus without causig the necessity of intubation and respiratory assistance.

Summary: Treatment of status epilepticus in multimorbid patients, in particular those with liver dysfunction, poses special problems that often lead to prolonged intensive care treatment. Levetiracetam may be a promising option in the treatment of status epilepticus and can help to minimize the risk of further liver dysfunction, prolonged intensive care or intubation as drugs such as valproate, phenytoin, phenobarbital or primidone can be avoided.

1 D. Neumann , 1 J.-P. Ernst , 1 P. Sager 1Epilepsy Centre Kork (Kork, D)

Evaluated are the datas from 40 children (18 female, 22 female) in the mean age of 5.9 y (0.8-18.8 y). The mean duration of epilepsy is 3.5 y (0.7-14 y). Before starting ketogenic diet they were treated with 9 (1–15) drugs in single or combination therapie.

20 children have a symptomatic epilepsy, for example complex malformation of the brain, or residual state after bleeding, infections and others.

20 children have an idiopathic epilepsy like Lennox-Gastaut-Syndrome, Grand Mal Epilepsy and others.

62.5% have generalized, 37.5% focal seizures.

67.5% of the children are doing classical 4:1 ketogenic diet. The were underlying a moderated John-Hopkins-protocoll, without fastning and with no restriction of liquid.

10% were without any other medication, 82.5% had 1 or 2 drugs under ketogenic diet. 37.5% have a comedication with valproin acid.

The children do ketogenic diet 18 weeks in the mean; the responder rate is 55% (n = 22). No effects were seen in grand mal epilepsy.

The 22 responder do ketogenic diet in the mean for 48 weeks (13–144 weeks). 30% with a seizure reduction over 50%, 17.5% over 90%, 3 children are complete seizure free.

Even after ending the ketogenic diet the responding children show a significant reduction of seizures.

The ketogenic diet can be used in the treatment of pharmacoresistent and inoperable epilepsies in childhood.

1 C. Wehinger , 1 P. Gallmetzer , 1 B. Mamoli 1Ludwig Boltzmann Institute for Epilepsy (Vienna, A)

Purpose: Photosensitivity is a common cause of seizure EEG pattern precipitation. Fixation-off sensitivity (FOS) which is suggested in routine EEG by abnormalities that consistently occur as long as the eyes are closed but not with eyes open is less common.

Methods: Case report

Results: A 66- year- old man with a known history of seizures and alcohol abuse was admitted to our department after three generalized tonic–clonic seizures (GTCS).He had been treated with oxcarbazepin (OXC) for three years following a single GTCS. Upon admission there was no visible seizure activity. A postictal confusionary state resolved within the first few hours. During the following days the patient showed an unimpaired mental state. Due to elevated hepatic parameters the patient was stepwise switched to gabapentin (GBP). A routine EEG performed on day 5 showed a generalized, 2 to 4 c/ s spike-wave and polyspike-wave pattern with complete suppression on eye opening and recurrence on eye closure. Intermittent photic stimulation (IPS) consistently elicited polyspike-wave complexes with a return to baseline activity in stimulation-free intervals. After switching to oral Levetiracetam(LEV) 500 mg bid and discontinuation of OXC and GBP treatment the EEG pattern completely resolved with only a slight amount of occipital polyspike activity following eye closure and no reaction to IPS.

Conclusion: LEV was effective in the treatment of nonconvulsive status epilepticus with FOS.

1 O. Debus , 1 H. Koch 1St. Marienhospital Vechta (Vechta, D)

Hypothalamic Hamartoma (HH) usually has a characteristic clinical course. Apart from polymorphic seizure types with typical gelastic seizures a psychomotor regression develops over the time. Mostly this epileptic syndrome is refractory to the currently available antiepileptic medications. Epilepsy surgery is chosen in rapidly deteriorating patients but its success is not predictable.

A 15-year-old girl suffering from severe psychomotor retardation is presented. Her seizure history began with a febrile status epilepticus in infancy. At the age of 3 she started with complex partial seizures followed by secondary generalized and gelastic seizures. A HH was diagnosed as the cause of the epilepsy which did not grow over the years. After multiple antiepileptic drug trials the situation was stable under a combination with lamotrigine, clobazam and sulthiame with only few and short complex partial seizures. The EEG showed sparce groups of bilateral precentrally commencing and rapidly generalizing spike wave and few tonic patterns. After the reduction of sulthiame the girl developed a clinical and electroencephalographical absence status that could be controlled by a reintroduction of sulthiame.

Clinical seizures and their EEG-correlate in HH can present as focal or generalized activity. The choice of an appropriate medication in this syndrome usually depends on the seizure type. Sulthiame is an antiepileptic drug commonly applied in focal epilepsies. Its introduction can be justified in generalized seizure types, when standard medications have been used unsuccessfully.

1 G. Krämer , 2 T.W. May , 3 B. Schmitz , 4 H. Stefan , 5 B.J. Steinhoff , 6 S. Stodieck , 7 E. Trinka 11 Swiss Epilepsy Centre (Zurich, CH) ; 2 Gesellschaft für Epilepsieforschung e.V. (Bielefeld, D) ; 3 University Hospital Charité (Berlin, D) ; 4 Epilepsy Center – University Erlangen (Erlangen, D) ; 5 Epilepsy Centre Kork (Kehl-Kork, D) ; 6 University Hospital of Neurology and Epileptology (Hamburg, D) ; 7 University Hospital of Neurology (Innsbruck, A)

Background: Even at the beginning of the 21st century carbamazepine and valproate remain the most commonly prescribed standard antiepileptic drugs (AEDs). Since the introduction of vigabatrin there have been eight other new AEDs with partial approval for monotherapy up to now. This has not only led to a broadening of the options of AED therapy but also to a kind of confusion for the prescribing physicians especially regarding the comparative value of newer AEDs. Because evidence based guidelines are mainly based on the regulatory studies and head-to-head comparisons are not available, these guidelines of neurological or other medical societies can only partially solve this problem. Therefore we adopted a questionnaire for experts which has already been used twice in the US (Karceski S, Morrell MJ, Carpenter D. Epilepsy Behav 2001; 2, Suppl.:A1-A50, and Epilepsy Behav 2005;7 Suppl 1: S1-64) to the situation in the three German speaking countries.

Methods: We sent a questionnaire to 45 experts in the field of adult epileptology (“opinion leaders”) in Germany, Austria, and Switzerland in the second half of December, 2006. All had been asked before whether they were willing to participate and had agreed. The questionnaire consisted of 28 case histories in the treatment of idiopathic as well as symptomatic or presumably symptomatic epilepsies. The experts were asked to give their opinion for 22 possible alternative treatments (in addition to all available AEDs as well ketogenic diet and vagus nerve stimulation) on a 9-point scale (with “9” most appropriate and “1” least appropriate). This should be done irrespective of the current approval status based on their personal experience and view.

Results and conclusions: Forty-three out of the 45 contacted experts answered until the end of January 2007. The statistical analysis of the results is currently ongoing. We expect the results to be of complimentary usefulness to existing guidelines for treatment of adult patients with epilepsy.

The study and the evaluation of the results was performed with unrestricted educational grants by (in alphabetical order) GlaxoSmithKline, Novartis, Sanofi-Aventis, and UCB. All analyses were done without any influence of any of these companies by the “Gesellschaft für Epilepsieforschung” in Bielefeld.

Abstract

  1. Top of page

Poster session 1.3. Epilepsy surgery and miscellaneous

1 C.G. Bien , 1 M. Szinay , 1 H. Urbach , 1 A. Becker , 1 J. Schramm , 1 C.E. Elger 1University of Bonn (Bonn, D)

Purpose: Presurgical workup in patients with nonlesional brain MRI is complex, intricate, and often unsuccessful. Surgically treated patients have an inferior outcome compared to lesional surgical candidates. It is therefore a matter of debate if presurgical evaluation is worthwhile in those patients. Only a few studies on the outcome of surgically treated cases have been published. A possibly even greater limitation lies in the fact that—to the best of our knowledge—there is no study available comparing the outcome of surgically treated non-lesional patients to those who have not been offered epilepsy surgery. The present was undertaken to provide such data.

Method: Review of the records of all patients undergoing presurgical assessment at this centre between 2000 und 2005 in whom high-resolution brain MRI on initial assessment was found to be negative for a potentially epileptogenic lesion. For all patients, the most recent available follow-up at our outpatient department or (if this was unavailable) through telephone interviews was noted.

Results: We identified a total of 176 patients. Twenty-one of them underwent resective surgery (12%). Between the surgery and the non-surgery groups, there were neither demographical differences nor such regarding the extent of noninvasive presurgical investigations.

Outcome after a mean follow-up of 2.2 ± 1.5 years was available for all operated and 121 nonoperated patients. Eleven (52%) of the surgery patients [histopathology: four FCDs, one cavernoma, four cases of mild hippocampal sclerosis Wyler I/II, and 12 with non-specific findings] vs. 19 (16%) of the non-surgery patients were seizure-free for at least one year (P < 0.001).

Supposing that no surgical treatment had been offered to anyone in the whole cohort, i.e., that the seizure-freedom rate of all patients with follow-up were that of the real nonoperated subgroup: The seizure-freedom rate of that fictive group (16%, 22/142) wouldn't have been inferior to that of the real one (30/142, 21%, P = 0.2).

Conclusion: The seizure-freedom rate of the operated patients was superior to that of their nonoperated counterparts. However, the seizure-freedom rate of the whole group was not significantly improved by the surgery subgroup. There is still need for reliable prognostic parameters to be applied in rejection and selection of nonlesional surgical candidates.

1 C. Bußmann , 2 H.-J. Meink , 2 H.H. Steiner , 3 W. Broxtermann , 4 C.G. Bien , 1 D. Rating , 1 T. Bast 11 Children's Hospital Heidelberg (Heidelberg, D) ; 2 University Hospital Heidelberg (Heidelberg, D) ; 3 Social Pediatric Center (Mosbach-Neckarelz, D) ; 4 University Clinic of Epileptology (Bonn, D)

Background: The therapeutic role of vagus nerve stimulation (VNS) in epilepsy was clearly demonstrated. We investigated the effects of VNS in a patient with epilepsia partialis continua (EPC), a condition usually refractory to antiepileptic treatment.

Case report: An 18-year-old male patient experienced EPC in the left foot, sometimes including the left leg/arm/face, which had lasted for 10 years. Epilepsy started at the age of 6 years with complex-partial seizures (CPS) with left sided cloni. At the age of 8 years epilepsy worsened acutely, including the onset of EPC and he developed a stable mild spastic hemiparesis. VNS was implanted at the age of 18 years (follow-up 24 months). Actual frequency of CPS was 6–8/month and he suffered from additional isolated and prolonged auras. Because of severe myoclonia of the left foot it took the patient up to three hours to fall asleep.

Diagnostic workup: MRI at the age of 13 years showed diffuse cerebral atrophy with right hemispheric maximum, which was progredient in further MRI controls, corresponding to the formal diagnostic criteria of Rasmussen encephalitis. FDG-PET showed a diffuse hypometabolism of the right hemisphere. Jerk-locked averaging of perimyoclonic MEG signals and source analysis revealed a right central generator with spikes.

Polygraphic neurophysiological investigations, including giant left tibialis SEP indicated a cortical myoclonus.

Follow-up: VNS reduced CPS frequency >50% and auras were shortened by magnet activation. EPC stopped when currents of 1mA were reached, although the patient still experienced several myoclonic jerks of the left leg per hour. However, the sleep problem was solved. Acute reduction of stimulation currents from 1.25 to 0.75 mA was necessary two times because of psychogenic panic attacks with sweating and dyspnea. EPC worsened after 6 to 12 hours. A prompt response was noted when currents were increased again 4 weeks after the first and 24 hours after the second episode.

Conclusion: The reported patient with Rasmussen encephalitis profits from VNS as former pharmacoresistant seizures are reduced. Especially the reduction of cortical myoclonus is a considerable improvement of quality of life. Therefore VNS may be an effective treatment in epilepsia partialis continua.

1 F. Kerling , 1 B. Fraunberger , 1 H. Stefan 1Epilepsy Center – University Erlangen (Erlangen, D)

Introduction: After successful epilepsy surgery, a discontinuation of antiepileptic drugs (AED) is desirable to achieve a state of complete seizure freedom without medication. However, prospective randomised trials of AED discontinuation are difficult because of their risk of seizure relapse. We report about a prospective withdrawal study with patients from our epilepsy centre.

Methods: All postsurgical patients were invited to follow up visits every year in the first five years after epilepsy surgery. Patients who were completely seizure free one year after surgery and without any severe EEG changes (frequent epileptiform discharges) participated in this study. The patients could decide whether medication should be discontinued (tapering group) or kept at the same dosage (control group). Discontinuation was realized in small tapering steps over a one year period for the first drug. When patients had more than one drug, the second drug could be discontinued in the next year and so on.

Results: Eighty-four patients could be included since 1997, 50 in the tapering group and 34 in the control group. Forty two patients in the tapering group and 31 patients in the control group already passed the two years follow up visit, 24 patients in the tapering group and 20 patients in the control group had passed the five years follow up visit.

93% (38 of 41) in the tapering group and 83% (26 of 32) in the control group were seizure free two years postoperatively. Five years after surgery, 83% (20 of 24 patients) in the tapering group and 60% (12 of 20 patients) in the control group were seizure free.

Conclusions: These preliminary data show no higher risk for discontinuation of antiepileptic drugs after successful surgery. Nevertheless, the number is small. We suggest to initiate a multicenter study with a standardised protocol.

1 S. Fauser , 2 T. Bast , 1 D.-M. Altenmüller , 1 J. Schulte-Mönting , 3 K. Strobl , 3 B.J. Steinhoff , 1 J. Zentner , 1 A. Schulze-Bonhage 1 University of Freiburg (Freiburg, D) ; 2 University of Heidelberg (Heidelberg, D);3 Epilepsy Centre Kork (Kehl-Kork, D)

Introduction: Focal cortical dysplasia (FCD) is a common cause of pharmacoresistant epilepsy in childhood and adulthood. Determination of prognostic factors for epilepsy surgery is important when counselling these patients.

Methods: 120 patients with histologically proven focal cortical dysplasia were retrospectively analysed. Multivariate logistic regression analyses were performed to assess the prognostic importance of the clinical variables duration of epilepsy, age at epilepsy surgery, age at epilepsy onset, location of FCD, and histology of FCD.

Results: Longer duration of epilepsy reduced the chance of becoming seizure-free after epilepsy surgery. One year postoperatively, 75% of the patients with epilepsy duration between 0 and 5 years were seizure-free, as compared to only 39% of patients with epilepsy duration > 20 years. This effect was independent of the duration of follow-up. Location or histological subtype of the FCD, however, had no statistically significant influence on the postoperative outcome.

Conclusion: This finding strongly suggests early consideration of epilepsy surgery in patients with FCD.

1 J. Wellmer , 1 F. von der Groeben , 1 C. Weber , 1 H. Urbach , 1 C. Schaller , 1 C.E. Elger 1University Hospital Bonn (Bonn, D)

Introduction: With the introduction of high resolution MRI into the presurgical work up of pharmacoresistant epilepsy patients, epilepsy surgery has become highly individualized. In this context, the chronic implantation of sudural grid electrodes may be considered to delineate the seizure onset from functionally relevant cortex. However, the decision for grid implantation has to be met in view of the risks that are linked to this procedure. We performed a retrospective evaluation of grid implantation-related complications.

Materials and methods: In the period from 2004 to 2006, 57 epilepsy patients (28 f; 3-60 ys, mean 34 ± 12) received grid electrodes. We defined grid electrodes as any electrode array with 16, 20, 32 or 64 electrode contacts which was inserted under sight via open craniotomy. With the exception of three patients, all patients had one grid. Two had two and one patient three grids (mean number of grid contacts: 44 ± 17). Additionally, one to eleven strip electrodes and one to two depth electrodes were implanted in 41 and 25 patients, respectively (total number of electrodes: 20 to 120, mean 60 ± 20). Complications occurring during the period of implantation (5 to 40 days, mean 16 ± 6) are classified as grade 1: complication visible on CT/MRI only; grade 2: clinically apparent complication, no surgical revision needed; grade 3: open revision or termination of invasive workup necessary.

Results: In 17 patients (30%) we observed the following complications: subdural or epidural hematoma (N = 11); brain swelling/edema (N = 5); cortical contusion (N = 5); interacerebal hemorrage (N = 3); olfactory nerve contusion (N = 1). Grade 1: N = 5 (9%); grade 2: N = 3 (5%); grade 3: N = 9 (16%). Following grade 3 complications, after hematoma evacuation the workup was continued in seven patients, removal of grids or other electrodes were necessary in one patient each. No patient exhibited a persistent neurological deficit. We found no correlation between the occurrence of complications and the number of implanted electrode contacts, grid size, or the duration of the implantation procedure. In all 57 patients invasive monitoring finally led to a definite result, either recommendation of surgery (N = 52) or rejection (N = 5).

Discussion: Invasive presurgical workup with grid electrodes is not free of complications. However, we did not observe implantation related permanent neurological deficits. The invasive workup with grid electrodes should be restricted to patients who can not be diagnosed otherwise.

1 J. Wellmer , 1 C. Weber , 1 F. von der Groeben , 1 C. Schaller , 1 C.E. Elger 1University Hospital Bonn (Bonn, D)

Introduction: Despite functional neuroimaging, invasive EEG recording and electrocorticostimulation (mapping) with subdural grid electrodes remain necessary tools in the presurgical workup of pharmacoresistant epilepsy patients. Because of higher risk of complication when compared with “acute” (intraoperative) electrocorticography, “chronic” grid implantation can only be justified by a gain of information relevant to planning and execution of surgery. To quantify usual efforts related to the creation of comprehensive surgical maps, we evaluated 57 extraoperative presurgical procedures.

Materials and methods: From 2004 to 2006, 57 epilepsy patients (28 f; 3-60 ys, mean 34 ± 12) received implantation of grid electrodes. We defined grid electrodes as any electrode array with 16, 20, 32 or 64 electrode contacts which was inserted under sight via open craniotomy. Indications for grid implantation were: localization of the epileptic focus (N = 57) and mapping eloquent cortex (N = 50). Depending on the grid location, we mapped for motor cortex, SMA, negative SMA, language (including counting, confrontation naming, repeating sentences, reading, token test and body commands), sensory functions, and visual phenomena. During mapping, all grid electrode contacts were stimulated at least twice in different bipolar relays.

Results: During a monitoring period of one to 36 days (8.0 ± 5.8) we recorded spontaneous seizures in 56 of 57 patients. Following reduction of antiepileptic drugs, first seizures occurred after one to 12 days (2.6 ± 2.7). Seizure onset was identified within the grid in 43 patients. The mapping period took one to six days (2.8 ± 1.4) or 30 to 850 minutes (252 ± 181), respectively. 77 distinct functions were found in 47 of 50 mapped patients. Despite performance under antiepileptic medication, electrostimulation elicited seizures in 21 patients. Grids remained implanted five to 40 days (16 ± 6). In all patients, sufficient information for the creation of comprehensive maps was collected.

Discussion: Creation of comprehensive and reliable maps is complex and tends to be time-consuming. Unlike acute electrocorticography and stimulation, extraoperative monitoring allows verification of doubtful results, adaptation of the examination velocity to the capacity of individual patients, and discontination of mapping in case of unpredicted events (such as seizures). From the neurosurgical point of view, preoperatively created maps allow for a natural work flow during surgery.

1 K. Strobl , 1 H. Mayer , 1 J.-P. Ernst 1Epilepsy Centre Kork (Kehl-Kork, D)

Introduction: Hypothalamic hamartomas with epilepsy still are a huge challenge in epilepsy surgery. Nowadays different strategies seem to solve this problem: Interstitial radiosurgery (ISR), LINAC- or gammaknife (GK)-radiosurgery and open or endoscopic microsurgery (MS) seem to compete for the best results.

Methods: We report about 8 patients treated in different centers with at least one of these methods:

1 open microsurgery (2x in the same patient) only,

1 ISR + open microsurgery,

1 ISR (twice) + LINAC + open microsurgery,

1 GK + endoscopic microsurgery,

4 GK only.

Results: Overall the best outcome was seen in the 4 patients who were operated in the end (Engel IA 2x, Engel IIA 1x, Engel IIIA 1x). Two of these patients developed postoperative complications: obesity + hypogonadism or obesity + transient hemiparesis respectively.

There was only little positive effect on seizure frequency in the 4 patients with only radiotherapy until now (follow up more than one year). Some of them might have profited concerning behavior and cognitive parameters.

Conclusion: Looking for the best treatment for 8 children with hypothalamic hamartoma and epilepsy different strategies in different centers were used. In our very small group it looks like as if microsurgery is the most effective procedure, although there are also very encoureging results about gammaknife or interstitial radiosurgery in literature.

Besides this discussion about the best option we should give this small group of our patients a chance and start to offer them early a proper device and additional radiosurgical or microsurgical treatment.

1 T. Falkenstetter , 1 T. Benke , 1 J. Dobesberger , 1 I. Unterberger , 1 M. Ortler , 1 T. Fiegele , 1 E. Trinka 1University Hospital Innsbruck (Innsbruck, A)

Introduction: Analysis of seizure semiology may help to lateralize and localize the seizure onset zone and symptomatogenic zone, respectively, and may contribute to our knowledge regarding functional neuroanatomy. In temporal lobe epilepsy (TLE), early head turning, mostly ipsilateral, and late contralateral versive head movements have been described.

Eye movements, visuospatial attention and visuomotor integration are controlled by a cerebral network presumably involving areas in the frontal lobe (e.g., the frontal eye field) and in the parietal lobe (e.g., the parietal eye field, gyrus supramarginalis, other parietal sites). According to known studies, head and eye movements are functionally linked to the prefrontal area.

We report a case of contralateral nonversive head deviation induced by electrical stimulation at the temporoparietal junction.

Case report: A 45-year-old man suffered from pharmacoresistent non-lesional left sided epilepsy with a presumed posterior temporal seizure onset zone. He underwent invasive EEG recording and stimulation to further delineate the seizure onset zone and potentially involved eloquent areas. Frameless stereotactic implantation of two hippocampal depth electrodes as well as subdural grid-electrodes (left frontal, temporal and parietal lobes) and subsequent video-EEG allowed the localization of the seizure onset zone to the left superior temporal gyrus. Electrical stimulation at the left temporoparietal junction repeatedly elicited contralateral nonversive head deviation. The seated patient, performing language tasks, was not aware of his head turning. The involuntary head turns resembled head turns intentionally conducted to look for something or to pay attention to something, but occurred without any external lateralized stimuli being present. The evoked movements exactly resembled the semiology of early head turning, which was in this case contralateral to the side of stimulation. Early head turning was not found in the patient's habitual seizure semiology.

Discussion: We hypothesize that stimulating the area at the temporoparietal junction where head turning was induced had an effect on the cerebral network controlling eye movements and visuomotor integration. This finding reemphasizes the challenges in accurately interpreting the lateralizing and localizing value of different types and sequences of head turning. Special caution should be paid in analyzing non-versive head turns in posterior neocortical TLE.

1 K. Wagner , 1 L. Frings , 2 T. Maiwald , 1 A. Schinkel , 1 C. Lehmann , 1 A. Buller , 1 R. Everts , 1 A. Carius , 1 A. Schulze-Bonhage 11 University Hospital Freiburg (Freiburg, D) ; 2 Center for Data Analysis and Modeling (Freiburg, D)

In order to assess titration effects of levetiracetam (LEV) in patients with focal or generalized epileptic seizures, we applied a novel method that allows for temporally fine-grained monitoring of cognitive performance as well as subjective well-being using hand held computers.

Twenty patients recruited in the outpatient clinic of the epilepsy center took part in this pilot study, 10 were newly medicated with LEV, the other 10 patients served as a control group in which medication was held constant. Differences between groups were analyzed regarding self-rated cognitive condition, psycho-physical condition, aggressiveness, and cognitive test performance by applying a 2-back working memory task assessed 3 times a day (morning, early afternoon, evening) over the course of 6 days. The first day served as a baseline with no add-on medication, on the subsequent days LEV was introduced as follows: 250 – 0 – 250 mg (day 1), 500 – 0 – 500 mg (day 2), 1000 – 0 – 1000 mg (during days 3 to 5).

Results revealed a significant group difference in self-rated aggressiveness (p < 0.05), which was higher in the LEV group in the early afternoon across days. Furthermore, control patients' cognitive performance significantly improved in the morning over the course of days compared to LEV patients, in which morning test performance was rather stable. Beyond proving feasibility of this novel method of patient monitoring, we were able to detect early effects during titration of LEV on (1) subjective aggressiveness depending on time of day and (2) changes in working memory performance over the span of days. We hereby emphasize the unique opportunities for patient monitoring and counseling provided by this novel method.

1 A. Carius , 1 A. Schulze-Bonhage 1Epilepsy Centre – University Hospital Freiburg (Freiburg, D)

Purpose: It is well known that an increase of body temperature may trigger epileptic seizures.

There is only little information in the literature which effect an artificial hyperthermia by bathing in thermal springs, by having a sauna or during therapeutic hyperthermia, e.g., by infrared radiation may have.

Methods: We present a retrospective summary of casereports of outpatients in whom seizures occured while having a sauna.

Results: 11 of our adult patients (6 m, 22-57 years) spontaneously reported to have suffered from seizures during sauna use between 2003 and 2006; we saw about 120 outpatients monthly. 9 of the seizures have been doubtless of epileptic origin. Usually these seizures were habitual, but more than the half were of an extended duration. In 6 of these 9 patients a brain tumor was the cause of their epilepsy. 2 patients incured their first seizure while having a sauna, both have got a brain tumor. 7 of 11 patients were under a stable treatment with 1 to 3 antiepileptic drugs, which had not been changed during the weeks before the seizure occured. The seizure frequency of all patients was 0 (in 7 patients!) to 5 habitual seizures monthly, the mean was one seizure in one month before the seizure during sauna has happened.

Conclusions: These results suggest that artificial hyperthermia may facilitates the occurence of seizures in epileptic patients especially in symptomatic focal epilepsy due to a brain tumor. This information may be of importance for this patient group. If the first seizure of a patient occurs during saunaing a brain tumor should be ruled out carefully. More information is needed with regard to the degree to which the risk of seizures is increased and which mechanisms play a role in hyperthermia-induced seizures

1 K. Hattemer , 1 S. Knake , 1 W. Oertel , 1 H. Hamer , 1 F. Rosenow 1University Hospital Giessen and Marburg GmbH (Marburg, D)

Seizures related to alcohol are a common finding and are usually attributed to alcohol withdrawal or a neurotoxic effect of ethanol leading to seizures that are unrelated to acute alcohol consumption or withdrawal. There is increasing evidence for a third kind of alcohol-related seizures: alcohol-induced seizures.

We describe a 36 year old patient who during the years from 1996 – 2006 was admitted 27 times with alcohol-induced seizures that occurred during alcohol consumption. Excluding one documented withdrawal seizure, serum ethanol levels at admission were 3.24o/oo ± .67o/oo. There were no seizures unrelated to alcohol consumption.

This case report supports the evidence of a seizure-provoking effect of ethanol that should be considered in the diagnosis and treatment of patients with alcohol-dependency and seizures.

1 M. Hoppe , 1 U. Specht , 1 R. Thorbecke , 1 T.W. May , 1 A. Ebner , 1 B. Pohlmann-Eden 1Epilepsy Center Bethel (Bielefeld, D)

Rationale: Suspected seizure provocation due to conditions at the workplace probably contributes to the high rates of unemployment in patients with epilepsy. The diagnosis of epilepsy frequently leads to restrictions concerning workplaces with word processing personal computers (PCs), even in patients without photosensitivity. Current knowledge about photosensitivity does not suggest an increased risk of seizures from office work with PCs because of the high refresh frequencies (≥ 75 Hz) of modern PC screens; however, studies with EEG registration in patients working at a PC have not been performed.

Methods: Patients with a documented photoparoxysmal response (PPR), i.e., epileptiform EEG discharges (ED) provoked by intermittent photic stimulation (IPS), were included. They underwent a standardized EEG procedure with simultaneous video recording. It comprised three evaluation periods in randomized order, each lasting 10 minutes: 1) baseline period (resting condition), 2) PC-test period (reading and editing a text at a personal computer using a word processing program), 3) control period (reading and editing a similar text printed on a white paper). For the PC-test period, a commercially available personal computer with a 17 inches cathode ray tube (CRT) monitor display (resolution 800×600 dpi, refresh rate of 85 Hz) was used. The number and duration of ED during each of the evaluation periods were assessed by two blinded examiners. In addition, a standard IPS was performed in order to document present PPR.

Results: Twenty-five patients were examined. Eight did not show PPR on IPS and thus were excluded. All remaining 17 patients had idiopathic generalized epilepsy. In 8 of them (47%), ED occurred at least during one of the three evaluation periods. There was no statistically significant difference between the three conditions (p = 0.25). Pairwise comparisons even indicated that during PC-test period the duration of ED was marginally shorter compared to baseline period (p = 0.094) and to control period (p = 0.063). None of the 17 patients showed an increase in duration of ED during condition “PC-test” compared to condition “control.” No seizures were recorded during the study.

Conclusion: In accordance with other studies using IPS, our results do not give evidence for an increased risk of seizure precipitation in photosensitive patients when using a word processing program displayed on a 17” CRT PC screen. Thus, restrictions concerning PC workplaces for office work are not warranted.

The study was supported by UCB Pharma GmbH, Kerpen, Germany

1 I. Coban , 1 R. Kretz , 1 S. Eibach , 1 V. Gaus , 1 B. Schmitz 1Charité Campus Virchow-Hospital (Berlin, D)

Background: The treatment with antiepileptic drugs (AED) during pregnancy is associated with the risk of teratogenic effects for the foetus. The data with respect to the teratogenicity of the older AED are conflicting, mainly due to inadequate sample size and methodological shortcomings. The teratogenic potential of newer AED is even less known.

To address this problem, a prospective registry for pregnancies exposed to AED has been established.

Methods: EURAP is a prospective and retrospective observational study. Women taking AED at the time of conception are included within the first 16 week of gestation for a prospective risk assessment. Cases ascertained later in pregnancy are recorded retrospectively.

Evaluation: At present more than 750 physicians from 39 countries have contributed 8785 cases to the central registry (Interim Report 11/2006). 3749 prospective cases have completed the study including the one-year follow-up after birth (age 29.6 ± 5.1 years (mean ± SD), ranging from 15 to 45 years. Epilepsy syndrome: generalized 41.2%, localization-related 52.7%; monotherapy 80.3%).

3252 live births, 279 spontaneous abortions, 125 induced abortions, 62 stillbirths and 31 perinatal deaths have been reported. There were 232 major congenital malformations (MCM) which represent a malformation rate of 6.7% and 30 chromosomcal monogenic abnormalities.

In 166 out of 2726 pregnancies with AED monotherapy one or more birth defects were observed (6.9%), as opposed to 64 out of 655 pregnancies with AED polytherapy (10.6%).

Until January 2007 170 neurologists, gynecologists and pediatricians reported 810 cases (prospective 793 = 89.3%) in Germany. Regarding age, epilepsy syndrome and monotherapy there were no significant differences compared to the international database.

The most frequently used anticonvulsive monotherapies are lamotrigine (43.2%), valproate (22.1%) und carbamazepine (20.1%).

There have been 489 live births, including 45 (8%) preterm deliveries and 36.1% caesarean sections. Spontaneous abortions occurred in 33 cases (4.7%), induced abortions in 16 cases (2.3%), including 4 cases with fetal anomalies. Furthermore 4 (0.7%) stillbirths and 5 (1%) perinatal deaths were observed.

There were 25 MCM, including 3 induced abortions (monotherapy 3.6%, polytherapy 5.7%). 4 more cases are not yet classified.

With respect to specific AED the malformation rates were: valproate 8 (8.2%), lamotrigine 6 (3.8%), carbamazepine 3 (3.5%).

1 P. Ratzka , 1 I. Schäfer , 1 M. Bartl , 1 A. Reeh , 1 C.D. Reimers 1St. Ansgar Hospital (Höxter, D)

We present the case of a 49-year-old woman who was initially admitted to our internal medicine department to investigate drop-attacks exsisting over the last 2 years. Remarkably, the drop-attacks only occurred during walking. During a typical episode, she stereotypically falls on her knees without loss of conciousness. On physical examination at admission, she presented with massive bruises at both knees. The extensive syncope evaluation revealed no pathological findings and the patient was referred to our neurological department for further diagnosis to exclude an epileptic cause. On neurological examination no pathological findings were found except weak tendon reflexes of the m. quadriceps femoris on both sides. The past medical history included a NIDDM since 8 years and a bronchial asthma since 12 month. She reported to be postmenopausal. The EEG showed intermittent frontotemporal theta-delta-activity on the left hemisphere, however no signs of epileptic activity. Because of the typical history, a normal MRI of the brain and after ruling out all other causes, we initially decided on the diagnosis of cryptogenetic drop-attack in climacteric (“maladie des genoux bleus”).

However, a subsequent EMG surprisingly revealed myopathic changes. We then performed a MRI of the thigh muscles which indicated a chronic neuromuscular disease of the ventral portion of the muscles. In a subsequent muscle biopsy of the left m. vastus lateralis we found typical signs of polymyositis.

In our poster presentation, we discuss the differential diagnostic procedure of the “maladie des genoux bleus.”

1 K . Anneken , 1 M. Fischera , 1 C. Kellinghaus , 1 G. Möddel , 1 S. Evers 1University Hospital Munster (Munster, D)

Introduction: The brainstem is supposed to be a component of the epileptic circuit in cortical seizures. However, the role of subcortical areas as a generator of epileptic seizures is discussed controversely (1). Here we report a case of focal status myoclonicus in a patient with transient brainstem ischemia.

Case: A 58-year-old woman was admitted to the emergency unit because of myoclonic jerks of her right arm, vertigo, anisocoria, diplopic images and left sided hemipareses for four hours before admission. An EEG was normal. The focal myoclonic jerks could be terminated by successive application of 3 mg lorazepam, all other symptoms ceased spontaneously within 48 hours. MRI imaging of the brain and cerebral fluid analysis did not indicate an ischemic lesion or inflammation. The diagnosis of transient brainstem ischemia was made.

Conclusion: Our case of focal status myoclonicus associated with brainstem symptoms suggests that special subtypes of epileptic seizures may originate from subcortical regions.

Literature: 1 Mesiwala AH, Kutarani JD, Avellino AM, Roberts TS, Sotero MA, Ellenbogen RG. (2002) Focal motor seizures with secondary generalization arising in the cerebellum. Case report and review of the literature. J Neurosurg 97:190–196.

1 M. Pfäfflin , 2 T.W. May , 1 H. Sudbrock , 1 S. Urban , 2 R. Neb , 3 A. Übelacker , 4 H. Kießling , 4 D. Dennig , 3 B. Schmitz 11 Epilepsy Center Bethel (Bielefeld, D) ; 2 Gesellschaft für Epilepsieforschung (Bielefeld, D) ; 3 University Hospital Charité (Berlin, D) ; 4 Schwerpunktpraxis Epileptologie (Stuttgart, D)

Aims: Epilepsy affects multiple aspects of life including everyday life, social relations, work, and leisure activities. Thus, comprehensive care for patients with epilepsy includes support and counseling specific to patients' needs. In outpatient clinics and private practices nurses are key persons for patients. To improve nurses' skills for counselling and patient management a modularized educational program was developed for nurses working in an outpatient setting of epilepsy care. The efficacy of this educational program will be evaluated in a controlled study.

Methods: Nurses are trained on the job including workshop-like modules every three months. Topics include interactive skills, epilepsy specific knowledge, and emotional aspects of coping, counseling (including regulations), self-management and patient (case) management.

The efficacy of the educational program will be evaluated in a multi-center, prospective, controlled study with respect to, e.g., epilepsy knowledge of the patients, coping, treatment outcome, compliance, quality of life and especially contentedness of the patients with counseling and treatment. The patients were randomly assigned (either counseled by the nurse or not in addition to standard treatment) and answered questionnaires before and six months after counseling / respectively standard treatment. Three outpatient care units are included. The evaluation study has started in June 2006; up to now 254 patients have been included.

Results: The modules and the organization of the educational program are introduced. Experiences and first outcome results of the evaluation will be presented. The systematic evaluation will allow a more profound assessment of the efficacy of the educational program for epilepsy nurses and improvement of the program.

Conclusion: Preliminary feedback of patients and physicians indicate that epilepsy nurses play an important role in counselling of patients with epilepsy and improve the contentedness of the patients in an outpatient setting of epilepsy care.

The project is supported by an unrestricted grant of the UCB Pharma GmbH, Kerpen, Germany and the German Chapter of the International League Against Epilepsy (Deutsche Gesellschaft für Epileptologie e.V. Bielefeld, Germany).

1 A. Strzelczyk , 1 J.P. Reese , 1 R. Dodel , 1 H. Hamer 1Philipps-University (Marburg, D)

Objective: To give an overview on published cost of illness studies and their methodological approaches.

Background: Epilepsy imposes a substantial burden on individuals and society as a whole. The mean prevalence of epilepsy is estimated at 0.52% in Europe and according to the WHO Epilepsy Atlas peaks up to 1.0% in developing countries. Estimation of the health economic burden of epilepsy is of pivotal relevance to enable a rational distribution of society's resources. Even more as the introduction of new antiepileptic drugs, the marketing of the vagal-nerve stimulator and the intensifying of surgical treatment options may have caused a considerable increase in costs of therapy.

Methods: A systematic literature review was performed to identify studies that evaluated direct and indirect costs of epilepsy. Using a standardized assessment form, information on the study design, methodological framework, and data sources was extracted from each publication and systematically reported.

Results: We identified 21 studies on costs of epilepsy worldwide. The majority of the studies reflected the costs of epilepsy in Europe [UK (3), I (3), GER (1), NL (1), CH (1), F (1) and EU (1)] and the US (4), but also studies were available from India (2), Hong Kong, Oman, Burundi and Mexico. The studies utilize different frameworks to evaluate costs. All of them used a bottom-up approach, however only 12 studies (57%) evaluated direct as well as indirect costs. The range for the mean annual direct costs lay between 48 Euros (adjusted for PPP 2003) in rural Burundi and 4040 Euros in a German epilepsy center. Anticonvulsant medication was the main contributor to direct costs in recent years. The indirect costs ranged between 12% (I) and 85% (US) of the total annual costs.

Conclusions: A reliable comparison of the different cost of illness studies in epilepsy is currently not feasible as the evaluated studies had substantial limitations in respect to their patient selection criteria, diagnostic stratifications, but also regarding methods and evaluated costs. Most of the studies did not adhere to the guidelines for performing health economic evaluations. Therefore there is an urgent need for studies which evaluate direct and indirect costs in a comparable fashion.

1 S. Novak , 1 R.A. Sälke-Kellermann , 2 E. Martin , 1 G. Krämer 11 Swiss Epilepsy Centre (Zurich, CH) ; 2 University Children's Hospital Zurich (Zurich, CH)

Introduction: Endosulfan is a chlorinated hydrocarbon insecticide with a high toxicity on the central nervous system. It alters the serotoninergic system as well as the GABAergic and cholinergic systems and inhibits the calmodulin dependent Ca-ATPase. Though forbidden in many countries, it is still broadly used and constitutes an important cause of poisoning worldwide. Acute neurological symptoms after intoxication are tonic–clonic seizures, headache, dizziness, and ataxia.

Case report: We report on a previous healthy boy who at the age of 2 3/4 years accidentally ingested an insecticide containing about 10 ml of Endosulfan. Shortly after ingestion he had a syncope followed by a convulsive status epilepticus and a metabolic and respiratory acidosis. He was intubated and referred to a pediatric intensive care unit. The status was stopped with intravenous diazepam. The post-status EEG showed a severe nonspecific disturbance without epileptiform activity. CT-scan of the brain was normal. He recovered well and was discharged after five days. Three years later he presented complex-focal seizures with left sided temporal lobe semiology, increasing learning difficulties, and behavioral disturbances. MRI at this time showed left sided hippocampal sclerosis (HS). At the age of 7 years the neuropsychological assessment showed an organic amnestic syndrome. At this time MRI demonstrated the suspicion of bilateral pathology. Two years later bilateral HS was proved beyond doubt. He still suffered from refractory seizures at this time.

Discussion: In the literature it has been shown with serial MRI studies that central nervous system lesions associated with acute Endosulfan intoxication affect specific structures localised to the caudate nucleus, putamen, and occipital cortex. These lesions were bilateral and reversible. To our knowledge there are no reports of residual hippocampal lesions as sequel of an intoxication with Endosulfan. Harmful events in childhood as prolonged febrile seizures, convulsive status epilepticus, insults, or infections of the central nervous system, play an important role in the pathogenesis of HS.

Conclusion: The evidence available in this case, especially the convulsive status and the following latency period until the manifestation of epilesy, would suggest that the Endolsufan poisoning was not the primary cause of the chronic epilepsy and hippocampal sclerosis but rather that it was due to the Endosulfan-induced convulsive status epilepticus.

1 P. Gallmetzer , 1 B. Glawar , 2 J. Ferrari , 3 W. Lalouschek , 1 B. Mamoli 11 Ludwig Boltzmann Institute for Epilepsy (Vienna, A) ; 2 Convent Hospital "Barmherzige Brüder” (Vienna, A) ; 3 University Hospital of Neurology (Vienna, A)

Purpose: To assess the incidence and predictive factors of poststroke epilepsy in the elderly.

Methods: A total of 1265 patients (aged 65 to 98 years) with recent ischemic cerebrovascular event, were prospectively enrolled at 9 neurological departments. The follow up was 19 ± 9.5 month (mean, ± SD). Seizure patients were followed for 39 ± 12.7 month (mean, ± SD).

Results: Nineteen of the 1265 patients (1.5%) developed poststroke epilepsy. Persistent neurological deficit (NIHSS > 12) at day 7 after stroke (odds ratio [OR] 7.1, 95% CI 2.1 to 23.9), cerebral complications like cerebral edema and secondary hemorrhagic transformation (OR 18.5, 95% CI 4.8 to 71.1) and previous early poststroke seizures (OR 56.8, 95% CI 16.9 to 190.7) were independently associated with poststroke epilepsy. Diabetic comorbidity showed inverse correlation to seizure recurrence (OR 0.1, 95% CI 0.01 to 0.9).

Conclusion: Epilepsy is occurring in a minority of patients after ischemic stroke. A Persistent neurological deficit (NIHSS > 12), cerebral complications like edema and hemorrhagic transformation and previous early poststroke seizures were predictors of poststroke epilepsy, whereas diabetic comorbidity showed an inverse correlation to seizure recurrence.

1 C. Pollo , 2 M. Seeck , 2 C. Boex , 2 L. Spinelli , 1 E. Pralong , 1 G. Foletti 11 CHUV (Lausanne, CH) ; 2 HUG (Geneva, CH)

Background and purpose: High-frequency (HF) amygdalo-hippocampal stimulation (AHS) has recently been proposed as a therapeutic option for intractable temporal lobe epilepsy. Although preliminary results are promising, long-term efficacy of this technique as well as stimulation paradigms or implantable device are still under evaluation. We report the Lausanne-Geneva experience regarding the safety and effect of AHS on seizure control.

Material and methods: 6 patients with intractable epilepsy unsuitable for resective surgery were selected for unilateral AHS. All cases but one showed strictly unilateral epileptogenic zone located in the mesial temporal lobe, 3 were right sided, one left, one bilateral predominantly left.

The electrode was implanted through a posterior approach at the junction between the hippocampus and the parahippocampal gyrus with the distal contact located within the inferior aspect of the amygdala. External extension was provided to perform deep EEG recordings for 3-4 days. The stimulator was then implanted under general anesthesia. Initially, stimulation was delivered simultaneously through the 4 contacts at a frequency of 130 Hz, pulse width 450μs, and amplitude up to 1–2 V and subsequent refinement was performed according to deep EEG recordings.

Results: Stimulation was performed in the amygdala in 1/6 case, anterior hippocampus in 3/6 cases and posterior hippocampus in 2/6 cases. Follow-up from 5 months to 3 years shows a decrease of seizure frequency in all the patients (30-90%). No stimulation-induced side effects were noticed. The whole device was changed in 1 case following dysfunction of the electrode.

Conclusion: HF-AHS is safe and effective as an alternative treatment for medically intractable temporal lobe epilepsy in selected cases. Further controlled trials and fundamental research are needed for the assessment of long-term efficacy and a refined understanding of mechanisms of action of electrical stimulation.

Abstract

  1. Top of page

Poster session 2.1. Basic sciences and experimental epileptology

1 N. Gueler , 2 M. Kukley , 2 M. Nikoulina , 2 D. Dietrich 11 Swiss Epilepsy Centre (Zurich, CH) ; 2 University Hospital Bonn (Bonn, D)

Objectives: Presynaptic group II mGluRs modulate transmitter release at many glutamatergic terminals in the CNS but it is unclear which patterns of presynaptic neuronal activity leads to a relevant feedback inhibition. To determine the requirements for synaptic activation of group II mGluRs and to elucidate the factors regulating their physiologic activation by released glutamate.

Material and methods: Hippocampal slices were prepared from male wistar rats (20-40 days old). Field excitatory postsynaptic potentials (fEPSP) were recorded in the middle molecular layer of the dentate gyrus. Repetitive stimulation of presynaptic fibers was used to activate mGluRs. We varied the number of stimuli as well as the frequency of stimulation. In addition we used different post-tetanic intervals to assess how long mGluRs may remain activated. We employed the following protocols: 1, 3, 7, 14, 28, 56, 100, 200 stimuli @ 3, 5, 20, 100, 200 Hz and 0.03, 0.05, 0.2, 0.3, 0.8, 1, 3, 6, 20 sec post-tetanic interval. To evaluate the degree of synaptic activation of group II mGluRs we applied the antagonist LY 341495 (3μM) and analysed the resulting increase of the test fEPSPs following tetanic stimulation.

Results: Tetanic stimulation leads to a modest, but significant activation of group II mGluRs. The broad spectrum glutamate transporter blocker DL-TBOA (100μM) facilitates synaptic activation of mGluRs. The results show that synaptic activation of group II mGluRs is dependent on stimulation frequency, number of stimulation and post-tetanic interval. More stimulations and a higher stimulation frequency yielded a more pronounced activation of the group II mGluRs. Strongest inhibition of the test fEPSP was observed between 0.1 and 0.3 sec following the tetanus. GABA receptor antagonists (Bicucullin, CGP) did not alter the magnitude of the test fEPSPs after application of LY341495.

Discussion: In contrast to ionotropic glutamate receptors, synaptic activation of group II mGluRs occurs only after high frequency stimulation. The optimal stimulation protocol is most likely determined by an interplay of glutamate removal mechanisms, replenishment of releasable vesicles and the activation/deactivaton kinetics of mGluRs. We propose that the slow deactivation kinetics of metabotropic glutamate receptors may provide longer lasting effects.

1 M. Tomka-Hoffmeister , 1 R. Schulz , 1 A. Ebner 1Epilepsy Center Bethel (Bielefeld, D)

Objective: Earlier studies demonstrated the efficacy of levetiracetam given to patients who did not become seizure-free after epilepsy surgery. Does this also apply to patients who did not respond to levetiracetam before surgical intervention?

Method: We included patients meeting the following criteria:

  • - 
    had epilepsy surgery from 2003 to 2006
  • - 
    had no response to levetiracetam before the operation
  • - 
    were not given levetiracetam during and immediately after the operation
  • - 
    did not become not seizure-free after surgical intervention
  • - 
    were postoperatively readministreted levetiracetam

N = 9 (1 child und 8 adults)

Mean age at time of operation = 31 yrs. (Range 4.75 to 47 yrs.)

Gender: 6 male und 3 female

Comparison was made of the seizure-situation:

  • - 
    prior to the operation
  • - 
    after the operation without levetiracetam for 6 months. The child alone was observed for only 4 months.
  • - 
    after the operation with levetiracetam for an average 17 months (Range 5 to 30 months)

Taken into consideration was the dosage-level of levetiracetam (and comedication) before and after the operation.

Result: 6 patients became seizure-free despite lower doses of levetiracetam; 5 of these became completely free of auras and seizures and 1 had only auras. In 2 further patients there was improvement in frequency. Only in 1 case was a worsening observed.

Conclusion: Our results confirm other studies showing that levetiracetam is effective in patients who underwent surgery but did not become seizure-free after the operation. Furthermore we show that the majority of patients who did not respond to levetiracetam before surgery, did so after that surgery.

1 R. Trollmann , 1 J. Schneider , 1 D. Wenzel , 1 W. Rascher , 2 O. Ogunshola , 2 M. Gassmann 11 University of Erlangen (Erlangen, D) ; 2 University of Irchel (Zurich, CH)

Background: Hypoxia-inducible transcription factors (HIF), the most important mediators of the molecular response to hypoxia, are ubiquitously expressed in human and rodent tissues, including brain. By transcriptional activation of specific target genes HIF-regulated systems are critically involved in adaptive mechanisms to hypoxic-ischemic brain lesions. As the antiepileptic drug levetiracetam (LEV) reduces cellular degeneration in focal cerebral ischemia, we aimed to prove its effects on endogenous adaptive factors in developing hypoxic brain.

Objectives: To study effects of LEV on endogenous neurotrophic HIF-regulated factors (HIF-1alpha subunit, VEGF, iNOS) in developing mouse brain at different developmental ages (P0, P7) as well as effects of LEV on hypoxia-inducible, but HIF-independent, NOS isoforms (nNOS, eNOS).

Material and methods: Fetal C57BL/6 mice (P0, P7) were treated with saline or LEV (i.p.; 50 mg/kg) 1 h before exposure to normoxia (21% O2; N) or severe systemic hypoxia (8% O2, 6 h; H). Brains were dissected immediately without reoxygenation. HIF-1alpha protein accumulation was investigated by Western blot and immunohistochemistry (cerebral cortex, hippocampus). Gene expression was quantified by TaqMan RT PCR.

Results (mean ± SEM): Under normoxia, no significant changes of cerebral HIF-1alpha, VEGF and NOS gene expression were observed in LEV-treated (P0, n = 6; P7 n = 7) compared to control mice (P0, n = 6; P7, n = 7). Similarly, HIF-1alpha staining did not show remarkable differences of protein accumulation in hippocampus (HC) or cerebral cortex between saline- and LEV-treated animals. Systemic hypoxia itself led to prominent accumulation of HIF-alpha protein at P0 and P7 as well as significant up-regulation of VEGF at P0 (N, 0.022 ± 0.001, n = 6; H, 0.031 ± 0.003, n = 6) and at P7 (N, 0.096 ± 0.032, n = 7; H, 0.873 ± 0.069, n = 7; p < 0.001). Interestingly, a significant decrease of iNOS mRNA levels in hypoxic brains at P0 (N, 1.001 ± 0.057; H, 0.733 ± 0.048, p < 0.01) and at P7 (N, 1.364 ± 0.083; H, 0.319 ± 0.047, p < 0.001) was observed. LEV treatment did not affect HIF-1alpha accumulation in hypoxic cerebral cortex or HC (P0, n = 9; P7, n = 7) compared to controls (P0, n = 6; P7, n = 7), or substantially alter gene expression of VEGF, iNOS, eNOS and nNOS, except, a normalization of iNOS decrease in hypoxic P0 brains.

Conclusions: Severe systemic hypoxia differentially affects expression of HIF- regulated vasoactive factors in fetal mouse brain in vivo during acute hypoxia. No evidence of toxic effects of LEV on HIF-1alpha and vasoactive HIF target genes was observed. Under hypoxia, expression of HIF-1alpha protein, HIF-dependent and independent genes was not significantly altered by LEV-treatment (P0, P7) indicating no significant influence of LEV on crucial endogenous, HIF-regulated neuroprotective mechanisms.

1 L. Müller , 1 T. Tokay , 2 C. Rüschenschmidt , 2 J. Chen , 2 A. Becker , 1 R. Köhling , 2 H. Beck , 1 T. Kirschstein 11 University of Rostock (Rostock, D) ; 2 University of Bonn (Bonn, D)

Chronic temporal lobe epilepsy is often accompanied by cognitive dysfunction such as declarative memory deficit. In order to find promising strategies to address these deficits, we study fundamental molecular mechanisms of learning and memory such as long-term potentiation (LTP) and long-term depression (LTD). These are long-lasting changes of synaptic strength due to the history of prior neuronal activity. By using long-term extracellular recording techniques, we asked how chronic epileptic seizures influence synaptic plasticity in the hippocampal CA1 region in the rat pilocarpine-induced status epilepticus model. Synaptic plasticity can be induced by activation of either NMDA-type glutamate receptors or metabotropic glutamate receptors (mGluR). First, we investigated the mGluR-dependent form of LTD and observed a significant reduction of this type in epileptic rats. This was found to be due to a specific transcriptional down-regulation of the mGluR5 subtype. Then we focused on NMDA receptor-dependent synaptic plasticity. Whereas NMDA receptor-dependent LTD is unaltered in chronic epilepsy, we found an unexpected increase of NMDA receptor-dependent LTP. Therefore, we asked whether this discrepancy is due to differential alteration of the most important NMDA receptor subtypes NR2A and NR2B. We found that application of low micromolar zinc (1-10 μM), which primarily modulates NR2A, virtually abolished LTP in controls, whereas in epileptic rats significant LTP was still observed. At a higher zinc concentration (100 μM), LTP was inhibited in both groups. In contrast, when NR2B was specifically inhibited, LTP remained inducible in controls, but was completely lost after status epilepticus. These results suggested a dominant role of NR2B in LTP induction in epileptic rats which may be caused by post-transcriptional and/or post-translational modification of these NMDA receptor subtypes.

In conclusion, status epilepticus leads to persistent changes of various forms of synaptic plasticity in the CA1 region. These changes could be in part accountable for cognitive impairment seen in many patients. A more precise understanding of the underlying mechanisms will be essential to develop new therapeutic strategies.

1 H. Feldwisch genannt Drentrup , 1 B. Schelter , 2 J. Wohlmuth , 2 J. Nawrath , 2 A. Brandt , 1 J. Timmer , 2 A. Schulze-Bonhage 11 University of Freiburg (Freiburg, D) ; 2 University Hospital Freiburg (Freiburg, D)

Concerning the assessment of the predictability of epileptic seizures, a continuous and reliable analysis of neurophysiological recordings of epilepsy patients is necessary. Recent work has indicated that methods from nonlinear dynamics can detect significant preseizure changes in the EEG at least for some patients. As a mean to increase seizure prediction performance we tested whether or not two different kinds of combination of seizure prediction methods yield superior prediction results. In this study a bivariate phase synchronization index (PSI) and the univariate “Dynamical Similarity Index” (SIM) were adapted for seizure prediction. The analysis of these prediction methods was based on long-term intracranial EEG data with continuous recordings of 8 patients for 7-11 days including 124 seizures. Approaches to combine methods were applied and compared to the individual methods. All results have been validated by a statistical test procedure (Schelter et al., Chaos 2006; 16: 013108). By evaluating the EEG data continuously and without knowledge of ‘future’ events, a prediction situation analogous to an online application was possible. For a fixed set of parameters, significant prediction results could be observed for five (PSI) and four (SIM) patients by evaluating each method individually, with average sensitivities of 51% (PSI) and 49% (SIM). A combination of both methods using a logical “and” showed improved results, with a mean sensitivity of 64%. A combination of both prediction methods applied shows an increase in sensitivity of about 25%. These findings represent a next step towards a clinical application of seizure prediction.

1 A. Capurro , 1 A. Aertsen , 2 A. Schulze-Bonhage 11 University of Freiburg (Freiburg, D) ; 2 University Hospital Freiburg (Freiburg, D)

Purpose: To assess the evolution of high frequency (HF) content in afterdischarges (ADs) during kindling.

Methods: We recorded ADs evoked by short stimulus trains delivered to the right hippocampus of 10 Wistar rats. The recording electrodes were placed bilaterally in hippocampi and frontal neocortices (4 in the right dorsal hippocampus, 2 in the left dorsal hippocampus and 1 in right and left frontal neocortex each). A common reference electrode was placed above the cerebellum. The stimulus consisted of a 60 Hz train of biphasic rectangular pulses lasting 1.6 s. Each biphasic pulse lasted 0.4 ms with an amplitude of ± 0.5 mA. The stimulation was performed daily in each rat over a period of 5-10 days. EEG was recorded at 10.4 kHz sampling rate and signals were split into physiologically meaningful periods, e.g., control period before stimulus and various AD periods with different waveforms. For each period, spectra and spectrograms with an optimized window width were computed to assess the HF content.

Results: After the stimulus we recorded an AD at all recording sites. On the first day the AD started with a series of slow spikes. In the subsequent days, HF ripples (100-300 Hz) were superimposed to these slow waves, the total duration of the AD increased, and the HF ripples extended to the contralateral hippocampus. We often observed a 2nd phase of AD containing HF separated from the initial AD by a period with low amplitude rhythmic alpha or mixed frequency activity. During the 2nd phase of the AD a behavioral seizure was observed in some cases. The HF content of the EEG (consisting of ripples superimposed on slow waves of high amplitude) was observed before and during the seizure, but behavioral seizure manifestation was not accompanied by a clear change in the EEG pattern.

Conclusions: During the kindling process, afterdischarges occurred from the 1st day on. They were not limited to the site of stimulation but appeared in both, ipsilateral and contralateral hippocampus and neocortical areas. HF discharges developed over subsequent days and increased in duration and amplitude when behavioral seizures occurred, and were not limited to the kindling site but also extended to the contralateral hippocampus.

Supported by BMBF (Grant 01GQ0420 to BCCN-Freiburg). The authors thank to Martin Müller, Ute Haeussler, Ralph Meier, Armin Brandt and Michael Jachan.

1 B. Schelter , 1 H. Feldwisch genannt Drentrup , 1 J. Wohlmuth , 1 J. Nawrath , 2 A. Brandt , 1 J. Timmer , 2 A. Schulze-Bonhage 11 Center for Data Analysis and Modeling (Freiburg, D) ; 2 University Hospital Freiburg (Freiburg, D)

Objective: A reliable prediction of epileptic seizures would offer new therapeutic options for treating the severe disease epilepsy. An abnormal synchronization of neurons leading to epileptic seizures proposes multivariate time series analysis techniques detecting synchronized dynamics as a promising approach for seizure prediction when applied to invasive EEG recordings of epilepsy patients. The proper assessment of these techniques is a prerequisite for a seizure prediction device. Apart from statistical significance the clinical relevance is investigated.

Methods: A quantity measuring phase synchronization (Mormann et al., Physica D 2000;144:358-369) has been applied to an invasive EEG data pool of 14 patients with long term intracranial continuous EEG data of more than 100 days duration. The seizure prediction performance has been assessed for long intervention times of up to 4 hours and occurrence periods of up to 2 hours, which are not unusual in epilepsy research. The obtained results have been compared to the performance of an unspecific random predictor (Schelter et al., Chaos 2006;16:013108). Moreover, additional measures have been used to quantify the strain of especially false predictions on patients. Thus, the fraction of false predictions with respect to the total number of predictions as well as the time a patient is awaiting a seizure that will never occur have been investigated to quantify the clinical relevance of the prediction techniques.

Results: For several patients a prediction performance superior to a random predictor could be shown. Sensitivities of up to 75% were observed when allowing 2 false predictions per day. To achieve this promising result the patient is awaiting seizures for 15%, i.e., 3–4 hours every day that will never occur. This fraction is rather low compared to other published results, which is the benefit of subdividing the prediction horizon in intervention time and occurrence period. A fraction of false alarms with respect to the total number of alarms of more than 50% on average was observed.

Conclusions: Preictal changes in the synchronization of the EEG dynamics may offer a chance for epileptic seizure prediction. Besides statistical significance other parameters have to be taken into account quantifying the clinical relevance. Especially avoiding false predictions will reduce the strain on patients and, thereby, lead to a much higher clinical relevance.

1 M. Bechstein , 1 M.C. Müller , 1 M. Kirsch , 1 A. Flubacher , 1 S. Tinnes , 1 C.A. Haas 1University of Freiburg (Freiburg, D)

The widening of the granule cell layer (granule cell dispersion, GCD) in the dentate gyrus is a characteristic feature of Ammon's horn sclerosis which often accompanies temporal lobe epilepsy (TLE). The dispersion is considered to be a local migration defect, but the mechanisms underlying its development are rather unclear. We have recently shown that GCD formation is paralleled by the appearance of a radial glial scaffold in the dentate gyrus, observable in TLE patients and in a mouse model of TLE (Fahrner et al., 2006; Heinrich et al., 2006). In order to investigate the role of glial cells in GCD development, we used a mouse model of TLE and combined it with preactivation of astrocytes. To this end, we injected ciliary neurotrophic factor (CNTF), a potent activator of glial cells, into the hippocampus of adult mice 48 hours prior to the injection of kainate (CNTF preconditioning). Injection of kainate (KA) alone is known to induce focal epileptic seizures and GCD. At sixteen days after KA injection we measured the width of the granule cell layer (GCL) in cresyl violet-stained hippocampal sections of CNTF-preconditioned mice and in a control group treated with saline prior to KA injection. While these mice showed a mean increase in GCL width by 102 micrometers, in CNTF preconditioned mice the increase was only 27 micrometers. This corresponds to a reduction of the GCD by 74%. In order to verify CNTF-mediated activation of hippocampal astrocytes, GFAP mRNA expression was measured by real time RT-PCR analysis. Two days after CNTF injection we found a strong up-regulation of GFAP mRNA expression. The effect was noticable even 18 days after CNTF injection by GFAP immunohistochemistry. Taken together our results show that CNTF-mediated activation of astrocytes strongly reduces GCD formation.

(Supported by the DFG: Transregio SFB TR3 and SFB 505).

1 M. Osswald , 1 M.C. Müller , 1 S. Huber , 1 C.A. Haas 1University of Freiburg (Freiburg, D)

The impairment of neurogenesis in Ammon's horn sclerosis, which is often associated with temporal lobe epilepsy (TLE), has recently been described in the resected hippocampi of patients suffering from this neurological disorder (Fahrner et al., Exp Neurol 2007). Unilateral intrahippocampal injection of kainic acid (KA) in adult mice induces an epileptic focus associated with the major histopathological features of TLE including the observed loss of neurogenesis. In order to elucidate the molecular mechanisms involved in the regulation of neurogenesis we investigated potential changes in the expression of vascular endothelial growth factor (VEGF), a potent neurogenic and angiogenic molecule, in the KA-injected hippocampus. We performed a postlesional time course analysis at two, four, eight, twenty-one and eighty days after KA injection. Using in situ hybridization we found constitutive expression of VEGF mRNA in pyramidal cells of the cornu ammonis and in the granule cell layer of the dentate gyrus. This expression pattern did not change during the whole time period. However, we observed a transient up-regulation of VEGF mRNA expression in the ipsilateral hilus and molecular layer with a peak at eight days after KA injection. In order to identify the phenotype of these VEGF mRNA-positive cells, we performed fluorescence in situ hybridization (FISH) for VEGF mRNA combined with immunolabeling for NeuN (as neuronal marker) or GFAP (as astrocytic marker): Most of the newly appearing VEGF mRNA-expressing cells were double labeled with GFAP showing that reactive astrocytes synthesize VEGF mRNA in the epileptic hippocampus. Since VEGF is also a strong angiogenic factor, possible alterations in blood vessel distribution were studied by immunolabeling for CD 31, a marker for adult and embryonic endothelial cells. We observed a permanent increase in blood vessel density in subregions of the epileptic hippocampus. Taken together, we show that epilepsy triggers a de novo synthesis of VEGF in astrocytes. We hypothesize that VEGF does not play a role in the down-regulation of neurogenesis, but rather in vessel formation in the epileptic hippocampus. (Supported by the DFG: TR-3).

1 Y.G. Weber , 1 J. Kempfle , 1 D. Blazevic , 1 S. Maljevic , 1 H. Lerche 1University of Ulm (Ulm, D)

The syndrome of benign familial neonatal convulsions (BFNC) is characterized by clusters of epileptic seizures within the first days of life disappearing spontaneously after weeks to months. BFNC is caused by loss-of-function mutations in the voltage-gated potassium channels KCNQ2 or KCNQ3, which can explain the occurrence of epileptic seizures via membrane depolarization. However, the underlying mechanism of the transient phenotype of BFNC still remains to be elucidated. One possible explanation might be a change in the expression pattern of these channels during development. We therefore examined the expression and localization of KCNQ3 channels in the developing mouse brain.

We used 5-10 μm frozen sections of mouse brain at postnatal (P) days P0-P40 and performed immunofluorescence staining using a polyclonal rabbit antibody against KCNQ3. A specific S35-labeled KCNQ3-probe was used for the in situ hybridization.

KCNQ3 is ubiquitously expressed in brain with predominance in hippocampus, cortex and thalamus. In the hippocampus, staining of axon initial segments was observed in pyramidal neurons in the CA1 to CA3 region starting from P3, whereas in the dentate gyrus a specific KCNQ3 signal at axon initial segments was first detected at P15. The delayed appearance of the staining in the dentate gyrus was confirmed in the in situ hybridization, in which the KCNQ3 signal appeared at P8.

The analysis of KCNQ3 expression at different developmental stages in mice brain revealed a delayed staining of the dentate gyrus which might contribute to the age dependence of BFNC.

1 S. Kovac , 1 Y. Sirin , 1 C. Kellinghaus , 1 G. Möddel , 1 M. Fischera , 1 M. Dogan , 1 S. Evers , 1 E.-J. Speckmann , 1 A. Gorji 1University Hospital Munster (Munster, D)

In the present study, the effects of adenosine on stimulus-induced bioelectric activity were investigated in the rat hippocampus and neocortex under epileptiform (n = 17) and non-epileptiform (n = 16) conditions.

Spatio-temporal distribution of activities was studied by using a voltage-sensitive dye (RH795) and a fast optical recording system. Epileptiform activity was induced by omission of extracellular Mg2+ before adenosine (1, 10, 50 and 100 μmol/l) was added to the bath medium. In hippocampal slices, stimulus (with an intensity of 1, 5 and 10 μA) was given in Schaffer collaterals, and dye-related fluorescence signals were detected by a 464-element honey-comb shaped photo diode in CA1 regions. In neocortical slices, stimulus was given either in the border zone between white and gray matter (afferent fibre stimulation) or in layer I/II (stimulation of parallel fibres), and signals were detected in layer I-VI of the cortical motor regions.

In the hippocampus, adenosine led to a dose dependant reduction of stimulus-induced activity. The reduction was most effective when adenosine was applied in concentrations of 10 and 50 μmol/l and when stimulus was given with middle (5 μA) or low strengths (1μA). Furthermore, a significant reduction of signal amplitude was observed primarily under epileptiform conditions. In the neocortex adenosine showed no effect on stimulus induced signal amplitude neither in non-epileptiform nor in epileptiform tissues.

In conclusion, the results demonstrate a different regional effect of adenosine on stimulus-induced activity. The neuroinhibitory effect of adenosine is predominant in hippocampal tissues where adenosine receptors are found in high densities. Furthermore, data here indicate that adenosine exerts its modulatory effects via synaptic network activity especially when the tissues are highly excited.

1 F. Klinker , 1 D. Hering , 1 R. Koch , 1 M.A. Nitsche , 2 H. Potschka , 2 W. Löscher , 1 W. Paulus , 1 F. Tergau , 1 D. Liebetanz 11 University of Gottingen (Gottingen, D) ; 2 School of Veterinary Medicine (Hannover, D)

Purpose: Cathodal transcranial direct current stimulation (tDCS) is able to induce a lasting, focal depression of cortical excitability. This may be a promising tool for the treatment of focal cortical hyperexcitability leading to focal epileptic seizures. In this study, we evaluated the anticonvulsant potential of cathodal tDCS in a transcranial ramp stimulation model of focal epilepsy.

Methods: To determine the threshold for localized seizures (TLS), we designed a new model of transcranial ramp stimulation (TRS). TLS was determined in freely moving rats by applying a single train of rising bipolar pulses through a unilateral epicranial electrode and was defined as the minimum current strength leading to a unilateral forelimb clonus. The unilateral electrode montage permitted the focal application of both tDCS and TRS over the same cortical area without trepanation. After tDCS, TLS was determined repeatedly for 120 min at an interval of 15 min. The first group of animals received 2 sessions of cathodal tDCS at 100μA, one for 30 and one for 60 min. A third session consisted of 60 min of anodal tDCS. A second group received cathodal tDCS at 200 μA for 15 and for 30 minutes, as well as anodal tDCS for 30 min.

Results: After application of cathodal tDCS at 100 μA for 60 min a significant TLS increase could be seen for 120 min. When the intensity was increased to 200 μA, the duration needed to achieve a lasting TLS elevation was reduced to just 30 min. In contrast, anodal tDCS at identical stimulation durations and current strengths had no significant effect on TLS, proving the polarity dependency of the antiepileptic effect.

Conclusion: Cathodal tDCS is able to induce a lasting elevation of the threshold for localized seizures. The duration of tDCS needed to induce a lasting effect is dependent on the current strength of the stimulation. These results imply a possible therapeutic use of tDCS for the focal treatment of drug refractory partial epilepsy. Cathodal tDCS could be of particular therapeutic value in patients with epileptic foci at the surface of the hemisphere such as epilepsia partialis continua and focal or secondary generalized epilepsy resulting from cortical dysplasia.

1 T.V. Wuttke , 1 F. Lehmann-Horn , 2 W. Paulus , 3 J. Penzien , 1 K. Jurkat-Rott , 1 H. Lerche 11 University of Ulm (Ulm, D) ; 2 University of Gottingen (Gottingen, D) ; 3 Hospital Augsburg (Augsburg, D)

Benign familial neonatal convulsions (BFNC) is caused by mutations in the voltage-gated potassium channels Kv7.2 and Kv7.3, being expressed in CNS and PNS. BFNC is characterized by frequent unprovoked seizures, typically beginning within the first days of life and spontaneously disappearing after several weeks to months. Peripheral nerve hyperexcitability (PNH, neuromyotonia, myokymia) is characterized by muscle overactivity due to spontaneous discharges of lower motor neurons. Antibodies against or mutations within different types of voltage-gated potassium channels can cause PNH. PNH could only be linked to one single Kv7.2 mutation (R207W) so far, in a family with PNH and BFNC. Here we describe two new mutations within the Kv7.2 channel, leading either to BFNC (E119G) or myokymia (R207Q).

Genomic DNA was amplified and sequenced. WT and mutant Kv7.2 channels were heterologously expressed in Xenopus oocytes and functionally characterized using two-electrode voltage-clamping.

Electrophysiological studies of E119G demonstrated that this mutation induces a slight depolarizing shift of the voltage dependence of activation. The steady-state activation curves for E119G and all co-expressions of E119G with Kv7.2 and/or Kv7.3 wild type (WT) channels were significantly shifted compared to Kv7.2 or Kv7.2/Kv7.3. These shifts resulted in significant reductions of the relative current amplitudes for mutant channels in the subthreshold range of an action potential. Furthermore, the voltage dependence of the time constants of activation was shifted by about +10 mV revealing a significant slowing of the activation time course. In contrast, functional characterization of R207Q and R207W revealed large depolarizing shifts of the conductance-voltage relationships and marked slowing of the activation time course for R207W > R207Q compared to wild type channels. Coexpression of both mutant channels with wild type channels revealed a dominant negative effect, reducing the relative current amplitudes in the physiologically relevant subthreshold range by 70–90%.

Many of the known BFNC mutations result in haploinsufficiency, but as our results suggest, very subtle changes in channel gating are already sufficient to cause neonatal seizures. In contrast, only a distinct reduction of channel function by the voltage-sensor mutations R207Q and R207W seems to be able to affect PNS, suggesting CNS being more vulnerable than peripheral motoneurons.

1 C. Kurth , 2 A. Schulze-Bonhage , 1 B.J. Steinhoff 11 Epilepsy Centre Kork (Kork, D) ; 2 University of Freiburg (Freiburg, D)

Objective: FDG-PET is done frequently in preoperative monitoring work up of drug resistant epilepsy patients. Therefore we evaluated the impact of this functional imaging method on preoperative diagnostic and therapeutic decisions.

Method: In this retrospective study 38 patients (17 men, age 18-63, mean age 38,5) were analyzed, who underwent presurgical monitoring with video-EEG with surface electrodes, MRI of the skull, neuropsychological testing and FDG-PET. Neurosurgery prior to the monitoring was an exclusion criterion.

Results: 9 patients were nonlesional in MRI. Decisions for further invasive diagnostics were made in 4 of them on the basis of semiology and EEG. In the other 5 patients no option for further diagnostics was seen, due to semiology and EEG. PET scans showed hypometabolism in all 9 patients, but decisions could be made without these informations. 10 patients showed isolated unilateral hippocampussclerosis (HS) with or without increased signal in the ipsilateral temporal pole. In these patients selective amygdalohippocampectomy (AHE) or AHE and 2/3 resection of the temporal lobe was performed on the basis of MRI, EEG and semiology. PET showed multiregional hypometabolism in most patients and did not contribute to the presurgical decisions.

In 3 patients with bilateral HS bitemporal hypometabolism was found in PET. This finding was not helpful for the decision to plan invasive recording with bilateral hippocampal depth electrodes. In 1 patient with unilateral HS no further evalution was planed due to an extensiv temproal dysplasia in MRI. In 6 patients with pure extratemporal lesions in MRI PET findings were inconsistent. In 4 patients resetion of the lesion was recommended independent from the PET scan. In 2 patients invasive diagnostic was planned according to MRI and EEG. 5 patients showed complex lesional patterns in the MRI including temporal and extratemporal structures. Invasive recording depending on semiology, EEG and MRI was planed for these patients. PET did not contribute significantly to the final judgment. Only in one patient with fairly poor quality of the first MRI, PET scan was taken into considerations for further diagnostics – but only as long as no better MRI scan was available.

Conclusions: Routine PET scans had only little impact on decision in the presurgical work up of drug resistant epilepsy patients. Due to this finding further evaluations with a larger number of patients are necessary to determine reasonable indications for this method in presurgical epilepsy diagnostics.

1 J. Larch , 1 I. Unterberger , 1 J. Dobesberger , 1 N. Embacher , 1 G. Walser , 1 A. Auckenthaler , 1 G. Luef , 1 G. Bauer , 1 E. Trinka 1Medical University Innsbruck (Innsbruck, A)

Background: Juvenile myoclonic epilepsy (JME) is is a common age-related epilepsy syndrome. Under appropiate antiepileptic drugs (AED) up to 85% of the patients are seizure-free but after AED withdrawal more than 90% of patients will have a relapse. Hence JME is known as “easy to treat but hard to cure.” Several authors believe it is a lifelong disease. Only few studies have adressed the issue of long term seizure freedom. The aim on this study is to analyse seizure outcome in a large hospital based cohort.

Methods: Retrospective hospital-based study. All pts. with JME (n = 242) treated at the Universitätsklinik für Neurologie, Innsbruck, Austria (1975-2006), were screened. Inclusion criteria: (1) JME diagnosis (based on definition of ILAE) and (2) medical treatment >2 years in outpatient clinic. We included 175 pts. and analysed age, sex, seizure onset age, first and last documented seizure type, EEG, photosensitivity, MRI/CT, febrile convulsions (FC), family history, and AED response (seizure-free >2years).

Results: Population: 175 pts (111w/64m), mean age 38 years (range 14–87), mean observation 8 years (range 2–38). Four percent (7/175) had FC. Photosensitivity was present in 14% (24/175). Seizure Outcome: Fifty four percent (94/175) were seizure-free of all seizure types >2 years. Seventeen percent (16/94) of them without AED. Eight percent (15/175) were seizure-free for >1 year and < 2 years. Thirty-eight percent (66/175) were not seizure-free on appropriate AED. Fifty-seven percent were seizure-free between 2 and 5 years, 28% between 6 and 10 years and 15% for > 10 years. Seizure types: Pts. not seizure-free had significantly more often all three seizure types within the first year of their disease than those who were seizure-free at last follow up (11% vs. 3%, Chi-square = 4.679, df = 1, p = 0.031). Persisting seizure types were most often myoclonia with (Chi-square = 9.297, p = 0.002) or without grand mal on awakening (Chi2 = 3.987; p = 0.046), while absences tend to remitt.

Conclusions: We cannot confirm the assumption that JME needs lifelong treatment. A substantial proportion of JME patients remain seizure free with or without AEDs.

1 M. Dogan , 1 A. Gorji , 1 E.-J. Speckmann 1University of Munster (Munster, D)

Spread of excitation in the dentate gyrus and in the hippocampal CA3 region with stimulation of perforant path and of stratum granulosum was investigated. Data were sampled from fluorescence changes of the voltage-sensitive dye RH 795 incorporated into cellular membranes and recorded with a 464-photodiode array in a neuronal network in vitro.

The experiments show that stimulation of the crest of the stratum granulosum evoked activity in this entire layer and the hilus communicating with both ends of the stratum granulosum. The inner edge of the endal limb adjacent to the presubiculum was omitted. Maximal activity spread to the endal limb and the hilus within 6.3 ms and reached the ectal limb within 12.6 ms. Whereas activity evoked by stimulation of the stratum granulosum in the endal limb comprised the entire layer in this limb only the end of this stratum and a small region adjacent to the prosubiculum of the entorhinal cortex in the ectal limb were included. Again activity spread within 6.3 ms. Stimulation of stratum granulosum in the ectal limb did not only show higher level of activity but also the activity pattern was inverse: activity could be found in the whole layer of the ectal limb but only at its end of the endal limb. Unlike the stimulation in the endal limb the level of activity in the dentate gyrus was higher and activity reached the area around the crest within 3.78 ms. Stimulation of the perforant path at the crest mainly evoked activity in the whole ectal limb and only in the distal part of the endal limb- including stratum granulosum and moleculare in both cases. Activity spread within 12.6 ms.

In conclusion, the extent of activity in the hilus, at both ends of the stratum granulosum and in the region adjacent to the prosubiculum was always existent in the same extent independent of position and amplitude of the stimulus.

1 E. Ziegler , 1 M. Bodusch , 1 K. Krampfl 1Medical School Hannover (Hannover, D)

Gamma-aminobutyric acid (GABA) is the neurotransmitter at most of the fast inhibitory synapses within the mammalian brain whereas glycine plays a major role in fast inhibitory synaptic transmission in brain stem and spinal cord. GABAA receptor channels mediate fast inhibitory neurotransmission throughout the CNS while the expression of ionotropic glycine receptors is mainly restricted to spinal cord and brain stem. Neuroactive steroids are well known as positive allosteric modulators of GABAA receptors and there are data showing an interaction of neuroactive steroids with ionotropic glycine receptors. In our study, we compared the modulatory actions of androsterone and progesterone at human recombinant homoligomeric α1 glycine receptors, heterooligomeric α1β glycine receptors and α1β2γ2 GABAA receptors.

Recombinant α1 glycine receptors, α1β glycine receptors, or α1β2γ2 GABAA receptors were transiently expressed in human embryonic kidney cells (HEK) 293. The effects of androsterone and progesterone were studied using the whole-cell patch-clamp technique in combination with tools for ultrafast test solution exchange.

At α1β2γ2 GABAA receptors, not at ionotropic glycine receptors, a direct activation by androsterone was observed with a sigmoid dose-response relationship. Upon pulse-wise application of progesterone however, no direct activation of GABAA or glycine receptor channels was observed.

Androsterone enhanced currents elicited by submaximal (EC20) agonist concentrations at GABAAα1β2γ2 receptors and α1 glycine receptors in a concentration dependent manner. The dose response-relationship oft his positive allosteric modulation was tent-like with an increase of the relative peak current amplitude up to a concentration of 30 micromolar androsterone an a decrease of the relative amplitudes at upon coapplication of higher concentrations with an EC20 of GABA or glycine. Progesterone showed a similar effect but enhanced currents elicited by EC20 agonist concentrations at α1 glycine receptors, α1β glycine receptors and α1β2γ2 GABAA receptors to an extent minor than androsterone.

The data show the differential interaction of steroids with the most important fast inhibitory synaptic receptors, the ionotropic glycine receptor channels and the GABAA receptor channels. Different molecular mechanisms of interaction could be identified and quantitatively described. A major direct activation of inhibitory ionotropic receptors was observed with androsterone at GABAA receptors. A coactivation of currents elicited by nonsaturating agonist concentrations was observed with both compounds at glycine and GABAA receptors. At concentrations of 0.1 mM and higher, there were also hints to a channel block like mechanism.

Abstract

  1. Top of page

Poster session 2.2.Genetics, Pediatric Epileptology, Clinical Neurophysiology and Imaging

1 U. Reuner , 1 B. Kunath , 1 S. Hallmeyer-Elgner 1University Hospital Dresden (Dresden, D)

Neonatal convulsions are a heterogeneous group of disorders with variable aetiology and outcome. Benign familial neonatal convulsions comprise a distinct entity characterised by familial occurrence, normal psychomotor development and low rate of consequent epilepsy manifestation. Multifocal or generalised tonic–clonic convulsions typically commence around day 3 after birth and disappear spontaneously after a few weeks or months. The major locus for these convulsions is localised on chromosome 20q13.3 and mutations of the K+ channel gene KCNQ2 have been identified as the underlying cause.

On the other hand, myokymia is characterised by spontaneous involuntary contraction of muscle fiber groups that can be observed as vermiform movement of the overlying skin. Electromyography typically shows continuous motor unit activity with spontaneous oligo- and multiplet – discharges of high intraburst frequency (myokymic discharges).

The conccurrence of neonatal convulsions and myokymia in the same patient has seldom been reported. We describe four members of a single family, in which a coincidence of benign familial neonatal convulsions and myokymia in later life was established. This syndrome was previously shown to correlate with an amino acid exchange within the putative voltage sensor of the K+ channel KCNQ2 (Dedek K. et al., Proc Natl Acad Sci USA, 2001; 12272 -7).

This link between benign familial neonatal convulsions and myokymia should be considered when diagnosing either of these syndromes, in view of their importance in genetic counselling and family planning. The concurrence of these clinical syndromes denotes the importance of K+ channels in regulating both central and peripheral nerve excitability.

1 J. Ullmann , 2 I. Morger , 3 F. Donati 11 University Berne (Berne, CH) ; 2 Children's Hospital Wildermeth (Biel, CH) ; 3 Hospital Center Biel (Biel, CH)

Objective: Recent studies showed a significantly increased frequency of epileptic seizures around full moon in adults. The aim of this study is to evaluate the correlation between moon phases and febrile seizures in children.

Methods: From our database we retrospectively collected the data of 200 children (age < 6 years) presenting their first febrile seizure lasting less than 15 minutes backward from September 2006. To evaluate the correlation between febrile seizures and moon phases, we introduced the exact date of the first febrile seizure in a moon calendar.

Results: We collected the data of 200 children (114 male (56%), 91 female (44%), mean age 1 year and 8 months). With the exception of a marginal, but significant reduction of the febrile seizures in girls during the week before the full-moon and in the 34 children with positive family history (first degree relatives), which had significant less febrile seizures around the full-moon days (full-moon plus/minus three days), we did not find any significant correlation between the febrile seizures and the moon phases.

Conclusion: Our results suggest that in children there is no influence of the lunar phases on febrile seizures.

1 C. Gumbinger , 1 A. Schulze-Bonhage , 1 R. Korinthenberg , 1 M. Häffner , 1 S. Fauser 1University of Freiburg (Freiburg, D)

Introduction: Focal cortical dysplasia (FCD) is a common cause of pharmaco-resistant epilepsy in childhood and adulthood. The etiology is so far unknown. Genetic and environmental factors are discussed. Because of the histological resemblance to tubera in tuberous sclerosis (TSC), the TSC1- and TSC2 genes are interesting candidate genes.

Patients and methods: 34 patients with FCD were investigated a) clinically as to dermatological stigmata, retinal hamartoma, cardial rhabdomyoma, and renal angiomyolipoma and b) genetically by SSCP electrophoresis and sequencing of the TSC1 and TSC2 genes. As control group served 5 patients with tuberous sclerosis and 40 healthy volunteers.

Results: In our patient group, we could not detect any stigmata of tuberous sclerosis, neither in patients with unifocal nor with multifocal dysplasia.

In the TSC1 gene we could find several sequence alterations in exon 6, 10, 14 and 22, and in intron 14 and 23a. In the TSC2 gene frequent sequence alterations were seen in exon 14, 15, 31, 31, 39, 40 and 41. However, we could not detect any mutation which has already been identified as causative for tuberous sclerosis. Moreover, most of the sequence alterations do not result in an exchange of amino acids. All detected polymorphisms could be detected in both, lesional brain tissue and in the blood of the patients. The described sequence alterations could be observed in similar frequency even in the normal population.

Conclusions: The phenotype-genotype analysis and the comparison to negative controls (normal population) give evidence that at least most FCDs are caused by mechanisms other than mutations in the TSC1 or TSC2 gene. Thus, FCD can not be regarded as a forme fruste of TSC.

1 C. Bußmann , 1 A. Allmendinger , 1 T. Bast , 2 G. Ramantani , 1 D. Rating 11 University Children's Hospital Heidelberg (Heidelberg, D) ; 2 University Children's Hospital Dresden (Dresden, D)

Background: Seizures in newborns occur with a frequency of 1,5 of 1000 living newborns. If a causative treatment ist not possible and the seizures do not stop after administration of pyridoxine, phenobarbital is the treatment of choice. In the past few years this treatment option has been extensively dicussed. An animal model of the newborn rat showed an increased rate of apotosis after treatment with phenobarbital. In addition the risk of cognitive impairment known in infants and toddlers treated with phenobarbital should be considered. Levetiracetam is a new antiepileptic drug. The administration of levetiracetam in adults and children above the age of 4 years shows good results. In animal model treatment with levetiracetam did not increase the rate of apoptosis of neuronal cells.

Inclusion criteria and pilot study: In a prospective, open, monocentric pilot study (proof of principle) levetiracetam was used as a single treatment regimen of neonatal seizures after exclusion of hypoglycaemia, hypomagnesaemia and pyridoxine dependency. Included were newborns and prematures (> 2.000 g and > 30 weeks of gestation) without preceding antiepileptic medication. If needed two prior doses of phenobarbital with a maximum dosage of 10 mg /kg/d were allowed. In total 6 neonates were included.

Results: Under treatment with levetiracetam (30-50mg/kg/d) all of the included newborns became seizure free within 6 days. Severe adverse events were not observed. Under the ongoing therapy 2/6 patients had a seizure relapse in the following months. At present 5/6 of the included newborns are seizure free. 2/6 Patient stayed seizures free after the medication by levetiracetam had been stopped.

Conclusion: Our observations show that the administration of levetiracetam in newborns is possible without problems.The small number of treated newborns does not allow a judgment of efficacy and potential adverse effects of this medication. Our results could be the basis for a controlled randomized study to further assess this medication as treatment for neonatal seizures.

1 F. Bösebeck , 2 M. Ritter , 2 C. Kellinghaus , 1 R. Kiefer 11 Diakonie Hospital Rotenburg (Rotenburg, D) ; 2 University Hospital Munster (Munster, D)

Cranial bone defects frequently lead to rhythmical EEG activity of minor clinical relevance. This “Breach-Rhythm” regularly ranges from 6-11/s and is nonreactive to motor activation, which make it easy to differentiate it from physiological central μ-Rhythm.

We present two patients with previously performed decompressive craniectomy after a malignant MCA infarction (case 1) or intracerebral bleeding (case 2). Several years after operation, the patients were admitted for impaired consciousness of initially unknown reason. An emergent EEG was performed to exclude nonconvulsive status epileptikus. In both patients, rhythmical high voltage Delta-Activity was documented over the site of lesion. The EEG activity was nonreactive to vigilance changes or intravenous lorazepam. In one pat. simultaneous EEG Doppler sonography proved the activity to be pulse-artifacts arising from meningeal arteries. Both patients improved after general medical support without any further anticovulsive treatment.

In our case report, we presented atypical pulse artifacts as another non-specific EEG-finding following cranial bone defects which should be taken into account especially after osteoclastic trepanation.

1 F. Müller , 1 K. Krakow , 1 U. Ziemann 1Johann Wolfgang Goethe-University (Frankfurt am Main, D)

Objectives: To study corticospinal excitability and intracortical inhibitory mechanisms in a patient with epilepsia partialis continua (EPC).

Methods: Focal transcranial magnetic stimulation (TMS) was applied to the left and right primary motor cortex (M1) in a 60-year-old patient with EPC affecting primarily his right middle finger flexor. EPC was caused by a focal tumor lesion at the bottom and posterior wall of the central sulcus adjacent to the hand/arm area of left M1. Muscle activity was recorded by surface EMG from the forearm flexor (FLEX) and first dorsal interosseus (FDI) muscles bilaterally. Resting (RMT) and active (AMT) motor thresholds were determined for all target muscles. The cortical silent period (CSP) was determined at 150% AMT in all muscles. Finally, EMG activity from the right FLEX and right FDI was recorded continuously over >30 min while TMS over the left M1 forearm representation was applied every 5 sec at an intensity of 120% RMT.

Results: At rest, myocloni occurred in the right FLEX and to a far lesser extent in the right FDI. The mean interval between two myoclonic bursts was 1.26 s. RMT and AMT were clearly higher in left M1 as compared to right M1. CSP duration of either target muscle was comparable between left and right M1, but MEP amplitude during maximal voluntary contraction was significantly reduced in left M1. At rest, corticospinal excitability in the left M1 was increased for approximately 100 ms after the end of a myoclonic burst, but decreased thereafter, with almost complete MEP suppression at around 200 ms after myoclonic activity. Thereafter, MEP amplitudes increased rapidly, and at intervals >450ms after the myoclonic burst, TMS always triggered myoclonic activity that directly followed the MEP and resembled the spontaneously occurring myocloni. These MEP-EMG burst complexes occurred significantly earlier than the expected timing of a spontaneous myoclonic burst (0.67 s vs. 1.26 s, p < 0.0001).

Conclusions: RMT, AMT and CSP/MEP amplitude ratio revealed increased inhibitory over excitatory cortical excitability in the left vs. right M1, presumably reflecting compensatory processes. The time course of corticospinal excitability in left M1 following a myoclonic burst provides evidence for a myoclonic burst generator that is accessible to TMS over M1.

1 B. Oehl , 1 C. Gierschner , 1 A. Buller , 1 A. Schulze-Bonhage 1University Hospital Freiburg (Freiburg, D)

Purpose: To investigate wether the dosage of antiepileptic medication corresponds to the degree of subclinical epileptic activity in patients with mesiotemporal focal epilepsy.

Methods: We retrospectively analyzed data from 28 patiens who suffered from mesiotemporal epilepsy and underwent invasive video EEG monitoring with depth electrodes to one or both hippocampi. 5 patients were excluded as antiepileptic drug regime remained constant during monitoring. 23 patients (mean age 33.8 y, 10 female) were analyzed regarding the frequency of clinically manifest and subclinical electrographic seizures during days of maximal and mimal antiepileptic drug intake. Statistical analyses were performed using the Wilcoxon rank sign text, and using pearson correlations.

Results: Distribution of seizures and subclinical event was skewed. Mean (median) frequency of clinically manifest seizures during maximal antiepileptic drug intake was 1.0/day (0.0/day), during minimal drug intake 1.5/day (1.0/day). Mean (median) frequency of subclinical electrophic seizures during maximal antiepileptic drug intake was 2.4/day (0.0/day), during minimal drug intake 8.3/day (0.3/day). Differences did not reach statistical significance. 3 patients developed clusters of up to 136 subclinical events per day during tapering of AED. There was a positive correlation between the frequency of subclinical events during AED reduction and under maximal AED dosage, and between the frequency of subclinical and clinical events during minimal AED dosage.

Discussion: Although mean rates of subclinical electrographic seizures were higher under AED reduction, this difference did not become significant in this sample of patients with mesiotemporal seizure origin. It is of note, however, that clusters of subclinical events occurred during tapering of AED, which may have neuropsychological correlates not determined here. Positive correlation between subclinical seizure frequency under minimal and maximal AED dosage suggests that factors other than AED have a major impact on the number of subclinical EEG patterns.

1 S. Schubert , 2 H.-M. Meinck , 3 W. Schrank , 1 D. Rating , 1 T. Bast 11 Children's Hospital Heidelberg (Heidelberg, D) ; 2 Neurological Clinic Heidelberg (Heidelberg, D) ; 3 Social Pediatric Center (Wiesbaden, D)

We report on a 3-year-old, normally developed girl with epilepsia partialis continua (EPC). After a complicated birth parents noticed a paresis of the left arm. The motor function normalized completely within the first four weeks of life. Jerking of the left hand was first seen at the age of 6 months. However, the patient developed normally with good motor function of the affected hand. Interpretation varied from monoparesis to tremor or myoclonia over time. At the age of 2.5 years clear epileptic seizures with tonic posturing and cloni of the left arm occurred. These seizures disappeared after administration of oxcarbazepine, but myoclonia of the left hand persisted. Repeated MRI was interpreted as normal. She was referred to our centre for further diagnosis. Polygraphic EMG revealed synchronous myoclonic jerks of left arm and hand muscles appearing at a frequency of 5.5 Hz, with duration of < 50 ms (to < 100 ms). This clearly corresponds to myoclonus and duration of jerks was suspect for cortical origin. Atypically for cortical myoclonus, N20 amplitude was decreased to 2μV after left median nerve stimulation compared to the right side (6.3 μV). Long-term video EEG demonstrated frequent myoclonia of the left hand when awake. Myoclonus infrequently appeared in sleep state 1 and 2 and was not observed in deeper sleep. Myoclonus was clearly exercise activated, but not in the sense of a reflex myoclonus. Minispikes were detected in EEG over the right centroparietal region, but did not always correlate with myoclonic jerks. EMG-triggered averaging of EEG segments (BESA software) resulted in a preceding sharp, rhythmic activity and proofed cortical myoclonus. Source analysis resulted in dipole localization within the right central region. Critical review of the previous MRI lead to the diagnosis of distinct, right central focal cortical dysplasia (FCD) with blurring of the grey-white matter boundary and slightly increased subcortical T2-signal. Since the patient was not impaired by EPC, no status treatment was initiated after at least 30 months of stable disease. We plan to optimize conventional anticonvulsant drug treatment, because risk for recurrence of impairing seizures and for deterioration of motor function seems high.

Conclusion: EPC is well known as a typical presentation of central FCD. The course was unusual in our patient, because EPC started immediately after birth and did not lead to a marked impairment at time.

1 T. Ehrenfried , 1 M. Kaltenhaeuser , 1 A. Paulini , 1 S. Rampp , 1 H. Stefan 1University of Erlangen (Erlangen, D)

Purpose: We retrospectively reviewed the clinical outcome of 38 patients with medically intractable epilepsy five years after epilepsy surgery. All patients underwent preoperative evaluation including magnetoencephalography (MEG). Goal of this study was to compare the dipole source localization of interictal spikes with the area of surgery in dependency of the five year postoperative outcome Engel.

Methods: We performed preoperative spike recording with MEG in 38 epilepsy surgery candidates (12 (32%) female), 26 (68%) male) with two 37-channel sensors. Dipole source localization of the epileptic activity was visualized on coregistered MRI scans. All patients underwent epilepsy surgery. Among them, 33 (87%) underwent temporal and 5 (13%) extratemporal resection. Mean age at surgery was 34 ± 10 years (range 20–53 years).

Results: Of all patients, 28 (74%) had an Engel outcome of 1, five (13%) of 2, three (8%) of 3 and two (5%) of 4. Thirteen (34%) patients stayed completely seizure free postoperatively.

MEG source localization and area of resection were concordant in 32 of 38 patients (p = .006). Of these 32 patients 24 (86%) received Engel outcome 1, five Engel outcome 2 (16%), one Engel outcome 3 (3%) and two Engel outcome 4 (6%).

Of the six nonconcordant patients, four received Engel outcome 1 and two Engel outcome 3. In the latter patients the dipole source was found in eloquent brain areas.

A marginal significance (p = 0.06) was observed between outcome and consistency of MEG source location and area of surgery.

Conclusion: To a high degree MEG source localization is concordant with the area of resection in epilepsy surgery. Five year postoperative clinical outcome was worthwhile in 30 of 32 concordant patients. Of 38 patients, only in six patients (19%) MEG localization and area of resection showed no concordance.

1 J. Kröll-Seger , 1 R.A. Sälke-Kellermann , 2 M. Weissert , 3 E. Martin , 1 H.-J. Huppertz 11 Swiss Epilepsy Centre (Zurich, CH) ; 2 Children's Hospital St. Gallen (St. Gallen, CH) ; 3 University Children's Hospital Zurich (Zurich, CH)

Background: Focal cortical dysplasia (FCD) is a frequent cause of medically refractory epilepsy. Identification of the epileptogenic lesion with its whole extension as well as complete removal are prerequisites for seizure-freedom after epilepsy surgery. The following case report demonstrates that morphometric MRI analysis can detect incompletely removed dysplastic cortex in a patient with postoperatively unfavorable seizure control.

Case report and methods: The 8-year-old girl suffers from seizures with right-sided myoclonias of the eyelids, left-sided cloni and autonomic symptoms (flush, gooseflesh) since babyhood. With MRI demonstrating cortical dysplasia of the right frontal lobe extending to the anterior insular region a partial resection of the right frontal lobe was performed at age 3 years. After surgery seizures persisted in unchanged semiology and increasing frequency (Engel class IV). Postsurgical MRI (2006) depicted the resection area with gliosis but no residual dysplastic cortex. For further investigation morphometric analysis of the pre-operative MRI data from 2001 was performed: from the original T1-weighted 3D dataset three new images were derived which encode in comparison with a normal database characteristic imaging features of FCD (cortical thickening, abnormal extension of grey matter into white matter and blurring of the grey/white matter junction). In addition co-registration with the postoperative MRI was carried out.

Results: The three new feature maps visualise abnormal differentiation of the grey-white matter junction of parts of the right frontal lobe extending to the anterior insular region. Coregistration with the postsurgical MRI demonstrates residual dysplastic, probably epileptogenic cerebral cortex.

Discussion: MRI post-processing and morphometric MRI analysis are useful additional diagnostic tools of conventional neuroimaging in the presurgical evaluation of epilepsy patients. As demonstrated the method may also be very valuable in the reevaluation of patients with failure of previous resective treatment and the planning of reoperation.

1 C. Kellinghaus , 1 M. Deppe , 1 T. Duning , 1 H. Schiffbauer , 1 S. Kovac , 1 S. Mohammadi , 1 H. Kugel , 1 S. Knecht 1University Hospital Munster (Munster, D)

Rationale: Juvenile myoclonic epilepsy (JME) is a syndrom of idiopathic generalized epilepsy (IGE). Although absence of structural brain lesions is mandatory for the diagnosis of JME., the presence of typical, bifrontal accentuated epileptiform discharges has resulted in the hypothesis of frontal dysfunction. The use of modern imaging methods and neuropsychological tests seem to support this hypothesis. We examined the integrity of the white matter in JME patients using whole head diffusion tensor imaging (DTI).

Methods and design: Ten patients with JME were recruited from the epilepsy outpatient clinic of the University Hospital Münster, Germany. Inclusion criteria were: 1) established diagnosis of JME according to the ILAE classification of 1989, 2) no other neurological disease, 3) normal conventional MRI (1.5-Tesla scanner), 4) no generalized tonic–clonic seizures in the last week. The patients underwent MRI in a 3-Tesla scanner (Philips) using the following sequences: volumetric T1-weighted, T2-weighted, Fluid attenuation inversion recovery (FLAIR), and DTI (20 gradient directions). Fractional anisotropy (FA) maps were calculated from DTI data to assess the white matter integrity. The results were compared on a voxel-by-voxel basis with data of an age-matched control group of 64 healthy volunteers (SPM5, two-sample t-test, treating the subjects' age as “effect of no interest”).

Results: The patient's mean age was 28 years (± 8 years), the control group proband's mean age was 27.9 years (range 21–42 years). Age of epilepsy onset was 13.7 ± 2.6 years. 7 of the patients had suffered 2-15 generalized tonic–clonic seizures (median: 9). All patients were under anticonvulsant therapy (5 VPA monotherapy, 3 LTG monotherapy). 4 patients were completely seizure free, 2 other patients had only myoclonic seizures since more than 2 years. The conventional imaging sequences (T1, T2, FLAIR) using the 3-Tesla scanner did not show any abnormalities. DTI revealed significantly reduced FA in basal ganglia, frontobasal regions, thalamocortical fibers as well as in the parahippocampal white matter in JME patients compared to the control group. The severity of the frontal abnormalities was correlated to age, disease duration, treatment duration and number of generalized tonic–clonic seizures. However, these variables were strongly correlated to each other.

Conclusion: Compared to healthy volunteers, JME patients seem to have reduced integrity of frontobasal, thalamo-cortical, basal ganglia and parahippocampal fibers. It remains to be determined whether these abnormalities are related to the cause, the result or the treatment of the disease.

1 J. Wellmer , 1 H. Urbach , 1 C. Quesada , 1 T. Kral , 1 O. Schijns , 1 A. Becker , 2 H.-J. Huppertz , 1 C.E. Elger 11 University Hospital Bonn (Bonn, D) ; 2 Swiss Epilepsy Centre (Zurich, CH)

Introduction: In patients with pharmacoresistant focal epilepsies, the likelihood of permanent seizure freedom following epilepsy surgery increases with the preoperative detection of epileptogenic lesions. Thus, much effort should be put in the search for such lesions. By presenting two patients classified as nonlesional after cMRI we illustrate the impact which morphometric MRI analysis (“postprocessing”) can have on the detection of subtle focal cortical dysplasias (FCD). Patients and methods: Two females, (AH, 40 ys, and AP, 15 ys) suffered from pharmacoresistant seizures since childhood. Both patients had previous MRI scans classified as nonlesional, AH only at 1.5 Tesla (due to vagal nerve stimulation), AP at both, 1.5 and 3 Tesla. Seizure semiology contained elements suspicious for frontal paramedian or interhemispheric seizure origin (i.e., supplementary motor area). EEG pointed to a frontal seizure generation without further specification. We performed a morphometric MRI analysis (MAP05, Huppertz et al., Epilepsy Res 2005;67:35-50), which compares T1-weighted volume data sets of individual brains with a normal database and creates new feature maps highlighting typical FCD signs (i.e., abnormal extension of gray matter into white matter, blurring of the gray-white-matter junction, and abnormal thickness of the cortical ribbon).

Results: In both patients, all feature maps pointed to a singular lesion which, in similar position, was located in the superior frontal gyrus, reaching the surface of the gyrus at its interhemispheric face. Only with this information, the lesions could be recognized on the morphological MRI scans. The following invasive monitoring confirmed the seizure onset in the detected lesions and its close vicinity. We performed extended lesionectomies with respect to the seizure onset zone. Histological examination confirmed FCD Palmini type 2b in both cases. According to last available outcome, both patients are seizure free (Engel 1a).

Discussion: MRI postprocessing with morphometric analysis can enhance the presurgical workup by helping to identify epileptogenic lesions in otherwise cryptogenic patients. Yet, the method is restricted to the identification of malformations of cortical development. Sensitivity and specificity of the method are subject to ongoing studies.

1 A. Holl , 1 M. Feichtinger , 1 F. Payer , 1 F. Fazekas , 1 E. Ott 1University Hospital of Neurology (Graz, A)

Introduction: Transient imaging abnormalities have been described in neuroimaging (MRI) studies of epileptic seizures, especially status epilepticus. These findings may mimick neoplastic or ischemic changes on MRI. The risk factors precipitating postictal imaging abnormalities and the underlying pathophysiology are still not completely understood.

Patients and methods: We report 3 patients with a history of oligoastrocytoma, incomplete surgical removal and subsequent radiation therapy, who were reexamined because of clinical worsening and the suspicion of focal seizures. Magnetic resonance imaging showed extensive hyperintense signal abnormalities involving predominantly the cortex with swelling and contrast enhancement. Given the proximity of the abnormalities to the surgical defect and their mass effect MR imaging findings were initially considered to indicate tumor growth. However, in the differential diagnosis an association with the seizures was also entertained and was supported by EEG changes showing focal frequent epileptiform discharges. FDG-PET scanning revealed widespread cortical hypermetabolism in correspondence to the MRI abnormalities which was felt to be atypical for highly malignant tumor growth but consistent with seizure activity.

All three patients received anticonvulsive therapy, their clinical condition improved and follow-up MRI examinations showed a remission of the MRI abnormalities within a few weeks.

Conclusion: It is known that transient MR imaging abnormalities showing predominantly cortical hyperintensities may be a consequence of epileptic activity. While these findings per se may raise suspicion of ischemic or neoplastic lesions they are also an important differential diagnostic consideration in patients with a history of brain tumor. Co-registration of structural and functional imaging and its correlation with EEG findings can help to corroborate the correct diagnosis.

1 G. Hagemann , 1 S. Rupprecht , 1 C. Fitzek , 1 O.W. Witte , 1 C. Terborg 1Friedrich-Schiller-University (Jena, D)

Background and objective: The clinical differentiation between stroke and seizure is usually straightforward but postictal neurological deficits can be mistaken for stroke in case no detailed medical history is available. In ischemic stroke cranial MRI depicts early diffusion changes and a mismatch between regions with abnormal diffusion and perfusion is often taken as an indicator of emerging stroke. This “tissue at risk”, can potentially be salvaged by thrombolyis. The MR-pattern of seizures is less clear and might depend on the time point investigated.

Case report: A 42-year-old male presented to the emergency room with somnolence, a dense hemiparesis on the left and eye deviation to the right. Later on we learned, that prior to admission he had experienced a complex partial seizure with secondary generalization. Nine month before he had right hemispheric stroke due to a severe stenosis of the middle cerebral artery which was treated with a stent. Consecutively he experienced a hyperperfusion syndrome with intracerebral hemorrhage and symptomatic seizures afterwards. A cranial MRI was obtained which showed the old lesions without new diffusion abnormality but extensive hypoperfusion of the complete right hemisphere. MR-angiography depicted an artificially discontinuing M1-segment of the middle cerebral artery because of the stent. Furthermore MR-angiography showed fetal origin of the right posterior cerebral artery but otherwise no abnormality and no change to earlier MRA of the patient. The EEG 6 hours after the episode showed right sided slowing only. The patient made a fast recovery and 10 hours later there were no perfusion changes and no new lesions on MRI follow-up.

Discussion: This case demonstrates that postictal paralysis (Todd's paralysis) can be accompanied by an extensive cerebral hypoperfusion without any permanent neurological or MRI sequelae. It is well known that ictal perfusion imaging tends to depict a hyperperfused zone which delineates the epileptogenic focus and can be used in presurgical evaluation. Long seizure activity, like in status, usually is accompanied by hyperperfusion and causes at least transient diffusion changes. Early postictal and interictal perfusion studies mostly demonstrated localized, region specific and time-dependent hypoperfusion. Although, in this case it is difficult to completely rule out a transient ischemic attack the clinical course as well as the missing diffusion changes and no additional recent change of the MRA, make it unlikely. It can only be speculated, whether both, the antecedent hyperperfusion syndrome and the episode described here point towards a pathological vasoregulation in this patient.

Conclusion: Postictal hypoperfusion can be extensive and should not be mixed up with emerging stroke. Further investigations are warranted to clarify whether this perfusion pattern is typical of Todd's paralysis.

1 J. Dobesberger , 1 G. Walser , 1 I. Unterberger , 1 N. Embacher , 1 G. Luef , 1 T. Gotwald , 1 M. Ortler , 1 E. Trinka 1Medical University Innsbruck (Innsbruck, A)

Purpose: A transient lesion in the splenium of the corpus callosum (SCC) has been described in patients with epilepsies as well as other clinical conditions, like encephalitis, malnutrition or high-altitude illness. So far, aetiopathogenesis is not clarified. Therefore, the aim of this study was to analyze frequency, possible causes and relevance of this focal lesion in an epilepsy centre.

Methods: All patients undergoing video-EEG monitoring between January 1999 and September 2006 were screened retrospectively. Patients were included, when an MRI was performed within 1 week after monitoring and a focal lesion in SCC found.

Results: Between January 1999 and September 2006 731 video-EEG sessions were recorded. In 50% (368/731), an MRI was done within 1 week after the session. A lesion in the SCC was found in 7% (24/368; 12 M, mean age 35 ± 9 years). Twenty-one of 24 patients had focal epilepsies (17 temporal lobe epilepsy, 4 frontal lobe epilepsy), two patients generalised epilepsies and one patient had syncopes. On the average, five seizures were recorded during video-EEG: the lesion occurred in two patients after a nonconvulsive status epilepticus and in one patient with epilepsy but without recorded seizures. The following antiepileptic drugs (AEDs) were reduced or discontinued: carbamazepine, levetiracetam, lamotrigine, topiramate, oxcarbazepine, valproic acid, primidone, phenytoin und clobazam. In 12 patients follow-up MRI scans were done on the average after 9 months, which showed the disappearance of the lesion in all of them.

Conclusions: Transient lesions in SCC after intensive video-EEG monitoring are not rare, the pathophysiology remains unclear. An association exclusively with occurring epileptic seizures before seems unlikely as one patient had no epilepsy an in one patient no epileptic seizures were recorded during monitoring. As all patients were treated with AEDs before monitoring, the tapering off and discontinuation of AEDs could play a role in pathogenesis. The prognosis is good with complete disappearance of the lesion after one year.

1 M. Bauer , 2 O. Langer , 3 A. Hammers , 1 R. Karch , 1 E. Pataraia , 1 F. Zimprich , 4 M. Koepp , 1 A. Abrahim , 5 G. Luurtsema , 1 K. Kletter , 1 C. Baumgartner , 1 M. Müller 11 Medical University Vienna (Vienna, A) ; 2 Medical University Vienna and Austrian Research Centers, Seibersdorf (Vienna, A) ; 3 Imperial College London and Hammersmith Hospital (London, UK) ; 4 University College London (London, UK) ; 5 VU University Medical Center Amsterdam (Amsterdam, NL)

Background: R-[11C]verapamil, a radiolabelled substrate of the multidrug transporter P-glycoprotein (P-gp), has been developed as a probe for in vivo evaluation of P-gp function at the human blood-brain barrier with PET (Lubberink et al., 2006). Because regional overexpression of P-gp in epileptic brain tissue is discussed as a factor that contributes to pharmacoresistance in epilepsy, we studied brain R-[11C]verapamil distribution in 7 patients with drug-resistant unilateral therapy-refractory temporal lobe epilepsy.

Methods: After i.v. bolus injection of about 400 MBq of R-[11C]verapamil (specific activity: >20 GBq/μmol) dynamic brain PET imaging and arterial blood sampling were performed for 60 min. Volumes of interest (VOIs) were defined on MRI scans by using a frequency-based anatomical atlas (Hammers et al., 2003). R-[11C]verapamil kinetics were compared in homologous brain VOIs located ipsilateral and contralateral to the seizure focus. PET data were modeled by a previously described 1-tissue compartment model using a partially metabolite-corrected arterial input function.

Results: Among different VOIs, radioactivity was highest in the choroid plexus. The hippocampal VOI could not be used for data analysis because it was contaminated by spill-in of radioactivity from the adjacent choroid plexus. In several other temporal lobe regions that are known to be involved in seizure generation and propagation ipsilateral influx rate constants K1 and efflux rate constants k2 of R-[11C]verapamil were increased as compared to the contralateral side. Parameter asymmetries were most prominent in parahippocampal and ambient gyrus (K1, range among patients: −3.8% to +22.3%; k2, range: −2.3% to +43.9%), amygdala (K1, range: −20.6% to +31.3%; k2, range: −18.0% to +38.9%) and medial anterior temporal lobe (K1, range: −8.3% to +14.5%; k2, range: −14.5% to +31.0%). In contrast to temporal lobe VOIs, asymmetries were minimal in a region not involved in epileptogenesis located outside the temporal lobe (superior parietal gyrus, K1, range: −3.7% to +4.5%; k2, range: −4.2% to +5.8%). In five of seven patients, ipsilateral efflux (k2) increases were more pronounced than ipsilateral influx (K1) increases, which resulted in prominent ipsilateral reductions (10–26%) of R-[11C]verapamil distribution volumes.

Conclusion: Our pilot data suggest that regionally enhanced P-gp activity might contribute to pharmacoresistance in some patients with temporal lobe epilepsy.

References:

Hammers A, et al. (2003) Hum Brain Mapp 19:224247.